Psi Chi Journal of Psychological Research – Spring 2023

ISSN: 2325-7342

Published by Psi Chi, The International Honor Society in Psychology ®

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH SPRING 2023 | VOLUME 28, NUMBER 1

EDITOR

STEVEN V. ROUSE, PhD

Pepperdine University

Telephone: (310) 506-7959

Email: steve.rouse@psichi.org

ASSOCIATE EDITORS

JENNIFER L. HUGHES, PhD

Agnes Scott College

STELLA LOPEZ, PhD

University of Texas at San Antonio

TAMMY LOWERY ZACCHILLI, PhD Saint Leo University

ALBEE MENDOZA, PhD

Delaware State University

KIMBERLI R. H. TREADWELL, PhD University of Connecticut

ROBERT R. WRIGHT, PhD

Brigham Young University-Idaho

EDITOR EMERITUS

DEBI BRANNAN, PhD

Western Oregon University

MANAGING EDITOR

BRADLEY CANNON

DESIGNER

JANET REISS

EDITORIAL ASSISTANTS

EMMA SULLIVAN

ADVISORY EDITORIAL BOARD

GLENA ANDREWS, PhD

RAF Lakenheath USAF Medical Center

AZENETT A. GARZA CABALLERO, PhD

Weber State University

MARTIN DOWNING, PhD Lehman College

HEATHER HAAS, PhD

University of Montana Western

ALLEN H. KENISTON, PhD

University of Wisconsin–Eau Claire

MARIANNE E. LLOYD, PhD Seton Hall University

DONELLE C. POSEY, PhD

Washington State University

LISA ROSEN, PhD

Texas Women's University

CHRISTINA SINISI, PhD

Charleston Southern University

PAUL SMITH, PhD

Alverno College

ABOUT PSI CHI

Psi Chi is the International Honor Society in Psychology, founded in 1929. Its mission: "recognizing and promoting excellence in the science and application of psychology." Membership is open to undergraduates, graduate students, faculty, and alumni making the study of psychology one of their major interests and who meet Psi Chi’s minimum qualifications. Psi Chi is a member of the Association of College Honor Societies (ACHS), and is an affiliate of the American Psychological Association (APA) and the Association for Psychological Science (APS). Psi Chi’s sister honor society is Psi Beta, the national honor society in psychology for community and junior colleges.

Psi Chi functions as a federation of chapters located at over 1,180 senior colleges and universities around the world. The Psi Chi Headquarters is located in Chattanooga, Tennessee. A Board of Directors, composed of psychology faculty who are Psi Chi members and who are elected by the chapters, guides the affairs of the Organization and sets policy with the approval of the chapters. Psi Chi membership provides two major opportunities. The first of these is academic recognition to all inductees by the mere fact of membership. The second is the opportunity of each of the Society’s local chapters to nourish and stimulate the professional growth of all members through fellowship and activities designed to augment and enhance the regular curriculum. In addition, the Organization provides programs to help achieve these goals including conventions, research awards and grants competitions, and publication opportunities.

JOURNAL PURPOSE STATEMENT

The twofold purpose of the Psi Chi Journal of Psychological Research is to foster and reward the scholarly efforts of Psi Chi members, whether students or faculty, as well as to provide them with a valuable learning experience. The articles published in the Journal represent the work of undergraduates, graduate students, and faculty; the Journal is dedicated to increasing its scope and relevance by accepting and involving diverse people of varied racial, ethnic, gender identity, sexual orientation, religious, and social class backgrounds, among many others. To further support authors and enhance Journal visibility, articles are now available in the PsycINFO®, EBSCO®, Crossref®, and Google Scholar databases. In 2016, the Journal also became open access (i.e., free online to all readers and authors) to broaden the dissemination of research across the psychological science community.

JOURNAL INFORMATION

The Psi Chi Journal of Psychological Research (ISSN 2325­7342) is published quarterly in one volume per year by Psi Chi, Inc., The International Honor Society in Psychology. For more information, contact Psi Chi Headquarters, Publication and Subscriptions, 651 East 4th Street, Suite 600, Chattanooga, TN 37403, (423) 756­2044. www.psichi.org; psichijournal@psichi.org

Statements of fact or opinion are the responsibility of the authors alone and do not imply an opinion on the part of the officers or members of Psi Chi.

Advertisements that appear in Psi Chi Journal do not represent endorsement by Psi Chi of the advertiser or the product. Psi Chi neither endorses nor is responsible for the content of third­party promotions. Learn about advertising with Psi Chi at http://www.psichi.org/Advertise

COPYRIGHT

Permission must be obtained from Psi Chi to reprint or adapt a table or figure; to reprint quotations exceeding the limits of fair use from one source, and/or to reprint any portion of poetry, prose, or song lyrics. All persons wishing to utilize any of the above materials must write to the publisher to request nonexclusive world rights in all languages to use copyrighted material in the present article and in future print and nonprint editions. All persons wishing to utilize any of the above materials are responsible for obtaining proper permission from copyright owners and are liable for any and all licensing fees required. All persons wishing to utilize any of the above materials must include copies of all permissions and credit lines with the article submission.

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

SPRING 2023 | VOLUME 28 | ISSUE 1

2 Conquering APA Style for the Seventh Edition: Advice From APA Style Experts

Jennifer L. Hughes1, Bradley Cannon2, Abigail A. Camden3, Kimberli R. H. Treadwell4, Joel G. Thomas1, and Bonnie M. Perdue1

1 Department of Psychology, Agnes Scott College

2 Psi Chi Headquarters

3 Department of Psychological Sciences, Auburn University

⁴ Department of Psychological Sciences, University of Connecticut

14 Coping During Crisis: An Intervention on Gratitude and Personal Well-Being During COVID-19

Isabel F. Gilbertson and Audrey A. Graves

Department of Psychology, Seattle University

26 Identification With All Humanity Predicts Perceptions of COVID-19 Safety Precautions

Morgan Ferqueron, Jonathan F. Bassett*, and Amanda J. Cleveland*

Department of Psychology, Lander University

38 Mask-Wearing and Emotional Intensity Perceptions

Brienna Dove and Teddi S. Deka*

Department of Psychology, Missouri Western State University

46 When Hints Hurt Memory: The Influence of the Number of Part-Set Cues on Free Recall

Deniz Akpinar1, Vaughan Bamford2, Sara Martinez Guzman3, Tracey Nassuna1, Madison Stevens3, and Matthew R. Kelley3*

1Neuroscience Program, Lake Forest College

2Department of Biology, Lake Forest College

3Department of Psychology, Lake Forest College

52 Psychological and Behavioral Predictors of Procrastination in Undergraduates

Bradley B. Gregory1* , Megan Golson2 , and McKay Larsen2

1 Department of Psychology, North Greenville University

2 Department of Psychology, Southern Utah University

67 Interpersonal Identity Cues: The Effect of Therapist Identity on Expectations for the Therapeutic Relationship

Jessica S. Philip and Melanie R. Maimon

Department of Psychology, Rutgers University

79 Replication of the Interpersonal Sunk Cost Effect

Tristan C. Bolinger, Mason A. Ostermiller, and Scott D. Martin*

Department of Psychology, Brigham Young University-Idaho

Conquering APA Style for the Seventh Edition: Advice From APA Style Experts

Jennifer L. Hughes1, Bradley Cannon2, Abigail A. Camden3,

Kimberli

R. H. Treadwell⁴, Joel G. Thomas1, and Bonnie M. Perdue1

1 Department of Psychology, Agnes Scott College

2 Psi Chi Headquarters

3 Department of Psychological Sciences, Auburn University

⁴ Department of Psychological Sciences, University of Connecticut

ABSTRACT. In this updated article, APA Style experts give advice about conquering APA Style based on the seventh edition of the Publication Manual of the American Psychological Association (APA, 2020; cf. Hughes et al., 2017). Learning and teaching APA Style can be difficult because of the many rules (Hughes et al., 2017), and the seventh edition added additional rules and changed existing rules that writers are expected to follow. This article is meant to be a resource for those teaching and learning APA Style. The first part of the article covers these new rules. The second part details common writing issues that APA Style tutors often see. The third part gives writing tips and tricks to help in a research methods course. The next part contains APA Style rules that many writers do not know, and the final part has style rules often missing from empirical research submissions to academic journals.

Keywords: APA writing style, research design and methods, psychology papers

In this updated article, APA Style experts give advice about conquering APA Style based on the seventh edition of the Publication Manual of the American Psychological Association (APA, 2020). APA Style was created to provide guidelines for authors when preparing articles for publication, and also for students preparing papers for course assignments (APA, 2020). Uniformity helps readers to find important information (e.g., key points, findings, sources) quickly and to focus on the ideas in the articles instead of the formatting (APA, 2020).

Learning APA Style takes time and practice especially because of the intricate details of the writing style (APA, 2020; Hughes et al., 2017). Students more effectively learn APA Style if it is presented in many formats (Franz & Spritzer, 2006). Examples mentioned in Hughes et al. (2017) include “checklists (Franz & Spitzer, 2006), templates (Franz & Spitzer, 2006; Stahl, 1987), games (Hughes, 2017), sample articles riddled with APA mistakes (Smith & Eggleston, 2001), peer review of APA Style in papers (Mandernach et al., 2016), and online tutorials (Mages & Garson, 2010)” (p. 154). More recent examples include games (Clark & Murphy, 2021) and video tutorials (Obeid & Hill, 2018).

In addition, teaching the meaning behind the rules for APA Style can help those learning APA Style (Hughes et al., 2017; Mandernach et al., 2016).

This article is meant to be an additional resource for those teaching and learning APA Style. The first part of this article has new rules from the seventh edition of APA Style and the second part details common writing issues that APA Style tutors often see. The third part gives writing tips and tricks to help in a research methods course. The next part contains APA Style rules that many writers do not know, and the final part covers elements that are commonly missing from empirical research submissions to academic journals.

Part 1: New Rules From the Seventh Edition of APA Style

Kimberli R. H. Treadwell, Associate Editor

To space or double space, that is the question that I typically look up first each time the new APA Style book is published (and yes, I have been writing through a few editions of these). This may seem trivial, and one might wonder why the standards change across time. APA Style standards provide guidance to promote

consistency across several fields of study in writing scientific reports. This increases accuracy and readability across the field. Our field evolves, and things have changed a bit since 1929 when psychologists, anthropologists, and business managers first collaborated to produce a 6­page paper with guidelines (Bentley et al., 1929) that evolved into the current publication manual (APA, 2020). The changes between the sixth edition guidelines published in 2009 and those of 2020 reflect expansion to more fields (e.g., behavioral sciences, medical fields, the humanities), expansion to instructors, students, and other consumers, changes in language use, expansion in research, and more. However, the spirit of the original authors remains the same–consistency and uniformity in the presentation of ideas.

Some of the updates to the Publication Manual (APA, 2020) are summarized below by chapter. The changes are varied and do not focus solely on mechanics of writing. A significant addition to the seventh edition is a focus on inclusive language and expression of ideas based on the APA’s inclusive language guidelines (APA, 2021). Other notable changes are highlighted below.

1. Professional Writing

APA (2020) has expanded the scope of its audience with this newest edition, as well as research designs and data sharing. High school students and those new to APA Style find guidelines for class papers, annotated bibliographies, and reaction essays, as well as expansions addressing graduate school milestones for masters’ theses and dissertations (APA, 2020, Section 1.10, p. 9–10). Adding to the information for quantitative research sections, there are now sections for qualitative research methods and mixed ­ method designs (APA, 2020, Sections 1.2–1.5, p. 5–8).

All professional writers will benefit from the expanded ethics in writing section with a convenient table (APA, 2020, Ethical Compliance Checklist, p. 26), summarizing key ethical issues from conception of a study, to ethics review, informed consent, maintaining data confidentiality, and key issues for publishing such as copyright and author agreements. As data sharing and open access to databases has increased in the scientific publishing world, new guidelines assist researchers in navigating these decisions.

2. Formatting a Scientific Paper

Changes for formatting a title page and the running head will be covered in detail in Parts 2 and 3 of this article. Levels of headings and font changes will also be covered in Parts 2 and 3 of this article. Other changes include expansions to the Author Note such as ORCID IDs, statements regarding any conflicts of interest, and

study registration information (APA, 2020, Section 2.3, p. 30–37). Clinical trials are typically registered with the National Institutes of Health. An increasing number of qualitative research studies are preregistering to distinguish between prediction and postdiction as a means to distinguish hypothesis generation and testing while improving the credibility of research. As this practice widens, publicly accessible platforms are becoming available, such as the Center for Open Science, and these steps are now captured on the title page. Finally, more font options and sizes are now acceptable (APA, 2020, Section 2.19, p. 44).

3. Standards in Reporting Research

In 2008, APA developed Journal Article Reporting Standards (JARS) for qualitative research that were adopted as an appendix to the sixth edition. The new seventh edition has a full chapter devoted to JARS based in part on two publications (Appelbaum et al., 2018; Levitt et al., 2018) with noted expansions (APA, 2020). First, qualitative research reporting reflects evolving standards and reflects expansions including dividing hypotheses and conclusions into the three sections of primary, secondary, and exploratory. New modules reflect standards for qualitative research and mixed­methods research, and meta­analyses and structural equation modeling standards are reported. These reporting standards are provided in detail at https://apastyle.apa. org/jars/quantitative

4. Effective Writing

The aim of the fourth chapter is for broader attention to the style of writing to convey meaning and tone. For instance, the term “they” can now be used to refer to a singular person to provide more inclusive language use (APA, 2020, Section 4.16, p. 121). As well it is advised to desist from using “he or she” as generic third person pronouns. A new section outlines helpful wording to avoid anthropomorphism for inanimate sources or animals used in research (APA, 2020, Section 4.11, p. 118). And although not new, authors are encouraged to use active voice as much as possible (APA, 2020, Section 4.13, p. 119).

5. Bias-Free Language

A good practice when proofing a student or professional paper is to read through the manuscript for preconceptions about groups of people, just as would be done for the mechanics of spelling, grammar, and writing style. A comprehensive chapter guides use of language to be as free of implied or irrelevant evaluations of a group as possible. Practices include focusing on relevant characteristics of a person or group and,

Hughes, Cannon, Camden, Treadwell, Thomas, and Perdue | APA Style

when differences that exist are noted, to acknowledge the meaning of the difference in the context of the target population (APA, 2020, Section 5.1, p. 132–133). To decrease biased language, specificity by topic is recommended, such as providing exact age ranges, naming specific conditions or disabilities, providing modifiers of gender, distinguishing nation or region of origin when identifying racial or ethnic groups, identifying specific ranges or socioeconomic status rather than general labels, and indicating the context of research when referring to people taking part in the study (e.g., participants vs. people; APA, 2020, Section 5.2, p. 133–135). These guidelines are in large part based on the newly published inclusive language guidelines (APA, 2021) that provide guidelines in language use and its continual evolution.

6.

Nuts and Bolts of Writing

APA (2020, Section 6.1, p. 154) dictates using one space after a period, but one space is also needed after other punctuation such as colons, commas, and semicolons. Do note however that no space is added after the periods in abbreviations such as e.g., or around the colon for ratios such as 2:6. Expanded guidelines for abbreviations, lists, and using quotation marks for emphasis within the text are provided, and these were also mentioned earlier in this article.

7. Presenting Ideas in Tables and Figures

Authors may now embed tables and figures in the text, or present tables and figures following the reference list per previous guidelines (APA, 2020, Section 7.6, p. 198). Authors format the titles, notes, and numbers for tables and figures using the same format: title, the table or image, and notes (APA, 2020, Section 7.4, p. 197). Color may be used so long as it is accessible (APA, 2020, Section 7.26, p. 228). Over 40 examples of tables and figures in the new Publication Manual (APA, 2020, Sections 7.1–7.36, p. 199–250) help authors to visualize formatting styles.

8. New Guidelines for Citations and References

Changes to citations and references will be covered in more detail in Part 2 of this article, but additional changes to the references will be covered next. APA mentions that the use of “retrieved from” is no longer necessary to use before hyperlinks (see Section 9.35, p. 299–300) unless the source is likely to frequently change and is unarchived (e.g., a website homepage; APA, 2020, Section 9.16, p. 290. For eBooks the platform or device is no longer needed, although the publisher is required (APA, 2020, Section 8.28, p. 274; Section 10.2, p. 322). When citing podcasts or television series, contributing author references are provided, such as the host or director (APA, 2020, Section 8.7, p. 259). Dozens of new reference examples have been

added, highlighting newer online sources that might be cited such as TED talks, podcast episodes, YouTube videos, and social media sources (APA, 2020, Section 10.15, p. 348–352). Guidelines for referencing hashtags and emojis are also addressed. Another new change to the reference list involves using the names of authors for the first 20 authors, rather than the first seven, prior to using ellipsis for the remaining authors (APA, 2020, Section 9.8, p. 286). In addition, authors are asked to include the issue number following the volume number for all journal citations APA, 2020, Section 9.25, p. 294). Part 4 of this article includes information about the publisher’s location no longer being listed for books and similar materials. Finally, over 100 examples demonstrate these guidelines in Chapters 8 and 9 (APA, 2020).

9. The How-to’s for Publishing Your Research

Wondering how to adapt a thesis or dissertation into a journal article? New help has arrived in this edition for students and early career researchers. Condensing a longer education milestone for publication is outlined, as well as guidance for selecting journals, prioritizing helpful journals, and avoiding predatory journals, as well as detailed advice and figures to guide authors through the publication process (APA, 2020, Sections 12.1–12.8, p. 372–381). A new section also assists researchers in promoting their work postpublication, such as sharing a summary of the article and the DOI on professional social media networks (APA, 2020, Section 12.24, p. 395).

10. Finally, You Can Be Part of the Process!

APA Style welcomes comments and feedback that could impact the next edition. Let the APA Style Committee know about how you are using the APA Style Guidelines and additional suggestions at Contact APA Style Many online resources are available to assist learning APA format. Tutorials and webinars are available at https://apastyle.apa.org/instructional­ aids/tutorialswebinars?_ga=2.149162422.1803984740.1665692971696592103.1665313413. Handouts and guides can also be found at https://apastyle.apa.org/instructional­aids/ handouts­guides.

Part 2: Common Writing Issues That Tutors Address

Abigail A. Camden, Former APA Style Tutor

The following section reviews APA Style errors commonly seen by APA Style tutors and teaching assistants. Hughes et al. (2017) outlined 10 problems commonly seen in these settings. These have been updated in response to the seventh edition of the Publication Manual of the American Psychological Association (APA, 2020) when relevant.

1. Missing Citations of Paraphrased Material

One common error seen in student writing is insufficient citations for paraphrased material (APA, 2020, Section 8.23–8.27, pp. 269–270). Any text that is paraphrased from a reference must be cited with an in­text citation, either parenthetically or in the narrative, even if no direct quotation is utilized. In cases where writers provide multiple sentences on one paraphrased source, it is not necessary to repeat the in­text citation if it is clear that the same source is being referred to, unless the paraphrase continues in a new paragraph. Although page numbers are not necessary for paraphrased material, authors may choose to include them if the reference is a long work, such as a book. If a direct quotation is used, the page number(s) should be included in the in­text citation (e.g., “Last Name et al., Date, p. Number). Information about long quotes with 40 or more words will be presented in Part 3.

2. Incorrect Use of “et al.” and In-Text Citations

In the seventh edition, there is a shift in the use of “et al.” in in­text citations (APA, 2020, Sections 8.10–8.22, pp. 261–269). When citing a paper with three or more authors, all in­text citations for that reference should include the first author followed by “et al.” (e.g., Last Name et al., Year).” (For two authors, list both authors each time.) In contrast to the sixth edition (2010), authors should no longer list all authors for citations with three to five authors in the first reference to that paper. The only exception to this is if it creates ambiguity between two or more references that would shorten into the same shortened form. In this case, the writer should list authors until there is a distinguishing name, then follow with et al. (APA, 2020, Section 8.18). Other common errors that arise with use of et al. and in­text citations are (a) the placement of the period (i.e., after “al.” not “et”), (b) use of commas (i.e., only between “al.” and the date), (c) placement of the period at the end of the sentence (i.e., after the ending parenthetical of the in­text citation), and (d) order of in­text citations. Regarding the latter, citations should be listed in alphabetical order with a semicolon between citations (e.g., “Last Name et al., Year; Last Name et al., Year; Last Name et al., Year”). Additionally, in contrast to some other citation styles such as Chicago, “ibid” should not be used to convey that a citation has already been mentioned. Rather, provide the full in­text citation each time. Finally, et al. may be used both in parenthetical citations and in narrative citations.

3. Common Errors With Parentheses

Parentheses are commonly used in APA Style (APA, 2020, Sections 6.8–6.9, pp. 159–160). Some common errors seen with parentheses are using two parentheticals beside each other rather than combining them with a semicolon.

Another common error is improper usage of “i.e.” (that is) or “e.g.” (for example) in parentheses, which should be followed by a comma (i.e., like this). Next, periods should be placed outside of the parenthetical, unless an entire sentence is contained in the parenthetical, in which the period would be placed inside. Last, rather than using a parenthetical within a parenthetical, brackets should be used inside instead.

4. Incorrect References

In the seventh edition of APA Style (2020), there are minimal changes to reference format, with the exception of book references, from which place of publisher is no longer included. The references section is one place of the manuscript where errors in APA Style are often found. See Table 1 for examples of commonly used references; also see APA (2020, Section 9, p. 281, and Section 10, p. 311) for more information on reference lists and reference examples. For example, in a journal article reference, writers should: (a) place the title of the article in sentence case, (b) place the name of the journal in title case and italicize it, (c) italicize the volume number but not the issue number, (d) use an en dash (i.e., rather than a hyphen) between page numbers (see Part 4 of this article for more information about en dashes), and (e) use a hyperlinked digital object identifier (DOI) at the end when possible. References derived from sources such as Google Scholar often do not include the DOI. The DOI can usually be found in the article or the article webpage on the journal website. If a DOI is not available, the URL may be utilized instead. APA now requires DOIs or URLs to be included as hyperlinks that directly link to the resource to simplify retrieval (APA, 2020, section 9.27). When formatting the references section, references should be presented in

TABLE 1

Examples of Common References

Journal Article

Last Name, A. B., Last Name, C. D., & Last Name, E. F. (Year). Title of article goes here. Name of Journal Goes Here, 1(2), 345–678. https://URL of DOI here, with no period at the end*

*The reason that there is not a period at the end of the URL is because a period could interfere with the functionality of the link (see Section 9.35, pp. 299–300)

Book Without Editors

Last Name, A. B. (Year). Title of book goes here (Edition number here if relevant).

Publisher Name Here. https://URL of DOI if relevant here, with no period at the end

Chapter in an Edited Book

Last Name, A. B., Last Name, C. D., & Last Name, E. F. (Year). Title of chapter goes here. In G. H. Last Name & I. J. Last Name (Eds.), Title of book goes here (Edition number here if relevant, pp. Page Number–Page Number).

Publisher Name

Here. https://URL of DOI if relevant here, with no period at the end Note. See APA (2020) for additional examples and guidance regarding references.

Hughes, Cannon, Camden, Treadwell, Thomas, and Perdue | APA Style

alphabetical order, with each reference presented with a hanging indent on the second line. If an author is cited multiple times, place the references with that first author in alphabetical order according to the second author; if that author is the single author on multiple references, order the reference in ascending date. Finally, if writers utilize references produced from sources such as Google Scholar, they should be cautious to check the APA Style. If authors do not have a copy of the APA manual, the APA Style website (https://apastyle.apa.org/) is a reputable source to refer to for the formatting of references (as well as other APA Style questions, of course!).

5. Formatting of Text

As mentioned in Part 1 of this article, one notable change in the formatting of text from the sixth edition to the seventh edition is the number of spaces following a sentence (APA, 2020, Section 6.1, p. 153). Writers should only use one space after a sentence rather than two. If authors are struggling to adjust to this change, they can utilize the “find and replace feature” of their word processing software to search for two spaces in their document (i.e., search for “ ” and replace with “ ”). Another common error seen regarding formatting of text is the spacing between lines, which should consistently be double spaced throughout the document (APA, 2020, Section 2.21, p. 45). Writers will want to check that the word processor software has not added additional spaces before or after paragraphs. This can be found by highlighting the entire document and then going to “line spacing options” under the “line and paragraph spacing” icon on the ribbon. Authors should then make sure that line spacing is set to “double” rather than “multiple” or “single” and that space before and after paragraphs is set to “0 pt.”

6. Italicize Most Statistical Coefficients

Another error commonly found in articles, particularly within the results section is failing to italicize statistical coefficients (APA, 2020, Section 6.44, pp. 182–187). Almost all statistical coefficients should be italicized (e.g., p , t , N , n , M , SD ). One exception to this rule, however, is Greek statistical coefficients (e.g., χ², partial η2) and subscripts or superscripts. When italicizing, be cautious not to italicize anything that may follow the coefficient, such as an equal sign.

7. Levels of Heading

As a tutor, I would often find that students underutilized subheadings in their writing. However, using subheadings within sections of the paper, such as the literature review, can add clarity and structure to writing, serving as signposts to guide the reader (Hughes et al., 2017).

With the seventh edition of the APA Publication Manual, levels of heading were updated (APA, 2020, Section 2.27, p. 47–49). The main change in the seventh edition is that there is an additional level of heading required before using headings that end in a period (i.e., now begins at the fourth level of heading), and that some section headings are now bolded (i.e., the title of the paper, “Abstract,” “References,” “Table,” and “Figure”). For a helpful table outline of paper headings, see the Publication Manual (APA, 2020, pp. 48–49). For example, writers may wish to use subheadings in a literature review. In this case, after the title page, authors should begin the literature review with the title of their paper, bolded and centered (i.e., the first level of heading). Levels of heading in the literature review should then begin with the second level of headings (i.e., flush left, bold, title case). Then, similar to an outline, the third heading (i.e., flush left, bold italic, title case) can be utilized under any second level heading that requires two or more subsections. Further levels of subheadings may be utilized in a similar fashion if warranted (e.g., fourth level: indented, bold, title case, ending with a period).

8. Running Heads

There is one main change to running heads for professional title pages in the seventh edition of APA (2020, Section 2.8, p. 37). In contrast to the sixth edition of APA (2010), authors no longer need to include the label “Running head” before the running head on the first page of the manuscript. Rather, the running head should be the same on each page. It should be formatted in all caps, flush left, while the page number should be flush right at the 1­inch left margin and half of an inch down from the top of the page within the header. The content of the running head should be no more than 50 characters and should serve as a summary of the title of the paper. One common mistake seen in manuscripts is for the running head and page number to be a different font and font size than the body of the manuscript. Authors should ensure that this font style is consistent with the rest of their manuscript.

9. Effective Abstracts

As noted by Hughes et al. (2017) and APA (2010, 2020), the abstract is one of the most important paragraphs in the manuscript, given that readers often decide on the utility of a paper based on this summary (Section 3.3, pp. 73–75). Given this, the abstract should be dense with information and cover the most important parts of the research paper, including the aim of the study, the sample utilized, important aspects of the methodology, key results accompanied by significance levels, confidence intervals, and effect size when space allows, and conclusions (e.g., implications).

Hughes et al. (2017) provided a helpful strategy for writing concise abstracts with all important information covered. To use this strategy, writers can create a structured abstract that includes headings for each important domain (e.g., objective, method, results, conclusions). The headings will prompt writers to include all important domains of the paper, and writers can aim to use one to two sentences per heading to keep the abstract concise. These headings can then be removed if needed. Finally, authors should also include carefully chosen key words at the end of the abstract that will aid readers in finding their article (APA, 2020, Section 2.10, p. 38–39).

10. Word Choice and Flow of Writing

As noted by Hughes et al. (2017), effective communication in writing is a challenge for all writers. The following section includes common problems outlined by Hughes et al. (2017) in addition to other considerations. Common writing problems seen by tutors include: (a) lack of transition words (APA, 2020, Section 4.2, p. 112), (b) not utilizing parallel construction (Section 4.24, p. 124–125), (c) not using an Oxford comma (Section 6.3, p. 155), (d) use of colloquialisms (e.g., “a lot,” “actually;” Section 4.8, p. 116), (d) improper use of words such as “since” or “like” (e.g., Section 4.22, p. 123) and (e) overuse of jargon (Section 4.9, p. 116). For example, to aid in the flow and readability of writing, writers should utilize transitionary words or phrases (e.g., “for example,” “likewise,” “although”) and should use parallel structure of elements (e.g., in compound sentences, lists). Additionally, when listing elements, authors should utilize an Oxford comma before the resultant “and” (e.g., incorrect: “depression, anxiety and mania” correct: “depression, anxiety, and mania;” ). To improve the comprehension of their writing, authors should avoid colloquialisms (i.e., informal words or expressions) and jargon (i.e., technical terms that are difficult to understand for nonspecialists). If jargon is used, terms should always be defined the first time that they are present. Finally, to improve one’s writing, authors should consider seeking out feedback from others, such as that from campus writing centers, advisors, and professors. For additional information on improving writing style, see Section 4 of the APA Style Manual (2020, pp. 111–127).

Part 3: Writing Tips and Tricks to Help in a Research Methods Course

Joel G. Thomas and Bonnie M. Perdue, Research Design and Methods Professors

Research design and methods courses play a special role in student development. In addition to providing a space in which students can explore the research process by learning about different methods of investigation, these courses

often offer students their first experience engaging directly with articles as well as reading and writing using scientific language. The skills they learn in these courses are essential to upper level course success because professors often expect that students at that stage can communicate like a researcher or professional. We commonly see the following APA writing errors in our research design and methods courses. The tips that follow address how papers can be written more effectively in APA format and are updates to the tips discussed in reference to the prior edition of the manual (APA, 2010, Hughes et al., 2017).

1. Use of “As cited in…”

Primary sources are ones that report original content, whereas secondary sources report content that is cited in another source. According to APA (2020), whenever possible, authors should try to find the primary source and use this for in­text citations and references. This includes citing original research directly rather than citing a professor’s lectures or textbook that cites a primary source. In cases in which the primary source is out of print or in a language the author does not understand, the secondary source can be cited using the following format: (Rabbitt, 1982, as cited in Lyon et al., 2014, APA, 2020, Section 8.6, p. 258). Note: If only the secondary source (i.e., Lyon et al., 2014) was read then it should be the only one listed in the reference section and not the primary source.

2. Direct Quotations

Direct quotations should be used sparingly in scientific writing. Typically, a quotation should only be used when the wording used in the quote is essential to the context in which the quote is discussed (e.g., when reproducing an exact definition, when an author has said something memorably or succinctly, or when an author wants to respond to the exact wording in the quote; APA, 2020, Section 6.22, p. 170). As mentioned in part 2, if the quote is less than 40 words, it can be included in the text with quotation marks around the quote followed by the parenthetical citation and page number(s). However, if the quote is 40 words or more, it should be presented in block form: indented 0.5 in. from both margins with no quotes, introduced with a colon from the preceding sentence, and should end with the punctuation for the quote followed by the parenthetical citation. If the author and year appeared in the sentence introducing the quote, only the page number(s) should appear in the parenthetical citation (APA, 2020, Sections 8.25–8.36, pp. 270–278).

3. Use of Contractions

As scholarly writing is typically technical and formal in nature, contractions should generally be avoided (APA,

Hughes, Cannon, Camden, Treadwell, Thomas, and Perdue | APA Style

SPRING 2023

2020, Section 4.8, p. 116). Exceptions include when reproducing a quote with a contraction, when using an idiom that contains a contraction, or when intentionally making an informal remark (Lee, 2022).

4. Semantics

Science is an open­ended activity that involves steps such as verification of hypotheses, refutation, reconceptualization, and replication (Cuttler et al., 2020). Authors should acknowledge this tentative context by using words such as “suggests,” “indicates,” or “demonstrates,” when introducing findings from the empirical literature. Definitive vocabulary such as “proves” should be avoided as this overstates the validity and reliability of findings in psychological science. For additional information on improving clarity in writing style, see Sections 4.7–4.11 of the APA Publication Manual (2020, pp. 115–117).

5.

Student Paper Versus Professional Paper

The seventh edition offers several clear distinctions between student papers and professional papers. These differences, and how to look for and identify the correct information, may need to be reiterated to students. One noteworthy difference is that student papers no longer require an abstract (APA, 2020, Section 2.2, p. 30). Professional papers continue to typically include an abstract page, author note, and running head.

6. Student Title Page

There are several items to note regarding title pages for student papers. A standard title page for a student paper will not include a running head, although it should contain the page number (APA, 2020, Section 2.3, p. 30). In addition to the title, student name, and affiliation, students should also include the course number and name, the instructor’s name, and the due date of the assignment on the title page (APA, 2020, Section 2.3, p. 30). We have found that students often leave off the blank line between the title and their name.

7. Page Numbers

As technology progresses, citation style will adapt to change. One example of this to emphasize to students in a research methods course relates to elements of a reference entry. As online publications become more common, page numbers are not always an available element to include in a reference entry. If an article comes from an online journal that does not have page numbers but has an article number, one should instead write “Article” followed by the article number (APA, 2020, Section 9.27, p. 294). An example is PLOS ONE, 17(8), Article e0270739.

8. Paragraph and Subsection Length

Hughes et al. (2017) described how paragraph length in student papers is sometimes too short and emphasized the need for an introductory sentence, body of paragraph, then concluding or transitional sentence in all paragraphs. A related trend is that students will sometimes create too many subsections within a paper. The number of distinctive sections in a paper should guide the number of headings, which averages three levels (APA, 2020, Section 2.27, pp. 47–48). A very brief introduction does not necessarily need many subheadings, but students in a methods course who are beginning to write in APA format will sometimes tend toward using a lot of unnecessary subsections.

9. Spacing in References

The spacing between elements in a reference list can sometimes be confusing with students initially including double spaces, or sometimes no spaces at all. An easyto­follow rule is that there should be one space between the different elements of a reference entry, including the first and middle initial of authors (APA, 2020, Section 9.8, p. 286).

10. Numbers in APA

Students often report numbers incorrectly (Hughes et al., 2017). The standard rule is to use words for numbers below 10 (e.g., one, two, three) and use numerals to express numbers 10 and above (e.g., 10, 11, 12). However, there are exceptions to this rule (see APA, 2020, Section 6.32–6.39, pp. 178–181), so writers will want to reference the Publication Manual

Part 4: APA Style Rules Many Writers Do Not Know

Jennifer L. Hughes, Associate Editor

In the original article, I selected the following 10 errors that are not often known by writers, which I found to be frequently committed in research papers (Hughes et al., 2017). The errors listed in the original paper have been updated to be in alignment with the new edition of the Publication Manual released in 2020. Information about the rules and how to avoid violating these rules is included.

1. Emphasizing Terms

On the first use of words or phrases, authors should use quotation marks for slang, ironic comments, or invented or coined expressions (APA, 2020, see Section 6.7, pp. 157–159). Although past versions of APA (2010) allowed authors to italicize terms for emphasis, APA (2020, see Section 6.22, pp. 170–171) suggests avoiding italics for emphasis. They suggest rewriting sentences in order to provide emphasis. This could include placing important

words or phrases at the beginning or end of a sentence. Exceptions include using italics if the material might be misread, emphasis lost, or the first time defining or introducing a key term or phrase (i.e., given with a definition; see Section 6.7, pp. 157–159).

2. Presentation of Lists of a Series of Items

APA (2020) gives directions about how authors should list a series of items (see Section 6.49, p. 189). If authors are using a series that is within a sentence or paragraph, they should use letters instead of numbers. For example, (a) first, (b) second, and (c) third. If authors want to display sentences or paragraphs in a series (i.e., itemized conclusion or steps in a procedure) they should use numbered lists by using Arabic numerals followed by a period (see section 6.51, pp. 189–190). The first word should be capitalized. APA cautions that using numbered lists can indicate ordinal position among the items and if authors do not want this to happen, they should use bulleted lists instead.

3. Spacing of Numbers in Text

When using numbers, authors should treat them like words (APA, 2020, see Section 6.45, pp. 187–188). The numbers should have spaces before and after them, which makes them easier to read (e.g., not 3+7, but 3 + 7). However, if a minus sign is used to indicate a negative value, a space is used before the minus sign, but not after it.

4. When to Use a Zero Before Decimals

APA (2020) suggests to “use a zero before the decimal point in numbers that are less than 1 when the statistic can exceed 1” (p. 180, see Section 6.36). Hughes et al. (2017) gave the example of Cohen’s d being reported as 0.70 because it can exceed 1. Authors should not use a zero before a decimal when the statistic being used cannot exceed 1 such as correlations and levels of statistical significance (APA, 2020).

5. Punctuation and Quotation Marks

Authors should put periods and commas inside of quotation marks (APA, 2020, see Section 6.7, pp. 157–159). For example, “. . . like this.” Other punctuation marks such as colons, semicolons, and ellipses are placed outside of the quotation marks.

6. Using En Dashes Between Page Numbers in References

Both academics and students often say they have never heard of an en dash (Hughes et al., 2017). APA (2020) uses two types of dashes, which include the em dash and the en dash (see Section 6.6, p. 157). The em dash is longer than a hyphen and is used to set off an element. The en

dash is shorter than an em dash but is longer and thinner than a hyphen. It is used between page or date ranges. The following includes examples of them: hyphen (­), em dash (—), and en dash (–). Once authors find out about the en dash, they often do not know how to create one because there is not a key on keyboards for them. These directions can help. Those using a Mac who want to add an en dash should hold down the option key while pressing the minus key at the top of the keyboard (McAdoo, 2010). Those using a PC should hold the control key and type the minus key at the top of the keyboard (McAdoo, 2010).

7. Personal Communications Need to Be Cited in the Text, but Not the References

Information that cannot be recovered by readers is cited as personal communications. However, it should be noted that if the original source of information can be found, then it should be cited instead. Personal communications “include emails, text messages, online chats or direct messages, personal interviews, telephone conversations, live speeches, unrecorded classroom lectures, memos, letters, messages from nonarchived discussion groups or online bulletin boards, and so on” (APA, 2020, Section 8.9, p. 260). This information is cited in the text as (J. L. Hughes, personal communication, March 6, 2023) or J. L. Hughes (personal communication, March 6, 2023) and is not included in the reference section (APA, 2020, Section 8.9, p. 260).

8. Article Titles That End With an Exclamation Point or Question Mark

When giving a title in the references and it already contains punctuation like an exclamation point or question mark, the authors should keep that punctuation and not add a period because two punctuation marks after the title are not needed (APA, 2020, see Section 9.19, pp. 291–292).

9. Listing Editors of Books in References

When writing references for books, editors’ initials go before their surnames (e.g., J. L. Hughes & A. A. Camden), which is the opposite of how authors are listed (APA, 2020, see Section 9.28, p. 295). This was shown in Part 2 of this article. However, most writers do not know that a comma is not used between two editors’ names (APA, 2020, see Section 10.3.38, p. 326), but it is used if there are three or more editors (APA, 2020, see Section 10.3.40, p. 326) or two or more authors (APA, 2020, see Section 10.3.43, p. 327).

10. Pronouns for People and Animals –“Who” vs. “That”

Authors should use “who” for people and “that” for nonhuman animals (i.e., rats) and for inanimate objects

Hughes, Cannon, Camden, Treadwell, Thomas, and Perdue | APA Style

(APA, 2020, see Section 4.19, p. 121). For example, “the professors that taught the psychology courses” would be incorrect. “The professors who taught the psychology courses” would be correct. APA (2020, Section 4.19, p. 121–122) also advises to use neuter pronouns when referring to animals such as “the dog” but they note that if the dog has a name and the sex is known, gendered pronouns are acceptable.

Part 5: Elements Commonly Missing From Empirical Research Submissions

Bradley Cannon, Managing Editor

In the original article (Hughes et al., 2017), I discussed 10 APA Style mistakes that authors frequently make when submitting to Psi Chi Journal. At that time, I had reviewed over 700 of these manuscripts, and that number is now closer to 1,300. Despite the many changes introduced in the new seventh edition of the Publication Manual of the American Psychological Association (APA, 2020), I find that most of those original tips hold up well, so this time around I focus on 10 errors that are all­too­common and also highly likely to impact editors’, peer reviewers’, and—if unaddressed during the review process—future readers’ views of a published article. Specifically, I wish to address important details that are frequently missing entirely from empirical manuscripts. When information is not just insufficient but absent, editors and reviewers must request that these details be added during the next round of revisions. As a result, the reviewers cannot give accurate feedback on this unseen content until the submission returns for revision, which greatly increases the likelihood that the authors will then have to endure an additional round of revisions after that. This may even be a best­case scenario, as opposed to the worst­case, in which the editor or reviewer may reject the manuscript entirely under the belief that the missing information (e.g., an effect size, participant demographics) was intentionally omitted in order to hide some sort of fatal flaw. So, what are those missing elements? Let us begin!

1. Missing Hypotheses

Hypotheses are the proposed suppositions made before a new study is conducted, generally based on prior research. Hypotheses are likely one of the first lessons taught to students about conducting science; however, hypotheses are sometimes forgotten months or even years later when a completed study is finally submitted for publication.

APA (2020) Table 3.1 (JARS­Quant, p. 78) outlines that authors should state specific hypotheses, aims, and objectives, any theories they are based on, and other planned analyses in their manuscript. Further, APA (2020, Section 3.6, p. 86) also refers to exploratory hypotheses, which are

suggested by authors based on data collected for the present study rather than being based on previously reported studies. If a study contained exploratory hypotheses or even had no specific hypotheses due to its exploratory nature, the authors should explain this in the manuscript body. Hypotheses are typically placed near the end of a manuscript’s Introduction section, as shown in the APA (2020) Sample Paper (pp. 52–53). They may be formatted in paragraph format or as a numbered list (APA, 2020, Section 6.51, pp. 189–190).

2. Unreported Participant Demographics

Authors often fail to report some or all relevant demographic information for participants. Not only does this cause challenges for reviewers when analyzing a paper’s methodology and interpretation of the results, it can also create questions later with regard to replications and who the results would most likely apply to. At minimum, authors should include age, gender identity and/or sex assigned at birth, ethnicity and race, and socioeconomic status and social class (Appelbaum et al., 2018), and they should also include any additional demographics relevant to the context of their study. Phrasing demographics questions can be tricky, so a recent Psi Chi Journal editorial (Hughes et al., 2022) provides 38 examples of how to write various questions. Researchers are also encouraged to review the Publication Manual (APA, 2020, Section 5: Bias­Free Language Guidelines). In the unfortunate event that the authors failed to collect demographics, explain this in the manuscript’s Limitations section (APA, 2020, Section 3.8, pp. 89–90). On a related note, most participants in psychology­related journals are WEIRD (Western, educated, industrialized, rich, and democratic), which can unfortunately affect the generalizability of study results (Brass & Charlton, 2022). Authors of all Psi Chi Journal articles are now encouraged to briefly address in their Limitations section how their participant characteristics might limit the extent to which the results can be generalized, and they should identify populations and subpopulations to whom generalizations might be unwarranted. For an excellent example of this, we often cite Goldie and O’Connor (2021).

3. Failure to Introduce Abbreviations

It is important to remember that not all readers and reviewers will be well­versed in an article’s niche subject area and its unique jargon. APA (2020, Section 6.24, p. 172) advises authors to introduce each abbreviation the first time that it is used and then consistently use the abbreviation only throughout the remainder of the manuscript. The abbreviations should only be reintroduced in each table or figure where they may appear, generally in the table or figure’s Notes section (APA, 2020, Section

7.15, p. 205). A few exceptions should be mentioned.

First, statistics (e.g., M, p) and abbreviations listed as terms in a dictionary (e.g., AIDS, LGBTQ) need not be introduced. Second, authors are encouraged to avoid overusing abbreviations because this can also be confusing for readers. To reduce the number of abbreviated terms, avoid abbreviating short terms (e.g., L for large) and never abbreviate a term that appears fewer than three times in a manuscript (APA, 2020, Section 6.24, p. 172).

4. Missing IRB Approval

Institutional review boards (IRBs) protect and manage risk to human participants; receiving their approval or exemption before any data collection begins is essential. The latest edition of the Publication Manual requires authors to “describe institutional review board agreements, ethical standards met, and safety monitoring” (APA, 2020, Table 3.1, p. 77–81). This information should be included in the Method section along with other sampling procedures such as any payments made to participants. Although a journal’s editors and peerreviewers will know IRB approval was required prior to submission, future readers may not be aware of a journal’s specific policies, making inclusion of this step highly important.

5. Absent Effect Sizes and Reliability Coefficients

Among the research reporting practices that have become increasingly prevalent are the inclusion of effect sizes and reliability coefficients. In general, all F and p statistics should be paired with an effect size (e.g., d; APA, 2020, Section 3.7, p. 89). Psi Chi Journal even requires this for nonsignificant p results. Whenever possible, a reliability coefficient (e.g., α) should also accompany the description of any multi­item measures used (APA, 2020, Table 3.1, p. 79). Before submission, authors should always access a journal’s submission guidelines and even browse a few recently published articles to identify whether any details such as these should be added to their manuscript.

6. Omitted Limitations Section

Readers unfamiliar with empirical research may be unaware that limitations are highly common across all published studies. Despite this, editors and peer reviewers will be expecting these details, and omission of such information may even cause them to view an article with a more skeptical eye. APA (2020) does not explicitly require a distinct section header for Limitations, but authors are welcome to include this. At minimum, authors should set aside a couple paragraphs in their Discussion section to identify any limitations. As examples, failure to collect adequate participant demographics information should be discussed as

well as participant constraints of generality (previously discussed in the second item of this list). Further, APA (2020, Section 3.9, p. 90) advises authors to include sources of potential bias, imprecision of measures, and more. On the other extreme of failing to include limitations, authors occasionally conclude their papers with a discussion of their limitations, but what a sour note to end on! When this happens, I often encourage authors to consider adding a brief conclusion paragraph to remind readers (a) what the study accomplished, (b) where future research should look next, and (c) why the present research is important.

7. No Permission to Reprint

A more unsettling title for this seventh section of the list would be “Plagiarism.” Although possibly well­meaning, authors sometimes fail to consider whether they own the copyright of all the materials they are seeking to publish. For example, they will occasionally include an appendix detailing all the questions of a previously published scale; photographs that participants viewed, which were acquired without permission from the internet; or in ­ text information lacking appropriate citations and references. It is the authors’ responsibility to obtain appropriate permission from such sources. See APA 12.17 (pp. 387–389) and 12.18 (pp. 389–390) for details on obtaining permission from the copyright owner. When authors wish to include information for which permission would be required, they should seek out permission as early as possible, because this can sometimes be a lengthy process depending on the copyright owner’s procedures, and of course, fees may be involved. It is also important to keep in mind that publishers (e.g., Wiley, SAGE) have their own unique rules regarding permissions to reprint. Although these guidelines must fall within the restrictions of U.S. copyright law, they may be slightly more or less strict than, for example, the APA’s permissions to reprint. For instance, a figure in one publication may be fully open access, whereas a similar figure elsewhere may require permission to reuse some or any of its contents. As a general rule, cite original sources for all content and materials used and also determine whether official permission is needed. Determining whether permission is needed is not always easy. Although some individuals will say that permission should be obtained any time 10% or more of an article is reproduced, the law does not identify any specific percentage of copied work that automatically requires permission (Misthal, 2017). APA (2020, Section 8.34, p. 277) says that permission may be needed for “lengthy quotations” (usually more than 800 words). Regardless, it stands to reason that the more content that is reused, the more likely it is to fall outside of fair use (i.e., permission would be needed). Further, even a

Hughes, Cannon, Camden, Treadwell, Thomas, and Perdue | APA Style

very short amount of an existing article can be considered beyond fair use if it reuses “the heart of the work” from the existing article, particularly if used for commercial purposes (Columbia University Libraries, n.d.). Shorter and more creative works often also have stricter copyright rules (Copyright Alliance, n.d.). Therefore, permission is likely needed when using any content from (a) an existing table or figure (which is often considered a standalone piece separate from its related article; APA, 2020, Section 12.15, p. 385); (b) a scale, vignette, questionnaire, or data (APA, 2020, Section 12.15, p. 385); (c) a song or poem (APA, 2020, Section 8.34, p. 277); or (d) artwork or photography, including photographs taken by the authors of identifiable people (APA, 2020, Section 12.17, p. 387–389). Please note that this is not legal advice, nor is it comprehensive.

8. Citations Missing in the References Section

This reviewer’s general experience is that authors have improved at writing references during the past five or so years, partly due to increased accessibility of online example documents (e.g., APA, 2020, Sample Professional Paper, pp. 50–60; also available at https:// apastyle.apa.org/style­grammar­guidelines/paper­format/sample­papers) and free online reference generator tools (e.g., APA Style Central, Scribbr, MyBib). Still, one problem that continues to occur are some citations missing entirely from the references section (and likewise, some references that are not cited in the manuscript body). Because most articles contain numerous pages of references, it is easy to imagine how this could happen. However, this can again result in accidental plagiarism (APA, 2020, Section 8.2, pp. 254–256). To ensure that all citations and references match up, best practice is to use Microsoft Word’s “Find” feature to search for the first author’s last name of each individual citation in the manuscript body and then do the same for each reference in the References list. In discussing the relationship between references and in­text citations, APA (2020, Section 8.4, p. 257) notes a few exceptions such as personal communications and general mentions of whole websites or periodicals, which do not require corresponding references. However, in general, all citations and references should match up. Importantly, last names and dates should use consistent spelling too!

9. No Statistics in the Abstract

According to APA (2020), abstracts should include “the basic findings, including effect sizes and confidence intervals and/or statistical significant levels” (Section 3.3, pp. 73–75). Many journals do not require this information; however, it can be extremely helpful for readers and even editors when they are deciding whether to dig

deeper into the full text of an article. Academic rigor is not consistent across all publications and authors’ works, so including this information will help readers “get a feel” for an article’s credibility. At a minimum, authors should include the sample size (e.g., N = 52) and statistics for a few of the primary findings (e.g., F[3, 44] = 12.79, p < .001,η2 = .47). Psi Chi Journal requires this information for quantitative studies; when conducting preliminary APA Style reviews, I request for the authors to add statistics to their abstracts for more than three quarters of papers submitted to our journal.

10. Unmasked Information

The final item in this list addresses articles that are improperly masked—a reverse item of sorts for this current list. When a document is masked, it means that the authors’ names and affiliations should be missing from the file to protect the authors’ identities during the review process. And yet all too often, authors fail to do this. Of the past 10 papers I reviewed, eight were not properly masked. Authors should remove their names and affiliations from the manuscript text. This is done correctly most of the time; however, it is only the first step. Hyperlinks generated through an author’s school may include the school’s name in the link, so authors should also be on the lookout for this. For example: https://doi­org.lib.pepperdine.edu/10.1037/h0025583 should be changed to https://doi.org/10.1037/h0025583 (Also of note: only readers with access to Pepperdine’s database subscriptions would be able to view the article at that first link, so it is advisable to replace the first link, even for papers intentionally left unmasked). Any school names in the DOI hyperlinks should be removed; in niche areas of research, this embedded affiliation can easily give away the authors’ identities to an astute reviewer. Further, authors should select the File tab in the upper left of Microsoft Word. On the right side of the page that will appear, “Author” and “Last Modified by” fields commonly feature the authors’ identities. If an author can see their name here, then any future reviewers can too. To remove this information, select “Inspect Document” on this same page and then choose to “Remove All” documents and personal properties from the file (due to numerous versions of Word, authors may have to search online for different instructions to remove this information).

Conclusion

In this article, we have highlighted key rules of APA Style as well as changes in the seventh edition of the Publication Manual (2020). We hope that this article can be a resource for both those learning APA Style and those teaching it. We encourage writers to continuously learn about and practice using APA Style (Hughes et al., 2017).

References

American Psychological Association. (2020). Publication manual of the American Psychological Association (7th edition). Author. American Psychological Association. (2021). Inclusive language guidelines

https://www.apa.org/about/apa/equitydiversity-inclusion/languageguidelines.pdf

Appelbaum, M., Cooper, H., Kline, R. B., Mayo-Wilson, E., Nezu, A. M., & Rao, S. M. (2018). Journal article reporting standards for quantitative research in psychology: The APA Publications and Communications Board Task Force report. American Psychologist, 73(1), 3–25.

https://doi.org/10.1037/amp0000191

Brass, T., & Charlton, S. R. (2022, Winter). Has psychology gotten any less WEIRD? Eye on Psi Chi, 27(2), 8.

https://www.psichi.org/blogpost/987366/480469/

Bentley, M., Peerenboom, C. A., Hodge, F. W., Passano, E. B., Warren, H. C., & Washburn, M. F. (1929). Instructions in regard to preparation of a manuscript. Psychological Bulletin, 26(2), 57–63. https://doi.org/10.1037/h0071487

Clark, D. A., & Murphy, W. (2021). The efficacy of a classroom game for teaching APA Style citation. Teaching of Psychology, 48(3), 209–214. https://doi.org/10.1177/0098628320977263

Columbia University Libraries, Copyright Advisory Services. (n.d.). Fair use. https://copyright.columbia.edu/basics/fair-use.html

Copyright Alliance. (n.d.) What is fair use? https://copyrightalliance.org/faqs/what-is-fair-use/

Cuttler, C., Jhangiani, R. S., & Leighton, D. C. (2020). Research methods in psychology. Open Textbook Library.

Franz, T. M., & Spitzer, T. M. (2006). Different approaches to teaching the mechanics of American Psychological Association Style. Journal of Scholarship of Teaching and Learning, 6(2), 13–20. https://www.semanticscholar.org/paper/InstructionalStrategies-to-Improve-College-APA-Mandernach-Zafonte/ c4877d87c19ae4f201d1e06a26c799a55a6eed37

Goldie, P. D., & O’Connor, E. E. (2021). The gender achievement gap: Do teacher–student relationships matter? Psi Chi Journal of Psychological Research, 26(2), 139–149. https://doi.org/10.24839/2325-7342.JN26.2.139

Hughes, J. L. (2017, March 8–11). Teaching APA Style by making it into a game. In J. L. Hughes, Keeping students' interest: Class activities for undergraduate psychology courses [Conference presentation]. Sixty-third annual meeting of the Southeastern Psychological Association, Atlanta, GA, United States.

Hughes, J. L., Brannan, D., Cannon, B., Camden, A. A., & Anthenien, A. M. (2017). Conquering APA Style: Advice from APA Style experts. Psi Chi Journal of Psychological Research, 22(3), 154–162. https://doi.org/10.24839/2325-7342.JN22.3.154

Hughes, J. L., Camden, A. A., Yangchen, T., Smith, G. P. A., Domenech Rodríguez, M. M., Rouse, S. V., McDonald, C. P., & Lopez, S. (2022). Guidance for researchers when using inclusive demographic questions for surveys: Improved and updated questions. Psi Chi Journal of Psychological Research, 27(4), 232–255. https://doi.org/10.24839/2325-7342.JN27.4.232

Lee, C. (2022, October 23). Contractions in formal writing: What’s allowed, what’s not. APA Style Blog. http://blog.apastyle.org/apastyle/2015/12/ contractions-in-formal-writing-whats-allowed-whats-not.html

Levitt, H. M., Bamberg, M., Creswell, J. W., Frost, D. M., Josselson, R., & SuárezOrozco, C. (2018). Journal article reporting standards for qualitative primary, qualitative meta-analytic, and mixed methods research in psychology: The APA Publications and Communications Board task force report. American Psychologist, 73(1), 26–46.

https://doi.org/10.1037/amp0000151

Mages, W. K., & Garson, D. S. (2010). Get the cite right: Design and evaluation of a high-quality online citation tutorial. Library and Information Science Research, 32(2), 138–146. https://doi.org/10.1016/j.lisr.2009.10.004

Mandernach, B. J., Zafonte, M., & Taylor, C. (2016). Instructional strategies to improve college students’ APA Style writing. International Journal of Teaching and Learning in Higher Education, 27, 407–412. https://www.semanticscholar.org/paper/InstructionalStrategies-to-Improve-College-APA-Mandernach-Zafonte/ c4877d87c19ae4f201d1e06a26c799a55a6eed37

McAdoo, T. (2010, September 23). Computer editing tip: En dashes. APA Style Blog. http://blog.apastyle.org/apastyle/2010/09/computer-editing-tip-en-dashes.html

Misthal, M. P. (2017, July). Copyright and the myth of the “ten percent rule.” Gottlieb, Rackman & Reisman, P.C. https://grr.com/publications/copyright-myth-ten-percent-rule/ Obeid, R., & Hill, D. B. (2018). Freely available instructional video and rubric can improve APA Style in a research methods classroom. Scholarship of Teaching and Learning in Psychology, 4(4), 308–314. https://doi.org/10.1037/stl0000123

Smith, G. E., & Eggleston, T. J. (2001). Comprehending APA Style through manuscript analysis. Teaching of Psychology, 28(2), 108–110. http://dx.doi.org/10.1207/S15328023TOP2802_08

Stahl, J. M. (1987). Facilitating use of the APA Publication Manual in the scientific writing Course. In V. P. Makosky, L. G. Whittemore, & A. M. Rogers (Eds.), Activities handbook for the teaching of psychology (vol. 2, pp. 311–314). American Psychological Association.

Author Note. Jennifer L. Hughes

https://orcid.org/0000­0002­7978­5650

Bradley Cannon

https://orcid.org/0000­0002­7724­5829

Abigail A. Camden

https://orcid.org/0000­0001­8231­836X

Kimberli R. H. Treadwell

https://orcid.org/0000­0002­6595­4193

Joel G. Thomas

https://orcid.org/0000­0002­3693­3879

Bonnie M. Perdue

https://orcid.org/0000­0002­1815­4353

We have no known conflict of interest to disclose.

Jennifer L. Hughes is the Charles A. Dana Professor of Psychology, Psi Chi Advisor at Agnes Scott College, and an Associate Editor of Psi Chi Journal of Psychological Research

Bradley Cannon works at the Psi Chi Headquarters and is a Psi Chi Writer and Managing Editor of the Psi Chi Journal of Psychological Research. Abigail A. Camden is a current doctoral student. Kimberli R. H. Treadwell is an Associate Professor of Psychological Sciences at the University of Connecticut and an Associate Editor of Psi Chi Journal of Psychological Research

Bonnie M. Perdue is an Associate Professor of Psychology at Agnes Scott College. Joel Thomas is an Assistant Professor of Psychology at Agnes Scott College.

Correspondence concerning this article should be addressed to Jennifer L. Hughes, Department of Psychology, Agnes Scott College, 141 E. College Ave., Decatur, GA, 30030, United States. Email: jhughes@agnesscott.edu

COPYRIGHT 2023 BY PSI CHI,

Coping During Crisis: An Intervention on Gratitude and Personal Well-Being During COVID-19

Isabel F. Gilbertson and Audrey A. Graves Department of Psychology, Seattle University*

ABSTRACT. For the current study, we investigated how a gratitude intervention of counterfactual thinking influenced COVID­19 experiences, dispositional gratitude levels, and psychological well ­ being. College students from a university in the Pacific Northwest (N = 152; Mage = 22.5, SD = 2.6) completed an online survey where they were randomly assigned into either a gratitude treatment group, where they practiced counterfactual thinking, or a control group, where counterfactual thinking was not practiced, and self­reported their COVID­19 experiences, gratitude levels, and psychological well ­ being. We sought to explore whether counterfactual thinking could mitigate the effects of COVID­19 by increasing gratitude and personal well­being. Statistical analyses indicated that the treatment group exposed to counterfactual thinking reported higher gratitude levels than the control group, t(115) = 2.81, p = .006; t(113) = 3.01, p = .003. A series of Pearson correlations found a small, statistically significant negative relationship between COVID ­ 19 experiences and personal well­being, r(125) = –.23, p = .02. It was found that gratitude and personal well ­ being were statistically significant and positive, r(125) = .77, p = .001; r(121) = .85, p = .001. Results showed a statistically significant difference between the gratitude control and gratitude treatment groups regarding personal well­being when controlling for the severity of COVID­19 experiences, F(2, 102) = 5.97, p = .004, partial eta squared = .11. This study contributes to existing literature by providing an empirical demonstration that counterfactual thinking can improve gratitude and personal well­being, as well as mitigate negative COVID­19 experiences.

Keywords: gratitude, counterfactual thinking, personal well­being, coronavirus, COVID­19

TRỪU TƯỢNG. Đối với nghiên cứu hiện tại, chúng tôi đã điều tra xem sự can thiệp về lòng biết ơn của tư duy phản thực tế đã ảnh hưởng như thế nào đến trải nghiệm COVID­19, mức độ biết ơn trong tính cách và sức khỏe tâm lý. Sinh viên đại học từ một trường đại học ở Tây Bắc Thái Bình Dương (N = 152; Mage = 22,5, SD = 2,6) đã hoàn thành một cuộc khảo sát trực tuyến, trong đó họ được phân ngẫu nhiên vào một nhóm được điều trị bằng lòng biết ơn, nơi họ thực hành suy nghĩ phản thực, hoặc một nhóm kiểm soát, trong đó tư duy phản thực tế không được thực hành và tự báo cáo về trải nghiệm COVID­19, mức độ biết ơn và sức khỏe tâm lý của

Replication badge earned for transparent research practices.

Gilbertson and Graves | An Intervention on Gratitude and Well-Being

họ. Chúng tôi đã tìm cách khám phá xem liệu tư duy phản thực tế

có thể giảm thiểu tác động của COVID­19 hay không bằng cách

tăng cường lòng biết ơn và hạnh phúc cá nhân. Các phân tích thống

kê chỉ ra rằng nhóm điều trị tiếp xúc với suy nghĩ phản thực tế báo

cáo mức độ biết ơn cao hơn nhóm đối chứng, t (115) = 2,81, p = 0,006; t(113) = 3,01, p = 0,003. Một loạt các mối tương quan Pearson đã tìm thấy mối quan hệ tiêu cực nhỏ, có ý nghĩa thống kê giữa trải nghiệm COVID­19 và sức khỏe cá nhân, r(125) = –.23, p = .02. Người ta thấy rằng lòng biết ơn và hạnh phúc cá nhân có ý nghĩa thống kê và tích cực, r(125) = .77, p = .001; r(121) = .85, p = .001. Kết quả cho thấy sự khác biệt có ý nghĩa thống kê giữa nhóm kiểm soát lòng biết ơn và nhóm đối xử với lòng biết ơn về sức khỏe cá nhân khi kiểm soát mức độ nghiêm trọng của trải nghiệm COVID­19, F(2, 102) = 5,97, p = 0,004, bình phương một phần eta = 0,11 . Nghiên cứu này đóng góp vào các tài liệu hiện có bằng cách cung cấp một minh chứng thực nghiệm rằng tư duy phản thực tế có thể cải thiện lòng biết ơn và hạnh phúc cá nhân, cũng như giảm thiểu những trải nghiệm tiêu cực về COVID­19.

Từ khóa: lòng biết ơn, suy nghĩ phản thực, hạnh phúc cá nhân, virus corona, COVID­19

Gratitude practices have been shown to improve individuals’ health behaviors and mental health, but there is limited literature on the effect of gratitude interventions during high stress times such as the novel coronavirus (e.g., COVID­19) pandemic (Dickens, 2017). A global pandemic is a stressful time for the general population, but certain groups are at higher risk of experiencing heightened psychological distress (Lv et al., 2020; Sun et al., 2020). Groups such as frontline and essential workers, healthcare professionals, students, and caregivers are at risk of increased psychological distress due to severe disruptions in work and their proximity to the virus (Boyraz & Legros, 2020; Firew et al., 2020; Luan et al., 2020; Schmits et al., 2021; Wang et al., 2020). Researchers have found that psychological distress related to the pandemic most commonly presents as mental health symptomology through increases in anxiety and depression, exposure to posttraumatic stress disorder, and lowered psychological well­being. Increases in behavioral health concerns and rises in mental health issues have been found in the general population, illustrating that the COVID­19 pandemic has exacerbated mental health concerns and issues across the globe, but also in high­risk groups (Hamza et al., 2021; Lopez ­ Castro et al., 2020; Passavanti et al., 2021; Torales et al., 2020; Xiong et al., 2020). These

findings beg the question: As the result of the pandemic, what cost­effective and accessible interventions exist as a starting point for addressing rises in mental health symptomology and lowered psychological well­being?

Gratitude Is Correlated to Positive Physical and Mental Health Behaviors

Gratitude is conceptualized in psychological literature as a state and a trait (Boggiss et al., 2020). As a state, individuals discover gratitude through moments or positive outcomes, whereas trait gratitude is oriented toward appreciating the world and life (Wood et al., 2010). Gratitude interventions are exercises and practices; they are distinct from other psychological interventions due to their easy­to­deliver mechanisms and small resource cost (Boggiss et al., 2020). Gratitude interventions may be one of many solutions, which offer an effective approach to addressing the current mental health crisis and unknown long­term effects of the pandemic for the general public and high­risk groups.

Gratitude interventions have shown consistent links to better physical and mental health behaviors (Hill et al., 2013; Jackowska et al., 2016). A study conducted on the relationship between dispositional gratitude and self ­ rated physical health by Hill et al. (2013) found that gratitude is positively correlated with self­reported

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

SPRING 2023

An Intervention on Gratitude and Well-Being | Gilbertson and Graves

physical health, propensity for healthy activities (e.g., nutrition and exercise), and willingness to seek help for health concerns. In addition, dispositional gratitude was a significant predictor when controlling for age and the Big Five traits, illustrating that more grateful individuals tend to report better physical health and positive health behavior. Further, Jackowska et al. (2016) found that a gratitude treatment improved study participants’ sleep quality and lowered their diastolic blood pressure. It was further discovered that the gratitude intervention helped reduce emotional distress and increase optimism and positive emotional style.

Gratitude and Psychological Well-Being and Life Satisfaction

Beyond health behaviors and outcomes, gratitude interventions have demonstrated positive associations with improved psychological well­being and life satisfaction. A meta­analysis on the effectiveness of gratitude by Dickens (2017) found that gratitude interventions have positive benefits such as improved well ­being, happiness, life satisfaction, and mood. The idea that gratitude can have a positive effect on life satisfaction was further explored by Robustelli and Whisman (2018). The researchers found that gratitude is uniquely associated with life satisfaction. Furthermore, the bivariate associations between gratitude and all three measures (e.g., work, health, relationships) were found to be statistically significant, illustrating that people who recorded higher levels of gratitude also reported higher levels of life satisfaction in the three measures. In addition, Wood et al. (2009) found that gratitude explained a significant variance in psychological well­being. From this research, it was discovered that the relationship between gratitude and psychological well­being were independent from the Big Five traits, which suggests that gratitude plays a significant role in an individual’s psychological wellbeing (Wood et al., 2009).

Gratitude’s Beneficial Effect on Psychological Symptoms and Trauma

In addition to aiding in better health outcomes and an association with increased psychological well­being, researchers have also found that gratitude can produce protective factors to deter against psychological trauma and negative psychological symptoms, such as suicidal ideation and depression. Kleiman et al. (2013) explored the influence of gratitude on hopelessness and depressive symptoms and found that higher gratitude levels buffer the effect of the two suicidal risk factors: hopelessness and depressive symptoms. The results show the beneficial effect that higher gratitude levels have on suicidal risk factors and the risk of suicide. Research conducted

by Lin (2015) discovered that gratitude has a direct effect on suicidal ideation and depression for undergraduate students. Grateful undergraduate students tend to possess lower levels of depression, decreasing their suicidal ideation; this supports the idea that gratitude can reduce depressive symptoms and protect against suicide. Additionally, gratitude has been found to promote positive outcomes and prevent adverse psychological effects following traumatic events. A study by Vieselmeyer et al. (2017) explored the role of gratitude on posttraumatic stress (PTS) and posttraumatic growth (PTG) following a campus shooting at Seattle Pacific University. The results found that individuals with high levels of gratitude have a higher chance of PTS developing into PTG, which suggests that people with higher levels of gratitude may be able to convert high PTS into PTG (Vieselmeyer et al., 2017). Therefore, gratitude can be a protective trait to help individuals cope with traumatic life events and experiences.

Research has consistently shown a strong association between gratitude and positive physical and mental health outcomes, but its determined effect on major depression disorder (MDD) and other psychological disorders are inconsistent and further research is required (Dickens, 2017; Kleiman et al., 2013; Lin, 2015; Robustelli & Whisman, 2018; Vieselmeyer et al., 2017). Research on the gratitude effect on psychological disorders such as MDD and others shows inconclusive effects, and there is not a generalized acceptance of gratitude as a major change on psychological disorders (Boggiss et al., 2020; Dickens, 2017). A systematic review on gratitude interventions by Boggiss et al. (2020) argued that more research is needed to make firm conclusions on clinical populations. These recommendations are important to acknowledge when studying gratitude and its possible effects on the general population.

Counterfactual Thinking and Gratitude

Counterfactual thinking is defined as the mental exercise of subtracting positive life events in one’s life (Koo et al., 2008). Individuals often look to what they do not have in life, falling into the mental trap of comparison or believing that the grass is greener on the other side. With an increase in social media usage and online communication, people today are exposed to more people than ever before. This increased exposure through social media allows individuals to see other people’s appearances, vacations, jobs, homes, lifestyles, families, material goods, and more. This information can make people long for what others have through self­comparison, which can lower their self­esteem (Hawi & Samaha, 2016; Lee & Lee, 2021). Self­comparison through social media and real ­ life situations can happen consciously and

subconsciously, and it can be detrimental for individuals’ mental health (Vogel et al., 2015). Counterfactual thinking combats self­comparison and other negative cognitive behaviors by helping individuals acknowledge the positive things or events in their lives and that their present situations are far better than other alternative scenarios. It is common for people to think about what they do not have rather than what they do have, and counterfactual thinking can increase awareness around positive aspects that people usually ignore. This mental exercise can encourage individuals to express gratitude and remember everything they have rather than long for what they do not have.

Research on counterfactual thinking has shown that mentally subtracting positive life events can improve affective states, relationship satisfaction, and gratitude level. A study by Koo et al. (2008) explored how counterfactual thinking impacts affective states, relationship satisfaction, and gratitude. It was found that participants who practiced counterfactual thinking reported more positive feelings and significantly greater positive effects than participants who were not exposed to or did not practice this thinking style. Further, results showed that romantic partners who practiced counterfactual thinking by imagining if they had never met their partners were more satisfied with their relationship than those who did not engage in counterfactual thinking. Another study conducted on counterfactual thinking and gratitude by Nicuță and Constantin (2021) found similar results to Koo et al. (2008). In the study, adolescent study participants who practiced counterfactual thinking reported higher gratitude levels and decreased levels of negative emotion.

Purpose of the Present Study

This study explored experiences of university students in the United States (U.S.) during COVID­19 and how a brief gratitude intervention of counterfactual thinking would impact students’ gratitude levels, psychological well­being, and COVID­19 experiences. Research has shown positive associations between gratitude and positive mental and physical health outcomes, but to our knowledge, there have been no studies investigating the impact of counterfactual thinking on the mitigation of COVID­19 experiences. To address these gaps, we examined the relationship between one’s experiences with COVID­19 and a gratitude intervention of counterfactual thinking, evaluating whether the intervention can increase gratitude and personal well­being and buffer negative COVID­19 experiences.

This investigation contributes to gratitude and COVID­19 research in two ways. First, this study focused on gratitude for undergraduate college students in the context of the COVID­19 pandemic. Limited literature

exists on gratitude regarding college students in the context of a global pandemic. Gratitude is primarily researched in clinical or occupational settings. Studies have been conducted that explore gratitude interventions during COVID­19, but few exist with undergraduates as the study population. Undergraduates are a significant population to examine during COVID­19 because young adulthood and emerging independence are challenges in a normal context, let alone during a pandemic. Routine changes brought about by COVID­19, such as moving home from college, learning online, and limited social interaction may have drastic implications for individuals' mental health. Research has found that student status may be a risk factor associated with elevated levels of psychological distress related to COVID­19 (Xiong et al., 2020). For these reasons, college students may be at particular risk for experiencing increases in anxiety, depression, and stress, especially in the context of a global pandemic.

Second, the present study was one of the first empirical research studies that explored the relationship between COVID­19 experiences and the gratitude intervention of counterfactual thinking. The vast majority of gratitude interventions that have been examined during the pandemic have not utilized counterfactual thinking. Furthermore, most published gratitude research has only investigated how different treatments can increase beneficial psychological variables; they have not explored whether these treatments mitigate negative experiences related to the pandemic. Beyond exploring if counterfactual thinking increases gratitude and personal well­being, the present study also went a step further by testing whether counterfactual thinking can act as a buffer against negative COVID­19 experiences by producing an increase in personal well­being.

In this study, we had four hypotheses. The first hypothesis was that the gratitude treatment group that utilizes counterfactual thinking would produce higher levels of gratitude than the gratitude control group that did not utilize counterfactual thinking. This hypothesis was based on previous research by Koo et al. (2008) and Nicuță and Constantin (2021). Koo et al. (2008) and Nicuță and Constantin (2021) found that counterfactual thinking can improve affective states and gratitude levels, which made us hypothesize that counterfactual thinking would produce greater gratitude levels for university students than students not exposed to counterfactual thinking. The second hypothesis for this study was that more severe COVID­19 experiences would be associated with lower personal well­being. Our second hypothesis was supported by Hamza et al. (2021) and Lopez­Castro et al. (2020)’s findings that U.S. college students who reported more traumatic COVID­19 experiences (e.g.,

An Intervention on Gratitude and Well-Being | Gilbertson and Graves

contracting the virus, hospitalization, loss of resources as a result of the pandemic, loss of a friend or loved one) experienced more psychological distress and poorer mental health than students who did not report severe COVID­19 experiences. The third hypothesis was that higher levels of gratitude would be associated with higher levels of personal well­being. This hypothesis was based on research findings from Dickens (2017), Robustelli and Whisman, (2018), and Wood et al. (2009). These researchers found a consistent relationship between gratitude and personal well­being, making us hypothesize that higher levels of gratitude would be associated with higher levels of personal well­being for university students during the pandemic. Our final hypothesis was that the gratitude treatment of counterfactual thinking would mitigate the effects of severe COVID­19 experiences through increased personal well­being based on previous research from Koo et al. (2008), Nicuță and Constantin (2021), and Vieselmeyer et al. (2017). Although Koo et al. (2008) and Nicuță and Constantin (2021) did not test whether counterfactual thinking would mitigate stressful or traumatic experiences, their findings that counterfactual thinking increase affective states and gratitude point to promising results that this style of thinking can buffer negative experiences. This hypothesis is further supported by Vieselmeyer et al.’s (2017) findings that high levels of gratitude can protect against traumatic life events, such as a campus shooting. Although a campus shooting is different than a pandemic, research has shown that gratitude can help mitigate severe experiences and adverse psychological impacts.

Method Design

This study utilized a between­subjects design and manipulates one factor, random assignment to either the gratitude control or the gratitude treatment group. The outcome variables included gratitude level and psychological wellbeing. This study’s design was based on the research design from Koo et al. (2008) by utilizing roughly the same study procedure for the gratitude treatment and control group in terms of random assignment and gratitude questions. Following the procedure from Koo et al. (2008), after participants were randomly sorted into one of the two groups and responded to the prompt questions, they were then asked to complete dependent study measures. However, although Koo et al. (2008) measured affective states, the present study’s dependent measures were gratitude levels measured by the Gratitude Questionnaire Six­Item Form (GQ­6), Gratitude Resentment and Appreciation Test (GRAT), and Psychological Well­Being (PWB) scales (McCollugh, 2000; Ryff, 1989; Stoddard & Kaufman, 2020; Watkins et al., 2003).

Participants

Participants in this study were U.S. undergraduate students. At the beginning of the survey, participants confirmed they were at least 18 years old and were current university students. The study criteria specified that participants did not need to be current college students at the university where the study was taking place, but they did need to confirm they were a current student at a U.S. institution during the study.

The study was administered through Qualtrics, and researchers published links to the survey through Sona, a university platform where undergraduate students at the university can take student surveys for class credit. The survey link was also published on Instagram and Facebook. In this study, 153 people participated. One participant’s responses were excluded from the study due to incomplete data. One hundred thirty­one participants were recruited through Sona, and 22 were recruited through social media. Sample differences between students and online participants were not tested since participation criteria being the same across Sona and social media.

In the sample, the average age was 22.5 years (SD = 2.6, range = 18–78). Participants identified as cisgender women (n = 74, 48%), cisgender men (n = 35, 22.7%), nonbinary (n = 3, 1.9%), transgender women (n = 2, 1.2%), transgender men (n = 1, 0.6%), and agender (n = 1, 0.6%). Additionally, 25% of participants (n = 38) did not disclose their gender identity. Participants identified as White or European American (n = 76, 49.3%), Asian or Asian American (n = 21, 13.6%), Latino/Latina/ Latinx/Latine ( n = 7, 4.5%), Biracial or Mixed Race (n = 5, 3.2%), and Black or African American ( n =3, 1.9%). Approximately 24.9% of participants (n = 39) chose not to disclose their race/ethnicity. In terms of year in college, participants identified as first­year or freshman (n = 13, 8.4%), second­year or sophomore (n = 7, 4.5%), third­year or junior (n = 49, 31.2%), and fourth­year or senior (n = 24, 15.6%). Some participants chose to not disclose their year in college (n = 6, 3.9%).

Measures COVID-19 Experience

Participants reported the severity of their COVID­19 experiences using the COVID­19 Impact Scale (Stoddard & Kaufman, 2021). Participants were asked to rate their COVID­19 experiences through 12 items on a 3­ and 4­point scale. The first nine items asked participants to rate changes in access to food, income, social support, healthcare services, and experiences of stress related to the pandemic on a 3­point scale from 1 (no change) to 3 (severe). Items 10 and 11 asked participants about the number of immediate and extended family members

and friends diagnosed with COVID­19 and to rate the symptoms of the person who was the most sick. Item 10 asked participants to list the number of immediate family members diagnosed with COVID­19 and to rate the symptoms of the person who was the most sick. If participants listed 0 as the amount of immediate family members diagnosed with COVID­19, they skipped rating the symptoms of the person who was the most sick and moved on to the next question. If a participant listed at least one person as an immediate family members diagnosed with COVID­19, the participant was asked to respond on an ordinal 4­point scale to rate symptoms from 1 (mild: Symptoms effectively managed at home) to 4 (immediate family member died from COVID-19). Item 11 asked participants to list the number of extended family members or close friends diagnosed with COVID­19 and to rate the symptoms of the person who was the most sick and followed the same response procedure as Item 10. Question 12 was a write­in question that allowed participants to discuss any other ways the COVID­19 pandemic impacted their life. The possible scores for this scale ranged from 2–35, with higher scores indicating restricted access to resources and social support, high stress, and that family and friends were diagnosed with COVID­19 and might have had moderate to severe symptoms. Stoddard et al. (2021) reported a Cronbach alpha measure of .82. This analysis found an alpha reliability coefficient of .72.

Gratitude Control

Participants were randomly assigned into the treatment or control group through Qualtrics. Of the 152 participants, 76 participants were randomly assigned to the gratitude control group. In the gratitude presence control group, participants were asked to discuss the presence of a positive life event using the gratitude presence procedure from Koo et al. (2008). Participants were asked to identify an event for which they felt grateful from one of seven categories: education, health, safety/security, possessions, break/ vacation/weekends/holidays, act of kindness/support from others, and achievement/performance. Once participants had identified an event, they were then asked to write a short response on the prompt, “Please describe the ways in which this thing or event happened easily or was certain to become part of your life.” Respondents were then asked to write a short response on the prompt, “Please describe the ways in which it is NOT AT ALL SURPRISING that this thing or event is part of your life.” Koo et al. (2008) found an alpha of .77 for the gratitude control (Koo et al., 2008). The current study found an alpha reliability coefficient of .78.

Gratitude Treatment

Participants were randomly assigned into the treatment or control group through Qualtrics. Of the 152 participants, 76 participants were randomly assigned

to the gratitude treatment group. The treatment asked participants to mentally subtract a positive life event using the gratitude treatment of counterfactual thinking from Koo et al. (2008). Participants were asked to identify an event for which they felt grateful from one of seven categories: education, health, safety/security, possessions, break/vacation/weekends/holidays, act of kindness/ support from others, achievement/performance. Once participants had identified an event, they were then asked to write a short response on the prompt, “Please describe the ways in which this thing or event might never have happened or might never have been part of your life.” Respondents were then asked to write a short response on the prompt, “Please describe the ways in which it is SURPRISING that this thing or event is part of your life.” Koo et al. (2008) found an alpha of .77 for the gratitude treatment. This study found an alpha reliability coefficient of .75.

Gratitude Level

Participants were asked to rate their current gratitude levels using the Gratitude Questionnaire Six­Item Form (GQ­6) and Gratitude Resentment and Appreciation Test (GRAT; McCullough et al., 2002; Watkins, 2003). Participants were first asked to assess their propensity to experience gratitude using the GQ­6, a six­item gratitude questionnaire that measures dispositional gratitude on a 7­point Likert scale from 1 (strongly disagree) to 7 (strongly agree). Example items in this scale are “I have so much in life to be thankful for” and “If I had to list everything I felt grateful for, it would be a long list.” Possible scores for this scale range from 6–42, with higher scores indicating a higher level of dispositional gratitude. McCullough et al. (2002) found the GQ­6 to have an alpha of .82. This analysis found an alpha reliability coefficient of .84.

After the GQ­6, participants were asked to rate their dispositional gratitude using the GRAT. Respondents were asked to respond to the short­form GRAT, which consists of 16­items on a 9­point Likert scale from 1 (strongly disagree) to 9 (strongly agree). Examples of items in this scale are “Life has been good to me” and “When I look at the world, I don’t see much to be grateful for.” Possible scores for this scale range from 16–144 with higher scores indicating higher dispositional gratitude. Watkins et al. (2003) found the GRAT scale to have an alpha of .93. This study found an alpha reliability coefficient of .92.

Psychological Well-Being

Participants were asked to assess their psychological well­being through the 18­item Psychological Well­Being (PWB) scale (Ryff, 1989). Respondents were asked to rate how strongly they agree or disagree with the 18­item scale using a 7­point Likert scale from 1 (strongly agree) to 7 (strongly disagree). Example items in this scale are

An Intervention on Gratitude and Well-Being | Gilbertson and Graves

“I like most parts of my personality” and “The demands of everyday life often get me down” (Ryff, 1989). The PWB scale has six subscores: autonomy, environmental mastery, personal growth, positive relations with others, purpose in life, and self­acceptance (Ryff, 1989). Possible scores for the PWB scale range from 18–68 with higher scores indicating higher psychological well ­ being. Participants’ total scores for PWB were used in test analyses. Akin (2008) reported the 18­item PWB scale to have an alpha of .78. The current study found an alpha reliability coefficient of .75.

Procedure

Before beginning participant recruitment or data collection, the study received institutional review board approval from Seattle University. The study was distributed through an online survey on Qualtrics and all responses recorded were anonymous. Before consenting to the study, participants confirmed that they were at least 18 years old and were a current U.S. university student. After confirming they were at least 18 and a current university student, participants were provided with study risks and informed consent. Participants were informed that they could stop the survey at any time without penalty. After confirming that they were at least 18 years old, a current college student, and consenting, participants read instructions regarding how to proceed. The first scale that participants completed was the COVID­19 Impact Scale. Next, participants were randomly assigned in a between­groups, double­blind experiment into either the gratitude control or treatment group. Respondents were asked to identify a specific life event and answer two short­answer questions regarding the event they picked. After completing either the treatment or control, respondents assessed

TABLE

1

Means and Standard Deviations for the COVID-19 Impact, GQ-6, and GRAT Scales Grouped by Full Sample, Gratitude Control Group, and Gratitude Treatment Group

their current gratitude level and disposition through the GQ­6 and GRAT scales. After the gratitude scales, respondents were then asked to assess their psychological well­being through answering questions using the PWB scale. After respondents completed the experiment and measures, they answered demographic questions. Participants who completed the study through Sona received class credit. No compensation was provided.

Results

Descriptive statistics for the measures for the full sample of participants and the gratitude control or the gratitude treatment groups are presented in Table 1. The full sample was found to have experienced a moderate change in access related to food, healthcare, education, and social support related to the pandemic. Personal well­being was not run as a descriptive statistic due to the variable being a factor analysis score. These results indicate that the treatment group displayed higher dispositional gratitude, but more severe COVID ­ 19 experiences compared to the control group.

The Gratitude Groups and Overall Gratitude

For the first hypothesis, independent­samples t tests were conducted to compare gratitude levels between the gratitude control and treatment groups. An independent ­ samples t test was conducted to compare the average gratitude levels through the GQ ­ 6 for participants who received the gratitude treatment and participants who received the gratitude control. There was a statistically significant difference in gratitude levels for the GQ­6, such that participants who received the gratitude treatment ( M = 36.15, SD = 5.02) had on average higher gratitude level scores than participants who received the gratitude control (M = 31.86, SD = 10.50), t(115) = 2.81, p = .006, one­tailed, d = 0.52. Figure 1 illustrates the difference in gratitude between experimental groups.

Note. GQ-6 = Gratitude Questionnaire Six-Item Form. GRAT = Gratitude Resentment and Appreciation Test. PWB = Psychological Well-Being. Higher scores for the COVID-19 Impact Scale indicate more severe COVID-19 experiences with this scale having possible scores ranging from 2–35 points. Higher scores for the GQ-6 and the GRAT indicate higher gratitude with possible scores ranging from 6–42 for the GQ-6 and 16–144 for the GRAT.

An independent ­ samples t test was conducted to compare the average gratitude levels through the GRAT for participants who received the gratitude treatment and participants who received the gratitude control. There was a statistically significant difference in gratitude levels for the GRAT, such that participants who received the gratitude treatment ( M = 121.18, SD = 14.94) had on average higher gratitude level scores than participants who received the gratitude control (M = 106.17, SD = 34.54), t(113) = 3.01, p = .003, one­tailed, d = 0.56. Figure 2 illustrates the difference in participants’ gratitude results between the experimental groups.

Relationship Between COVID-19 Experience and Psychological Well-being

Regarding the second hypothesis, a Pearson correlation was used to assess the relationship between the severity of COVID­19 experiences and personal well­being. A statistically significant and small negative correlation was found between the severity of COVID­19 experiences and personal well­being r(125) = –.23, p = .02. Despite being statistically significant, COVID­19 experience shares only 5.29% of the variability in personal well­being. In other words, COVID­19 experiences can only account for 5.29% of variation in personal well­being scores.

Relationship Between Gratitude Level and Personal Well-being

A Pearson correlation was used to assess the third hypothesis regarding the relationship between gratitude and personal well­being using the GQ­6, GRAT, and PWB. For the GQ­6 and PWB scales, a statistically significant positive correlation was found, r(125) = .77, p = .001. For the GRAT and PWB scale, a statistically significant positive correlation was found, r(121) = .85, p = .001. These analyses indicated that there is a strong positive relationship between gratitude and personal well­being, such that, as one’s level of gratitude increases, so does one’s personal well­being. This means that gratitude accounts for 59.29% of variation in personal well­being scores.

Regarding the fourth hypothesis, a one ­ way between ­ groups analysis of covariance (ANCOVA) was conducted to determine a statistically significant difference between the gratitude control and gratitude treatment groups regarding personal well­being when controlling for the severity of COVID­19 experiences. The independent variable was the type of intervention (e.g., treatment, control), and the dependent variable was personal well­being. Participants’ scores of level of severity on the COVID­19 Impact Scale were used as the covariate in this analysis. After adjusting for COVID­19 experiences, there was a significant difference between the two groups for post intervention scores on the PWB scale, F(2, 102) = 5.97, p = .004, partial eta squared = .11. Controlling for COVID­19 experiences, a mean of 29.65 was found for the control group and a mean of 42.02 for the intervention group. A partial eta squared value of .02 was found. This result indicates that the gratitude treatment mitigated COVID­19 experiences, and was found to elicit higher gratitude and personal well­being scores among participants.

Discussion

This study was one of the first empirical demonstrations of how counterfactual thinking can be utilized to increase gratitude level and psychological well­being and mitigate severe pandemic experiences in the context of the spring 2020. The purpose of the present research was to explore

the gaps in research related to the effect of the gratitude treatment of counterfactual thinking on gratitude, personal well­being levels, and the ability to mitigate severe experiences in the context of the COVID­19 pandemic. Overall results indicate that counterfactual thinking as a brief gratitude treatment has the ability to improve individuals’ gratitude and personal well­being levels and mitigate severe experiences from the pandemic. Consistent with the study’s first hypothesis, the gratitude treatment group produced higher levels of gratitude than the gratitude control group; both independent ­ samples t tests using the GRAT and GQ­6 scales were found to have statistically significant results. This illustrates that engaging in counterfactual

Gratitude Scores From the Gratitude Questionnaire Six-Item Form for the Control and Treatment Group

Gratitude Scores From the Gratitude Resentment and Appreciation Test for the Control and Treatment Group

FIGURE 2

FIGURE 1

An Intervention on Gratitude and Well-Being | Gilbertson and Graves

thinking did increase participants’ gratitude levels. The analyses found medium effect sizes of 0.52 and 0.56 for both independent­samples t tests run, indicating the practicality and significance behind the results. These statistics illustrate that the average GQ­6 gratitude level of participants who received the gratitude treatment was 0.52 standard deviations greater than participants who did not receive the gratitude treatment. Similarly, the average GRAT gratitude levels of participants who received the gratitude treatment was 0.56 standard deviations greater than participants who did not receive the gratitude treatment (Cohen, 1988).

Our second hypothesis was that stronger and more extreme COVID­19 experiences would be associated with lower personal well­being. We tested this hypothesis by running a series of Pearson correlations that found a weak, statistically significant correlation between COVID ­ 19 experiences and personal well ­ being, indicating that stronger and more extreme COVID­19 experiences were associated with having lower personal well­being. This correlation is consistent with the second hypothesis despite the relationship between the two variables being weak. A possible explanation for this result is that more severe COVID­19 experiences contribute to decreased states of emotional, mental, and physical health, which influence personal well­being.

Results supported our third hypothesis that higher gratitude levels will produce higher personal well­being. A positive statistically significant relationship revealed that higher gratitude levels produced higher personal well­being levels. Both the GQ­6 and the GRAT, when tested with the PWB, were found to have a strong and positive statistically significant relationship, illustrating that as gratitude increases, so does personal well­being.

The fourth and final hypothesis was also supported by data. After running a one ­ way between ­ groups ANCOVA, controlling for COVID ­ 19 experiences, researchers found there was a significant difference between the two groups on post ­ intervention scores regarding psychological well­being. Participants in the gratitude treatment group reported higher psychological well­being mean scores and accounted for more variance in well­being, suggesting that the gratitude treatment helped to mitigate severe COVID­19 experiences and negative effects of the pandemic.

For both the gratitude control and treatment group, participants were asked to identify something or an event they were grateful for. Figure 3 depicts a Word Cloud of the most popular words listed by participants. On average, participants reported that they were the most thankful for their family. After family, words with the highest frequency of use were participants’ homes, jobs, education, and health.

Our findings are aligned with prior research and offer a new perspective on counterfactual thinking as a gratitude treatment in the context of a pandemic. Our first hypothesis echoes the findings from Koo et al. (2008) and Nicuță and Constantin (2021) by demonstrating that counterfactual thinking increases gratitude levels. While Koo et al. (2008) did not measure gratitude, researchers found improved affective states, which are positively associated with gratitude levels. Results are consistent with Nicuță and Constantin (2021) as well, as their results showed that counterfactual thinking improves gratitude levels. Our second hypothesis is in alignment with research from Hamza et al. (2021) and Lopez­Castro et al. (2020) insofar as it illustrates that more severe COVID­19 experiences lower personal wellbeing. The third hypothesis is also supported by prior literature from Dickens (2017), Robustelli and Whisman (2018), and Wood et al. (2009) because our results found that higher levels of gratitude are associated with higher levels of personal well­being. Our last hypothesis was supported by research from Koo et al. (2008), Nicuță and Constantin (2021), and Vieselmeyer et al. (2017). This study took place in spring 2020, a time when COVID­19 was surging and some of the most extensive health protocols and restrictions were in place. Many people, including university students, did not know whether the pandemic would last months or years, and this period of uncertainty was a time of high stress for the general population. Although various gratitude interventions’ effectiveness have been explored during the pandemic, this study is one of the few that investigates the effectiveness of counterfactual thinking’s relationship to COVID­19 experiences, dispositional gratitude, and personal well­being. Furthermore, this study is one of the first that has explored how counterfactual thinking

FIGURE 3

Things or Events Participants Were Grateful for

Note. Participants in the gratitude control and treatment groups were asked to identify something they were grateful for. This Word Cloud depicts the most popular words participants listed.

can mitigate severe COVID­19 experiences. Our study furthers research on the effectiveness of counterfactual thinking related to gratitude and the impact of COVID­19 on U.S. university students. Although Koo et al. (2008) found a difference in gratitude levels after implementing the treatment of counterfactual thinking, they did not analyze gratitude levels during a mass crisis event such as a global pandemic. For this reason, the context of our results makes them especially significant. Our results also expand on previous literature by illustrating that counterfactual thinking can be a protective factor that increases psychological well­being through gratitude and can control for severe COVID­19 experiences. Our findings contribute importantly to existing literature by demonstrating that counterfactual thinking can be beneficial to improving gratitude levels and that higher gratitude levels produce higher levels of personal well­being during a pandemic.

Limitations and Strengths

This study has several limitations. One limitation is that college students who were taking a leave of absence (i.e., students who did not enroll in classes for an academic quarter or semester) during spring 2020, when this study was conducted, were excluded from the study criteria. One of the requirements to participate in the study was to be a current U.S. university student to help researchers capture the accurate population at the time. However, after this study was complete, the researchers learned that many U.S. college students had paused their studies during the spring of 2020 for a variety of reasons, ranging from only wanting to pay for college if they were able to be in­person to needing to help financially support their family. Financial instability during a recession and job loss during the pandemic made continuing higher education too expensive for many families, causing many university students to take a leave of absence. Since a requirement of the study was that participants had to be current university students, this situation naturally weeded out students who were less privileged and financially secure than students that continued their studies during this time. Excluding these individuals and their resulting COVID­19 experiences and personal well­being might have severe and significantly impacted study results. College students who were taking a break from their studies during this study were not meant to be excluded, but the language of the study, which listed current U.S. university student status as a prerequisite (or requirement) for participation was not precise and excluded these individuals from participating.

The study also had a lack of diversity in the sample with most participants identifying as female and White. Our sample did not accurately reflect the diverse

demographic of college students in the United States due to the lack of racial and gender diversity. The pandemic has impacted individuals of color at a much higher rate in terms of virus infection, mortality, and unemployment, and our sample lacked racial representation, making it difficult to capture COVID­19 experiences and personal well­being of different racial demographics. The sample was also mainly comprised of young adults in their early twenties, which is not representative of the different ages of individuals that attend college. In addition, the study utilized a convenience sample of students, so the results cannot be generalized to the general population.

Although the study had limitations, it also had several strengths. The study used a well ­ researched experimental design to explore university students’ COVID ­ 19 experiences, psychological well ­ being, and how counterfactual thinking can impact gratitude and mitigate the negative effects of severe COVID­19 experiences. To our knowledge, this study is one of the first empirical demonstrations that counterfactual thinking can help buffer severe COVID­19 experiences. The study also had acceptable Cronbach alpha values for all measures used. All alpha values were at least .7 as recorded by original researchers, and our analyses also found each of our measures to meet the .7 cutoff.

Implications and Future Research

The study findings highlight the importance of gratitude interventions such as counterfactual thinking on undergraduate students in the context of a natural disaster such as a pandemic. This study’s population is important because undergraduate students are coming of age during a global health crisis and will be future leaders, making it important to address any adverse psychological or mental health issues they have incurred as a result of the pandemic. Although there has been an increased focus on mental health during the pandemic, more research and work are needed to understand the long­term impact of the pandemic on students’ mental health and lives.

This study prompts three recommendations for further research. First, future research should explore the effect of socioeconomic status and family life on gratitude level because these two variables strongly contribute to maturation and available resources. Second, research should investigate the impact of gratitude interventions on other demographic variables, such as level of formal education, race, and sex, to see if counterfactual thinking produces varied effects among different demographics. Third, research should also examine the effect of counterfactual thinking on symptoms such as depression and anxiety because these symptoms have increased during the COVID­19 pandemic.

An Intervention on Gratitude and Well-Being | Gilbertson and Graves

By utilizing the aforementioned recommendations, future studies on gratitude interventions could inform positive psychology practices and mental health practitioners in the context of traumatic, long­term crises.

Conclusion

The current study contributed a new perspective on the behavioral health outcomes of university students, particularly focused on how a brief gratitude intervention of counterfactual thinking can improve students’ gratitude, personal well­being, and buffer against negative COVID­19 experiences. Results add to existing research related to counterfactual thinking and COVID­19, and researchers hope gratitude treatments such as counterfactual thinking will be further researched and explored in hopes of providing support to vulnerable groups that have experienced trauma or loss during the pandemic. This study underscores how inexpensive and accessible interventions such as counterfactual thinking, which can be practiced daily by any individual, can help improve individuals’ mental health.

Although further research is needed to more fully understand the effectiveness and long­term benefits of counterfactual thinking as a gratitude treatment, this study offers a starting point in regard to the efficacy of counterfactual thinking for university students during the height of the COVID ­ 19 pandemic. The relationship between severe COVID­19 experiences and counterfactual thinking offers a promising perspective on this thinking style’s benefits. Due to the easy delivery method and accessibility of counterfactual thinking, this intervention could be used by a variety of health professionals such as social workers, clinicians, community health workers, and public health policymakers to help improve individuals’ gratitude levels and mitigate severe COVID­19 experiences.

References

Akin, A. (2008). The scales of psychological well-being: A study of validity and reliability. Educational Sciences: Theory and Practice, 8(3), 741–750.

Boggiss, A. L., Consedine, N. S., Brenton-Peters, J. M., Hofman, P. L., & Serlachius, A. S. (2020). A systematic review of gratitude interventions: Effects on physical health and health behaviors. Journal of Psychosomatic Research, 135, 110–165. https://doi.org/10.1016/j.jpsychores.2020.110165

Boyraz, G., & Legros, D. N. (2020). Coronavirus disease (COVID-19) and traumatic stress: Probable risk factors and correlates of posttraumatic stress disorder. Journal of Loss and Trauma, 25(6–7), 503–522. https://doi.org/10.1080/15325024.2020.1763556

Capra, F. (Director). (1946). It's a wonderful life [Film]. Liberty Films.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd Ed.). https://doi.org/10.4324/9780203771587

Dickens, L. R. (2017). Using gratitude to promote positive change: A series of meta-analyses investigating the effectiveness of gratitude interventions. Basic and Applied Social Psychology, 39(4), 193–208. https://doi.org/10.1080/01973533.2017.1323638

Firew, T., Sano, E. D., Lee, J. W., Flores, S., Lang, K., Salman, K., Greene, M. C., & Chang, B. P. (2020). Protecting the front line: A cross-sectional survey analysis of the occupational factors contributing to healthcare workers’ infection and psychological distress during the COVID-19 pandemic in the

USA. BMJ Open, 10(10), e042752. https://doi.org/10.1136/bmjopen-2020-042752

Hamza, C. A., Ewing, L., Heath, N. L., & Goldstein, A. L. (2021). When social isolation is nothing new: A longitudinal study on psychological distress during COVID-19 among university students with and without preexisting mental health concerns. Canadian Psychology/Psychologie Canadienne, 62(1), 20–30. https://doi.org/10.1037/cap0000255

Hawi, N. S., & Samaha, M. (2016). The relations among social media addiction, self-esteem, and life satisfaction in university students. Social Science Computer Review, 35(5), 576–586. https://doi.org/10.1177/0894439316660340

Hill, P. L., Allemand, M., & Roberts, B. W. (2013). Examining the pathways between gratitude and self-rated physical health across adulthood. Personality and Individual Differences, 54(1), 92–96. https://doi.org/10.1016/j.paid.2012.08.011

Jackowska, M., Brown, J., Ronaldson, A., & Steptoe, A. (2016). The impact of a brief gratitude intervention on subjective well-being, biology and sleep. Journal of Health Psychology, 21(10), 2207–2217. https://doi.org/10.1177/1359105315572455

Koo, M., Algoe, S. B., Wilson, T. D., & Gilbert, D. T. (2008). It's a wonderful life: Mentally subtracting positive events improves people's affective states, contrary to their affective forecasts. Journal of Personality and Social Psychology, 95(5), 1217–1224. https://doi.org/10.1037/a0013316

Kleiman, E. M., Adams, L. M., Kashdan, T. B., & Riskind, J. H. (2013). Grateful individuals are not suicidal: Buffering risks associated with hopelessness and depressive symptoms. Personality and Individual Differences, 55(5), 595–599. https://doi.org/10.1016/j.paid.2013.05.002

Lee, M., & Lee, H. (2021). Social media photo activity, internalization, appearance comparison, and body satisfaction: The moderating role of photo-editing behavior. Computers in Human Behavior, 114, 106579. https://doi.org/10.1016/j.chb.2020.106579

Lin, C. (2015). The relationships among gratitude, self-esteem, depression, and suicidal ideation among undergraduate students. Scandinavian Journal of Psychology, 56(6), 700–707. https://doi.org/10.1111/sjop.12252

Lopez-Castro, T., Brandt, L., Anthonipillai, N., Espinosa, A., & Melara, R. (2020). Experiences, impacts and mental health functioning during a COVID-19 outbreak and lockdown: Data from a diverse New York City sample of college students. PsyArXiv. https://doi.org/10.31234/osf.io/nyght

Luan, R., Pu, W., Dai, L., Yang, R., & Wang, P. (2020). Comparison of psychological stress levels and associated factors among healthcare workers, frontline workers, and the general public during the novel coronavirus pandemic. Frontiers in Psychiatry, 11, 583971. https://doi.org/10.3389/fpsyt.2020.583971

Lv, Z., Yang, J., Jiang, H., & Nan, J. (2020). Psychological impacts of the COVID-19 epidemic on Chinese people: Exposure, post-traumatic stress symptom, and emotion regulation. Asian Pacific Journal of Tropical Medicine, 13(6), 252–259. https://doi.org/10.4103/1995-7645.281614

McCullough, M. E., Emmons, R. A., & Tsang, J. (2002). The grateful disposition: A conceptual and empirical topography. Journal of Personality and Social Psychology, 82(1), 112–127. https://doi.org/10.1037/0022-3514.82.1.112

Nicuță, E. G., & Constantin, T. (2021). Take nothing for granted: Downward social comparison and counterfactual thinking increase adolescents’ state gratitude for the little things in life. Journal of Happiness Studies, 22(8), 3543–3570. https://doi.org/10.1007/s10902-021-00382-5

Passavanti, M., Argentieri, A., Barbieri, D. M., Lou, B., Wijayaratna, K., Foroutan Mirhosseini, A. S., Wang, F., Naseri, S., Qamhia, I., Tangerås, M., Pelliciari, M., & Ho, C. (2021). The psychological impact of COVID-19 and restrictive measures in the world. Journal of Affective Disorders, 283, 36–51. https://doi.org/10.1016/j.jad.2021.01.020

Robustelli, B. L., & Whisman, M. A. (2018). Gratitude and life satisfaction in the United States and Japan. Journal of Happiness Studies, 19(1), 41–55. https://doi.org/10.1007/s10902-016-9802-5

Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57(6), 1069–1081. https://doi.org/10.1037/0022-3514.57.6.1069

Schmits, E., Dekeyser, S., Klein, O., Luminet, O., Yzerbyt, V., & Glowacz, F. (2021). Psychological distress among students in higher education: One year after the beginning of the COVID-19 pandemic. International Journal of Environmental Research and Public Health, 18(14), 7445.

https://doi.org/10.3390/ijerph18147445

Stoddard, J., & Kaufman, J. (2020). Coronavirus Impact Scale https://www.kennedykrieger.org/sites/default/files/library/documents/ faculty/Coronavirus%20Impact%20Scale.pdf

Stoddard, J., Reynolds, E. K., Paris, R., Haller, S., Johnson, S., Zik, J., Elliote, E., Maru,

Gilbertson and Graves | An Intervention on Gratitude and Well-Being

M., Jaffe, A., Mallidi, A., Smith, A., Hernandez, R. G., Volk, H. E., Brotman, M. A., & Kaufman, J. (2021). The Coronavirus Impact Scale: Construction, validation, and comparisons in diverse clinical samples.

https://doi.org/10.31234/osf.io/kz4pg

Sun, L., Sun, Z., Wu, L., Zhu, Z., Zhang, F., Shang, Z., Jia, Y., Gu, J., Zhou, Y., Wang, Y., Liu, N., & Liu, W. (2020). Prevalence and risk factors for acute posttraumatic stress disorder during the COVID-19 outbreak. Journal of Affective Disorders, 283, 123–129. https://doi.org/10.1101/2020.03.06.20032425

Torales, J., O’Higgins, M., Castaldelli-Maia, J. M., & Ventriglio, A. (2020). The outbreak of COVID-19 coronavirus and its impact on global mental health. International Journal of Social Psychiatry, 66(4), 317–320. https://doi.org/10.1177/0020764020915212

Vieselmeyer, J., Holguin, J., & Mezulis, A. (2017). The role of resilience and gratitude in posttraumatic stress and growth following a campus shooting. Psychological Trauma: Theory, Research, Practice, and Policy, 9(1), 62–69. https://doi.org/10.1037/tra0000149

Vogel, E. A., Rose, J. P., Okdie, B. M., Eckles, K., & Franz, B. (2015). Who compares and despairs? The effect of social comparison orientation on social media use and its outcomes. Personality and Individual Differences, 86, 249–256. https://doi.org/10.1016/j.paid.2015.06.026

Watkins, P. C., Woodward, K., Stone, T., & Kolts, R. L. (2003). Gratitude and happiness: Development of a measure of gratitude, and relationship with subjective well-being. Social Behavior & Personality: An International Journal, 31(5), 431–452. https://doi.org/10.2224/sbp.2003.31.5.431

Wang, X., Hegde, S., Son, C., Keller, B., Smith, A., & Sasangohar, F. (2020).

Investigating mental health of US college students during the COVID-19 pandemic: Cross-sectional survey study. Journal of Medical Internet Research, 22(9), e22817. https://doi.org/10.2196/22817

Wood, A. M., Froh, J. J., & Geraghty, A. W. (2010). Gratitude and well-being: A review and theoretical integration. Clinical Psychology Review, 30(7), 890–905. https://doi.org/10.1016/j.cpr.2010.03.005

Wood, A. M., Joseph, S., & Maltby, J. (2009). Gratitude predicts psychological well-being above the Big Five facets. Personality and Individual Differences, 46(4), 443–447. https://doi.org/10.1016/j.paid.2008.11.012

Xiong, J., Lipsitz, O., Nasri, F., Lui, L. M. W., Gill, H., Phan, L., Chen-Li, D., Iacobucci, M., Ho, R., Majeed, A., & McIntyre, R. S. (2020). Impact of COVID-19 pandemic on mental health in the general population: A systematic review. Journal of Affective Disorders, 277, 55–64. https://doi.org/10.1016/j.jad.2020.08.001

Author Note.

Isabel F. Gilbertson

https://orcid.org/0000­0002­1829­0785

Audrey A. Graves

https://orcid.org/0000­0002­0363­491X

This study was not preregistered. We have no conflict of interest to disclose. This study was supported by the Department of Psychology at Seattle University and Lê Xuân Hy. Correspondence regarding this article should be addressed to Isabel Gilbertson at gilbertsonis@seattleu.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

Identification With All Humanity Predicts Perceptions of COVID-19 Safety Precautions

Morgan Ferqueron, Jonathan F. Bassett*, and Amanda J. Cleveland* Department of Psychology, Lander University

ABSTRACT. Factors related to compliance with and perceptions of public health measures concerning COVID­19 were examined in 2 studies. Study 1 was administered in spring 2020 via a university list serve and social media (n = 512, Mage = 34.51). Study 2 was administered in fall 2020 via Amazon Mechanical Turk, a university list serve, and social media (n = 523, Mage = 31.55). In both studies, participants completed measures of identification with all humanity (IWAH), the 10­item personality inventory (TIPI), social dominance orientation (SDO), right­wing authoritarianism (RWA), political orientation, and religion. In Study 1, participants also completed measures regarding how concerned they were about the effects of COVID­19 on society, how much they supported canceling events to promote social distancing, and how much they altered their behavior to avoid gatherings. In Study 2, participants were asked about the extent to which they engaged in recommended behaviors intended to reduce the spread of COVID­19 and their compliance with mask guidelines. Over and beyond the influence of any other factors, IWAH was associated with greater concern about the health of society with regard to COVID­19 and more altering of behavior to avoid gatherings in Study 1 and with more mask compliance in Study 2. These data suggest that a fruitful tactic for increasing compliance with public health measures for COVID­19 might be framing the issue as a way to contribute to welfare of broader society.

Keywords: COVID ­ 19 pandemic, masks, social distancing, identification with all humanity, social dominance orientation, right wing authoritarianism

As of August 13, 2022, there have been 92,560,911 COVID­19 infections, 1,031,426 total deaths (CDC COVID Data Tracker, 2022) and 5,134,157 hospitalizations in the United States (CDC COVID Data Tracker, 2022). Recommendations made in August 2020 by Anthony Fauci, Director of the National Institute of Allergy and Infectious Diseases (NIAID), stated there should be a “universal wearing of masks,” people should “stay away from places… where people congregate,” stay six feet apart, and wash their hands (O’Reilly, 2020). According to the CDC, in order to slow the spread of coronavirus, people should “wear a mask that covers your nose and mouth,” “stay 6 feet apart from others,” “avoid crowds,” and “wash your hands often with soap and water” (Protect Yourself, March 2021). These recommendations were based on the public health precedent of previous success in the use of such non ­ pharmacological interventions in reducing the spread of infectious outbreaks such as influenza and

Severe Acute Respiratory Syndrome (SARS). Previous research demonstrated that wearing masks (Liang et al., 2020) and social distancing (Chu et al., 2020) were effective at reducing the spread of SARS. Similarly, in a systematic review of available literature, Ayouni et al. (2021) documented empirical support for the effectiveness of public health initiatives in reducing the spread of COVID­19. Effective measures included restrictions on travel and large gatherings, promoting social distancing, and mask mandates. Given the potential effectiveness of public health measures in preventing the spread of infectious disease, it is important to understand the factors associated with compliance with these measures.

In the United States, support for and compliance with measures to prevent the spread of COVID ­ 19 continues to be divided along political and ideological lines. For example, Florida’s governor, Ron DeSantis, commented on other states’ approach to handling the COVID­19 pandemic by saying that they “are letting

hysteria drive them” (Fineout & Sarkissian, 2022). Alternatively, California’s governor, Gavin Newsom, led the way for adopting COVID­19 policies such as requiring that masks be worn indoors at schools and requiring all school children to be vaccinated (Blume et al., 2021). Less anecdotally, previous survey research has demonstrated that political liberalism is associated with more compliance with mask wearing and social distancing in the US (Cheng, 2020) and that political liberals tend to perceive more risks related to COVID­19 and to engage in more effort to minimize the spread of COVID­19 than political conservatives (Kerr et al., 2020).

In addition to political orientation, personality traits have previously been linked to attitudes about COVID19. For example, in a large scale cross­cultural study of people in the United States, Europe, Australia, and New Zealand, Lippold et al. (2020) found a relation between COVID­19 attitudes and the big five personality factors as measured by a brief version of the Big Five Inventory (Rammstedt & John, 2007). Specifically, higher levels of neuroticism were associated with more fears about the pandemic (Lippold et al., 2020). Further, in an online survey of American adults, high levels of conscientiousness and openness, as measured using the 44 ­ item Big Five Inventory (John et al., 2008), were related to greater adherence to public health recommendations to minimize the spread of COVID­19 (Bogg & Milad, 2020). In addition, in a sample of Polish adults recruited via social media platforms, low levels of agreeableness, as measured by the 20 Item Personality Item Pool (Donnellan et al., 2006), was tied to less compliance with COVID­19 restrictions (Zajenkowski et al., 2020).

Religiosity has also been examined to see the impact it may have on COVID­19 guideline adherence. For example, Defranza et al. (2021) found less adherence to COVID­19 guidelines, such as shelter­in­place directives, in regions with higher concentrations of religious communities relative to population. Shelter­in­place directives restrict religious gatherings, and this can result in communities with higher religiosity feeling that their religious freedom is being restricted making them less likely to comply (Defranza et al., 2021).

Additional factors that have received research attention in relation to predicting variability in responses to the pandemic are Social Dominance Orientation (SDO; Pratto et al., 1994) and Right Wing Authoritarianism (RWA; Altemeyer, 2007). SDO refers to how comfortable one is with group ­ based inequality and social hierarchies that perpetuate privilege (Pratto et al., 1994). RWA refers to a dogmatic and inflexible support for existing mainstream social practices and institutions, an unwillingness to consider the flaws or limitations of those practices and institutions, and antipathy towards

any group or movement perceived as threatening to undermine the validity of these traditional values (Altemeyer, 2007). RWA and SDO have been previously linked with a variety of undesirable social attitudes such as generalized prejudice towards outgroups and less commitment to human rights (McFarland, 2010; McFarland & Mathews, 2005), as well as less support for refugees (Lyall & Thorsteinsson, 2007). RWA has been found to negatively correlate with concerns about COVID­19 at the individual and societal levels (Peng, 2022). Similarly, SDO predicts increased dismissal of the dangers posed by COVID­19 as well as less support for government measures aimed at minimizing its spread, such as mask wearing and social distancing (Peng, 2022).

One potential strategy used for promoting public health measures has been to frame compliance in terms of an act of compassion to protect others who are more vulnerable. For example, Pfattheicher et al. (2020) found that experimentally inducing empathy for the elderly and vulnerable increased compliance with COVID­19 health policies, such as mask wearing and physical distancing, more than simply providing information about the effectiveness of these steps in preventing the spread of the virus. However, the effectiveness of this strategy would seem to depend on the extent to which people are genuinely concerned about others’ welfare.

One relevant variable in the political and social psychology literature related to concern about the welfare of others is the concept of identification with all humanity (IWAH; McFarland, 2011). IWAH refers to the perception of connection to and common identity with people regardless of ethnic, cultural, and religious classifications typically used to differentiate group status. IWAH involves being concerned about and wanting to help other people because of their intrinsic value and worth as fellow human beings. IWAH has previously been linked to prosocial attitudes and behaviors such as giving charitable donations for international causes and having more desire to learn about humanitarian concerns on a global scale (McFarland et al., 2012). IWAH has also been associated with more compassion towards refugees and less callousness towards civilian war causalities (Bassett & Cleveland, 2021). Further, IWAH has recently been linked to attitudes about those impacted by COVID­19. For example, McCutchen et al. (2022) reported that among Irish participants IWAH was associated with more favorable attitudes about the need for increased resources and initiatives to deal with the growing mental health issues stemming from COVID19. Similarly, Zagefka (2021) found that among British participants a tendency to conceptualize COVID­19 as a shared global problem requiring international cooperation was associated with a greater willingness to give to

Ferqueron, Bassett, and Cleveland | COVID-19 Safety Precautions

COVID-19 Safety Precautions | Ferqueron, Bassett, and Cleveland

charities assisting in­group (those in Great Britain) and also out­group (those in Syria and China) COVID­19 victims but that this effect was partially mediated by higher levels of IWAH.

IWAH has also been studied in relation to behavioral responses to COVID ­ 19. Kaplan et al. (2021) found that higher scores on IWAH were associated with less endorsement of the ideas that masks are ineffective, limit personal freedom, and are signs of weakness and conformity. Kaplan et al. did not find a statistically significant association between IWAH and compliance with public health measures designed to prevent the spread of COVID­19 among American participants. In contrast, Barragan et al. (2021) did find IWAH to be a significant predictor of greater adherence to COVID­19 guidelines, such as hand washing and social distancing in a multinational sample including participants from America, Asia, South America, and Europe. Further investigation into the relation of IWAH to attitudes and responses to COVID­19 seems warranted.

Overview of the Present Studies

The goal of the present research was to further assess factors that predict attitudes and reactions to the COVID­19 pandemic. We examined political orientation, religiosity, personality, SDO, and RWA because of the previously established role of these variables in the literature. Personality traits like neuroticism might predict greater adherence to public safety measures out of a personal concern about avoiding the dangers associated with contracting the virus. In contrast, variables like political orientation, SDO, and RWA might predict less adherence to public safety measures out of the desire to make an ideological statement or to show in­group loyalty. However, we were most interested in examining the ability of IWAH to predict reactions to COVID­19 over and beyond the influence of these other factors. IWAH should predict greater compliance with measures to reduce the spread of COVID­19 out of an altruistic concern for all other people who it could potentially harm.

Study 1 was conducted in the early days of the pandemic (March 2020) and assessed the relation of IWAH to worries about the societal health consequences of COVID­19 as well as support for canceling events and changing patterns of social activity to reduce the spread of the virus. Study 2 attempted to replicate the relation of IWAH to these same variables in a sample obtained via different recruitment methods. Further, the data for Study 2 were collected approximately seven months later and included a measure of compliance with mask wearing recommendations, which were not yet in place at the time of Study 1. We hypothesized that IWAH would be associated with more concern for the overall health of

society at large, more support for canceling events and activities to help prevent the spread of the virus, and more compliance with mask wearing recommendations.

Study 1

Method Participants

Participants were 512 people who completed an online survey between March 23 and May 11, 2020. Participants were recruited through email using a university student and faculty list serve as well as through social media where the authors posted links to the study on Facebook pages. Of the participants, 384 were female, 115 were men, 7 were transgender, 4 were non ­ binary, and 2 preferred not to answer. Of the participants, 443 indicated their racial/ethnic background as White, 24 as Black or African American, 19 as Hispanic or Latino, 5 as Asian or Asian American, 1 as American Indian or Alaskan Native, 1 as Native Hawaiian or other Pacific Islander, and 19 as other or unspecified. Of the participants, 189 were university students. Participants from 33 different states completed the survey. The majority of participants were from South Carolina (75.2 %), followed by Indiana (4.9 %) and North Carolina (3.1 %). Representation from each of the remaining states was about 1% or less of the sample. Participants ranged in age from 18 to 77 years (M = 34.51, SD = 15.65).

Materials

Identification With All Humanity. Participants completed the Identification with All Humanity Scale (McFarland et al., 2012). This scale contained nine questions rated on a 5­point Likert scale. Each question was answered in reference to people in the community, Americans, and people all over the world. A representative example was “How close do you feel to each of the following (people in my community, Americans, people all over the world)?” They responded using a scale from 1 ( not close at all ) to 5 ( very close ). The scale yields separate scores for how closely people identify with their community, America, and all of humanity. Across multiple samples, McFarland et al. reported good internal consistency for community (alphas ranged from .78–.89), America (alphas ranged from .78–.83), and all humanity (alphas ranged from .62–.83). The scales also had good internal consistency in the present study: community (α = .88), America (α = .85), and all humanity (α = .84).

Ten-Item Personality Inventory. Participants completed the Ten Item Personality Inventory (TIPI; Gosling et al., 2003). This scale included 10 items, two for each of the five TIPI characteristics. This scale measures levels of openness, conscientiousness, extraversion, agreeableness, and neuroticism. Gosling et al. reported

low internal consistencies for these scales (αs ranging from .43 to .73); however, they also reported strong correlations between scores on the brief measure with scores on the corresponding dimension based on longer measures (rs ranged from .65 to .87). Based on the data from the current study, the internal consistencies were somewhat low. The extraversion (α = .78) and emotional stability (α = .73) scales had good internal consistency. However, the agreeableness (α = .40), conscientiousness (α = .67), and openness (α = .29) scales had suboptimal internal consistency. However, the TIPI traits were not the main focus of our research and participants were already completing a lengthy questionnaire, so we chose to sacrifice psychometric excellence for expediency by choosing to use the TIPI for its brevity.

Social Dominance Orientation. Participants completed the Social Dominance Orientation Scale which contains 13 items (Pratto et al., 1994). The participants rated ideas stated on a scale of 1 ( very negative) to 7 (very positive). Seven of the items were worded such that higher values indicated more SDO, whereas the other six items were worded such that higher scores indicated less SDO. Before averaging the score, the six items had to be reverse scored to be on the same scale as the other seven items, so that higher scores reflected greater social dominance orientation. Pratto et al. reported good internal consistency for the measure (α = .83). The measure also had good internal consistency in the current study, as evidenced by a Cronbach’s α of .88.

Right Wing Authoritarianism. Participants completed the Right ­ Wing Authoritarianism Scale (Altemeyer, 2007). This scale has 22 items on a 9­point scale anchored at ­4 (very strongly disagree), 0 (neutral), and 4 (very strongly agree). Ten items were worded to reflect less RWA, and 12 items were worded to reflect more RWA. The 10 items had to be reverse scored before summing them with the other 12 items. Higher scores reflected more RWA. Altemeyer reported good internal consistency for this measure (α = .90). The measure had good reliability in the current study as well (α = .96).

Political Orientation. Participants were asked to respond to the question “How would you characterize your political orientation?” using a scale anchored at 1 (conservative), 4 (moderate), and 7 (liberal). Such single item measures have been used successfully in previous research examining differences in political ideology (Carney et al., 2008).

Religion. Participants were asked to rate the importance of religion in their lives on a scale from 1 (not at all important) to 7 (extremely important). Participants were asked to indicate the frequency of their attendance at religious services on a scale from 1 (never) to 7 (more than twice a week). The average religious attendance

was 3.61 (SD = 2.03) and the average rating for religious importance was 4.47 (SD = 2.15). These two items were strongly correlated (r = .79), therefore we averaged them in order to create a combined religiosity variable. This approach to measuring religion was used in previous research on the correlates of IWAH (Bassett & Cleveland, 2019).

COVID Concern. Participants rated the extent to which they were concerned about the COVID ­ 19 pandemic in regards to the health of society at large. Ratings were made on a 5­point scale from 1 (not at all) to 5 (very much).

Support for Canceling. Participants rated the extent to which they believed eight types of events should be canceled in order to prevent the spread of COVID­19. These events included large events such as concerts and sporting events and more routine events such as school and church events. Ratings were made on a scale from 1 (unnecessary), 2 (not sure), and 3 (necessary). A total support for canceling score was created using the sum of the eight items (Cronbach’s α = .85).

Behavioral Support for Canceling. Participants indicated whether they had limited their social interaction in seven different contexts. These contexts included canceling or postponing travel, visiting family, and trips to the grocery store. The behavioral support for canceling score was created using the sum of these seven items.

Procedure

Before conducting the study, researchers acquired approval from the Lander University Institutional Review Board (IRB application number 2019­15). The first survey was sent out in March 2020. Participants were excluded from the research if they were under 18 or lived outside of the United States and U.S. territories. Participants were informed in the consent form of this study that they would be asked to complete measures of attitudes and social values, share perceptions of social distancing as well as their level of participation, and provide some demographic information. They completed the materials in the order of demographics, support for canceling, behavioral support for canceling, COVID concern, IWAH, SDO, TIPI, and RWA.

Results

Descriptive statistics, reliabilities, and simple order correlations are presented in Table 1. More concern about the health of society was associated with being more politically liberal, less social dominance, less right wing authoritarianism, stronger identification with community, stronger identification with America, stronger identification with all humanity, and more openness. Stronger support for canceling was associated

Ferqueron, Bassett, and Cleveland | COVID-19 Safety Precautions

COVID-19 Safety Precautions | Ferqueron, Bassett, and Cleveland

with less religiosity, being more politically liberal, less social dominance, less right wing authoritarianism, and more identification with all humanity. More behavioral support for canceling was associated with being more politically liberal, less social dominance, less right wing authoritarianism, more identification with community, stronger identification with America, stronger identification with all humanity, greater extraversion, greater emotional stability, and greater openness.

The concern for society’s health, support for canceling, and behavioral support for canceling measures were each subjected to separate hierarchical multiple regression analyses. Combined religion, political orientation, SDO, RWA, identification with community, identification with America, extraversion, agreeableness, conscientiousness, emotional stability, and openness were entered in Step 1. Identification with humanity was entered in Step 2. Multicollinearity diagnostics showed that predictor variables were moderately correlated but not to the extent to adversely impact the interpretation of the regression

results. Variance inflation factors (VIFs) ranged from 1.14 to 3.52 and were all below the problematic threshold of 5. In the analysis of concern for society’s health, the regression model accounted for 18.6% of the variance in concern for society’s health and was statistically significant, F(11, 510) = 10.39, p < .001. Step 2 of the model, which included IWAH, now accounted for 19.5% of the variance and was statistically significant, F(12, 510) = 10.08, p < .001. As can be seen in Table 2, political orientation, SDO, IC, IA, and openness were all significant in Step 1. Political liberalism, IC, IA, and openness were all associated with more concern for the health of society at large; whereas SDO was associated with less concern for the health of society at large. IWAH was significant in Step 2 even when controlling for the other variables in the model, indicating that higher scores on IWAH accounted for unique variance in the tendency to express more concern for how COVID­19 impacted the health of society.

In the analysis of support for canceling, the

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, IWAH = identification with all humanity, EXTRA = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, CPH = concern for personal health, CFH = concern for family health, CCH = concern for community health, CSH = concern for society health, SC = support for canceling, BSC = behavioral support for canceling.

TABLE 1

Simple Order Correlations Among All Variables in Study 1

2

Predictors of Concern for the Health of Society at Large in Study 1

regression model accounted for 11.3% of the variance in support for canceling and was statistically significant, F(11, 511) = 5.78, p < .001. As can be seen in Table 3, only political orientation was significant in Step 1. More political liberalism was associated with more support for canceling. Step 2 of the model, which included IWAH, accounted for 11.5% of the variance and was statistically significant, F(12, 511) = 5.38, p < .001, but IWAH did not contribute any additional predictive value Δr2 = .002.

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity.

Predictors of Support for Canceling in Study 1

In the analysis of behavioral support for canceling, the regression model accounted for 14.4% of the variance and was statistically significant, F(5, 232) = 7.50, p < .001. Step 2 of the model, which included IWAH, accounted for 17.3% of the variance and was statistically significant, F(12, 500) = 8.53, p < .001. As can be seen in Table 4, in Step 1, RWA, IC, and emotional stability were all associated with levels of behavioral support for canceling. Higher scores in RWA were associated with less behavioral support for canceling. In contrast, higher scores in IC and emotional stability were associated with more behavioral support for canceling. In Step 2, even when controlling for other variables, IWAH showed a statistically significant relationship with behavioral support for canceling. The higher the score for IWAH the higher the behavioral support for canceling.

4

Predictors of Concern for the Health of Society at Large in Study 1

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity

p < .01. * p < .05.

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity.

** p < .01. * p < .05.

Ferqueron, Bassett, and Cleveland | COVID-19 Safety Precautions

COVID-19 Safety Precautions | Ferqueron, Bassett, and Cleveland

Discussion

The results of Study 1 further validate the fact that reactions to COVID­19 are highly polarized along political lines. In the present study, political liberalism was one of several factors associated with greater concern for how COVID19 was impacting the health of society at large and with a greater likelihood of changing personal behavior to reduce social contact and interaction in order to prevent the spread of the disease. Further, political liberalism was the only measured factor associated with support for canceling events and activities to prevent the spread of COVID­19. These findings are consistent with previously documented effects of political orientation on reactions to COVID­19 (Cheng, 2020; Kerr et al., 2020) and provide further support to the prospect that increasing compliance with public health measures related to COVID­19 may be more difficult among political conservatives who may see such actions as a sign of disloyalty to their political in­group, as governmental overreaching, or as an infringement on personal rights and freedoms.

In the current study, SDO was associated with less concern for society at large. This finding is consistent with previous research that found that higher SDO scores were associated with dismissal of the seriousness the threats posed by COVID­19 (Peng, 2022). One possible interpretation of this finding is that SDO might signify a generalized callousness to suffering, misfortune, and social problems, and consequently a muted reaction to the COVID­19 pandemic might be just another example of this tendency. This interpretation is speculative given the correlational nature of the current data but is logically consistent with previous findings relating high levels of SDO to a lack of concern for others (Bassett & Cleveland, 2019; Lyall & Thorsteinsson, 2007).

In the present study, RWA was associated with less behavioral support for canceling, such that authoritarians were less likely to have changed their routines or altered their attendance and participation in activities in an effort to prevent the spread of COVID­19. This finding might initially appear counterintuitive given the conceptual expectation that high authoritarianism would be associated with more not less rule following. One possible explanation of this finding revolves around the construct of perceived legitimacy. The tendency of people high on authoritarianism to follow rules might not have applied in this situation because the CDC and other government agencies making the public health recommendations might have been viewed as illegitimate. Their legitimacy was potentially undermined by right wing political claims that such policies represented an attempt to overextend the power of the government in order to advance a liberal agenda that would threaten their freedom.

The results of the present study revealed that IWAH predicted both more concern for how COVID­19 was impacting the health of society at large and reduced attendance and participation in events and activities as a way of trying to reduce the spread of the virus. However, IWAH did not predict the attitudes participants held toward canceling events and activities in general as COVID­19 related safety precautions. These results further validate the utility of IWAH as a concept that predicts prosocial attitudes and behaviors aimed at improving the well­being of others. These results also provide additional support to the idea that promoting empathy and concern for the welfare of others could be a means of increasing compliance with public safety measures aimed at preventing the spread and impact of COVID ­ 19. However, the current link is correlational and such a causal possibility must await the results of future experimental research. The generalizability of these findings was limited due to the fact that the sample was composed primarily of people from South Carolina. Study 2 sought to gain additional evidence on this topic with a more diverse sample surveyed at a time later in the pandemic. Study 2 also included questions about compliance with mask recommendations, which were not in effect at the time of the first survey.

Study 2 Method Participants

Participants were 523 people who completed an online survey between October 12 and November 15, 2020. Participants were recruited through email using a university student and faculty list serve, through social media where the authors’ posted links to the study on Facebook pages, and through Amazon’s Mechanical Turk. Of the participants, 275 were from Amazon’s Mechanical Turk, and they were paid $0.75 to complete the survey. Of the participants, 264 were women, 245 were men, 7 were nonbinary, 5 were transgender, and 2 preferred not to answer. Of the participants, 392 indicated their racial/ ethnic background as White or European American, 50 as Black or African American, 24 as Hispanic or Latino/a/x, 35 as Asian or Asian American, 10 as American Indian or Alaskan Native, 2 as Native Hawaiian or other Pacific Islander, and 10 as other or unspecified. Of the participants, 294 were university students. Participants from 44 different states completed the survey. Most participants were from South Carolina (47.4%), followed by California (8.6%), Texas (6.1%), Florida (4.6%), New Your (4.4%), North Carolina (4.4%), and Illinois (2.5). Representation from each of the remaining states was about 1% or less of the sample. Participants ranged in age from 18 to 78 years (M = 31.55, SD = 13.11).

Materials

IWAH, personality, SDO, RWA, religion, political orientation, COVID concern, and support for canceling were all measured in the same way as described in Study 1.

Behavioral Change . An index of the extent to which participants had changed their behavior to prevent the spread of COVID­19 was calculated as the sum of responses to several items. First, they were asked to indicate whether they had canceled, postponed, or declined the following as a result of the COVID ­ 19 pandemic: planned travel, in­person meetings, visiting a friend, seeing family, grocery store trips and/or social gatherings for yourself or your children. Participants were then asked to indicate whether they had altered their behavior because of the COVID­19 pandemic by staying home more and limiting contact with others or if they were acting as normal on a binary scale (0 = no, 1 = yes). Then participants were asked to answer the question, “Have you engaged in nonessential travel more than 50 miles from home since Friday, March 13?” (0 = yes, 1 = no). Participants were asked to answer the question “Have you increased how frequently you wash or sanitize your hands because of this virus?” (0 = no, 1 = yes).

Mask Compliance . Participants were asked to answer yes or no to two questions about whether they wear masks in buildings that do require them and whether they wear masks in buildings that do not require it. They were also asked to indicate whether they wore masks in the following situations: around immediate family, close friends, coworkers, peers, roommates, and large groups outdoors. Participants were also asked to indicate whether they regularly disposed of disposable masks and whether they regularly washed cloth masks after use. Responses to these questions were summed. The resulting score could range from 0 to 10 with higher scores indicating greater mask compliance.

Procedure

Before the second survey was sent out, the researchers got approval from the Lander University Institutional Review Board (IRB amendment application number 2020­21). The second survey was sent out in October 2020. The only compensation was given to the participants from Amazon’s Mechanical Turk in the second survey. The 275 Mechanical Turk participants were given $0.75 for completing the survey. Participants were excluded from

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, IWAH = identification with all humanity, EXTRA = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, CSH = concern for health of society at large, SC = support for canceling, BC = behavioral change, MCS = mask compliance score.

COVID-19 Safety Precautions | Ferqueron, Bassett, and Cleveland

the research if they were under 18 or lived outside of the United States and U.S. territories. Participants were informed in the consent form of this study that they would be asked to complete measures of attitudes and social values, share perceptions of COVID­19 safety precautions as well as their level of participation, and provide some demographic information. They completed the materials in the order of demographics, COVID concern, support for canceling, behavioral change, mask compliance, IWAH, SDO, TIPI, and RWA.

Results

Descriptive statistics, reliabilities, and simple order correlations are presented in Table 5.

COVID concern was associated with more political liberalism, less SDO, less RWA, more IC, more IA, more IWAH, more agreeableness, and more openness. Support for canceling was associated with more political liberalism, less RWA, more IWAH, less extraversion, less conscientiousness, and less openness. Behavioral change was associated with less religiosity, more political liberalism, less SDO, less RWA, more IC, more IWAH, more agreeableness, and more openness. Mask compliance was associated with more political liberalism, less SDO, less RWA, more IC, more IA, more IWAH, less extraversion, more agreeableness, and less conscientiousness.

COVID concern, support for canceling, behavioral

Predictors of Concern for the Health of Society at Large in Study 2

change, and mask compliance were each subjected to separate hierarchical multiple regression analyses. Combined religion, political orientation, SDO, RWA, identification with community, identification with America, extraversion, agreeableness, conscientiousness, emotional stability, and openness were entered in Step 1. Identification with humanity was entered in Step 2. Multicollinearity diagnostics showed that the predictor variables were moderately correlated but not to the extent to adversely impact the interpretation of the regression results. Variance inflation factors (VIFs) ranged from 1.28 to 3.42 and were all below the problematic threshold of 5.

In the analysis of concern for how COVID­19 impact the health of society, the regression model accounted for 27.9% of the variance and was statistically significant, F(11, 510) = 17.92, p < .001. As can be seen in Table 6, SDO, RWA, and emotional stability were associated with less concern; whereas, identification with community was associated with more concern. Step 2 of the model, which included IWAH, now accounted for 28.4% of the variance and was statistically significant, F(12, 509) = 16.85, p < .001. In Step 2, IWAH was associated with more concern even when controlling for the other variables in the model.

In the analysis of support for canceling, the regression model accounted for 24.7% of the variance and was statistically significant, F(11, 511) = 15.21, p < .001. As can be seen in Table 7, political orientation, identification

Predictors of Support for Canceling in Study 2

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity. ** p < .01. * p < .05.

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity.

Table 7

with America, and conscientiousness are all significant in Step 1. Greater support for canceling was associated with higher scores of political liberalism and identification with America and lower scores for consciousness. Step 2 of the model, which included IWAH, now accounted for 27.5% of the variance and was statistically significant, F(12, 510) = 16.09, p < .001. In Step 2, IWAH was associated with more support for canceling even when controlling for other variables in the model.

In the analysis of behavioral change, the regression model accounted for 20.2% of the variance and was statistically significant, F(11, 511) = 11.74, p < .001. As can be seen in Table 8, SDO, RWA, and identification with America were all significant in Step 1. Greater behavioral change was associated with less RWA, less SDO, and more identification with America. In Step 2, SDO and RWA were both significant, and lower scores predicted less behavioral change. Step 2 of the model, which included IWAH, now accounts for 20.4% of the variance and was statistically significant, F(12, 510) = 10.88, p < .001, but entering IWAH in Step 2 did not account for significantly more variance in behavioral change, Δr2 = .00.

In the analysis of mask compliance, the regression model accounted for 13.1% of the variance and was statistically significant, F(11, 511) = 7.02, p < .001. As can be seen in Table 9, political orientation, IA, and conscientiousness were all significant in Step 1. With

Predictors of Behavioral Change in Study 2

more liberalism, more IA, and more conscientiousness being associated with more mask compliance. Step 2 of the model, which included IWAH, now accounts for 16.1% of the variance and was statistically significant, F(12, 510) = 8.13, p < .001. In Step 2, IWAH was associated with more mask compliance even when controlling for the other variables in the model.

Discussion

The results of Study 2 replicated the finding from Study 1 that political liberalism is a strong predictor of support for canceling events to prevent the spread of COVID­19 and extended this finding to also show greater mask compliance among political liberals. These results are consistent with previous findings in the literature (Cheng, 2020; Kerr et al., 2020) and reinforce the idea that one of the main factors determining how Americans interpret and respond to the pandemic is their political orientation. The results of Study 2 also further validate the importance of SDO and RWA in predicting attitudes about COVID19. In Study 2, both SDO and RWA were associated with being less concerned about the health effects of COVID19 on society at large and with less behavioral change to try and reduce the spread of the virus.

The results of Study 2 also further demonstrate the utility of IWAH in predicting reactions to COVID­19. In Study 2, IWAH predicted more concern for society

9

Predictors of Mask Compliance in Study 2

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity.

p < .01. * p < .05.

Note. REL = combined religion, PO = political orientation, SDO = social dominance orientation, RWA = right wing authoritarianism, IC = identification with community, IA = identification with America, EXTR = extraversion, AGR = agreeableness, CON = conscientiousness, ES = emotional stability, OPEN = openness, IWAH = identification with all humanity.

** p < .01. * p < .05.

Ferqueron, Bassett, and Cleveland | COVID-19 Safety Precautions

at large, more support for canceling, and more mask compliance. It is noteworthy that unlike the results of Study 1, IWAH was not a significant predictor of behavioral change in Study 2. Although multiple factors could have been different in the lives of participants in Study 2 compared to those in Study 1, we would like to explore a few possible explanations for the discrepant results. The first is that the sample for Study 2 contained greater geographic diversity. People are experiencing the pandemic differently depending on the state in which they live. Some areas have more restrictive guidelines in place than others. In states with those more restrictive policies in place, there may be more situational pressure that makes it harder to see the impact of individual differences. The second possible explanation is related to the fact that the second study was conducted later in the course of the pandemic. Although people might know that changing their behavior is important, they might be tired of the pandemic. Higher levels of fatigue might result in more personal risks taken as far as COVID­19 safety precautions, in spite of the belief that others should be compliant with these precautions. That might explain why there was a possible behavior intention gap in Study 2, such that IWAH predicted support for canceling events in general but not a personal alteration in travel, social engagement, and participation in activities. Similarly, a perceived necessity of meeting the requirements of work and school as well as a possible growing need for social interaction might have motivated people to personally engage in behaviors that increased risk of COVID­19 transmission even though in general for most people they endorsed these behaviors as undesirable.

General Discussion

The results of the present research revealed that, in two different samples, IWAH was associated with greater participation in prosocial behaviors, such as social distancing and mask wearing. In both studies, IWAH was also found to be associated with a greater level of concern for society as a whole. These results are consistent with previous findings implicating IWAH as a relevant factor in predicting attitudes about COVID­19, such as greater support for charities helping the victims of COVID­19 (Zagefka, 2021), and greater advocacy for better mental health resources in response to COVID­19

(McCutchen et al., 2021). The previous literature on the relationship of IWAH to self­reported behavioral responses to COVID­19 is more ambiguous. In a survey of Americans, Kaplan et al. (2021) failed to find a relation between IWAH and compliance with public health measures to reduce the spread of COVID­19; whereas Barragan et al. (2021) did find IWAH to be a predictor

of compliance with such measures in a multinational study including respondents from the United States. The results of the present study were also somewhat ambiguous. IWAH predicted more self ­ reported behavioral change in response to COVID­19 in Study 1 but not in Study 2.

The findings of the present study are limited due to the characteristics of the samples used. Study 1 included participants primarily from the state of South Carolina, which has favored individual decision making over mandates and requirements in its policies and regulations related to COVID ­ 19. For example, the South Carolina state legislature passed a law prohibiting state agencies from issuing mask requirements (South Carolina General Assembly, 2021). The sample in Study 2 was slightly more diverse geographically but still contained responses primarily from South Carolina and primarily from university students. The data from both samples were collected relatively early in the course of the ongoing response to the pandemic. Additional research is needed to further explore the role of IWAH in predicting more recent attitudes and reactions to COVID­19 among a wider swath of participants. Future research might also benefit from using more measures of actual behavior to circumvent the problems associated with self­report measures.

People’s attitudes and responses to public health crises such as COVID­19 are influenced by multiple factors. These decisions can be based on the risk­benefit assessment of consequences to personal health but also on more ideological factors such as signaling political allegiance or evidencing adherence to valued moral principles. Among the many factors predicting these responses, one potentially important trait to consider is individual differences in the tendency to feel a shared sense of belonging to the collective human race and a sense of duty or obligation to feel compassion and offer help to those in need regardless of divisive group classifications such as gender, religion, or nationality. The results of the present study suggest the value of the construct of IWAH to researchers interested in investigating responses to COVID­19 as well as other global issues requiring international cooperation.

An additional implication from the present research is the importance of political orientation in shaping people’s attitudes about and reactions to COVID­19. Across both studies, political orientation was the factor most consistently and most strongly related to COVID19 attitudes. Political liberals were more likely to support the cancelation of public events and gathering and more likely to comply with mask wearing recommendations compared to political conservatives. These findings highlight the challenges facing health experts and policy

makers as they work to reduce the spread of COVID­19 and prepare for future public health threats related to infectious diseases. Maximizing future compliance with public health interventions aimed at preventing the spread of infectious diseases will likely be contingent on the ability to craft messaging and communication strategies that are persuasive to people across the political spectrum.

References

Altemeyer, B. (2007). The Authoritarians. University of Manitoba. https://theauthoritarians.org/

Ayouni, I., Maatoug, J., Dhouib, W., Zammit, N., Fredj, S. B., Ghammam, R., & Ghannem, H. (2021). Effective public health measures to mitigate the spread of COVID-19: A systematic review. BMC Public Health 21(1), 1015. https://doi.org/10.1186/s12889-021-11111-1

Barragan, R. C., Oliveira, N., Khalvati, K., Brooks, R., Reinecke, K., Rao, R. P. N., & Meltzoff, A. N. (2021). Identifying with all humanity predicts cooperative health behaviors and helpful responding during COVID-19. PLoS ONE, 16(3), e0248234. https://doi.org/10.1371/journal.pone.0248234

Bassett, J. F., & Cleveland, A. J. (2019). Identification with all humanity, support for refugees and for extreme counter-terrorism measures. Journal of Social and Political Psychology, 7(1), 310–334. https://doi.org/10.5964/jspp.v7i1.678

Blume, H., Lin, R.-G., & Luna, T. (2021, October 1). Newsom orders Covid vaccines for eligible students, the first state mandate for K-12 Schools. Los Angeles Times. https://www.latimes.com/california/story/2021-10-01/newsom-sets-covidvaccine-mandate-across-california-schools

Bogg, T., & Milad, E. (2020). Slowing the spread of COVID-19: Demographic, personality, and social cognition predictors of guideline adherence in a representative US sample. Health Psychology, 39(12), 1026–1036. http://dx.doi.org/10.1037/hea0000891

Carney, D. R., Jost, J. T., Gosling, S. D., & Potter, J. (2008). The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology, 29(6), 807–840. https://doi.org/10.1111/j.1467-9221.2008.00668.x

CDC COVID Data Tracker. (n.d.). Retrieved August 13, 2022, from https://covid.cdc.gov/covid-data-tracker/#datatracker-home

Centers for Disease Control and Prevention. (n.d.). CDC Covid Data tracker. Centers for Disease Control and Prevention. Retrieved August 13, 2022, from https://covid.cdc.gov/covid-data-tracker/#new-hospital-admissions

Cheng, J. (in press). The psychology and political orientation of social distancing compliance and attitude toward mask-wearing during the COVID-19 outbreak in the US. Personality and Individual Differences

Chu, D. K., Akl, E. A., Duda, S., Solo, K., Yaacoub, S., & Schünemann, H. J. (2020). Physical distancing, face masks, and eye protection to prevent person-toperson transmission of SARS-CoV-2 and COVID-19: A systematic review and meta-analysis. The Lancet, 395(10242), 1973–1987.

https://doi.org/10.1016/S0140-6736(20)31142-9

DeFranza, D., Lindow, M., Harrison, K., Mishra, A., & Mishra, H. (2021). Religion and reactance to covid-19 mitigation guidelines. American Psychologist, 76(5), 744–754. https://doi.org/10.31234/osf.io/zvyc2

Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP scales: Tiny-yet-effective measures of the Big Five factors of personality. Psychological Assessment, 18(2), 192–203. https://doi.org/10.1037/1040-3590.18.2.192

Fineout, G., & Sarkissian, A. (2022, January 4). GOP governors play down Omicron 'hysteria' while also asking for help. Politico Florida. https://www.politico.com/states/florida/story/2022/01/04/gop-governorsplay-down-omicron-hysteria-while-also-asking-for-help-1403885

Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A very brief measure of the Big Five personality domains. Journal of Research in Personality, 37(6), 504–528. https://doi.org/10.1016/S0092-6566(03)00046-1

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big-Five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (pp. 114–158). Guilford Press.

Kaplan, J., Vaccaro, A., Henning, M., & Christov-Moore, L. (2021, June 30). Moral reframing of messages about mask-wearing during the COVID-19 pandemic. PsyArXiv. https://doi.org/10.31234/osf.io/gfa5r

Kerr, J., Panagopoulos, C., & van der Linden, S. (2021). Political polarization on COVID-19 pandemic response in the United States. Personality and Individual Differences, 179, 110892. https://doi.org/10.1016/j.paid.2021.110892

Liang, M., Gao, L., Cheng, C., Zhou, Q., Uy, J. P, Heiner, K., & Sun, C. (2020). Efficacy of face mask in preventing respiratory virus transmission: A systematic review and meta-analysis. Travel Medicine and Infectious Disease, 36, 101751. https://doi.org/10.1016/j.tmaid.2020.101751

Lippold, J. V., Laske, J. I., Hogeterp, S. A., Duke, É., Grünhage, T., & Reuter, M. (2020). The role of personality, political attitudes and socio-demographic characteristics in explaining individual differences in fear of coronavirus: A comparison over time and across countries. Frontiers in Psychology, 11 https://doi.org/10.3389/fpsyg.2020.552305

Lyall, H., & Thorsteinsson, E. (2007). Attitudes to the Iraq war and mandatory detention seekers: Associations with authoritarianism, social dominance, and mortality salience. Australian Journal of Psychology, 59(2), 70–77. https://doi.org/10.1080/00049530601148421

McCutchen, C., Hyland, P., & Féich, P. Ó. (2022). Attitudes towards mental health services during the COVID-19 Pandemic: Findings from a nationally representative sample of Irish adults. The Journal of Behavioral Health Services & Research, 49(3), 397–405. https://doi.org/10.1007%2Fs11414-021-09785-x

McFarland, S. (2010). Authoritarianism, social dominance, and other roots of generalized prejudice. Political Psychology, 31(3), 453–477. https://doi.org/10.1111/j.1467-9221.2010.00765.x

McFarland, S. (2011). The slow creation of humanity. Political Psychology, 32(2), 1–20. https://doi.org/10.1111/j.1467-9221.2010.00801.x

McFarland, S., & Mathews, M. (2005). Who cares about human rights? Political Psychology, 26(3), 363–385. http://dx.doi.org/10.1111/j.1467-9221.2005.00422.x

McFarland, S., Webb, M., & Brown, D. (2012). All humanity is my ingroup: A measure and studies of identification with all humanity. Journal of Personality and Social Psychology, 103(5), 830–853. https://doi.org/10.1037/a0028724

O'Reilly, K. B. (2020, August 04). Dr. Fauci outlines 5 ways to blunt COVID-19 pandemic's resurgence. Retrieved April 05, 2021, from https://www.ama-assn.org/delivering- care/public-health/dr-fauci-outlines5-ways-blunt-covid-19-pandemic-s-resurgence

Peng, Y. (2022). Give me liberty or give me covid-19: How social dominance orientation, right-wing authoritarianism, and libertarianism explain Americans' reactions to COVID-19. Risk Analysis. https://doi.org/10.1111/risa.13885

Pfattheicher, S., Nockur, L., Böhm, R., Sassenrath, C., & Petersen, M. B. (2020). The emotional path to action: Empathy promotes physical distancing and wearing of face masks during COVID-19 pandemic. Psychological Science, 31(11), 1363–1373. https://doi.org/10.1177/0956797620964422

Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67(4), 741–763. https://doi.org/10.1037/0022-3514.67.4.741

Protect Yourself. (2021, March 8). Retrieved April 05, 2021, from https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/prevention.html

Rammstedt, B., & John, O. P. (2007). Measuring personality in one minute or less: A 10-item short version of the big five inventory in English and German. Journal of Research in Personality, 41(1), 203–212. https://doi.org/10.1016/j.jrp.2006.02.001

South Carolina General Assembly. (2021). H. 4507 Status information. https://www.scstatehouse.gov/sess124_2021-2022/bills/4507.htm

Zagefka, H. (2022). Prosociality during COVID‐19: Globally focused solidarity brings greater benefits than nationally focused solidarity. Journal of Community & Applied Social Psychology, 32(1), 73–86. https://doi.org/10.1002/casp.2553

Zajenkowski, M., Jonason, P. K., Leniarska, M., & Kozakiewicz, Z. (2020). Who complies with the restrictions to reduce the spread of COVID-19?: Personality and perceptions of the COVID-19 situation. Personality and Individual Differences, 166, 110199. https://doi.org/10.1016/j.paid.2020.110199

Author Note

Morgan Ferqueron https://orcid.org/0000­0001­8859­1009

Morgan Ferqueron is now in the Clinical Psychology master’s program at the University of South Carolina Aiken. Correspondence concerning this article should be addressed to Morgan Ferqueron. Email: morganferqueron@gmail.com

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR

Mask-Wearing and Emotional Intensity Perceptions

Brienna Dove and Teddi S. Deka* Department of Psychology, Missouri Western State University

ABSTRACT . The widespread use of sanitary face masks due to the COVID­19 pandemic renewed interest in facial emotion perception while wearing masks. Usage of face masks during this time was considered a nonpharmaceutical intervention to mitigate the spread of the virus. We examined whether wearing face masks affects the intensity of emotion perception and judgments of approachability while also considering the sex and age of the rater. Emerging adults (ages 18–25, M = 19.5, 16 men, 44 women, n = 60) and adults (ages 26–65, M = 44.8, 10 men, 31 women, n = 41) viewed photos of faces of a young adult man, middle­aged man, young adult woman, and a middle­aged woman, masked and unmasked, with happy, sad, neutral, and angry expressions. ANOVAs repeated on the face mask and no face mask conditions showed significant reductions in emotion intensity for happy (p < .001, η2 = .22) and sad faces (p < .001, η2 = .35), no differences for angry faces (p = .16, η2 = .02), and the opposite (increased intensity) with neutral faces (p = .03, η2 = .02). Unmasked happy faces were rated as more approachable than masked happy faces. Unmasked angry faces were rated as less approachable than masked angry faces but only by emerging adults. No differences appeared for sad emotions. Neutral faces again showed an unexpected pattern, with masks increasing approachability.

Keywords: facial expressions, emotions, face masks, emerging adults

Faces play several important roles in social interaction. One’s facial expressions, and interpretations of other peoples’ expressions often determine social interaction. Happy faces are friendly and welcoming. Sad faces may evoke avoidance or empathy. Angry faces can signal displeasure with another person. The six basic emotions—disgust, fear, joy, surprise, sadness, and anger—elicit distinct characteristic facial expressions which are consistent anthropologically and across cultures (Schmidt & Cohn, 2001). Emotion recognition, valence, trustworthiness, and willingness to approach or avoid are just some of the factors determined by facial expression.

The year 2020 marked the onset for widespread use of face masks by ordinary people. Face masks were nonpharmaceutical interventions employed during the COVID­19 pandemic. Face masks, in combination with other nonpharmaceutical interventions such as city lockdowns, isolation, quarantine of positive cases, and avoiding close contact, effectively reduced the spread of COVID­19 transmission (Ayouni et al., 2021). We wonder whether wearing face masks also affected social interactions. The following study builds upon the foundations of research on emotion recognition

from facial expression and newer research on emotion recognition, intensity, and approachability when wearing sanitary face masks.

Emotion Recognition

Different regions of the face contribute to emotion recognition. Covering or hiding the upper face (eye, brow) or the lower face (lower nose, mouth) allow examination of which regions contribute to which emotion identifications. The preponderance of research indicates that the lower face impacts the recognition of happiness, disgust, and surprise the most, and the upper face affects recognition of sadness, anger, and fear, but some emotions show contributions from both regions, namely sadness, disgust, and fear (see Table 1; e.g., Calder et al., 2000; Beaudry et al., 2014).

Younger adults pay most attention to the upper face and eyes on emotion recognition tasks, but older adults (persons 65 years and older) focus more on the lower face and mouth (Circelli et al., 2013; Sullivan et al., 2007; Wong et al., 2005). Older adults show less accuracy identifying sadness, anger, and fear, which rely on the upper face for recognition (Gonçalves et al., 2018; Ruffman et al., 2008). Recognition of anger and fear shows linear

decline from early adulthood to late adulthood, but recognition of disgust (dependent on lower face) shows linear improvement (Calder et al., 2003).

Emotion recognition studies vary on how parts of faces are covered, which alter how much of the region becomes obscured. Altering these coverings may explain some of the variances in facial recognition outcomes. Faces can be obscured without using any object to cover areas by presenting only one part of a face (Beaudry et al., 2014; Guarnera et al., 2017), showing limited areas of the face through round windows called “bubbles” (Blais et al., 2012; Gosselin & Schyns, 2001), or presenting composite or morphed expressions (Schurgin et al., 2014). Objects that cover the upper or lower parts of the face are often employed such as scarves (Calbi et al., 2021), sunglasses (Roberson et al., 2012), niqab (Fischer et al., 2012), and sanitary masks (Calbi et al., 2021; Carbon, 2020; Marini et al., 2021).

The use of sanitary masks has become more common in the United States following Centers for Disease Control guidelines for reducing the spread of COVID19, covering the mouth, cheeks and most of the nose (CDC Newsroom, 2020). Accuracy of identifying happy, sad (Carbon, 2020; Marini et al., 2021) fearful (Marini et al., 2021), angry (Carbon, 2020), and disgusted faces (Carbon, 2020) significantly declines when comparing faces covered with sanitary masks to unmasked faces. The outcomes align with the literature regarding the importance of the lower face for recognizing happy

References for Emotion Recognition by Facial Area

Emotion Upper face contributes Lower face contributes

Happy

Beaudry et al., 2014

Blais et al., 2012

Calder et al., 2000

Nusseck et al., 2008

Roberson, 2012

Shurgin et al., 2014

Sullivan et al., 2007

Sad Beaudry et al., 2014

Calder et al., 2000

Nusseck et al., 2008

Sullivan et al., 2007

Anger Beaudry et al., 2014

Calder et al., 2000

Shurgin et al., 2014

Disgust Nusseck et al., 2008

Beaudry et al., 2014

Nusseck et al., 2008

and disgusted faces, and both upper and lower face for sadness. Wearing sanitary masks should not affect anger and fear recognition, because these emotions rely mostly on the upper face (Beaudry et al., 2014; Calder et al., 2000; Schurgin et al., 2014). However, wearing sanitary masks seems to reduce recognition accuracy, even with these upper­face emotions when compared to unmasked faces (Carbon, 2020; Marini et al., 2021).

Emotion Intensity

Emotion recognition while wearing sanitary masks built upon a significant history of facial region research. The prolonged sanitary mask use during the COVID ­ 19 pandemic brought new inquiries into how covering faces affects the interpretation of emotional intensity, trustworthiness, and willingness to approach or avoid, all of which affect social interactions. These emerging investigations suggest that masks “dull” emotion perception. For instance, Kastandieck et al. (2021) compared perceptions of masked and unmasked happy and sad faces. Unmasked happy faces were perceived as happier than masked faces, and unmasked sad faces were perceived as sadder than masked faces. Hareli et al. (2013) also found reduction in perceived emotional intensity while wearing a sanitary face mask and surgical hat compared to wearing a casual scarf and hat or wearing a niqab. All three presentations occluded roughly the same amount of the face, but only the medical covering reduced perceived emotional intensity. Their findings suggest that the association of the sanitary face mask to the medical community contributes to the reduced emotional intensity judgments, which is a stereotype associated with medical professionals (Hareli et al., 2013).

Fear Beaudry et al., 2014

Calder et al., 2000

Schurgin et al., 2014

Beaudry et al., 2014

Calder et al., 2000

Nusseck et al., 2008

Sullivan et al., 2007

Beaudry et al., 2014

One’s facial expressions signal social information that influences interaction. Looking at unmasked emotional expressions, happy faces are judged as more trustworthy and angry faces as less trustworthy (Oosterhof & Todorov, 2009; Renard et al., 2016). Judging the valence of the emotional expression as positive or negative leads to approach or avoidance responses, respectively (Chen & Bargh, 1999). Limited explorations into masked faces and trustworthiness suggest that masks affect the rating of untrustworthy faces, so they are perceived as less untrustworthy (Grundmann et al., 2021; Marini et al., 2021), but trustworthy faces are perceived as trustworthy whether unmasked or masked (Marini et al., 2021). Masked faces, regardless of valence, may even appear more trustworthy than unmasked faces (Cartaud et al., 2020).

Surprise

Nusseck et al., 2008

Interpersonal distance—the comfortable distance between people—is an indicator of willingness to approach and socialize with others. Valence contributes to interpersonal distance, with happy faces eliciting lower distance ratings than angry faces (Calbi et al., 2021).

Dove and Deka | Mask and Emotional Intensity

Wearing face masks seems to decrease participants’ interpersonal distance, lowering the amount of perceived distance needed to feel comfortable (Cartaud et al., 2020). Such reduction may occur because people have a harder time judging the severity of angry faces when they are partially covered with a mask (misjudgment). Alternatively, masks provide safety for closer social interaction. People who associate face masks with protection against threat (spreading of the Coronavirus) award higher closeness ratings to others wearing face masks (Grundmann et al., 2021).

Purpose of Present Study

The current research examined differences in participants’ perceived strength of emotion and approachability when targets are wearing sanitary face masks or no face masks, using a novel approach for rating faces on contrasting­emotion scales. Participants rated unmasked and masked faces showing the expressions of happy, sad, neutral, and angry using three 9­point scales anchored at happy­sad, calm­angry, and approach­avoid. This technique did not focus on identifying the accuracy of emotion judgment, but rather the strength of the rating between two opposing anchors. Raters were expected to rate unmasked happy and sad faces closer to the happy or sad anchor, respectively, when compared to masked happy and sad faces. Doing so would indicate greater perceived emotional intensity for unmasked faces. Raters were expected to rate masked faces further from an anchor indicating lesser perceived emotional intensity, supporting Kastandieck et al. (2021). Raters were expected to show no difference in rating unmasked and masked neutral and angry faces on the calm­angry scale, because neither emotional expression derives information from the lower face. Similarly, raters were expected to rate unmasked faces closer to the approach anchor for happy faces, and closer to the avoid anchor for sad faces when contrasted to masked faces. Such findings would support reduction of ability to derive social information from masked faces with emotions that rely on the lower face (Grundmann et al., 2021; Marini et al., 2021). No differences were expected for masked versus unmasked neutral or angry faces, because additional information about the expression should not be revealed from uncovering the mouth. Although no formal hypotheses were put forth regarding age, two age groups were included to investigate the impact of age on facial judgments, because older individuals concentrate more on the lower part of the face and could show larger differences between unmasked and masked ratings (Calder et al., 2003; Circelli et al., 2013; Sullivan et al., 2007; Wong et al., 2005).

Participants

Method

Participants (N = 101) were divided into two age groups: emerging adults (ages 18 ­ 25, M = 19.5, 16 males, 44 females, n = 60) and adults (ages 26­65, M = 44.8, 10 males, 31 females, n = 41), following Berger’s (2020) age ranges and classifications. The majority of participants were White (83%), followed by Black or African American (10%), multi ­ racial (4%), Asian American (1%) and did not indicate (1%). Hispanic, Latinx, or Spanish origins were reported by 10% of the participants. According to Data USA’s 2020 information (https://datausa.io/), the sample was reflective of the racial/ethnic makeup of both the university and city.

Most emerging adults that took part in this study were fulfilling a course requirement for their introductory psychology class at a mid­sized, midwestern university, and received course credit for their participation. Adult participants were obtained from solicitations on social media and emails from college students enrolled in an upper­ division psychology course, reaching family, friends, colleagues, and community members, and did not receive an incentive for their participation. Participation was voluntary, and procedures followed the university’s Institutional Review Board (IRB) procedures for obtaining informed consent and conducting research.

Measures and Procedure

The MPI Faces Database (Ebner et al., 2010) contains pictures of 61 White young adults, 60 White middle­aged adults and 58 White older adults expressing happy, angry, fearful, sad, disgusted, and neutral facial expressions. Each image was validated by 48 to 84 raters equally divided by age group (young adult, middle­aged adult, older adult) and gender (male, female) who identified the emotional expression and the age of the person shown in the image (Ebner et al., 2010). Happy faces were correctly identified by 96% (SD = 9%) of the raters, sad 73% (SD = 14%), neutral 87% (SD = 12%) and angry 81% (SD = 13%). Ebner et al. (2010) indicated that these numbers are comparable to other facial database validation studies. All faces were also judged for perceived age. The average age judgment for young adult faces was 28.5 (SD = 3.5), middle­aged adult 49.2 (SD = 3.3) and older adult 68.6 years (SD = 4.1) with actual average ages of 24.2, 49, and 73.2 years. Ebner et al. (2010) described interclass correlation between raters as .88, reflecting good interrater reliability.

The picture numbers used and specific emotion validities and age judgments are presented in the following format: person (image number), happy, sad, neutral, angry ratings as H, S, N, A followed by a or b to indicate which image from a choice of two was used in the current research, and low­to­high age estimates for the four

emotion images. For young adult woman (image 140), Hb = 100%, Sb = 95%, Na = 87%, Ab = 94% with age estimate of 24.8 to 26.6. For young adult man (image 066), Hb = 100%, Sb = 96%, Nb = 97%, Ab = 88% with age estimate of 23.2 to 25.3. For middle­aged woman (image 168), Ha = 98%, Sb = 63%, Na = 52%, Ab = 91% with age estimate of 45.9 to 50.1. For middle­aged man (image 116), Hb = 100%, Sb = 94%, Nb = 91%, Ab = 97% with age estimate of 47.7 to 50.7. These ratings are available on the FACES website (https://faces.mpdl.mpg.de/imeji/).

Digital face masks were added to the 16 images using Kapwing’s free mask tool (Kapwing, 2020). The authors obtained permission to modify these publicly available images with face masks through the registration and release process established by the MPI FACES database ( https://faces.mpdl.mpg.de/imeji/register). Thus, 32 pictures (16 masked and 16 unmasked) were equally balanced among the four emotional expressions, age and sex (see Figure 1).

Participants rated each unmasked and masked face three times using nine­point scales anchored with happysad, calm­angry, and approach­avoid. The term “calm” was used instead of “neutral” for the calm­angry scale because “calm” is a near­antonym of angry (along with “placid” and “serene”) which would be understandable to raters, yet also seemed appropriate to anchor a scale opposite to “angry” (Merriam­Webster, n.d.). Each presentation consisted of one face and one rating scale. Half of the participants rated randomly ordered unmasked faces first (48 faces), and the other half rated randomly ordered masked faces first (48 faces). Altogether, there were 96 face presentations reflecting four targets (young adult woman, young adult man, middle­aged woman, middleaged man), with four emotional expressions (happy, sad, neutral, angry), two mask conditions (no mask, mask) and three ratings (happy­sad, calm­angry, approach­avoid).

Participants completed the first 13 items from the Rotter Locus of Control scale (Rotter, 1966) in between rating the masked face set and the unmasked face set and the scale was not included in the analysis. Participants completed a demographic questionnaire including age, race, ethnicity, and gender. Participants completed all questionnaires within 30 minutes.

Results

Emotion Ratings of Unmasked and Masked Faces

Emotional expression and strength were assessed with two 9­point scales anchored with happy­sad, and calmangry. Both scales were used for all the face images, so that participants could not easily identify the emotion. On the happy­sad scale, a score closer to 1 meant that the participant perceived the face as happy, and a score closer to 9 meant the participant perceived the face as sad. Likewise,

with the calm­angry scale, a score closer to 1 meant the participant perceived the face as calm, and a score closer to 9 meant the participant perceived the face as angry. There were four faces representing each emotional expression. The happy­sad scale was evaluated for the happy and sad faces, and the calm­angry scale was evaluated for the neutral and angry faces. Each rating ranged from 4 to 36, with low scores representing either happy or calm, and high scores representing either sad or angry. The happy­sad scale showed good reliability for the happy faces (unmasked Cronbach’s α = .86, masked α = .86) and sad faces (unmasked α = .81, masked α =.77). The calm­angry scale showed good reliability for the neutral faces (unmasked Cronbach’s α = .73, masked α = .83) and angry faces (unmasked α = .82, masked α = .86). For each of the four emotion expressions, a 2 (sex) by 2 (rater age: emerging adult, adult) mixed­design ANOVA compared participants’ ratings on the repeated variable, which was unmasked faces versus masked faces. Ratings of happy faces and sad faces supported hypotheses. Participants rated unmasked happy faces as significantly happier than masked happy faces (unmasked M = 5.61, SD = 3.52, vs. masked M = 8.31, SD = 5.85), F(1, 97) = 26.58, p < .001, η2 = .22. For sad, participants rated unmasked sad faces as significantly sadder than masked sad faces (unmasked M = 32.85, SD = 3.90, vs. masked M = 29.85, SD = 5.06), F(1, 99) = 54.22, p < .001, η2 = .35 (see Figure 2). There were no significant between­subjects effects.

Ratings of neutral faces and angry faces partially supported expectations. As expected, there were no significant differences between angry unmasked and masked faces (unmasked M = 33.04, SD = 3.65, vs. masked M = 32.64, SD = 4.20), F(1, 97) = 2.04, p = .16, η2 = .02. However, neutral faces showed a significant difference.

Examples of Unmasked and Masked Faces

Dove and Deka | Mask and Emotional Intensity

Figure 1

Happy Sad Neutral Angry

Unmasked Masked

Note. The figure shows 140b – young woman, 168b – middle-aged woman, 066b – young man, and 116b –middle-age man without masks from the MPI FACES database (Ebner et al., 2010), then with masks applied using the Kapwig application tool (Kapwig, 2020). The authors obtained permission to modify and publish these publicly available images with face masks through the registration and release process established by the MPI FACES database (https://faces.mpdl.mpg.de/imeji/register).

Mask and Emotional Intensity | Dove and Deka

Although all the neutral faces were rated closer to the calm pole than the angry pole, masked neutral faces were rated as calmer than unmasked neutral faces (unmasked M = 15.75, SD = 6.19, masked M = 14.07, SD = 6.60), F (1, 99) = 5.05, p = .03, η2 = .02 (see Figure 2). There were no significant between­subjects effects.

Approach-Avoid Ratings of Unmasked and Masked Faces

For approach ­ avoid, participants rated pictures of masked and unmasked happy, sad, neutral or angry faces for whether participants would approach or avoid the targets on a 1 ( approach ) to 9 ( avoid ) scale. Four images represented each emotion, so the minimum approach ­ avoid rating for each emotion was 4 (approach) and the maximum was 36 (avoid). The approach ­ avoid scale’s reliability for happy (unmasked Cronbach’s α = .91, masked α = .88) sad (unmasked α = .86, masked α = .81), neutral (unmasked α = .66, masked α = .80), and angry faces (unmasked α = .92, masked α = .92) was good to excellent except for neutral ­ unmasked faces. The young adult man image was perceived as more approachable (M = 3.63) than the other images (young adult woman M = 5.23, middle­aged man M = 5.87, middle­aged woman M = 6.28). However, eliminating any of the four images reduced the reliability coefficient.

For each of the four emotion expressions, a 2 (sex) by 2 (rater age: emerging adult, adult) mixeddesign ANOVA compared participants’ ratings on the repeated variable, which was unmasked faces versus masked faces. As hypothesized, participants rated

happy unmasked faces as significantly more approachable than happy masked faces (unmasked M = 8.62, SD = 6.17, vs. masked M = 10.81, SD = 7.49), F(1, 97) = 15.55, p < .001, η2 = .14. Participants were expected to show similar patterns for sad faces, however, no significant differences emerged when comparing unmasked and masked ratings for sad faces (unmasked M = 23.70, SD = 7.14, vs. masked M = 22.62, SD = 7.59), F(1, 97) = 0.36, p = .55, η2 = .004 (see Figure 3). There were no significant between­subjects effects.

Ratings for neutral and angry faces partially supported expectations. As expected, there were no significant differences in the ratings between the unmasked and masked angry faces (unmasked M = 32.15, SD = 5.37, vs. masked M = 31.73, SD = 5.82,) F(1, 96) = 1.03, p = .31, η2 = .01. However, there was a significant age group by mask interaction, F(1, 96) = 5.12, p = .03, η2 = .05. Separate sex by mask/no mask analyses for each age group indicated a significant difference for the emerging adult raters (n = 59), who rated unmasked angry faces as slightly more avoidable than masked angry faces (unmasked M = 32.93, SD = 4.71, vs. masked M = 32.08, SD = 5.28), F(1, 58) = 4.79, p = .03, η2 = .08. The adult group (n = 41) did not show any differences in their approach/avoid ratings for angry faces (unmasked M = 31.02, SD = 6.08, masked M = 31.22, SD = 6.54). Although no significant differences were expected for neutral faces, participants perceived unmasked neutral faces as significantly less approachable than neutral masked faces (unmasked M = 21.07, SD = 5.24, vs. masked M = 17.81, SD = 6.38), F(1, 98) = 34.94, p < .001, η2 = .26) (see Figure 3). There

Approach/Avoid Ratings of Facial Expressions

FIGURE 2

Emotion Ratings of Facial Expressions

Note. The figure shows the happy-sad ratings for happy and sad facial expressions, and the calm-angry ratings for the neutral and angry facial expressions. Symbols at the lowest or highest points on the figure represent stronger emotion ratings. Results for happy, sad, and neutral emotions were significant.

FIGURE 3

Note. Symbols at the lowest or highest points on the figure represent a stronger approach or avoid rating. Results for happy and neutral emotions were significant.

were no other significant between­subjects effects. Discussion

The current investigation examined whether face masks interfered with the perceived intensity and approachability of targets’ emotional expressions, using anchored rating scales. The first hypothesis was supported. Happy and sad emotions, which are dependent on the lower face for recognition, solicited higher intensity ratings when faces were unmasked, and lower intensity ratings when faces were masked. The second hypothesis was partially supported. Angry and neutral emotions, of which the lower face contributed little additional information, were expected to elicit no differences in intensity ratings whether unmasked or masked. Although ratings of angry faces showed no differences when unmasked or masked, neutral faces were rated more intensely when masked, toward the calm pole on a calm­angry scale.

The third hypothesis, that happy and sad emotions would elicit ratings closer to the approach or avoid poles when unmasked versus masked, was partially supported. Although happy unmasked faces were rated closer to approach than masked faces, no significant differences emerged for sad faces. The fourth hypothesis, that angry and neutral faces would show no unmasked versus masked differences in approach­avoid ratings, was partially supported. The within­subjects effect for angry faces was not significant, indicating no differences with rating the unmasked and masked angry faces. However, the significant age by facemask interaction indicated that the younger age group (emerging adults) rated unmasked angry faces as more likely to avoid than the masked angry faces. Participants, regardless of age, judged neutral masked faces as more approachable than unmasked neutral faces, contradicting hypotheses but aligning with the emotional intensity ratings found in this investigation.

Overall, ratings of happy and sad faces aligned with expectations. Both emotions are reliant on the lower face for identification of the emotion (see Table 1), and sanitary masks interfere with happiness and sadness recognition (Carbon, 2020; Marini et al., 2021). The current research supports the Kastandiek et al. (2021) study, which found that emotional intensity ratings for happy and sad faces were dulled with face­masked images. Happiness is most reliant on the lower face (see Table 1), which may explain why the approach­avoid ratings were significantly more intense (toward approach) for happy unmasked faces but not for sad (toward avoid) unmasked faces. The literature shows mixed expectations regarding facemask wearing and approachability and comfort (Cartaud et al., 2020; Kastedieck et al., 2021) possibly because face masks remove information by covering part of the face which could

contribute to misjudging whether to approach or avoid someone, but face masks also serve to protect individuals from contracting infections from others, thus encouraging approachability (Grundmann et al., 2021).

Ratings of angry and neutral faces partially aligned with hypotheses and opened possibilities for further investigation. Angry and neutral faces were not expected to show differences in intensity ratings, because neither depends predominantly on the lower face for recognition (see Table 1). Angry faces, for the most part, elicited no differences in intensity or avoidance ratings, whether masked or unmasked, and were rated toward angry and avoid.

Ratings of neutral faces contradicted hypotheses, with masked neutral faces eliciting ratings closer to calm and approach than unmasked neutral faces. Although Marini et al. (2021) found no differences in masked/ unmasked neutral face recognition, our results suggest some contribution from the lower face, at least with emotional intensity, when adding or removing a mask with neutral faces. The lower face contributes most strongly to the perception of happiness and disgust (see Table 1). Given the anchors of calm­angry, the masked face may have prompted anticipation of a happier expression hidden under the mask, nudging the masked ratings toward calm. Examining the approach­avoid ratings for the neutral faces indicated that raters perceived the masked face as more approachable. As with the emotional intensity judgments, covering the lower part of a neutral face may have led to an anticipation of positivity or happiness under the mask, eliciting higher approach ratings. Likewise, raters could have interpreted a person wearing a mask as more approachable because face masks are protective (Grundmann et al., 2021).

Raters’ age is a variable of interest, because older adults may focus more on the lower face to identify emotions (Circelli et al., 2013; Sullivan et al., 2007; Wong et al., 2005). Our results showed only one significant age difference for avoidance ratings of angry faces with emerging adults. Our sample was not adequate to explore age as a variable, which is one of several limitations discussed below.

Limitations and Future Directions

This study’s limitations include sample acquisition and characteristics, the rating scale, and the face targets. For sample acquisition, we used a convenience sample of college students for the emerging adult group, and word ­ of ­ mouth for the adult group. Broadening the sample to the surrounding community and using a more systematic approach could have allowed more opportunity to explore age as an independent variable by having at least three groups of adults: emerging adults (18­25), adults (26­64), and older adults (persons 65 years

Dove and Deka | Mask and Emotional Intensity

and older). Older adults focus more on the lower face than other adult age groups (Circelli et al., 2013; Sullivan et al., 2007; Wong et al., 2005), so face mask wearing may affect emotion judgment and strength more than younger adults.

Regarding sample characteristics, our sample lacked racial, cultural, educational, and gender diversity, which reduced generalizability. Reaching out to the greater community and adults living in larger cities would increase both diversity and the application of the research. Sample diversity could indicate whether such characteristics impact the interpretation of emotion strength and approachability.

Third, the face targets were all White men and women, as were all the images from the database. Using target images that diversified race, gender, and age could contribute to understanding facial expression consistency and universality (Schmidt & Cohn, 2001). Using a larger and more varied database should increase target diversity and improve the chances of selecting strong targets. In the current study, the database ratings for the middle­aged woman’s sad and neutral expressions were not as strong as the other face targets. Only using the publicly available faces further limited our selections.

Conclusion

Research on face mask wearing has just begun to flourish beyond the ability to recognize emotion into perceptions of emotional intensity, approachability, trustworthiness, and safety. Perceptions of happy faces were most clearly affected by the presence of face masks, reducing intensity and approach ratings. Neutral faces also showed differences in intensity and approach ratings, but in the opposite direction, and masked faces were rated as calmer and more approachable relative to unmasked counterparts. This research built upon Carbon’s (2020) examination of whether face masks affect emotion identification, and Kastandieck et al.’s (2021) investigation of whether facemasks change raters’ perception of emotion strength, by using a novel approach for rating faces on contrasting­emotion and approach­avoid scales. The findings expanded upon Kastandieck et al. (2021), by including neutral and angry faces in addition to happy and sad faces, and concentrated on approach ­ avoid perceptions instead of trustworthiness (Cartaud, 2020; Grundmann et al., 2021, Martini et al., 2021, and perceived closeness (Calbi et al., 2021; Cartaud et al., 2020).

References

Ayouni, I., Maatoug, J., Dhouib, W., Zammit, N., Fredj, S. B., Ghammam, R., & Ghannem, H. (2021). Effective public health measures to mitigate the spread of COVID-19: A systematic review. BMC Public Health, 21(1), 1–14. https://doi.org/10.1186/s12889-021-11111-1

Beaudry, O., Roy-Charland, A., Perron, M., Cormier, I., & Tapp, R. (2014). Featural processing in recognition of emotional facial expressions. Cognition and

Emotion, 28(3), 416–432. https://doi.org/10.1080/02699931.2013.833500

Berger, K. S. (2020). The developing person through the lifespan (11th ed.). Worth. Blais, C., Roy C, Fiset, D., Arguin, M., & Gosselin, F. (2012). The eyes are not the window to basic emotions. Neuropsychologia, 50(12), 2830–2838. https://doi.org/10.1016/j.neuropsychologia.2012.08.010

Calbi, M., Langiulli, N., Ferroni, F., Montalti, M., Kolesnikov, A., Gallese, V., & Umiltà, M. A. (2021). The consequences of COVID-19 on social interactions: An online study on face covering. Scientific Reports, 11, Article 2601. https://doi.org/10.1038/s41598-021-81780-w

Calder, A. J., Keane, J., Manly, T., Sprengelmeyer, R., Scott, S., Nimmo-Smith, I., & Young, A. W. (2003). Facial expression recognition across the adult life span. Neuropsychologia, 41(2), 195–202.

https://doi.org/10.1016/s0028-3932(02)00149-5

Calder, A. J., Keane, J., Young, A. W., & Dean, M. (2000). Configural information in facial perception expression. Journal of Experimental Psychology: Human Perception and Performance, 26(2), 527–551. https://doi.org/10.1037//0096-1523.26.2.527

Carbon, C. (2020). Wearing face masks strongly confuses counterparts in reading emotions. Frontiers in Psychology, 25, Article 566886. https://doi.org/10.3389/fpsyg.2020.566886

Cartaud, A., Quesque, F., & Coello, Y. (2020). Wearing a facemask against Covid-19 results in a reduction of social distancing. PLoS ONE 15(12), Article e0243023. https://doi.org/10.1371/journal.pone.0243023

CDC Newsroom. (2020, July 14). CDC calls on Americans to wear masks to prevent COVID-19 spread. U.S. Department of Health and Human Services.

https://www.cdc.gov/media/releases/2020/p0714-americans-to-wear-masks.html

Chen, M., & Bargh, J. A. (1999). Consequences of automatic evaluation: Immediate behavioral predispositions to approach or avoid the stimulus. Personality and Social Psychology Bulletin, 25(2), 215–224.

https://doi.org/10.1177/0146167299025002007

Circelli, K. S., Clark, U. S., & Cronin-Golomb, A. (2013). Visual scanning patterns and executive function in relation to facial emotion recognition in aging. Aging, Neuropsychology, and Cognition, 20(2), 148–173.

https://doi.org/10.1080%2F13825585.2012.675427

Ebner, N. C., Riediger, M., & Lindenberger, U. (2010). FACES-- A database of facial expressions in young, middle-aged, and older women and men: Development and validation. Behavior Research Methods, 42(351–362). https://doi.org/10.3758/BRM.42.1.351

Fischer, A. H., Gillebaart, M., Rotteveel, M., Becker, D., & Vliek, M. (2012). Veiled emotions: The effect of covered faces on emotion perception and attitudes. Social Psychological and Personality Science, 3(3), 266–273. https://doi.org/10.1177/1948550611418534

Gonçalves, A., Fernandes, C. Pasion, R., Ferreira-Santos, F., Barbosa, F., & Marques-Teixeira, J. (2018). Effects of age on the identification of emotions in facial expressions: A meta-analysis. PeerJ, 6, Article e5278. https://doi.org/10.7717/peerj.5278

Gosselin, F., & Schyns, P. G. (2001). Bubbles: A technique to reveal the use of information in recognition tasks. Vision Research, 41(17), 2261–2271. https://doi.org/10.1016/s0042-6989(01)00097-9

Grundmann, F., Epstude, K., & Scheibe, S. (2021). Face masks reduce emotionrecognition accuracy and perceived closeness. PLoS ONE, 16(4), Article e0249792. https://doi.org/10.1371/journal.pone.0249792

Guarnera, M., Hichy, Z., Cascio, M., Carrubba, S., & Buccheri, S. L. (2017). Facial expressions and the ability to recognize emotions from the eyes or mouth: A comparison between children and adults. The Journal of Genetic Psychology, 178(6), 309–318. https://doi.org/10.1080/00221325.2017.1361377

Hareli, S., David, S., & Hess, U. (2013). Competent and warm but unemotional: The influence of occupational stereotypes on the attribution of emotions. Journal of Nonverbal Behavior, 37(4), 307–317. https://doi.org/10.1007/s10919-013-0157-x

Kapwing. (2020). #MaskOn your profile picture for COVID-19 safety. https://www.kapwing.com/explore/maskon-your-profile-picture-forcovid19-safety

Kastendieck, T., Zillmer, S., & Hess, U. (2021). (Un)mask yourself! Effects of face masks on facial mimicry and emotion perception during the COVID-19 pandemic. Cognition and Emotion, 36(1), 59–69. https://doi.org/10.1080/02699931.2021.1950639

Marini, M., Ansani, A., Paglieri, F., Caruana, F., & Viola, M. (2021). The impact of facemasks on emotion recognition, trust attribution and re-identification. Scientific Reports, 11, Article 5577. https://doi.org/10.1038/s41598-021-84806-5

Merriam-Webster (n.d.). Merriam-Webster.com thesaurus. Retrieved March 15,

2022, from https://www.merriam-webster.com/thesaurus/ Oosterhof, N. N., & Todorov, A. (2009). Shared perceptual basis of emotional expressions and trustworthiness impressions from faces. Emotion, 9(1), 128–133. https://doi.org/10.1037/a0014520

Renard, S. B., deJong, P. J., & Pijnenborg, G. H. M. (2016). Approach–avoidance of facial affect is moderated by the presence of an observer-irrelevant trigger. Motivation and Emotion, 41(2), 265–272. https://doi.org/10.1007/s11031-016-9595-1

Roberson, D., Kikutani, M., Döge, P., Whitaker, L., & Majid, A. (2012). Shades of emotion: What the addition of sunglasses or masks to faces reveals about the development of facial expression processing. Cognition, 125(2), 195–206. https://doi.org/10.1016/j.cognition.2012.06.018

Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), 1–28. https://doi.org/10.1037/h0092976

Ruffman, T., Henry, J. D., Livingstone, V., & Phillips, L. H. (2008). A metaanalytic review of emotion recognition and aging: Implications for neuropsychological models of aging. Neuroscience and Biobehavioural Reviews, 32(4), 863–881. https://doi.org/10.1016/j.neubiorev.2008.01.001

Schmidt, K. L., & Cohn, J. F. (2001). Human facial expressions as adaptations:

Dove and Deka | Mask and Emotional Intensity

Evolutionary questions in facial expression research. Yearbook of Physical Anthropology, 44(1), 3–24. https://doi.org/10.1002/ajpa.20001

Schurgin, M. W., Nelson, J., Iida, S., Ohira, H., Chiao, J. Y., & Franconeri, S. L. (2014). Eye movements during emotion recognition in faces. Journal of Vision, 14(13), 1–16. https://doi.org/10.1167/14.13.14

Sullivan, S., Ruffman, T., & Hutton, S. B. (2007). Age differences in emotion recognition skills and the visual scanning of emotion faces. Journal of Gerontology, 62(1), 53–60. https://doi.org/10.1093/geronb/62.1.p53

Wong, B., Cronin-Golomb, A., & Neargardner, S. (2005). Patterns of visual scanning as predictors of emotion identification in normal aging. Neuropsychology, 19(6), 739–749. https://doi.org/10.1037/0894-4105.19.6.739

Author Note

Teddi S. Deka https://orcid.org/0000­0002­0870­1755 There are no known conflicts of interest. Correspondence concerning this article should be addressed to Teddi S. Deka, Missouri Western State University, 4525 Downs Drive, Saint Joseph, MO 64507. Email: deka@missouriwestern.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

When Hints Hurt Memory: The Influence of the Number of Part-Set Cues on Free Recall

Deniz Akpinar1, Vaughan Bamford2, Sara Martinez Guzman3, Tracey Nassuna1, Madison Stevens3, and Matthew R. Kelley3*

1Neuroscience Program, Lake Forest College

2Department of Biology, Lake Forest College

3Department of Psychology, Lake Forest College

ABSTRACT. The present study explored the effects of part­set cues on retention in a free recall task across 5 experiments. In the part­set cueing literature, researchers typically provide half of the to­be­remembered items as cues at test; accordingly, little is known about the effect of the number of part­cues on retention, particularly when very few or very many cues are presented at test. These experiments examined the effects of very few cues (2, 3, 4, 5, and 6 cues), half cues (15), and very many cues (25, 27, 28, and 29 cues) out of a list of 30 words. Put differently, the percentage of part­set cues in the very few condition ranged from 7–20% of the list, the half­cue condition was 50% of the list, and the very many cue condition ranged from 83–97% of the list. The very few cue conditions demonstrated null effects (all ps > .26), which suggests that there might be a minimum percentage of cues required before part­set cueing influences memory performance. In contrast, significant part­set cueing impairment occurred in all conditions where at least half of the cues were present (all ps < .03, except one marginally significant, p = .06, when comparing the 0 vs. 27 cue conditions). Generally, these results are consistent with predictions derived from the retrieval strategy disruption hypothesis (Basden & Basden, 1995). Furthermore, the patterns of results reported in the half­cue and very many cue conditions were consistent with similar conditions in Slamecka (1968), but the results of these two studies differed when few cues were present.

Keywords: part­set cueing, part­list cueing, free recall, strategy disruption, hints

Imagine that the students in an introductory psychology course have been asked to recall Schachter’s (2001) seven sins of memory on a test. All students receive a free recall prompt where they can write the sins down in any order, but half of the students receive three of the sins as hints (i.e., “To help you, three sins are: blocking, bias, and persistence. Now recall the remaining four.”), whereas the others receive no hints and must recall all seven sins. Which group of students will perform better on this free recall test—those with hints or those without? If you were a student in this course, would you rather have hints or no hints? Although one might intuit that hints should be helpful, research on part­set cueing has shown that, depending on the item material, encoding and testing conditions, such hints (i.e., part­set cues), often impair memory performance, although there are situations

where these hints can either facilitate or have no effect on performance (Basden, 1973; Bäuml & Aslan, 2006; Bäuml & Samenieh, 2012; Nickerson, 1984).

In the laboratory, the typical part ­ set cueing procedure is similar to that outlined in the example above: participants are given a set of information to remember (e.g., a list of words) and then they are asked to remember that information either in the presence or in the absence of part­set cues (Nickerson, 1984; Roediger, 1973; Slamecka, 1968). As indicated by the name, part­set cues are part of the set of information that the participants are asked to remember (usually half of the items), which tend to be chosen either randomly or systematically (e.g., alternating items) and then are provided at test. Historically, free recall tests have been the most common retention measures employed in the part­set cueing literature (see Nickerson, 1984, for

a review) and the most common result with free recall tests is the counterintuitive finding of part­set cueing impairment (previously termed inhibition), which reflects poorer performance following a cued test relative to an uncued test.

In contrast, the more intuitive result of part­set cueing facilitation—where hints improve performance—have been demonstrated when memory for order or location information are tested, specifically with tests of serial recall, reconstruction of order, and reconstruction of spatial location (Basden et al., 2002; Cole et al., 2013; Serra & Nairne, 2000). On these tests, when the part­set cues are placed in their original positions or locations, participants perform significantly better in the presence of part­set cues than in their absence (Drinkwater et al., 2006; Fritz & Morris, 2015; Kelley & Bovee, 2007; Kelley et al., 2016; Kelley & Parihar, 2018; Kelley et al., 2014).

Arguably the most robust theoretical explanation of part­set cueing comes in the form of the retrieval strategy disruption (RSD) hypothesis (Basden et al., 2002; Basden & Basden, 1995; Basden et al., 1977). The RSD hypothesis suggests that, when people encode a set of information, they will develop individual retrieval plans for use on the upcoming test (Basden & Basden, 1995; Basden et al., 1977). Part­set cueing impairment should occur when the experimenter provides part­set cues in a manner that is incongruent with the individual’s retrieval plan, which will disrupt the individual’s strategy and lead to suboptimal performance. In contrast, when no cues are provided, individuals can use their retrieval plan optimally and there will be no impairment. Furthermore, on order and location tasks, part­set cueing facilitation should occur because these part­set cues tend to be presented in a manner that is consistent with an individual’s encoded retrieval strategy (e.g., the person tried to learn the items in sequence and then received the correct items in their correct orders/positions). In other words, because the cues are congruent with the strategy, the individuals can perform optimally and performance will improve (Basden et al., 2002).

As indicated earlier, researchers typically provide half of the to­be­remembered stimuli as part­set cues, although there are two notable exceptions to this. Recently, Kelley et al. (2021) systematically examined the influence of the number of part­set cues on reconstruction of order and serial recall tests across seven experiments. On reconstruction of order tests for 16­item lists, they reported that part­set cueing facilitation was present when at least half of the list items were presented as part­set cues and the magnitude of facilitation tended to increase as the number of cues increased. However, when only a few part­set cues were presented (i.e., when 10­30% of the list was provided as cues), there was no

effect of cueing on the reconstruction tests. In contrast, on serial recall tests with 10­item lists, part­set cueing facilitation was only evident in just two of the “few cues” conditions (i.e., when 30–40% of the list was provided as cues). Taken together, these results are interesting because they showed that there are limits to the facilitative effect of part­set cues on order tests.

The other exception came from Slamecka’s (1968) original investigation of part­set cueing using free recall, where participants were asked to remember a 30­word list, either in the presence of 5, 15, 25, or 29 cues (i.e., 17%, 50%, 83%, or 97%, of the list was provided as cues). Across two experiments, Slamecka reported that the magnitude of part­set cueing impairment in free recall appeared to increase as the number of provided part­set cues increases (15%, 20%, 22%, 30–35% for 5, 15, 25, and 29 cues, respectively). Although suggestive, the generality of these results is limited because only four cue­number sets were used and the original article was light on key methodological, scoring, and statistical detail. As such, many questions have remained. Will Slamecka’s results replicate? What will happen when there are fewer than five cues out of 30 (i.e., less than 17% of the list provided as cues)? Are fewer cues sufficient to produce part­set cueing impairment or are a certain number of cues required before impairment becomes evident? Is the magnitude of part­set cueing impairment relatively stable across most cue­number sets, as suggested by Slamecka’s results with 15 and 25 cues (i.e., 50–83% of the list)? Was the particularly strong part­set cueing impairment in the 29­cue condition (i.e., 97% of the list) a special case or does the magnitude of impairment increase regularly from 25–29 items (out of 30)? The goal of the present article is to enhance our understanding of the influence of the number of part­set cues on free recall tasks by systemically varying more cue­number sets.

Experiments 1–5

Over the course of five experiments, participants viewed lists of 30 words and completed free recall tests with no cues, very few cues (2–6 cues; or 7–20% of the list), half cues (15; or 50% of the list), or very many cues (25–29; or 83–97% of the list). With regard to the questions posed above, one can make predictions based on past data from Kelley et al. (2021) and Slamecka (1968) and by incorporating the principles of the RSD hypothesis. For instance, one might imagine that there exists a minimum number of cues required before part­set cueing impairment becomes evident. After all, in terms of the RSD hypothesis, the presence of only two or three cues (or 7–10% of the list) might not be particularly disruptive to a person’s strategy. Indeed, experimental evidence from Kelley et al. (2021) showed that part­set cueing did not

Akpinar, Bamford, Guzman, Nassuna, Stevens, and Kelley | When Hints Hurt Memory

exert an influence on reconstruction tests until at least half of the list was presented as cues.

One also might predict relative stability in the magnitude of part­set cueing impairment for most cuenumber sets given that only a 2% difference in magnitude separated the 15­ and 25­cue and only a 5% difference separated the 5 ­ and 15 ­ cue conditions in Slamecka (1968). Finally, one might expect that the most extreme number of cues (29 of 30; or 97% of the list) might lead to the largest magnitude of part­set cueing impairment based on both Slamecka’s (1968) result and Kelley et al.’s (2021) finding with reconstruction that the condition with the largest number of cues (14 of 16; or 88% of the list) yielded stronger facilitation than the condition with the penultimate number of cues (13 of 16; or 81% of the list). In the end, these are empirical questions that were addressed across the following five experiments.

Method

Participants

Based on the effect sizes of past part­set cueing research with reconstruction of order tasks in the lab, a sample size calculator provided an estimated sample size that ranged between 50 and 70 participants. Given the counterbalances required, in each of Experiments 1–5, we recruited 54 distinct participants through Prolific at a fair market rate of $9/hr for their participation (270 unique participants in all). All Prolific participants were from the United States; no further demographic information was collected, which represents a limitation of this study that should be considered in future research.

We chose to use Prolific because, due to the COVID19 pandemic, online experimentation was our sole option for data collection. Given our small college budget, we could not afford to add demographic screeners to restrict our sample. Although we intended to add demographic questions to our experiment, in our haste, we failed to do so. We acknowledge that this limits the generalizability of our study and does not allow us to examine whether demographic characteristics influenced the results.

Materials and Design

In Experiments 1–5, participants viewed three lists of 30 words and received a free recall task immediately following each list. In each experiment, the within­subjects independent variable was the cue condition during the free recall task and there were three levels per experiment: Experiment 1 (0 cues, 2 cues, and 27 cues), Experiment 2 (0 cues, 3 cues and 28 cues), Experiment 3 (0 cues, 4 cues, or 29 cues), Experiment 4 (0 cues, 5 cues, or 15 cues), and Experiment 5 (0 cues, 6 cues, or 25 cues). In each of the three conditions within each experiment, participants

were shown a series of words on a computer screen and were asked to type as many words as they could recall in a blank box. On cued trials, the cues were provided above the blank box.

Ninety nouns were drawn from the Paivio et al. (1968) norms and were matched on imageability (M = 6.27, SD = 0.17), concreteness (M = 6.59, SD = 0.39), meaningfulness (M = 6.71, SD = 0.78), and word length ( M = 6.43, SD = 0.48). Each experiment had different randomized lists of these 90 nouns. Across the experiments, the cues were pseudo­randomly placed in the following positions: 2 cues (positions 5, 26), 3 cues (positions 6, 20, 26), 4 cues (positions 5, 10, 21, 26), 5 cues (positions 5, 9, 15, 21, 25), 6 cues (positions 5, 9, 13, 20, 23, 26), 15 cues (odd positions), 25 cues (all except positions 4, 7, 22, 25, 27), 27 cues (all except positions 7, 23, 27), 28 cues (all except positions 8, 22), and 29 cues (all except position 23).

Within each experiment, the three lists were always displayed in the same order (L1, L2, L3) and the words within the list always appeared in the same order. Using Latin squares, the cue conditions were counterbalanced such that across participants each cue condition appeared equally often with each list. The experiments were programmed and administered with Qualtrics.

Procedure

Institutional review board approval was received prior to data collection. At the beginning of each experiment, participants were randomly assigned to one of the three counterbalanced orders of cue conditions. After consenting to participate, the participants were instructed that there would be three study­test trials, each of which included presentation of the list (word by word, in the center of the screen, for 3 seconds each) and a free recall task (type recalled words into the blank box in any order). In other words, in immediate succession, participants viewed list one (90­seconds), took test one, viewed list two (90­seconds), took test two, viewed list three (90­seconds), and took test three. They also were informed that, on some trials, some of the words would

Statistics for the Experiments 1–5

When Hints Hurt Memory | Akpinar, Bamford, Guzman, Nassuna, Stevens, and Kelley

be given back as cues. They were told that the words would appear above the blank box and that the participants should read the cues prior to starting the free recall test. The experiment was self­paced and participants typically required 10–15 minutes to finish the entire task. Upon completion of the study, participants were debriefed and thanked for their participation.

Data Analysis

To be counted as correct, the items from word lists must have been typed into the blank box in any random order; typos were allowed, provided that the intended word’s identity was still clear and obvious to two raters. Each participant yielded a proportion correct for each of the three cue conditions. To calculate the proportion correct, we used the following equation: [number of words recalled] / [number of words on list – number of cues]. To illustrate, if a person correctly recalled 10 words from the list of 30 words when four cues were presented, the proportion correct would be .38 (i.e., 10 / [30­4] = 10 / 26 = .38).

Results

Separate one ­ factor repeated measures analyses of variance (ANOVAs) with Greenhouse ­ Geisser corrections were performed for Experiments 2–5, but the Greenhouse­Geisser correction was not required in Experiment 1. All five ANOVAs were statistically significant (see Table 1 for the F, p, and ηp2 values) and, accordingly, were followed by Sidak post­hoc comparisons. Figure 1 displays a cross­experiment comparison, where the mean proportion correct for each cue­number condition in each experiment is depicted together in a single graph.

Figure 1 clearly illustrates that there were null effects

of part­set cueing in all the very few cue conditions (2, 3, 4, 5, and 6 cues; or 7–20% of list) compared to their corresponding no cue conditions, with all ps > .26. These results suggest that there exists a minimum number (or percentage) of cues required before part­set cueing impairment becomes evident. Given that six (of 30) cues represent 20% of the list, the present experiments seem to suggest that more than 20% of the list must be presented as part­set cues before performance will be impaired by those cues. Despite the consistency of these results, we must keep in mind, however, that these findings deviate from those reported in Slamecka’s 5­cue condition (i.e., 17% of the list), which showed part­set cueing impairment of 15% relative to the no cue condition.

Figure 1 also clearly displays that the very many cue conditions (25, 27, 28, 29; or 83 ­ 97% of the list) showed significant part ­ set cueing impairment compared to their corresponding no­cue conditions, with all ps < .03 except for the marginally significant 0 vs. 27 comparison (p = .06). The 15­cue condition (50% of the list) also showed significant part­set cueing impairment ( p = .01). Generally, this pattern of results is similar to the pattern reported by Slamecka (1968), although the magnitude of impairment was much higher in Slamecka’s study (15%, 20%, 22%, and 30–35% in the 5, 15, 25, and 29 cue conditions) as compared to the current study (7%, 10%, 10%, 13%, and 16% in the 15, 25, 27, 28, and 29 cue conditions). Nominally, both 15­cue conditions yielded impairment of the smallest magnitude and the 29­cue conditions yielded impairment of the highest magnitude, with the intermediate numbers of cues producing similar levels of impairment to one another.

Akpinar, Bamford, Guzman, Nassuna, Stevens, and Kelley | When Hints Hurt Memory

FIGURE 3

Mean Proportion Correct as a Function of Cue-Number Condition (Expressed as Number of Cues and Percentage of Cues) for Experiments 1 Through 5 Together.

Discussion

The primary goal of these five experiments was to further explore the influence of the number of part­set cues on free recall tests, in a manner similar to how Kelley et al. (2021) systematically examined the influence of the number of part­set cues on serial recall and reconstruction of order tests. These experiments also allowed us to conceptually replicate and extend Slamecka’s (1968) original investigation on part­set cueing, in which he reported part­set cueing impairment for a 30­item word list when 5, 15, 25, and 29 cues (or 17%, 50%, 83%, and 97% of the list) were provided at test. Of particular interest were whether very few part­set cues (2, 3, 4, 5, 6; or 20% of the list or less) would lead to impairment or whether a minimum proportion of cues were required before effects of part­set cueing would become evident. Also, of interest, was whether the magnitude of part­set cueing impairment would change as the number of cues increased, especially with very many cues (15, 25, 27, 28, 29; or 50–97% of the list).

With respect to the very few cue questions of interest, we used the RSD hypothesis to reason that the presence of very few cues (or a small proportion of cues) might provide only minimal disruption to an individual’s retrieval strategy, which in turn could reduce or even eliminate part­set cueing impairment. Indeed, using reconstruction of order tests, Kelley et al. (2021) showed that part­set cueing did not exert an influence until at least half of the list was presented as cues. That said, with a free recall test, Slamecka (1968) showed significant part­set cueing impairment with five part­set cues (out of 30; or 17% of the list). As indicated earlier, Figure 1 shows that there were no significant effects of part­set cueing for any of the very few”cue conditions (2, 3, 4, 5, and 6 cues; or 7–20%) when compared to their respective no cue conditions. This suggests that more than 20% of the list must be presented as part­set cues before part­set cueing impairment manifests itself. Although it is difficult to explain the discrepancy in the 5­cue condition results between the present study (no impairment) and Slamecka’s study (impairment), we must note that Slamecka never replicated his study, whereas the current experiments essentially provide five replications of the very few cue conditions, each with a slightly different number of cues but all with very similar results.

With respect to the very many cue questions of interest, we reasoned that we might see relative stability in the magnitude of part­set cueing impairment for most cue­number sets because the differences in magnitude reported by Slamecka (1968) for the 15­ and 25 cue conditions was only 2% and only 5% for the 5 ­ and 15­cue conditions. We also reasoned those conditions

with the most extreme number of cues (29 of 30; or 97% of the list) might lead to the largest magnitude of part­set cueing impairment because both Slamecka’s (1968) and Kelley et al.’s (2021) reported the strongest effects of part­set cueing with the most extreme numbers of cues. As noted earlier, Figure 1 demonstrates that the half cue (15; or 50% of the list) and all the very many cue conditions (25, 27, 28, 29; or 83­97% of the list) showed significant part­set cueing impairment. Although the overall level of impairment was lower in the present experiments than in Slamecka’s (1968) study, both showed a similar pattern of results (there was only a 3–5% difference between 15, 25, and 27 cues, which shows relative stability in the magnitude of impairment; the most extreme number of cues yielded the largest magnitude of impairment). It appears that, once a critical mass of part­set cues is provided, an individual’s strategy will be sufficiently disrupted and performance will suffer. When considering why the most extreme number of part­set cues might be more disruptive than other large sets of cues, it might be helpful to imagine what confronts individuals in these situations. In the current situation, they viewed a list of 30 words, were given 29 of those words back, and had to recall the one missing word; that is a rather restrictive search set, with only one opportunity to be correct, and in the context of presumably massive disruption from the cues.

Although the present experiments helped to clarify the influence of the number of part­set cues on free recall, some questions remain and future research is required. For instance, what percentage of the list must be provided as part­set cues before we see impairment? The present data suggest that the answer lies between 20% and 50% for a 30­item list, but where is the tipping point? Also, is that tipping point (percentage) consistent across different list lengths (e.g., 20­item list vs. 40­item list)? Further, it might be interesting to examine part­set cueing in more of a real­world context, as described in the opening vignette. Imagine, for instance, a wellmeaning teacher who provides hints on an exam; it certainly would be important to know when those hints might start to do harm. Fortunately, the present data suggests that small numbers or percentages of hints are unlikely to impair memory.

Could we learn anything by collecting qualitative descriptions from participants regarding their strategy usage both in the presence and absence of cues? With such data, we might gain insight into what it feels like when participants are trying to recall information when hints are and are not available. Indeed, the RSD hypothesis is particularly vague about what specific strategies will use when attempting to retrieve information on these various tests. To make matters more complicated,

When Hints Hurt Memory | Akpinar, Bamford, Guzman, Nassuna, Stevens, and Kelley

participants might routinely change and adapt their strategies across situations. In any case, such explorations could provide important insights into the effects of partset cueing on memory performance.

Finally, as noted earlier, the generalizability of our results might be limited because we did not collect demographic information from our samples. It is entirely possible that we might see differences in performance as a function of different demographic variables. The field could also benefit from a direct replication of all these experiments, but with the cues in different positions to further enhance the generalizability of these experiments. Despite these limitations, we are comforted by facts that these results are generally consistent with the predictions of the RSD hypothesis (Basden & Basden, 1995), most of Slamecka’s (1968) results, and Kelley et al.’s (2021) findings that very few and very many cues can impact memory performance differently. Indeed, there seems to be a minimum number (or percentage) of cues that are required before part­set cueing impairment manifests itself and, once that threshold is met, one should expect significant part­set cueing impairment going forward.

References

Basden, B. H., Basden, D. R., & Stephens, J. P. (2002). Part-set cueing of order information in recall tests. Journal of Memory and Language, 47(4), 517–529. https://doi.org/10.1016/S0749-596X(02)00016-5

Basden, D. R. (1973). Cued and uncued free recall of unrelated words following interpolated learning. Journal of Experimental Psychology, 98(2), 429–431. https://doi.org/10.1037/h0034367

Basden, D. R., & Basden, B. H. (1995). Some tests of the strategy disruption interpretation of part-list cueing inhibition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(6), 1656–1669. https://doi.org/10.1037/0278-7393.21.6.1656

Basden, D. R., Basden, B. H., & Galloway, B. C. (1977). Inhibition with part-list cuing: Some tests of the item strength hypothesis. Journal of Experimental Psychology: Human Learning and Memory, 3(1), 100–108. https://doi.org/10.1037/0278-7393.3.1.100

Bäuml, K. H., & Aslan, A. (2006). Part-list cuing can be transient and lasting: The role of encoding. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32(1), 33–43. https://doi.org/10.1037/0278-7393.32.1.33

Bäuml, K. T., & Samenieh, A. (2012). Influences of part-list cueing on different

forms of episodic forgetting. Journal of Experimental Psychology: Learning, Memory, and Cognition, 38(2), 366–375. https://doi.org/10.1037/a0025367

Cole, S. M., Reysen, M. B., & Kelley, M. R. (2013). Part-set cueing facilitation for spatial information. Journal of Experimental Psychology: Learning, Memory, and Cognition, 39(5), 1615–1620. https://doi.org/10.1037/a0032424

Drinkwater, K., Dagnall, N., & Parker, A. (2006). Effects of part-set cuing on experienced and novice chess players’ reconstruction of a typical chess midgame position. Perceptual and Motor Skills, 102(3), 645–653. https://doi.org/10.2466/PMS.102.3.645-653

Fritz, C. O., & Morris, P. E. (2015). Part‐set cuing of texts, scenes, and matrices. British Journal of Psychology, 106(1), 1–21. https://doi.org/10.1111/bjop.12058

Kelley, M. R., & Bovee, J. C. (2007). Part-set cuing and memory for order. In A. Columbus (Ed.), Advances in Psychology Research. (Vol. 51, pp. 133–148). Nova Science Publishers.

Kelley, M. R., & Parihar, S. A. (2018). Part-set cueing impairment and facilitation in semantic memory. Memory, 26(7), 1008–1018. https://doi.org/10.1080/09658211.2018.1428993

Kelley, M. R., Parasiuk, Y., Salgado-Benz, J., & Crocco, M. (2016). Spatial part-set cueing facilitation. Memory, 24(6), 737–745. https://doi.org/10.1080/09658211.2015.1046382

Kelley, M. R., Pentz, C., & Reysen, M. B. (2014). The joint influence of collaboration and part-set cueing. The Quarterly Journal of Experimental Psychology, 67(10), 1977–1985. https://doi.org/10.1080/17470218.2014.881405

Kelley, M. R., Strejc, M., Walts, Z. L., Uribe, Y., Gonzales, L., Tcaturian, E., Keller, A. E., Bronswick, J. K., & Stephany, S. E. (2021). The influence of the number of part-set cues on order retention. Quarterly Journal of Experimental Psychology, 74(5), 928–943. https://doi.org/10.1177/1747021820977047

Nickerson, R. S. (1984). Retrieval inhibition from part-set cueing: A persisting enigma in memory research. Memory & Cognition, 12(6), 531–552. https://doi.org/10.3758/BF03213342

Paivio, A., Yuille, J. C., & Madigan, S. A. (1968). Concreteness, imagery, and meaningfulness values for 925 nouns. Journal of Experimental Psychology, 76(1, Pt.2), 1–25. https://doi.org/10.1037/h0025327

Roediger, H. L., III. (1973). Inhibition in recall from cueing with recall targets. Journal of Verbal Learning and Verbal Behavior, 12(6), 644–657. https://doi.org/10.1016/S0022-5371(73)80044-1

Schacter, D. L. (2001). The seven sins of memory: How the mind forgets and remembers. Houghton, Mifflin and Company.

Serra, M., & Nairne, J. S. (2000). Part-set cueing of order information: Implications for associative theories of serial order memory. Memory & Cognition, 28(5), 847–855. https://doi.org/10.3758/bf03198420

Slamecka, N. J. (1968). An examination of trace storage in free recall. Journal of Experimental Psychology, 76(4, Pt. 1) 504–513. https://doi.org/10.1037/h0025695

Author Note

Matthew R. Kelley https://orcid.org/0000­0003­2893­8706

We have no known conflicts of interest to disclose. Correspondence concerning this article should be addressed to Matthew Kelley, Department of Psychology, Lake Forest College, Lake Forest, IL 60035. Email: kelley@lakeforest.edu

Akpinar, Bamford, Guzman, Nassuna, Stevens, and Kelley | When Hints Hurt Memory

Psychological

and Behavioral

Predictors of Procrastination in Undergraduates

Bradley B. Gregory1*, Megan Golson2 , and McKay Larsen2

1 Department of Psychology, North Greenville University

2 Department of Psychology, Southern Utah University

ABSTRACT. Utilizing previous work in personality theory, implicit theory of intelligence, goal orientation, and self­efficacy theory, we conducted this exploratory study to identify predictors of general procrastination tendencies among undergraduates. We analyzed a sample of 267 undergraduate students from introductory psychology courses at a public rural university. A standard multiple regression analysis using IBM SPSS Statistics was performed on 16 psychological variables and 1 behavioral variable to identify presence of predictive influence on the dependent variable of procrastination as measured by Lay (1986) for college student populations. Regression analysis revealed that the model achieved significance at predicting procrastination in undergraduate students, F(17, 249) = 14.73, p < .001, r² = .50, adj r² = .47. The resultant model identified 4 significant predictors and 9 additional significant correlates, 7 of which were significant at p < .01. Positive predictors included growth mindset beliefs (β = .16, p = .003) and academic entitlement beliefs (β = .12, p = .023). Negative predictors included conscientiousness (β = –.55, p < .001) and college student efficacy beliefs (β = –.17, p = .011). These findings are consistent with previous work and further support the roles and directional influences of conscientiousness, college student efficacy beliefs, and implicit theory of intelligence beliefs on procrastination, and add to the growing literature on academic entitlement beliefs.

Keywords: procrastination, conscientiousness, efficacy beliefs, growth mindset, academic entitlement beliefs

Procrastination is defined as “the act of needlessly delaying tasks to the point of experiencing subjective discomfort” (Solomon & Rothblum, 1984, p. 503). Most individuals who engage in procrastination wish to reduce this behavior in themselves (Knaus, 1998; Solomon & Rothblum, 1984) as many understand it to be a time waster associated with harmful consequences (Beck et al., 2012; Blatt & Quinlan, 1967; Lay & Schouwenburg, 1993; Przepiorka et al., 2016; Wolters, 2003).

As defined, procrastination remains widespread among college students (Blunt & Pychyl, 1998; Reinecke et al., 2018), with past indications that an estimated 80 to 95% of college students reported engaging in procrastination (O’Brien, 2002) and near 50% indicated consistent and problematic procrastination (Day et al., 2000; Rozental et al., 2015). Historically, procrastination has been a prevalent and serious problem among college students (Blatt & Quinlan, 1967; Blunt & Pychyl, 1998;

Ellis & Knaus, 1977; Green, 1982; Pychyl, Morin, et al., 2000; Rabin et al., 2011), with more recent research indicating this behavioral phenomenon may be on the rise (Hinsch & Sheldon, 2013; Kachgal et al., 2001).

Within the college student population, academic procrastination, which positively correlates with general procrastination (Sirin, 2011), is characterized as intentionally delaying the completion of academic­related tasks (Burns et al., 2000). This form of procrastination has been reported in roughly 70% of college students (Schouwenburg, 1995), and is associated with pathological perfectionism (Burns et al., 2000), depression (Solomon & Rothblum, 1984), higher rates of course withdrawals (Semb et al., 1979), missing deadlines (Wolters, 2003), low course grades (Rothblum et al., 1986; Wesley, 1994), guilt (Pychyl, Lee, et al., 2000), anxiety (Rothblum et al., 1986), neuroticism (Watson, 2001), low self­esteem (Ferrari, 2000), and poor academic performance overall (Steel et al., 2001). Additional

research has revealed that academic performance in particular can suffer when the individual is under stress (Tice & Baumeister, 1997). Given that 60% of college students report a desire to decrease their academic procrastination (Solomon & Rothblum, 1984), this highlights their understanding that outcomes associated with procrastination can negatively affect their lives through some influence on behavioral, emotional, or academic functioning (Fee & Tangney, 2012). Despite numerous studies that have explored the consequences of procrastination, a disproportionate share has failed to explore predictors and causes of procrastination (Katz et al., 2014).

Meta-Analytic Correlates and Causes of Procrastination

In her meta­analysis of procrastination across multiple settings, van Eerde (2003) assessed the correlations from 121 studies that examined relationships between procrastination and personality, motivation, affect, and performance variables. Her findings revealed that the largest negative effect sizes were associated with the personality trait of conscientiousness and self­efficacy beliefs, with the largest positive effect size associated with self­handicapping. The effect size associated with the motivation of perfectionism, although significant, was small.

A subsequent meta ­ analysis by Steel (2007), informed by van Erde’s review (2003), sought to review the considerable amount of empirical work done on procrastination dating back to the 1930s. This newer research aimed to review and ultimately summarize the relevant conceptual, theoretical, and empirical work, which was collectively developed from correlational, experimental, and qualitative findings. Steel (2007) further confirmed several key findings of van Eerde (2003) and thus provided a new way to conceptualize the correlates and predictors of procrastination by dividing them into separate categories: task characteristics, individual differences, demographics, and a fourth category he labeled “outcomes,” which he describes “indicate the proximal effects of procrastination” (Steel, 2007, p. 67). Findings from these works provided the foundation for the following literature review.

There are individual differences in proneness to procrastination (Bridges & Roig, 1997). Trait/domain theory by Costa and McCrae (1992) describes the relationships between personality correlates and procrastination to explain behavioral tendencies. Some researchers have utilized the traits themselves according to the five­factor model to examine these correlations (Digman, 1990), whereas others have instead chosen to focus on specific facets within traits. The latter approach has led to some degree of uncertainty as no general consensus seems to exist at the facet level within the

field of personality theory (John & Sanjay, 1999). For the purposes of the current exploratory study, we chose to assess predictors of procrastination using the traditional five­factor model traits, whose respective hypothesized outcomes are discussed below.

Across multiple studies, the strong correlation between conscientiousness and procrastination has been reliably documented in the literature (Johnson & Bloom, 1995; Lay, 1997; O’Connor & Paunonen, 2007). Defined as “socially prescribed impulse control that facilitates task­ and goal­directed behavior, such as thinking before acting, delaying gratification, following norms and rules, and planning, organizing, and prioritizing tasks” (John et al., 2008, p. 120), conscientiousness has emerged as the most essential source trait of procrastination (Lay, 1997; van Eerde, 2003). Specifically, it has been shown that procrastination represents, conceptually, low conscientiousness and a failure to self­regulate (Park & Sperling, 2012; Steel, 2007), as well as a tendency toward heightened distractibility (Grunschel et al., 2013; Steel, 2007). Conversely, it has been shown that possessing a combination of high standards and high conscientiousness is correlated with lower levels of procrastination (Frost et al., 1990). Dewitte and Schouwenburg (2002) claim conscientiousness “explains the lion’s share of the variation in procrastination items” (p. 470) regardless of measure used to assess procrastination, a claim supported by the resultant effect size from Steel’s meta­analysis (2007), further validating the conclusions of van Eerde (2003).

“Outcomes” emerged as another summary category described by Steel (2007), in which he included mood and performance. High conscientiousness is positively correlated with performance in occupational (Hurtz & Donovan, 2000) as well as educational tasks (Barrick & Mount, 2003; van Eerde, 2003). Thus, individuals who procrastinate should tend to suffer both in terms of how they feel as well as what they personally achieve. The construct of mood shares some overlap with the construct of neuroticism, which by nature includes aspects of both anxiety and depression (Weinstock & Whisman, 2006). Individuals high in neuroticism may be prone to procrastinating due to the presence of depression among these individuals. Procrastination, which tends to negatively affect performance, can subsequently further suppress mood, blurring a possible cause­effect relationship as a poor mood may not only be the result of procrastination, but also a cause (Steel, 2007). The latter has been supported by work in which negative mood emerged as an important precursor to procrastination (Wohl et al., 2010).

Additional recent work on emotional regulation has shown that efforts to improve short­term mood are often prioritized over accomplishing long­term goals,

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

evidencing a failure to self­regulate which ultimately manifests as procrastination (Sirois & Pychyl, 2013). Furthermore, individuals in poorer moods tend to indicate that they procrastinate, even if their actual behavior does not support their claims of procrastination (Steel et al., 2001). Steel attempts to explain this paradoxical finding by hypothesizing that self­reported procrastination may include a self­assessment shaped by actual behavior as well as a tainted self­concept. Although Steel (2007) does not state or imply that individuals in poorer moods are thus no more likely to procrastinate than individuals in better moods, his earlier findings (Steel et al., 2001) obscure the true relationship between self­reported procrastination and mood. Given these differing findings, mood may potentially correlate with procrastination, but no known definitive evidence concerning the directionality of mood exists (Steel, 2007). In agreement with the articulated stance of Steel (2007), the current authors believe that a poor mood may lead to procrastination as well as be a result of procrastination, indicating a bidirectional influence.

Like the overlap between the roles of mood and neuroticism on procrastination, performance may correlate with procrastination through conscientiousness. Past research indicates that high conscientiousness, when combined with low neuroticism, best predicts success in academic activities (Ross et al., 2003). To the degree that procrastination is representative of low levels of conscientiousness, last­minute efforts should result in less success than efforts made in advance of a deadline (Steel, 2007; Tice & Baumeister, 1997). Such failure to succeed can lower one’s self­efficacy, defined as “the belief in one’s capabilities to organize and execute the courses of action required to manage prospective situations” (Bandura, 1995, p. 2). Indeed, low academic self­efficacy stemming from poor performance has been indicated as a possible cognitive predictor of procrastination (Chow, 2011; Ferrari et al., 1992). Additionally, as can be observed in mood, poor performance allows for the possibility of two possible cause ­ effect relationships between procrastination and self­efficacy: Procrastination may lead to poor performance thereby lowering confidence in one’s ability to successfully organize and execute a task, but low self­efficacy may be what leads an individual to procrastinate (Lindsley et al., 1995). This blurred distinction between procrastination and performance, given the intervening role of efficacy beliefs, prevents a clearer understanding of their true relationship.

Cohort Effects

Cohort effects are conceptually defined as attitudinal variations in the collective that result from incoming

generations being different from outgoing generations due to differences in socialization during early life (Glenn, 2005). More simply, cohort effects are variations in characteristics of an observed phenomenon (e.g., cannabis use among adolescents) over time among individuals with a shared experience (e.g., decade of birth). Such effects can influence the perceived acceptability of thoughts and behaviors across individuals of similar age across time (Ekstam, 2021). Cohort effects in personality may affect the collective perceptions of particular thoughts or behaviors that result in increasing or decreasing frequency of these thoughts and behaviors across members of the same demographic (Roberts et al., 2006). For example, Hamilton et al. (2018) examined frequent cannabis use across three cohorts (those born between 1991 and 2000, those born between 2001 and 2010, and those born between 2011 and 2018). They found those born in the earliest cohort experienced the greatest cohort effect, as this cohort demonstrated the greatest collective decrease in frequent cannabis use compared with the latter two cohorts.

The cohort analysis described by Steel (2007) indicates that procrastination increased globally from 1982 through 2003, but no consistent significant effects of year of publication on procrastination were found during or after 2004. This indicates that although rates of procrastination no longer significantly increased as of 2004, they also did not significantly decrease during this time. Recent work has confirmed that overall rates of procrastination remain high regardless of the absence of an overall directional trend (Tibbett & Ferrari, 2018), so the current cohort is likely statistically similar to earlier cohorts who experienced high rates of procrastination (Ferrari et al., 1995).

In addition to possible cohort effects, there are other indicators of self­regulation failure that help to understand and explain changes over time in procrastination. Procrastination has been described as a fundamental failure to self­regulate (Park & Sperling, 2012; Steel, 2007). Research on self­regulation reveals it is a multifaceted process (le Roux & Parry, 2021) that, when exerted effectively, allows for the self­control needed to inhibit an impulse that conflicts with a subjective desire or goal, such as engaging in planful deliberation to begin a necessary but aversive task, even when more appealing tasks can be undertaken instead (Ferrari & Pychyl, 2012; Senecal et al., 1997). The strength required for effective self­regulation is limited in capacity within the individual (Baumeister, 2018) and allows one to focus awareness beyond the immediate stimuli which is a precursor of self­regulatory behavior (Baumeister & Heatherton, 1996). At the time of publication, his metaanalytic analysis on self­regulatory failure indicated

that other self­regulatory failures such as obesity and excessive personal debt had been increasing since 1982 (Steel, 2007), suggesting a possible recent cohort effect on the self­regulatory failure of procrastination, much like the cohort effect seen concerning procrastination between 1982 and 2003.

The rise in social media consumption among younger consumers may be another example of a cohort effect, helping to explain why these individuals experience greater delays in beginning and completing intended tasks (Hinsch & Sheldon, 2013; Reinecke et al., 2018). This is reasonable to hypothesize, given that mean daily Internet use among those aged 10 to 19 has recently been reported as 3.11 hours (SD = 2.59), with a majority using the Internet daily (25.4%) or several times each day (46.0%), and that these behaviors directly related to decreased school performance through an irrational delay of homework or preparing for assessments (Reinecke et al., 2018). Regression analyses within the same study showed that trait procrastination positively correlated with Internet multitasking and inadequately controlled Internet use, further supporting the hypothesis of a cohort effect. Additional recent research among participants aged 18 to 58 demonstrated that Facebook use in particular becomes a viable escape for procrastinators through its entertainment and stress relief values (Przepiorka et al., 2016), which supports similar recent findings that social media use in general provides users a break from distasteful tasks, thereby enhancing subjective well­being (Reinecke et al., 2014). Thus, a correlation and predictive relationship between social media use and procrastination may exist among college students, given the frequency of both behaviors within the current cohort.

Another possible predictor of procrastination in college students may arise from academic entitlement beliefs. Commonly associated with the millennial generation of college students (Kopp et al., 2011), there is an observed trend indicating increasing academic entitlement—a propensity to possess an expectation of academic success without having to assume personal responsibility to achieve this success—among recent cohorts compared with previous cohorts (Chowning & Campbell, 2009; Keener, 2019; Kopp et al., 2011; Wasieleski et al., 2014). The increasing consumer mindset of students (Greenberger et al., 2008) and influences of technology and the media, including Facebook and YouTube, which allow and reinforce self­glorification (Jeffres et al., 2014, have been cited as possible causes of academic entitlement. Underscoring their widespread ability to self­promote made possible through the advent of these and similar forms of social media, the recent concept known as “Generation Me” characterizes this

cohort of college students as the most narcissistic and entitled generation thus far (Twenge & Campbell, 2009).

Achievement Goal Orientation and Implicit Theories

Procrastination, often assumed to be a failure of self­regulation (Ferrari & Pychyl, 2012), leads to low achievement motivation (Brownlow & Reasinger, 2000) and disorganization (Howell & Watson, 2007). Thus, achievement goal orientation may be a robust correlate and predictor of procrastination. First conceptualized by Nicholls (1983) and Dweck (1990), goal orientation theory attempts to account for and explain student differences in achievement behavior, noting that such differences are related to self­efficacy beliefs (McKinney, 2014) as well as various motivational, emotional, cognitive, and behavioral outcomes (Pintrich, 2000). More recent conceptualizations resulted in a 2 x 2 achievement goal framework which includes mastery ­ and performance­based goals according to an approach vs. avoidance paradigm. This newer framework includes mastery­approach goal orientation, where one seeks to learn all there is to learn; mastery­avoidance goal orientation, where one seeks to avoid not learning what one is otherwise able to learn; performance­approach goal orientation, where one seeks to perform better than one’s classmates; and performance­avoid goal orientation, where one simply seeks to avoid performing worse than classmates (Howell & Watson, 2007).

Multiple studies have found that goal orientation type determines a student’s learning strategies, cognitive and behavioral reactions, satisfaction with the student experience, and educational performance (Ames 1992; Barron & Harackiewicz, 2000; Benita et al., 2014 Roebken, 2007; Valle et al., 2003), yet few known studies have examined the connection between procrastination and the four achievement goal orientations within the 2x2 taxonomy. Negative associations have been found between procrastination and the mastery­approach goals (Seo, 2009), with a positive association found between procrastination and mastery­avoidance goals (Howell & Buro, 2009; Seo, 2009; Valle et al., 2003). Work by Valle et al. (2003) revealed no consistent connection between procrastination and performance­based goals, though prior studies demonstrated positive correlations between procrastination and performance­avoidance goals (Seo, 2009) and, at times, performance­approach goals (McGregor & Elliot, 2002; Wolters, 2004).

Related to achievement goal orientations are implicit theories (“mindset beliefs”), which are important in understanding motivation, learning, and intelligence as these outcomes are subjectively viewed through either a fixed or malleable lens (Dweck, 1999).

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

Individuals who endorse a fixed (“entity theory”) mindset belief perceive that an attribute, such as intelligence, is relatively stable and thus unchangeable. Individuals who endorse a growth (“incremental theory”) mindset belief perceive that such an attribute is malleable, and therefore open to influence through effort or practice. Howell and Buro (2009) found that fixed mindset beliefs correlated significantly and positively with procrastination, whereas incremental mindset beliefs correlated negatively. Furthermore, fixed mindset beliefs correlated significantly and negatively with the mastery­approach goal orientation, and positively with all three other goal orientations, further supporting the relatedness of achievement goal orientation and mindset beliefs.

Attitude and Satisfaction With College Life

Based on its documented connection to goal orientation, attitude is another suspected correlate and predictor of procrastination whose role can become further clarified through additional study. Defined as “a settled opinion” and “behavior reflecting this” (Abate, 1999, p. 44), attitude is believed to possess a cognitive and affective component which can lead to the behavioral response (Altmann, 2008). In their measurement of student attitude, Ames and Archer (1988) asked students “How would you rate your liking for this class?” (p. 262) on a 5­point scale. Students who felt encouraged by their instructors to learn for the sake of learning had a more positive attitude toward the class due to the mastery goal orientation approach adopted by these instructors. Combined with the documented negative correlation between procrastination and mastery goal orientation (Valle et al., 2003), a possible connection between procrastination and attitude thus appears reasonable. To assess this, Curtis and Trice (2013) conducted a factor analysis on academic locus of control among college students which identified four factors (hopelessness, distractibility, poor student attitude, impaired planning). Though they did not measure attitude directly, their work supports a plausible relationship between attitude and procrastination, as poor student attitude significantly correlated with academic entitlement and procrastination, further clarifying the connection between student attitude and procrastination.

Although little known work has examined the direct association between student attitude and procrastination, an experimental study assessing student performance and attitudinal differences between a lecture and an online course revealed a negative association between procrastination and attitude among the students who were randomized into the online class, but the same association was not seen for students assigned to the lecture course (Elvers et al., 2003). A strong association

was also found between student course satisfaction and procrastination, but this was only true for students assigned to the online course. Additional correlational findings from earlier work on procrastination revealed that college students who perceive greater control over how they structure and manage their time reported significantly greater satisfaction with academic life (Macan et al., 1990). Combined, these findings suggest that procrastination may be predicted, in part, by the attitude college students hold as well as their satisfaction with college life in general.

Purpose of the Present Study

In the present exploratory study, we attempted to identify psychological and behavioral predictors of general procrastination in undergraduates to further the understanding of researchers, educators, and students alike. Much of the prior research done on procrastination has examined its correlates and consequences in adults. The preponderance of research in undergraduates has focused on identifying correlates and consequences of academic procrastination. To bridge a gap between these two areas, the current study was uniquely aimed in its target focus as well as in its target population. Specifically, it identified predictors of general procrastination in undergraduates, thus filling a gap in the existing literature on procrastination among this demographic. Our research was guided by the following question: What are significant psychological and behavioral predictors of procrastination in undergraduates? Based on the documented and plausible correlates and predictors of procrastination in adults and those of academic procrastination in undergraduates, we hypothesized that conscientiousness, self­efficacy beliefs, mastery goal orientation, and attitude toward learning would negatively predict procrastination, and that daily social media usage would positively predict procrastination within the resultant regression model.

Method Participants

Participants were recruited through class visits, flyers containing the research details, and Sona Research Systems at a medium­sized southwestern university in the United States where data collection took place. Two hundred sixty­seven undergraduate students enrolled in in ­ person introductory psychology courses were included in the current sample (N = 267). Within the sample, 171 participants (64%) were women and 96 (36%) were men. The year­of­study distribution was 119 (44.5%) first­year students (71 women; 48 men), 90 (33.7%) second­year students (61 women; 29 men), 38 (14.2%) third­year students (25 women; 13 men),

and 20 (7.5%) fourth­year students (11 women; 9 men).

Demographic data on age and race/ethnicity were not collected. Participants received two credits toward their four required research participation credit assignments.

Procedure

Before conducting the study, approval was granted by the Southern Utah University institutional review board (#23­092015). After voluntarily opting to participate, each participant was provided with the informed consent form, and subsequent survey access was provided through a link to a survey­hosting website. The online self­report survey included 16 psychological variables and one behavioral variable (daily social media usage) to be used in the regression analyses. Student ID numbers were entered at the beginning of the survey to ensure that no student participated more than once. Upon completing the survey (mean completion time = 27 minutes, SD = 12.21, range = 117.0), course credit was awarded to each participant who showed proof of completion to the course instructor, and the ID number of the student was removed.

Instrumentation

Procrastination was measured by the Lay Procrastination Scale (Lay, 1986), a 20­item scale consisting of Likerttype responses that measure how characteristic tasks typically associated with procrastination are for the participant (e.g., “I often find myself performing tasks that I had intended to do days before.”). Answer items ranged from 1 ( extremely uncharacteristic ) to 5 ( extremely characteristic) and scores ranged from 20 to 100 with higher scores indicating greater procrastination. Lay (1986) indicated reliability through a Cronbach’s α of .82, with a test­retest reliability statistic of .80 (Ferrari, 1989) observed among college samples. In our sample, Cronbach’s α of .78 was observed.

Psychological Correlates & Predictors

Big 5 Personality Traits. Traits of extraversion, agreeableness, conscientiousness, neuroticism, and openness were each measured with the Big Five Inventory (BFI­44) by John and Srivastava (1999). Its 44 items, 16 of which are reverse coded, provided statements regarding how the participants views themselves, with responses indicated on a 5­point Likert scale from 1 (strongly disagree) to 5 (strongly agree). A higher score within a trait indicates greater trait prevalence. BFI­44 test­retest reliability has been demonstrated to be .83 in English­speaking samples (Rammstedt & John, 2006). In our sample, reliability of .70 was observed using Cronbach’s α.

Goal Orientation. Achievement goal orientation

was measured through the Goal Orientation Scales (Midgley et al., 1998), a 7 ­ point Likert ­ type inventory that assesses what a student is attempting to accomplish in a course. Answer ratings vary from 1 (not at all true of me) to 7 (very true of me). The scoring key includes 18 total items, with six items assessing each of the following goal types: task goal orientation (mastery goal orientation), ability ­ approach goal orientation (performance­approach goal orientation), and ability­avoid goal orientation (performance­avoid goal orientation). Sample statements corresponding to each respective domain above include “I prefer course material that really challenges me so I can learn new things,” “It is important to me to do better than the other students,” and “I worry about the possibility of getting a bad grade in this class.” It should be noted that the Mastery­Avoidance goal orientation is not included on this scale, as this orientation was completely neglected in research on achievement goals until relatively recently (Van Yperen, 2006; Van Yperen et al., 2009). Internal consistency as measured by Cronbach’s α for these items has been demonstrated to be above .70 and .80 (Midgley et al., 1998). Our sample demonstrated reliability at .87 using the same metric.

Mindset. The Mindset Survey, adapted by Dweck (2006), was used to assess growth mindset beliefs (e.g. “No matter how much intelligence you have, you can always change it quite a bit.”) vs. fixed mindset beliefs (e.g., “Your intelligence is something very basic about you that you can’t change very much.”) according to implicit theories of intelligence as discussed by Dweck and colleagues (1995). The 20 total items are scored according to a 4 ­ point Likert ­ scale format (strongly agree, agree, disagree, strongly disagree ), with scores for each belief type ranging from 10 to 40. Across six studies, such measures have shown very high internal consistency (α = .94 to .98) as indicated in Dweck et al. (1995). Internal reliability in our sample was also high at .94 using Cronbach’s α. Perfectionism. Due to its relationship with procrastination among college students (Burns et al., 2000) and its association with motivation (van Eerde, 2003), particularly motivation among women (Brownlow & Reasinger, 2000), the current researchers measured perfectionism with the 45 ­ item Multidimensional Perfectionism Scale (MPS) by Hewitt and Flett (1991). The MPS assesses self­oriented perfectionism (e.g. “One of my goals is to be perfect in everything I do.”), otheroriented perfectionism (e.g., “I have high expectations for the people who are important to me.”), and socially prescribed perfectionism (e.g., “My family expects me to be perfect.”). For the purposes of our study, a summed perfectionism score from 45 to 315 was computed

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

and analyzed for each participant. Adequate reliability (Cronbach’s α = .88) and validity have been previously demonstrated for the MPS (Hewitt & Flett, 1991). Our sample achieved internal reliability at α = .96.

Academic Entitlement. To measure academic entitlement beliefs, the 45­item Academic Entitlement Questionnaire (Kopp et al., 2011) was utilized. This Likert­type questionnaire included six answer options per item ranging from 1 (strongly disagree) to 6 (strongly agree). Example items included such statements as “If I don’t do well on a test, the professor should make tests easier or curve grades” and “Because I pay tuition, I deserve passing grades.” Scores on this questionnaire ranged from 26 to 126 with higher scores indicating greater academic entitlement. Our sample demonstrated acceptable internal consistency through a Cronbach’s α of .70.

Self-Efficacy.

General Self-Efficacy. To assess overall confidence in ability successfully accomplish tasks (e.g., “It is easy for me to stick to my aims and accomplish my goals”), the General Self­Efficacy Scale (Schwarzer & Jerusalem, 1995) was administered. Consisting of 10 statements measured through a 4­point Likert­scale format, it asked participants to indicate how true each statement is (not at all true, hardly true, moderately true, exactly true), with scores ranging from 10 to 40. Schwarzer and Jerusalem (1995) indicate high internal consistency (Cronbach’s α = .82 to .93) within this scale across multiple samples. The current study observed a similarly high internal consistency value of α = .85.

College Student Self-Efficacy. To obtain efficacy data more specific to our sample, the College Self­Efficacy Inventory (CSEI) by Solberg et al. (1993) was administered. This 20­item inventory, measured on a 10­point scale, asked participants to rate their confidence (1 = not at all confident; 10 = extremely confident) in successfully completing tasks associated with various aspects of college life (making new friends at college, managing time effectively, doing well on exams, etc.). Scores ranged from 20 to 200. Previous work has shown the CSEI to have very high internal consistency through a Cronbach’s α of .93 (Solberg et al., 1993). Our sample also demonstrated high reliability (Cronbach’s α = .91).

Additional Psychological Predictors. Building on previous work that suggests connections between procrastination and attitude (Ames & Archer, 1998; Curtis & Trice, 2013; Valle et al., 2003) and procrastination and school life satisfaction (Macan et al., 1990), two additional measures created by the principal investigator were included. To examine the connection between procrastination and attitude, a 5 ­ point Likert ­ scale

Attitude Toward Learning item was included which assessed personal level of agreement with the following statement: “I find myself to possess a positive attitude toward learning” (1 = strongly disagree , 5 = strongly agree ). This seeks to determine predictive ability of the negative correlation found between attitude and procrastination among college students (Curtis & Trice, 2013). Additionally, a 7­point Likert­type School Life Satisfaction item was included which assessed level of satisfaction with life as a college student based on the following question: “In general, how satisfied are you with your life as a college student?” (1 = very dissatisfied, 7 =very satisfied). The measure of school life satisfaction was included to assess predictive ability of the positive correlation found between perceptions of control over time management and satisfaction with school in their sample of college students (Macan et al., 1990).

Behavioral Predictors

To capture behavioral data that we hypothesized would positively predict procrastination, a 5­item social media usage scale was included in the survey. This scale asked participants to indicate the number of daily minutes spent using each of the following sites: Twitter, Facebook, Instagram, Pinterest, Tumblr. Answer options for each site consisted of six ranges of daily time spent on social media: 0 minutes, 1 to 30 minutes, 31 to 60 minutes, 61 to 90 minutes, 91 to 120 minutes, greater than 120 minutes. Results from this scale were reported as a sum of daily hours spent using social media.

Results

Multiple Regression Analysis

A standard multiple regression analysis was performed between the dependent variable and all independent predictor variables, with all 17 predictors analyzed simultaneously. This approach was used as our purpose and research questions were aimed at identifying possible significant predictors and not the “best” predictors, which would have required a different (e.g., forward or stepwise) regression model approach.

Preliminary results indicated no correlations within the independent variables themselves exceeded r = .70, satisfying the multicollinearity assumption. Additional assumptions were tested by examining normal probability plots depicting residuals and scatter diagrams of residuals versus predicted residuals. No major violations of normality, linearity, or homoscedasticity of residuals were detected as only one outlier (defined as +/­ 3 SD) was observed.

Regression analysis revealed that the model significantly predicted procrastination, F(17, 249) = 14.734, p < .001. Overall r2 for the model was .50, and adjusted r2 was .47 indicating a large effect size for the model.

Sufficient power (.81) was obtained, allowing the model to detect true effects. Table 1 displays the intercept of the model, and the unstandardized regression coefficient (B), standardized regression coefficient (β), t statistic, and significance (p) value for each predictor variable.

The resultant model indicated that two of the psychological variables positively predicted procrastination in our sample of undergraduates: growth mindset beliefs (β = .16, p = .003) and academic entitlement beliefs (β = .11, p = .023), neither of which was initially hypothesized. The model also identified two psychological variables as negative predictors of procrastination, conscientiousness (β = –.55, p < .001) and college student efficacy beliefs (β = –.17, p = .011), with conscientiousness demonstrating the strongest predictive influence from the variables included in the regression model. This hypothesized finding was expected based on prior work in this area.

Psychological variables expected to predict procrastination that did not reach significance were mastery goal orientation (β = –.01, p = .84) and attitude toward learning (β = .09, p = .14). The behavioral variable of daily social media use that was hypothesized to predict procrastination also failed to reach significance

(β = –.01, p = .82). Additionally, we hypothesized that self ­ efficacy beliefs would predict procrastination. Although college student efficacy beliefs emerged as a significant predictor as indicated above, general efficacy beliefs did not (β = .09, p = .16).

Descriptive statistics between each of the 17 predictor variables and the dependent variable are reported in Table 2. Of the 13 nonsignificant predictors, Pearson r values showed procrastination was significantly related to nine of these (three positively, six negatively), with seven of the nine being significant at p < .01 as described below.

The three positive correlates of procrastination at p < .01 each demonstrated a small effect size. These included attitude toward learning (r = .25, p < .001), neuroticism (r = .24, p < .001), and daily social media use (r = .19, p = .002). Although none of these variables significantly predicted procrastination, there were still significant positive associations between procrastination and attitude toward learning, the Big 5 trait of neuroticism, and daily time spent on social media in our sample of undergraduates.

The four negative correlates of procrastination at p < .01 demonstrated medium to small effect sizes and included mastery goal orientation ( r = –.35,

2

Descriptive Statistics and Correlations With Procrastination for Study Variables

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

p < .001), general efficacy beliefs (r = –.27, p < .001), school life satisfaction (r = –.27, p < .001), and fixed mindset beliefs (r = –.19, p = .003). Although none of these three variables significantly predicted procrastination, these findings revealed that higher procrastination was still significantly associated with a lower mastery goal orientation, lower general efficacy beliefs, lower satisfaction with the undergraduate experience, and lower fixed mindset beliefs.

Discussion

The purpose of this exploratory study was to identify psychological and behavioral predictors of general procrastination in undergraduates. The resultant significant regression model accounted for 47% of the variance in procrastination and identified four psychological predictors. Growth mindset beliefs and academic entitlement beliefs were identified as positive predictors, neither of which was included in our initial hypothesis. This key finding about growth mindset beliefs indicates that students who believe attributes such as intelligence to be changeable through deliberate practice are significantly more likely to procrastinate. This is surprising, given that previous research has shown growth mindset beliefs to be negatively correlated with procrastination (Howell & Buro, 2009). The differing findings obtained in our study may be attributed to the nature of growth mindset beliefs themselves, which allow individuals to see themselves as capable of changing due to increased agency over the self. Mandeville et al. (2015) speculate that individuals with greater agency over their self­concept as a result of growth mindset beliefs may likewise feel greater agency over their circumstances and experiences. If true, this implies that students who possess growth mindset beliefs may feel themselves capable of reducing or avoiding undesired behaviors such as procrastination, but simply choose to engage in these behaviors due to their perceived control over their own outcomes. This may help explain our unexpected finding of a significant positive predictive relationship between growth mindset beliefs and procrastination.

The second positive predictor of procrastination among our sample involved academic entitlement beliefs. This key finding reveals that students who feel entitled to receiving good grades or special academic treatment without necessarily having to put forth the effort that warrants such grades or treatment are significantly more likely to procrastinate. This predictive relationship adds to the growing body of knowledge on entitlement beliefs, and further clarifies their connection to procrastination. Procrastination itself includes oppositional behavior and resentment, both of which have been linked with entitlement beliefs (Ellis & Knaus, 1977). Newer research has

also revealed that procrastination significantly correlates with anger, revenge, and entitlement beliefs (Ferrari & Emmons, 1994). Thus, the findings of our study strengthen the connection between entitlement beliefs and procrastination by uncovering a significant predictive relationship between academic entitlement beliefs in particular and procrastination in college students. To our knowledge, no work has previously identified or documented this effect.

Conversely, conscientiousness and college student efficacy beliefs were identified as negative predictors, supporting part of our initial hypothesis. These additional key findings reveal that highly conscientious students are less likely to procrastinate, as are students who possess confidence in their abilities to successfully navigate the undergraduate experience through, among other things, effectively managing time, communicating with professors, and earning good assignment and course grades. These findings align with results from prior work in both conscientiousness (Park & Sperling, 2012; Steel, 2007) and academic self­efficacy beliefs (Chow, 2011; Ferrari et al., 1992).

The behavioral variable of daily social media use in our hypothesis did not significantly predict procrastination, but it did significantly positively correlate with procrastination. Past research has shown that procrastinators are inclined to use social media sites such as Facebook to reduce or eliminate the negative affect related with distasteful tasks (McCown et al., 2012), so our correlational findings between these variables were expected. However, the failure of daily social media use to predict procrastination was surprising, particularly when viewed in light of the amount of time each day undergraduates in our study reported engaging in social media (M = 4.48 hours, SD = 2.30 hours) across the five platforms included in the survey. Reinecke et al. (2018) revealed that trait procrastination was associated with entertaining online content, which included social media as well as online video and online gaming sites. Based on these findings, different forms of online content (social media vs. online gaming, for example) may vary in their ability to predict procrastination. It may also be true that collapsing the five platforms into one measure caused a loss of explanatory precision in our study which may have otherwise allowed for significant predictive effects of a specific social media site to emerge. Additionally, the current study assessed general procrastination as the dependent variable, but it may be that procrastination itself predicts social media use instead. Such results were obtained by Przepiorka et al. (2016) who found that general procrastination is a significant predictor of Facebook use. Thus, although the significant correlation between daily social media use and general procrastination was

seen in the current study, it is possible that no predictive effect emerged due to (a) a failure to capture data on different forms of online content, (b) a loss of precision created by an aggregated measure of daily social media that included five platforms, and/or (c) an incorrect assumption of the direction of effect between daily social media use and procrastination.

Mastery goal orientation also unsuccessfully predicted procrastination, failing to support this component of our hypothesis. Prior work examining the effects of goal orientation on procrastination revealed that mastery ­ approach goal orientations predict procrastination in college students (Howell & Watson, 2007; Seo, 2009). However, the measure used to assess mastery goal orientation in our study did not separate mastery­approach goals from masteryavoidance goals, but instead collapsed both mastery goal orientations into a combined set of questions. Specifically, the mastery goal section on the survey consisted of six total questions, whereas the sections on performance ­ approach goals and performanceavoid goals consisted of six questions each. Therefore, the goal orientation measure used did not allow us to capture the same level of specificity for mastery goals as it did with performance goals. Thus, it is possible that our results would have been more comparable to those of prior researchers had our measure assessed the two types of mastery goals separately. Furthermore, fixed mindset beliefs approached significance (β = .11, p = .06) as a positive predictor of procrastination.

Importance of the Study and Practical Implications

Based on the significant predictors of procrastination identified by our model, four important findings emerged from this study. First, the predictive role of academic entitlement adds a new dimension to the extant literature on predictors of general procrastination in college students. Our findings reveal higher academic entitlement beliefs predict greater procrastination, which supports the conclusions of Kopp et al. (2011) that claim a growing proportion of students expect good grades despite not having to expend much effort in earning these grades. As a result of this trend, it seems reasonable that entitlement beliefs may exist and even persist when students engage in procrastination, yet still expect to receive high marks despite delaying intended courses of action. Baumeister et al. (1994) stated that “Even the most talented students often seem to think that the route to success is less a matter of hard work, good study habits, and meeting deadlines than of doing extra­credit projects, being creative, and circumventing authoritarian rules with clever excuses and well­phrased requests for special treatment” (p. 5). These new and

unique findings from the current study may help college professors and higher education administrators to more fully understand similar beliefs and dynamics that form the basis of students’ academic entitlement and its subsequent predictive impacts on procrastination. Second, the role of conscientiousness as a key predictor of procrastination further confirms decades of previous work on this personality trait (Steel, 2007; van Eerde, 2003). Of the four significant predictors identified by our model, conscientiousness demonstrated the greatest association with and impact on procrastination as shown by its associated correlational and beta values, respectively, indicating that lower conscientiousness is associated with—and predictive of—greater procrastination. This is an important and encouraging finding because the trait of conscientiousness can be increased through effort and repeated practice (Javaras et al., 2019), practically implying that college students can lessen procrastination tendencies and the negative outcomes associated with these behaviors through deliberate, repeated practice of self­regulatory processes aligned with conscientiousness (Corker et al., 2012). As many colleges and universities offer or mandate a first­year experience course for freshmen (Gardner & Schroeder, 2003), such curricula should include direct and repeated instruction on goal­directed metacognitive skills and techniques such as planning, organization, time management, and other self ­ regulatory skills associated with conscientiousness (Pintrich, 2000). Such curricula would specifically provide students opportunities to strengthen these important academic and life skills during their first year of higher education. Skills developed and refined in such conscientiousness­based freshman­year coursework may also lead to improved retention rates, given the consequences of procrastination on academic performance.

Third, the significant predictive effect of college efficacy beliefs on procrastination coincides with previous findings (Haycock et al., 1998; Rabin et al., 2011), as our results similarly show that higher college efficacy beliefs predict lower procrastination. In addition to being able to improve self­regulatory skills associated with conscientiousness during first­year experience courses, students can also benefit through increasing their confidence in utilizing these skills (Grunschel et al., 2013), suggesting that focus also be placed on increasing student confidence in the ability to manage the broader college experience. This can occur through helping students identify and access campus­ and communitybased resources designed to support students as they adjust to, and journey through, an undergraduate learning environment. Tutoring services, campus mental health services, student success and accessibility centers,

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

writing centers, vocational rehabilitation services, and various academic clubs and organizations are resources that can provide opportunities to strengthen college student confidence in utilizing the self­regulatory skills needed for successfully navigating the undergraduate experience.

Finally, the significant predictive role of growth mindset beliefs diverges from findings of the limited known previous work on implicit theories of intelligence. Howell and Buro (2009) revealed that among 397 undergraduates, entity/fixed mindset beliefs positively predicted academic procrastination whereas a growth mindset emerged as a negative predictor. Our results showed that although fixed mindset beliefs approached significance (p = .06) as a positive predictor, growth mindset beliefs significantly positively predicted procrastination. This divergence from previous findings might be attributed to the nature of their chosen measure of procrastination, which is a 16­item scale that measures academic procrastination (Tuckman, 1991) whereas the Lay (1986) scale used in the current study measures general procrastination. Our correlational findings align with those of prior work in which mastery­approach goals were negatively correlated with procrastination (Howell & Buro, 2009; Howell & Watson, 2007; Valle et al., 2003).

Limitations

Despite the strengths associated with the current study, there were several limitations that make generalization of findings somewhat difficult. First, all predictor variables and the dependent variable were measured using a selfreport method. Although this approach has commonly been used in most research on procrastination (van Eerde, 2003), participants may have reported inaccurate information. However, data were anonymized, which increased the likelihood that participants provided honest answers on the survey, increasing validity of our results. Regardless, future researchers should create and validate other data collection methods through which to complement the self­reporting scales used here. A second limitation is that our sample included only undergraduates, so any attempt to generalize to nonacademic settings or to other student populations is not possible. A third possible limitation is that students who are more prone to procrastinating may have been over­represented in our study due to personal interest in the topic being researched, thereby skewing our results via selection bias. However, as indicated earlier, no major violations of normality, linearity, or homoscedasticity of residuals were observed, so this type of systematic skew is unlikely to have occurred. Another limitation stems from the use of the two researcher­created scales included in the survey: the Attitude Toward Learning scale and School

Life Satisfaction scale. As these were created for use in the current study, neither had been piloted or validated previously, thus possibly rendering these data suspect. Additionally, because the survey was presented in the same order to each participant, our findings may be limited due to carryover, in particular to fatigue effects.

It must also be kept in mind that participant characteristics likely limit the extent to which the current findings can be generalized. Such characteristics include—among others—age, race, and ethnicity, none of which were directly measured in the current study. Although the preponderance of students in the psychology classrooms from which our participants were drawn appeared to be traditional­age undergraduates, there is likely greater variability in age than initially was assumed due to the presence of some non­traditional students who returned to the classroom later in life. Even though these individuals were classified as undergraduates by credit hour status at the time of study completion, their data may diverge from those of the traditional­age undergraduates due to quantitative differences in overall life experience. Additionally, the university from which all participants were drawn historically represents a relatively homogenous racial, ethnic, and religious culture, making our sample perhaps unique, albeit internally more consistent than most universities, in its lack of variability within each of these characteristics. Therefore, it would be unjustifiable to assume that these findings would necessarily be equally true for other communities outside of the sample used here, which came exclusively from one specific university student participation pool.

Future Recommendations

Although gender (Steel, 2007; van Eerde, 2003), age (Howell et al., 2006), ethnicity (Kachgal et al., 2001), and intellectual ability (van Eerde, 2003) have shown no consistent directional relationship with general procrastination across previous studies, future procrastination researchers may wish to include these as possible additional exploratory predictors, as these data were not included in our analyses out of efforts to omit predictors that we did not feel would uniquely contribute to the model based on prior findings. Future researchers in college student procrastination should consider conducting confirmatory analyses to pinpoint which specific predictors identified in the current study contribute the greatest proportion of variance within regression models. Doing so would bring additional clarity to the findings presented in this study by accounting for the variance explained by each unique predictor identified through our exploratory analyses. Additionally, future researchers should conduct similar confirmatory analyses on academic procrastination (“the intentional delay in

the beginning or completion of important and timely academic activities” (Rabin et al., 2011, p. 344) in college students to reveal areas of overlap between general procrastination as was measured in the current study and procrastination behaviors specific to academic tasks.

It has been shown that certain facets of conscientiousness (self­control, order, industriousness, and responsibility) are significantly related to objective health markers (Sutin et al., 2018). To broaden the current understanding of the role and impacts of conscientiousness on general as well as academic procrastination, researchers should assess each of the six identified conscientiousness facets: self­control, order, industriousness, traditionalism, virtue, and responsibility (Costa & McCrae, 1995) to determine which of these lower­order traits that make up the domain of conscientiousness are most predictive of procrastination in these two contexts. As the current study only measured levels of overall conscientiousness as a domain, facet­level analysis would highlight the specific lower­order traits that are best suited for predicting general procrastination as well as academic procrastination.

Finally, future researchers should attempt to clarify the positive predictive connection between growth mindset beliefs and general procrastination, noting that a growth mindset negatively predicts academic procrastination. Additional work may clarify the opposing roles growth mindset beliefs play across these two behavioral domains. Prior work found that goal orientation, particularly mastery­avoidance goals, mediates the relationship between fixed mindset beliefs and procrastination (Howell & Buro, 2009). As the goal orientation measure in the current study did not include mastery­avoidance goals specifically, future work informing this area should focus on the role of mastery­avoidance goals as a way to provide additional explanatory power.

Conclusion

This study contributes to the research on general procrastination in undergraduates through its identification of significant predictors of this behavior, and in doing so broadens the extant body of knowledge on correlates and consequences of general procrastination where much prior focus has been placed. Although the hypothesized predictors of attitude toward learning and daily social media use did not emerge as significant predictors of procrastination, the current findings support previous research examining the predictive influence of conscientiousness, college efficacy beliefs, and mastery goal orientation on procrastination in college students. Additionally, and perhaps most importantly, current findings highlight the significant predictive effect of

academic entitlement beliefs on procrastination in undergraduates, opening the door to future research in which personality, mindset, efficacy beliefs, and goal orientation constructs can be further assessed for their contributions to student entitlement. No known published work has found or documented this predictive effect, so our results in this area are unique and noteworthy. Researchers should continue to study academic entitlement along with the personal qualities, characteristics, beliefs, and social factors that promote entitlement and its resultant impacts on procrastination, as present findings provide emerging evidence that this connection exists and, given recent cohort trends documented by Keener (2019), academic entitlement is likely to be increasingly observed in higher education classrooms.

References

Abate, F. R. (Ed.). (1999). The Oxford American dictionary of current English. Oxford University.

Altmann, T. K. (2008). Attitude: A concept analysis. Nursing Forum, 43(3), 144–150. https://doi.org/10.1111/j.1744-6198.2008.00106.x

Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84(3), 261–271. https://doi.org/10.1037/0022-0663.84.3.261

Ames, C., & Archer, J. (1988). Achievement goals in the classroom: Students’ learning strategies and motivation processes. Journal of Educational Psychology, 80(3), 260–267. https://doi.org/10.1037/0022-0663.80.3.260

Bandura, A. (1995). Self-efficacy in changing societies. Cambridge University Press.

Barrick, M. R., & Mount, M. K. (2003). Impact of meta-analysis methods on understanding personality-performance relations. In K. R. Murphy (Ed.), Validity generalization: A critical review (pp. 197–221). Erlbaum.

Barron, K. E., & Harackiewicz, J. M. (2000). Achievement goals and optimal motivation. In C. Sansone & J. M. Harackiewicz (Eds.), Intrinsic and extrinsic motivation: The search for optimal motivation and performance (pp. 229–254). Academic Press.

Baumeister, R. (2018). Self-regulation and self-control. Routledge.

Baumeister, R. F., & Heatherton, T. F. (1996). Self-regulation failure: An overview. Psychological Inquiry, 7, 1–15. https://doi.org/10.1207/s15327965pli0701_1

Baumeister, R. F., Heatherton, T. F., & Tice, D. M. (1994). Losing control: How and why people fail at self-regulation. Academic Press.

Beck, B. L., Koons, S. R., & Milgrim, D. L. (2012). Correlates and consequences of behavioral procrastination: The effects of academic procrastination, self-consciousness, self-esteem, and self-handicapping. Journal of Social Behavior and Personality, 15(5), 3–13.

Benita, M., Roth, G., & Deci, E. (2014). When are mastery goals more adaptive? It depends on experiences of autonomy support and autonomy. Journal of Educational Psychology, 106(1), 258–267. https://doi.org/10.1037/A0034007

Blatt, S. J., & Quinlan, P. (1967). Punctual and procrastinating students: A study of temporal parameters. Journal of Consulting Psychology, 31(2), 169–174. https://doi.org/10.1037/H0024413

Blunt, A., & Pychyl, T. A. (1998). Volitional action and inaction in the lives of undergraduate students: State orientation, procrastination, and proneness to boredom. Personality and Individual Differences, 24, 837–846. https://doi.org/10.1016/S0191-8869(98)00018-X

Bridges, K. R., & Roig, M. (1997). Academic procrastination and irrational thinking: A re-examination with context controlled. Personality and Individual Differences, 22, 941–944. https://doi.org/10.1016/S0191-8869(96)00273-5

Brownlow, S., & Reasinger, R. D. (2000). Putting off until tomorrow what is better done today: Academic procrastination as a function of motivation toward college work. Journal of Social Behavior & Personality, 15(5), 15–34.

Burns, L. R., Dittmann, K., Nguyen, N., & Mitchelson, J. K. (2000). Academic procrastination, perfectionism, and control: Associations with vigilant and avoidant coping. Journal of Social Behavior and Personality, 15(5), 35–46.

Chow, H. P. (2011). Procrastination among undergraduate students: Effects of

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

SPRING 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

emotional intelligence, school life, self-evaluation, and self-efficacy. Alberta Journal of Educational Research, 57(2), 234–240.

https://doi.org/10.11575/ajer.v57i2.55479

Chowning, K., & Campbell, N. J. (2009). Development and validation of a measure of academic entitlement: Individual differences in students’ externalized responsibility and entitled expectations. Journal of Educational Psychology, 101(4), 982–997. https://doi.org/10.1037/a0016351

Corker, K. S., Oswald, F. L., & Donnellan, M. B. (2012). Conscientiousness in the classroom: A process explanation. Journal of Personality, 80(4), 995–1028. https://doi.org/10.1111/j.1467-6494.2011.00750.x

Costa, P. T., & McCrae, R. R. (1992). The NEO Personality Inventory—Revised Psychological Assessment Resources.

Costa, P. T., & McCrae, R. R. (1995). Domains and facets: Hierarchical personality assessment using the revised NEO personality inventory. Journal of Personality Assessment, 61(1), 20–50.

https://doi.org/10.1207/S15327752JPA6401_2

talks to John Gardner. About Campus, 8(4), 9–16. https://doi.org/10.1177%2F108648220300800402

Glenn, N. D. (2005). Cohort analysis: Quantitative applications in the social sciences. Sage.

Green, L. (1982). Minority students’ self-control of procrastination. Journal of Counseling Psychology, 29(6), 636–644.

https://psycnet.apa.org/doi/10.1037/0022-0167.29.6.636

Greenberger, E., Lessard, J., Chen, C., & Farruggia, S. P. (2008). Self-entitled college students: Contributions of personality, parenting, and motivational factors. Journal of Youth and Adolescence, 37(10), 1193–1204. http://dx.doi.org/10.1007/s10964-008-9284-9

Grunschel, C., Patrzek, J., & Fries, S. (2013). Exploring reasons and consequences of academic procrastination: An interview study. European Journal of Psychology of Education, 28, 841–861. http://dx.doi.org/10.1007/s10212-012-0143-4

Hamilton, A. D., Jang, J. B., Patrick, M. E., Schulenberg, J. E., & Keyes, K. M. (2018). Age, period and cohort effects in frequent cannabis use among U.S. students: 1991–2018. Addiction, 114, 1763–1772.

https://doi.org/10.1111/add.14665

https://doi.org/10.2466/08.03.PMS.116.3.817-829

Curtis, N. A., & Trice, A. D. (2013). A revision of the academic locus of control scale for college students. Perception & Motor Skills: Physical Development & Measurement, 116(3), 817–829.

Day, V., Mensink, D., & O’Sullivan, M. (2000). Patterns of academic procrastination. Journal of College Reading and Learning, 30, 120–134. https://doi.org/10.1080/10790195.2000.10850090

Dewitte, S., & Schouwenburg, H. C. (2002). Procrastination, temptations, and incentives: The struggle between the present and the future in procrastinators and the punctual. European Journal of Personality, 16, 469–489. https://doi.org/10.1002/per.461

Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology, 41, 417–440. https://doi.org/10.1146/annurev.ps.41.020190.002221

Dweck, C. S. (1990). Self-theories and goals: Their role in motivation, personality, and development. In R. Dienstbier (Ed.), Nebraska symposium on motivation (Vol. 38, pp. 199–235). University of Nebraska Press.

Dweck, C. S. (1999). Self-theories: Their role in motivation, personality, and development. Psychology Press.

Dweck, C. S. (2006). Mindset: The new psychology of success. Ballantine Books.

Dweck, C. S., Chiu, C. Y., & Hong, Y. Y. (1995). Implicit theories and their role in judgements and reactions: A word from two perspectives. Psychological Inquiry, 6(4), 267–285. https://doi.org/10.1207/s15327965pli0604_1

Ekstam, D. (2021). The liberalization of American attitudes to homosexuality and the impact of age, period, and cohort effect. Social Forces, 100(2), 905–929. https://doi.org/10.1093/sf/soaa131

Ellis, A., & Knaus, W. (1977). Overcoming procrastination. Institute for Rational Living.

Elvers, G. C., Polzella, D. J., & Graetz, K. (2003). Procrastination in online courses: Performance and attitudinal differences. Teaching of Psychology, 30, 159–162. https://doi.org/10.1207%2FS15328023TOP3002_13

Fee, R. L., & Tangney, J. P. (2012). Procrastination: A means of avoiding shame or guilt? Journal of Social Behavior and Personality, 5(5), 167–184.

Ferrari, J. R. (1989). Reliability of academic and dispositional measures of procrastination. Psychological Reports, 64, 1057–1058.

https://doi.org/10.2466/pr0.1989.64.3c.1057

Ferrari, J. R. (2000). Procrastination and attention: Factor analysis of attention deficit, boredomness, intelligence, self-esteem, and task delay frequencies. Journal of Social Behavior and Personality, 15, 185–196.

Ferrari, J. R., & Emmons, R. A. (1994). Procrastination as revenge: Do people report using delays as a strategy for vengeance? Personality and Individual Differences, 17, 539–544.

https://doi.org/10.1016/0191-8869(94)90090-6

Ferrari, J. R., & Pychyl, T. A. (2012). “If I wait, my partner will do it”: The role of conscientiousness as a mediator in the relation of academic procrastination and perceived social loafing. North American Journal of Psychology, 14(1), 13–24.

Ferrari, J. R., Johnson, J. L., & McCown, W. G. (1995). Procrastination and task avoidance: Theory, research, and treatment. Plenum Press.

Ferrari, J. R., Parker, J. T., & Ware, C. B. (1992). Academic procrastination: Personality correlates with Myers-Briggs types, self-efficacy, and academic locus of control. Journal of Social Behavior and Personality, 7, 495–502.

http://dx.doi.org/10.4236/psych.2012.31003

Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449–468.

https://doi.org/10.1007/BF01172967

Gardner, J., & Schroeder, C. (2003). The first year and beyond: Charles Schroeder

Haycock, L. A., McCarthy, P., & Skay, C. L. (1998). Procrastination in college students: The role of self-efficacy and anxiety. Journal of Counseling & Development, 3(76), 317–324. https://doi.org/10.1002/j.1556-6676.1998.tb02548.x

Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: Conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60, 456–470. https://doi.org/10.1037/0022-3514.60.3.456

Hinsch, C., & Sheldon, K. M. (2013). The impact of frequent social internet consumption: Increased procrastination and lower life satisfaction. Journal of Consumer Behaviour, 12, 496–505. https://doi.org/10.1002/CB.1453

Howell, A. J., & Buro, K. (2009). Implicit beliefs, achievement goals, and procrastination: A mediational analysis. Learning and Individual Differences, 19, 151–154. http://dx.doi.org/10.1016/j.lindif.2008.08.006

Howell, A. J., & Watson, D. C. (2007). Procrastination: Associations with achievement goal orientation and learning strategies. Personality and Individual Differences, 43, 167–178. https://doi.org/10.1016/J.PAID.2006.11.017

Howell, A. J., Watson, D. C., Powell, R. A., & Buro, K. (2006). Academic procrastination: The pattern and correlates of behavioural postponement. Personality and Individual Differences, 40, 1519–1530. https://doi.org/10.1016/j.paid.2005.11.023

Hurtz, G. M., & Donovan, J. J. (2000). Personality and job performance: The Big Five revisited. Journal of Applied Psychology, 85, 869–879. https://doi.org/10.1037/0021-9010.85.6.869

Javaras, K. N., Williams, M., & Baskin-Sommers, A. R. (2019). Psychological interventions potentially useful for increasing conscientiousness. Personality Disorders: Theory, Research, and Treatment, 10(1), 13–24. http://dx.doi.org/10.1037/per0000267

Jeffres, M. N., Barclay, S. M., & Stolte, S. K. (2014). Academic entitlement and academic performance in graduating pharmacy students. American Journal of Pharmaceutical Education, 78(6), 1–9. https://doi.org/10.5688%2Fajpe786116

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big Five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (3rd ed., pp. 265–286). Guilford Press.

John, O. P., & Sanjay, S. (1999). The big five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality (pp. 102–138). Guilford Press.

John, O. P., & Srivastava, S. (1999). The Big-Five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (Vol. 2, pp. 102–138). Guilford Press.

Johnson, J. L., & Bloom, A. M. (1995). An analysis of the contribution of the five factors of personality to variance in academic procrastination. Personality and Individual Differences, 18, 127–133.

https://doi.org/10.1016/0191-8869%2894%2900109-6

Kachgal, M. M., Hansen, L. S., & Nutter, K. J. (2001). Academic procrastination prevention/intervention: Strategies and recommendations. Journal of Developmental Education, 25, 14–24.

Katz, I., Eilot, K., & Nevo, N. (2014). “I’ll do it later”: Type of motivation, selfefficacy, and homework procrastination. Motivation and Emotion, 38, 111–119. https://doi.org/10.1007/s11031-013-9366-1

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

Gregory, Golson, and Larsen | Predictors of Undergraduate Procrastination

Keener, A. (2019). An examination of psychological characteristics and their relationship to academic entitlement among millennial and nonmillennial college students. Psychology in the Schools, 57(2), 572–582. https://doi.org/10.1002/pits.22338

Knaus, W. J. (1998). Do it now! Break the procrastination habit. Wiley.

Kopp, J. P., Zinn, T. E., Finney, S., & Jurich, D. P. (2011). The development and evaluation of the academic entitlement questionnaire. Measurement and Evaluation in Counseling and Development, 44(2), 105–129. https://doi.org/10.1177%2F0748175611400292

Lay, C. (1986). At last, my research article on procrastination. Journal of Research in Personality, 20, 474–495.

https://doi.org/10.1016/0092-6566(86)90127-3

Lay, C. H. (1997). Explaining lower-order traits through higher-order factors: The case of trait procrastination, conscientiousness, and the specificity dilemma. European Journal of Personality, 11, 267–278. https://doi.org/10.1002/(SICI)1099-0984(199711)11:4%3C267::AIDPER281%3E3.0.CO;2-P

Lay, C. H., & Schouwenburg, H. C. (1993). Trait procrastination, time management, and academic behavior. Journal of Social Behavior and Personality, 8, 647–662.

le Roux, D. B., & Parry, D. A. (2021). Off-task media use in academic settings: Cycles of self-regulation failure. Journal of American College Health, 69(2), 134–141. https://doi.org/10.1080/07448481.2019.1656636

Lindsley, D. H., Brass, D. J., & Thomas, J. B. (1995). Efficacy-performance spirals: A multilevel perspective. The Academy of Management Review, 20, 645–678. https://doi.org/10.2307/258790

Macan, T. H., Shahani, C., Dipboye, R. L., & Phillips, A. P. (1990). College students' time management: Correlations with academic performance and stress. Journal of Educational Psychology, 82(4), 760–768. https://doi.org/10.1037/0022-0663.82.4.760

Mandeville, D., & Stoner, M. (2015). Assessing the effect of problem-based learning on undergraduate student learning in biomechanics. Journal of College Science Teaching, 45(1), 66–75. http://dx.doi.org/10.2505/4/jcst15_045_01_66

McCown, W. G., Blake, I., & Keiser, R. (2012). Content analyses of the beliefs of academic procrastinators. Journal of Rational-Emotive & CognitiveBehavior Therapy, 30, 213–222. https://doi.org/10.1007/s10942-012-0148-6

McGregor, H. A., & Elliot, A. J. (2002). Achievement goals as predictors of achievement-relevant processes prior to task engagement. Journal of Educational Psychology, 94, 381–395. https://doi.org/10.1037/0022-0663.94.2.381

McKinney, A. P. (2014). Impact of student goal orientation and self-regulation on learning outcomes. Journal of Organizational Psychology, 14(2), 66–77.

Midgley, C., Kaplan, A., Middleton, M., & Maehr, M. L. (1998). The development and validation of scales assessing students’ achievement goal orientations. Contemporary Educational Psychology, 23, 113–131. https://doi.org/10.1006/ceps.1998.0965

Nicholls, J. G. (1983). Conceptions of ability and achievement motivation. In R. Ames & C. Ames (Eds.), Research on motivation in education (Vol. 3., pp. 185–218). Academic Press.

O’Brien, W. K. (2002). Applying the transtheoretical model to academic procrastination. Unpublished doctoral dissertation, University of Houston.

O’Connor, M. C., & Paunonen, S. V. (2007). Big Five personality predictors of post-secondary academic performance. Personality and Individual Differences, 43, 971–990. https://psycnet.apa.org/doi/10.1016/j.paid.2007.03.017

Park, S. W., & Sperling, R. A. (2012). Academic procrastinators and their selfregulation. Psychology, 3(1), 12–23. https://psycnet.apa.org/doi/10.4236/psych.2012.31003

Pintrich, R. R. (2000). Multiple goals, multiple pathways: The role of goal orientation in learning and achievement. Journal of Educational Psychology, 92, 544–555. https://doi.apa.org/doi/10.1037/0022-0663.92.3.544

Przepiorka, A., Blanchnio, A., & Dias-Morales, J. F. (2016). Problematic Facebook use and procrastination. Computers in Human Behavior, 65, 56–64. https://doi.org/10.1016/j.chb.2016.08.022

Pychyl, T. A., Lee, J. M., Thibodeau, R., & Blunt, A. (2000). Five days of emotion: An experience sampling study of undergraduate student procrastination. Journal of Social Behavior and Personality, 15, 239–254.

Pychyl, T. A., Morin, R. W., & Salmon, B. R. (2000). Procrastination and the planning fallacy: An examination of the study habits of university students. Journal of Social Behavior and Personality, 15(5), 135–150.

Rabin, L. A., Fogel, J., & Nutter-Upham, K. E. (2011). Academic procrastination in college students: The role of self-reported executive function. Journal of

Clinical and Experimental Neuropsychology, 33(3), 344–357. https://doi.org/10.1080/13803395.2010.518597

Rammstedt, B., & John, O. P. (2006). Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German. Journal of Research in Personality, 41, 203–212. https://doi.org/10.1016/J.JRP.2006.02.001

Reinecke, L., Meier, A., Beutel, M. E., Schemer, C., Stark, B., Wolfling, K., & Muller, K. W. (2018). The relationship between trait procrastination, internet use, and psychological functioning: Results from a community sample of German adolescents. Frontiers in Psychology, 9, 1–12. https://doi.org/10.3389/fpsyg.2018.00913

Reinecke, L., Vorderer, P., & Knop, K. (2014). Entertainment 2.0? The role of intrinsic and extrinsic need satisfaction for the enjoyment of Facebook use. Journal of Communication, 64(3), 417–438. https://doi.org/10.1111/jcom.12099

Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course: A meta-analysis of longitudinal studies. Psychological Bulletin, 132, 1–25. https://doi.org/10.1037/0033-2909.132.1.1

Roebken, H. (2007). Multiple goals, satisfaction, and achievement in university undergraduate education: A Student Experience in the Research University (SERU) project research paper. Center for Studies in Higher Education, 2(7), 1–20.

Ross, S. R., Rausch, M. K., & Canada, K. E. (2003). Competition and cooperation in the Five-Factor Model: Individual differences in achievement orientation. The Journal of Psychology, 137(4), 323–337. https://doi.org/10.1080/00223980309600617

Rothblum, E. D., Solomon, L. J., & Murakami, J. (1986). Affective, cognitive, and behavioral differences between high and low procrastinators. Journal of Counseling Psychology, 33, 387–394. https://doi.org/10.1037/0022-0167.33.4.387

Rozental, A., Forsell, E., Svensson, A., Forsstrom, D., Andersson, & Carlbring, P. (2015). Differentiating procrastinators from each other: A cluster analysis. Cognitive Behaviour Therapy, 44(6), 480–490. https://doi.org/10.1080/16506073.2015.1059353

Schwarzer, R., & Jerusalem, M. (1995). Generalized Self-Efficacy scale. In J. Weinman, S. Wright, & M. Johnston, Measures in health psychology: A user’s portfolio. Causal and control beliefs (pp. 35–37). NFER-Nelson.

Schouwenburg, H. C. (1995). Academic procrastination: Theoretical notions, measurement, and research. In J. R. Ferrari, J. L. Johnson, & W. G. McCown, Procrastination and task avoidance: Theory, research, and treatment (pp. 71–96). Plenum Press. https://psycnet.apa.org/doi/10.1007/978-1-4899-0227-6_4

Semb, G., Glick, D. M., & Spencer, R. E. (1979). Student withdrawals and delayed work patterns in self-paced psychology courses. Teaching of Psychology, 6, 23–25. https://doi.org/10.1207%2Fs15328023top0601_8

Senecal, C., Lavoie, K., & Koestner, R. (1997). Trait and situational factors in procrastination: An interactional model. Journal of Social Behavior and Personality, 12(4), 889–903. https://doi.org/10.1016/j.paid.2005.11.023

Seo, E. H. (2009). The relationship of procrastination with a mastery goal versus an avoidance goal. Social Behavior and Personality, 37(7), 911–920. https://doi.org/10.2224/sbp.2009.37.7.911

Sirin, E. F. (2011). Academic procrastination among undergraduates attending school of physical education and sports: Role of general procrastination, academic motivation, and academic self-efficacy. Educational Research and Reviews, 6(5), 447–455.

Sirois F., & Pychyl T. (2013). Procrastination and the priority of short-term mood regulation: Consequences for future self. Social and Personality Psychology Compass, 7, 115–127.

Solberg, V. S., O’Brien, K., Villareal, P., Kennel, R., & Davis, B. (1993). Self-efficacy and Hispanic college students: Validation of the college self-efficacy instrument. Hispanic Journal of Behavioral Sciences, 15(1), 80–95. https://doi.org/10.1177/07399863930151004

Solomon, L. J., & Rothblum, E. D. (1984). Academic procrastination: Frequency and cognitive-behavioral correlates. Journal of Counseling Psychology, 31(4), 503–509. https://doi.org/10.1037/0022-0167.31.4.503

Steel, P. (2007). The nature of procrastination: A meta-analytic and theoretical review of quintessential self-regulatory failure. Psychological Bulletin, 133, 65–94. https://doi.org/10.1037/0033-2909.133.1.65

Steel, P., Brothen, T., & Wambach, C. (2001). Procrastination and personality, performance, and mood. Personality and Individual Differences, 30, 90–106. https://doi.org/10.1016/S0191-8869(00)00013-1

Sutin, A. R., Stephan, Y., & Terracciano, A. (2018). Facets of conscientiousness and objective makers of health status. Psychology & Health, 33(9), 1100–1115. https://doi.org/10.1080%2F08870446.2018.1464165

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

SPRING 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

Predictors of Undergraduate Procrastination | Gregory, Golson, and Larsen

Tibbett, T. P., & Ferrari, J. R. (2018). The U.S. as a “ProcrastiNATION”: Assessing indecision on life satisfaction and regret. North American Journal of Psychology, 20, 111–120.

Tice, D. M., & Baumeister, R. F. (1997). Longitudinal study of procrastination, performance, stress, and health: The costs and benefits of dawdling. Psychological Science, 8(6), 454–458.

http://dx.doi.org/10.1111/j.1467-9280.1997.tb00460.x

Tuckman, B. W. (1991). The development and concurrent validity of the procrastination scale. Educational and Psychological Measurement, 51(2), 473–480. https://doi.org/10.1177%2F0013164491512022

Twenge, J. M., & Campbell, W. K. (2009). The narcissism epidemic: Living in the age of entitlement. Free Press.

Valle, A., Cabanach, R. G., Nunez, J. C., Gonzalez-Pienda, J., Rodriguez, S., & Pineiro, I. (2003). Multiple goals, motivation, and academic learning. British Journal of Educational Psychology, 73, 71–87.

https://doi.org/10.1348/000709903762869923

van Eerde, W. (2003). A meta-analytically derived nomological network of procrastination. Personality and Individual Differences, 35, 1401–1418.

https://doi.org/10.1016/S0191-8869(02)00358-6

Van Yperen, N. W. (2006). A novel approach to assessing achievement goals in the context of the 2x2 framework: Identifying distinct profiles of individuals with different dominant achievement goals. Personality and Social Psychology Bulletin, 32, 1432–1445.

https://doi.org/10.1177%2F0146167206292093

Van Yperen, N. W., Elliot, A. J., & Anseel, F. (2009). The influence of masteryavoidance goals on performance improvement. European Journal of Social Psychology, 39, 932–943. http://dx.doi.org/10.1002/ejsp.590

Wasieleski, D. T., Whatley, M. A., Briihl, D. S., & Branscome, J. M. (2014). Academic entitlement scale: Development and preliminary validation. Psychology

Research, 4(6), 441–450. https://doi.org/10.17265/2159-5542%2F2014.06.004

Watson, D. C. (2001). Procrastination and the Five-Factor Model: A facet level analysis. Personality and Individual Differences, 30, 149–158. https://doi.org/10.1016/S0191-8869(00)00019-2

Weinstock, L. M., & Whisman, M. A. (2006). Neuroticism as a common feature of the depressive and anxiety disorders: A test of the revised integrative hierarchical model in a national sample. Journal of Abnormal Psychology, 115(1), 68–74. https://doi.org/10.1037/0021-843x.115.1.68

Wesley, J. C. (1994). Effects of ability, high school achievement, and procrastinatory behavior on college performance. Educational and Psychological Measurement, 54, 404–408.

https://doi.org/10.1177%2F0013164494054002014

Wohl, M. J. A., Pychyl, T. A., & Bennett, S. H. (2010). I forgive myself, now I can study: How self-forgiveness for procrastinating can reduce future procrastination. Personality and Individual Differences, 48, 803–808. https://doi.org/10.1016/j.paid.2010.01.029

Wolters, C. A. (2003). Understanding procrastination from a self-regulated learning perspective. Journal of Educational Psychology, 95(1), 179–187. https://doi.org/10.1037/0022-0663.95.1.179

Author Note

Bradley Gregory https://orcid.org/0000­0002­5524­5684

Bradley Gregory is now at the Department of Psychology at North Greenville University, Tigerville, SC.

We have no known conflicts of interest to disclose. Special thanks to Mary Mahan for insightful tips during final formatting. Correspondence concerning this article should be addressed to Bradley B. Gregory, North Greenville University, 7801 N. Tigerville Road, Tigerville, SC 29688. Email: brad.gregory@ngu.ed

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

Interpersonal Identity Cues: The Effect of Therapist Identity on Expectations for the Therapeutic Relationship

Jessica S. Philip and

ABSTRACT. I dentity cues can impact levels of comfort for marginalized individuals in various contexts, including STEM fields and medical spaces. In this study, we examined whether a therapist’s personality and race can serve as identity cues for racial/ethnic minority clients and affect the therapeutic relationship. We recruited Asian American and Hispanic/Latinx women (N = 260) for a 3 (trait: high agreeableness, low agreeableness, control) x 2 (race: Black, White) between­subjects experimental study to test the effects of a therapist’s race and personality traits on women’s expectations for a therapeutic relationship and anticipated prejudice from the therapist. We found that racial/ethnic minority women anticipated a more genuine relationship, F(1, 252) = 10.36, p = .001, ηp2 = .04, with a Black female therapist and perceived her to be more culturally competent, F(1, 252) = 20.04, p < .001, ηp2 = .07, and less likely to be racist, F(1, 252) = 12.68, p < .001, ηp2 = .05, than a White female therapist. We found no significant differences in perceived prejudice based on the therapist’s personality. Similarly, there were no significant differences in expectations for the therapeutic relationship based on therapist personality.

Keywords: identity cues, client­therapist relationship, social identity threat, prejudice, personality

ABSTRACTO. Las señales de identidad pueden afectar la comodidad de los individuos marginados en varios contextos, incluyendo los campos de STEM y los espacios médicos. En este estudio, examinamos si la personalidad y la raza de una terapist pueden servir como señales de identidad para las clientes de minorías raciales/étnicas y pueden afectar la relación terapéutica. Reclutamos mujeres asiáticas e hispanas/latinas (N = 260) para un estudio experimental interindividual de 3 (la característica: alta amabilidad, baja amabilidad, condición de control) por 2 (la raza: raza negra, raza blanca) para probar los efectos de la raza y la personalidad de una terapista en las expectativas de las mujeres para una relación terapéutica y el prejuicio percibido de la terapeuta. Encontramos que las mujeres de minorías raciales/étnicas anticipaban una relación más genuina con una terapista negra, F(1, 252) = 10.36, p = .001, η p 2 = .04, y la percibían como más competente culturalmente, F (1, 252) = 20.04, p < .001, ηp 2 = .07, y menos propensa a ser racista, F(1, 252) = 12.68, p < .001, ηp2 = .05, que una terapista blanca. No encontramos diferencias significativas en

*Diana T. Sanchez is the faculty mentor.

Melanie R. Maimon Department of Psychology, Rutgers University*

Diversity badge earned for conducting research focusing on aspects of diversity. Open Data and Open Materials badges earned for transparent research practices. Data and materials are available at https://osf.io/sjecw/

el prejuicio percibido basado en la personalidad de la terapista. Del mismo modo, no hubo diferencias significativas en las expectativas de la relación terapéutica basada en la personalidad de la terapeuta.

Las palabras claves: las señales de identidad, la relación entre la terapeuta y la cliente, la amenaza a la identidad social, el prejuicio, la personalidad

Although mental illness is similarly prevalent among both racial/ethnic minority Americans and White Americans (Turner et al., 2006), individuals with racial/ethnic minority identities are less likely than White individuals to seek out and receive mental healthcare (Benuto et al., 2018). Racial/ethnic minorities who do seek out treatment are more likely to be misdiagnosed and tend to be unsatisfied with the treatment that they receive (Cook et al., 2017; Liang et al., 2015). Only 12% of practitioners in the health service psychology workforce have racial/ethnic minority identities (Lin et al., 2018), which may contribute to lower levels of mental healthcare­seeking behaviors among individuals with racial/ethnic minority identities. Research on identity cues suggests that a variety of environmental and interpersonal factors, including lack of representation of people with a shared identity, can impact levels of comfort and expectations of treatment in various settings, including healthcare facilities (Cipollina & Sanchez, 2019; Pietri et al., 2018). In the current study, we examined whether the race and personality of a therapist could serve as identity cues for racial/ethnic minority women. We assessed whether the identity and traits of a therapist impact racial/ethnic minority women’s comfort with and desire to see the therapist among other therapeutic relationship outcomes that influence treatment effectiveness.

Identity Cues and Perceptions of Prejudice

Individuals with stigmatized identities rely on identity cues to determine how their identity will be valued in certain situations (e.g., Sanchez et al., 2017). There are two main types of identity cues. Identity threat cues indicate that marginalized group members may be treated poorly or are unsafe in the social environment, whereas identity safety cues indicate that marginalized group members will be treated well. Identity cues can exist within a social environment (e.g., gender­inclusive bathrooms; Chaney & Sanchez, 2018) or can come from other people (e.g., ingroup role models can alleviate identity threat; Chaney et al., 2018).

Identity cues can impact people with stigmatized identities in various contexts. For example, STEM fields tend to be unwelcoming for racial minorities and

women who are underrepresented in the field due, in part, to stereotypes relating to intelligence and math and science abilities (Beede et al., 2011). However, numerous studies have suggested that same­race and same­gender role models can serve as identity safety cues to help buffer against identity threats in the environment (Pietri et al., 2018; Stout et al., 2011). For example, Black female STEM students reported higher expectations of trust and belonging in STEM when they were exposed to Black male and female professors (Johnson et al., 2019). Even role models who do not share the same identity but are similarly stereotyped have proven to be beneficial for women in STEM, as White women perceive Black men as being less likely to endorse negative stereotypes about women’s intelligence than White men (Chaney et al., 2018).

Although the impact of same­race role models in the STEM fields has been studied by numerous scholars, minimal work has examined identities as identity safety cues in the therapeutic context. Research on medical contexts has demonstrated that the presence of identity safety cues can reduce expectations of encountering stereotypes and prejudice; improve expectations of medical visits and feelings of belonging; improve communication quality between the provider and the patient; and reduce vigilance to identity threat cues for patients with a stigmatized identity (Cipollina & Sanchez, 2019). Minority representation identity cues (i.e., diverse staff and clientele) signal to sexual minorities that the medical provider is more culturally competent and less likely to be biased against sexual minorities than medical providers lacking minority representation (Cipollina & Sanchez, 2021).

Maimon et al. (2021) suggested that personality traits can serve as interpersonal identity cues and influence perceptions of prejudice for individuals with a stigmatized identity, such that women expect a disagreeable White man to be more prejudicial and discriminatory than an agreeable White man. Although this work provided preliminary evidence that personality traits can serve as identity cues, it is unknown whether these findings would hold for women and racial minority targets. Both Black and White individuals perceive White men to be more prejudicial than White women on average (Babbitt et al., 2018), and people are more likely to perceive ambiguous behaviors as racist or sexist when perpetrated by White

men (Inman & Baron, 1996). Racial/ethnic minority individuals may rely on expectations of prejudice when interpreting other people’s behaviors (Mills & Gaia, 2012). Therefore, it is important to extend past work and examine targets with different identities (e.g., racial/ethnic minority group members, women, LGBTQIA+ individuals) to learn more about traits and identities that can potentially serve as identity safety cues for marginalized groups. In the current study, we aimed to examine whether interpersonal cues from a therapist (i.e., personality and race) could serve as identity safety cues for racial/ethnic minority women in a therapeutic context.

Therapy Outcomes for Racial/Ethnic Minority Group Members

Although White individuals have historically had a higher reported rate of lifetime prevalence of mental disorders overall than do racial/ethnic minority individuals (Kessler et al., 2005), racial/ethnic minority individuals tend to have worse outcomes, such as a longer course of illness (Breslau et al., 2006). This might have changed during the COVID­19 pandemic, as racial/ethnic minority individuals reported higher rates of depression, substance use, and suicidal thoughts/ideation than did White individuals (McKnight­Eily et al., 2021). Additionally, racial/ethnic minorities might have higher prevalence rates of specific disorders than White individuals. For example, Black/African American and Hispanic/Latinx adults may develop PTSD at higher rates than White adults (Marshall et al., 2009) and may experience more severe PTSD symptoms than their White counterparts (Ortega & Rosenheck, 2000). Black/African American and Hispanic/ Latinx adults are also less likely than White adults to achieve full recovery from PTSD during five years, even when receiving treatment (Sibrava et al., 2019). Additionally, people with racial/ethnic minority identities are more likely than White individuals to be misdiagnosed by their clinicians and are less likely to be satisfied with treatment (Cook et al., 2017; Liang et al., 2015). For example, there is an overpathologizing bias with clients of color, with Black patients being overdiagnosed with psychotic disorders three to four times more than White patients and Latinx patients being overdiagnosed three times more than White patients (Schwartz & Blankenship, 2014). This race ­ based overpathologizing bias is also seen in the diagnoses of PTSD, eating disorders, and schizotypal personality disorder, among other disorders (Cicero, 2016; Garb, 2021). This overpathologizing bias is more common in therapist–client racial/ethnic mismatches than when clients share the same racial/ethnic identity with their therapist. Racially/ethnically matched therapists judged African American, Asian American, and Mexican

American clients to have higher psychological functioning than mismatched therapists did (Russell et al., 1996). Racial/ethnic minority clients also report higher levels of satisfaction and rapport with therapists from a similar racial background than with White therapists (Chang & Yoon, 2011; Meyer & Zane, 2013). Additionally, racial/ ethnic minority clients are more likely to stay in therapy with therapists with a similar racial/ethnic identity and experience better client outcomes than when there is a mismatch in racial/ethnic identity with the therapist (Farsimadan et al., 2007; Naser, 2019).

However, the research examining clients’ preferences for therapist ascribed race has mixed findings. Some studies found that racial/ethnic minority clients prefer therapists with whom they share a racial/ ethnic identity (Farsimadan et al., 2007; Ilagan & Heatherington, 2022), whereas other studies have found that most clients express no preference for a therapist with a particular ascribed race (Sue et al., 1994). Notably, this preference for racial/ethnic identity matching is strongest among Black individuals (Ilagan & Heatherington, 2022). Although there has not been as much recent work on preferences for racial/ethnic matching, a meta­analysis by Cabral and Smith (2011) found that preferences for racial/ethnic matching is more salient for racial/ethnic minorities than for White individuals. Additionally, they found that, although Asian Americans might not express strong initial preferences for a therapist with whom they share a racial/ ethnic identity, they perceive Asian American therapists more favorably than therapists with a different racial/ ethnic identity. On the other hand, Hispanic/Latinx Americans express a strong preference for Hispanic/ Latinx therapists, but their perceptions of therapists do not differ significantly based on whether or not there is a racial/ethnic match (Cabral & Smith, 2011). Therefore, in the present study, we aimed to examine racial/ethnic minority women’s preferences for therapist race, and the impact of therapist race on expected therapeutic relationship outcomes that are central for effective relationship dynamics and treatment outcomes.

Therapeutic Relationship Outcomes

In the therapeutic context, it is particularly important that people feel comfortable disclosing personal and sensitive information. Positive disclosure experiences are linked to better well ­ being for individuals with stigmatized identities, so positive disclosure experiences and disclosure comfort can have implications for therapeutic outcomes (Chaudoir & Quinn, 2010). Additionally, therapists use the information that clients disclose to make diagnoses, plan treatments, and create goals with the client (Carkhuff & Truax, 1965). Therefore,

discomfort with disclosing this information could lead to clients being misdiagnosed or being dissatisfied with the treatment that they are receiving. In fact, greater levels of client self­disclosure are related with better therapy outcomes (Drinane et al., 2018; Love & Farber, 2019). This includes a better therapeutic relationship, with both clients and therapists reporting a weaker working alliance when the client reported keeping a relevant secret (Kelly & Yuan, 2009). Additionally, clients report experiencing relief, pride, safety, and a sense of authenticity after disclosing personal and sensitive information, and most clients believe that withholding this information reduces the effectiveness of therapy (Farber et al., 2004).

In healthcare settings, cultural competence can have important implications for clients and patients. For example, when patients perceive their medical provider to be more culturally competent, they also anticipate a higher quality visit with their provider (Cipollina & Sanchez, 2020). In terms of mental healthcare, studies have shown that therapist effectiveness and treatment outcomes vary significantly between White and racial/ ethnic minority clients in their caseload, which is likely influenced by the cultural competence of the therapist (Imel et al., 2011). Studies have also shown that there is a positive relationship between a therapist’s level of cultural competence and therapy outcomes (Anderson et al., 2019; Rasheed, 2011). Therefore, we were interested in examining cultural competence as both an outcome and as a mediator between the race of the therapist and other anticipated therapeutic relationship outcomes. Relationship genuineness is another contributing factor to therapy outcomes for diverse clients. Clients indicate that perceived genuineness and competence are important for the therapeutic relationship (Jung et al., 2015), which is consistently related to positive therapy outcomes (Flückiger et al., 2012; Horvath et al., 2011). Additionally, researchers have found a direct relationship between relationship genuineness and client change, such that genuineness generally predicts positive therapy outcomes, and can be a cause for client improvement (Orlinsky & Howard, 1987). Therefore, because research has shown the importance of these therapeutic relationship outcomes for therapy outcomes, in the current study, we aimed to examine whether the race and personality of a therapist would influence these therapeutic relationship outcomes (i.e., desire to see the therapist, disclosure comfort, perceived cultural competence, and relationship genuineness) for racial/ethnic minority women.

The Present Study

In the present study, we examined whether the personality traits and ascribed race of a therapist would influence Asian American and Hispanic/Latinx women’s

expectations about the therapeutic relationship and about the therapist’s likelihood of engaging in discriminatory behavior. Given findings from past work that disagreeable men are perceived as more prejudicial and discriminatory than agreeable men, we hypothesized that participants would view a disagreeable therapist as being more prejudiced (i.e., higher in racism, sexism, and heterosexism) and higher in social dominance orientation (SDO) than an agreeable therapist. We predicted that this effect would be stronger when the therapist is a White woman than when she is a Black woman. We also hypothesized that perceived SDO would mediate the relationship between a therapist’s personality and target outcomes and between a therapist’s race and target outcomes. We hypothesized that participants would expect more cultural competence from, report greater disclosure comfort with, expect a more genuine therapeutic relationship with, and indicate more interest in seeing a Black female therapist than a White female therapist. Finally, we predicted that perceived cultural competence would mediate the relationship between a therapist’s race and target outcomes.

Method

Data, syntax, and supplemental materials pertaining to the current study can be found on the Open Science Framework (OSF) at https://osf.io/sjecw/

Participants

After approval was given by Rutgers Institutional Review Board (Pro2019000463), we recruited Asian American and Hispanic/Latinx women (N = 275) through Prolific Academic. Prolific Academic is an online survey recruitment platform that researchers use to recruit people to participate in online studies in exchange for compensation (~$8/hr). Only adult women who identified as Asian American or Hispanic/Latinx were able to participate in the study. We excluded participants from the data who did not meet the inclusion criteria (n = 7), had repeating IP addresses (n = 4), and failed both manipulation checks (n = 4). We had also planned to exclude any participants who viewed the experimental manipulation for less than ⅓ of the median viewing time, had recaptcha verification scores suggesting they were “bots,” or completed the study in less than ⅓ the median completion time, but no participants met these exclusion criteria. The final sample included 260 participants (Mage = 27.61 years, SDage = 9.29 years; 68.8% Asian American, 31.5% Latinx/Hispanic; 76.2% heterosexual, 15.4% bisexual, 3.1% queer, 2.7% lesbian, 2.7% another sexual orientation). We conducted an a priori power analysis for a small to medium effect size and 80% power using G*power to determine a minimum sample size of 244 participants for this study (Faul et al., 2009).

Procedure

Participants who consented to participate in the study were randomly assigned to one of six conditions. The present study employed a 3 (trait: high agreeableness, low agreeableness, control) x 2 (target race: White, Black) between ­ subjects design. Participants read a Psychology Today profile and reviews from previous clients that were manipulated to depict a target therapist as either a Black woman or a White woman. The Psychology Today profiles listed the race/ethnicity of the therapy as “Black” in the Black therapist condition, and as “White” in the White therapist condition. The profiles can be found in the supplemental files on OSF.

We also manipulated the personality traits of the therapist as described in the profile and reviews (i.e., agreeable, disagreeable, no information). Participants viewed a review from a previous client who described the therapist as “attentive and energetic” in all conditions. The review also described the therapist as “disagreeable” in the disagreeable condition, and “agreeable” in the agreeable condition. All client reviews included in the study can be found in the supplemental files on OSF.

Participants then completed a manipulation check to gauge understanding of the profile and reviews they read, followed by a series of open­ended questions about how they believe the described therapist would treat them in a hypothetical interaction. Participants then completed measures of likeability, perceived racism, perceived sexism (Sanchez et al., 2017), perceived heterosexism, perceived SDO (Ho et al., 2015), clientele demographics, and perceived identity of the therapist in a randomized order. Participants also completed measures of therapy interest, relationship genuineness, disclosure comfort, and perceived cultural competence for a hypothetical therapy session with the therapist (Kelley et al., 2010). Finally, participants completed an assessment of their own agreeableness and reported on their therapy experiences before completing demographic measures. At the end of the study, participants read a debrief form, and were compensated $1.50. Although there was no time limit imposed on the study, participants on average completed the survey in 9 mins 50 seconds.

Measures

We reverse­coded items when appropriate and averaged all scale items to create one average score for each measure with higher values indicating greater endorsement of the construct. To keep the study length short and avoid participant fatigue, we created three brief measures for the present work. These measures are similar to existing scales but were adapted for the design of this study.

First Impressions of the Therapist

Open Response Impressions. We asked participants two open­ended questions: “What is your first impression of the therapist?” and “How do you think a therapy session with the therapist would go?”

Perceived Identity. We measured perceived identity with two items. Participants indicated what they believed the therapist’s political ideology and sexual orientation would be.

Likeability. On a scale from 1 (strongly disagree) to 7 (strongly agree), participants completed five items indicating how much they like or would enjoy meeting the therapist depicted in the profile (e.g., “I would get along well with the therapist;” Sanchez et al., 2017).

Anticipated Prejudice

Perceived Prejudice. Participants completed three items indicating the perceived likelihood that the therapist depicted in the profile would discriminate based on race (e.g., “How likely is this person to discriminate based on race/ethnicity?”), three items indicating how likely they believed the therapist would be to discriminate based on sex (e.g., “How likely is this person to discriminate based on sex?”), and three items indicating how likely they believed the therapist would be to discriminate based on sexual orientation (e.g., “How likely is this person to discriminate based on sexual orientation?”) on a scale from 1 (very unlikely) to 7 (very likely; Sanchez et al., 2017).

Perceived SDO. We measured perceived SDO on a scale from 1 (they would strongly oppose) to 7 (they would strongly favor) with eight items (Ho et al., 2015). Participants reported their perceptions that the therapist would favor or oppose hierarchy­endorsing statements like, “an ideal society requires some groups to be on top and others to be on the bottom.”

Therapy-Specific Measures

Therapy Interest. We measured therapy interest with three items on a scale from 1 (strongly disagree) to 7 (strongly agree). Participants indicated their desire to participate in therapy with the described therapist, how much they believed they would benefit from a therapy session with the described therapist, and how likely they would be to recommend the therapist to a friend or family member (e.g., “I would choose to see the therapist over other therapists”).

Relationship Genuineness. We measured relationship genuineness using a four­item modified version of the Real Relationship Inventory (RRI­C) measure (Kelley et al., 2010). On a scale from 1 (strongly disagree) to 7 (strongly agree), participants indicated whether they believed they would be able to be themselves and be open and honest with the therapist during a therapy session (e.g., “I would be able to be myself with the therapist”).

Disclosure Comfort. We measured disclosure comfort on a scale from 1 ( strongly disagree ) to 7 ( strongly agree ). Participants completed four items indicating how comfortable they would be sharing sensitive information with the therapist depicted in the profile (e.g., “I would feel comfortable disclosing personal information to the therapist;” adapted from Landes et al., 2013).

Perceived Cultural Competence. We measured perceived cultural competence with three items on a scale from 1 (strongly disagree) to 7 (strongly agree). Participants indicated how cognizant they believed the described therapist would be of the participant’s culture and cultural differences (e.g., “The therapist would be aware of cultural differences when diagnosing and treating me;” modified from LaFromboise & Coleman, 1991).

Clientele Demographics. Participants indicated what percentage of the therapist’s clientele they believed were White, Black/African American, Hispanic/Latinx, Asian American, or another race/ethnicity.

Participant Information

Participant Agreeableness. Participants completed a measure of their own level of agreeableness using six items (e.g., “I am compassionate, I have a soft heart;” 1 = strongly disagree, 7 = strongly agree; α = 0.76) from the BFI­2­S (; Soto & John, 2017).

Participant Therapy. Participants indicated whether they had attended therapy before (dichotomous response) and when they had their last therapy session (< 6 months ago, 6 months – 1 year ago, 1 year – 2 years ago, > 2 years ago).

Bivariate Correlations of Target Outcomes

4.

Demographics. Participants completed demo

graphic measures of their gender identity, racial/ethnic background, age, and sexual orientation.

Additional Measures. There were a few additional measures included in the study that can be seen in the supplemental materials on OSF.

Results

Preliminary Analyses

Through bivariate correlations, we found that many of the target outcomes were related to one another (see Table 1). All the target outcomes measuring perceived prejudice (i.e., perceived racism, sexism, heterosexism) and perceived SDO were significantly positively correlated to one another, ps < .001. These target outcomes were also significantly negatively correlated with therapy interest, relationship genuineness, perceived cultural competence, and disclosure comfort, ps < .001. Therapy interest, relationship genuineness, perceived cultural competence, and disclosure comfort were all significantly positively correlated with one another, ps < .001. Finally, likeability was negatively correlated with the target outcomes measuring perceived prejudice and meta­SDO, ps < .001, and positively correlated with the anticipated therapeutic relationship variables, ps< .001.

We conducted a 3 (trait: high agree vs. low agree vs. control) x 2 (target race: Black woman vs. White woman) ANOVA to examine differences in likeability by condition. There were significant main effects of trait, F(2, 254) = 18.41, p < .001, ηp2 = .13, and target race, F (1, 254) = 13.01, p < .001, η p 2 = .05, on likeability. However, there was no significant interaction between trait and target race on likeability, F (2, 254) = 0.86, p = .42, ηp2 = .01. Participants perceived both the control therapist (M = 5.56, SD = 0.81) and the highly agreeable therapist (M = 5.39, SD = 0.90) as significantly more likeable than the disagreeable therapist ( M = 4.77, SD = 01.06), p < .001. Overall, participants viewed the Black therapist ( M = 5.44, SD = 0.93) as being significantly more likeable than the White therapist (M = 5.04, SD = 1.01). Therefore, we controlled for likeability in subsequent analyses examining differences in the dependent variables by condition.

Note. SDO = social dominance orientation. All bivariate correlations were significant at p < .01. Cronbach’s alpha values for the measures are reported on the diagonal.

Similarly, we conducted a 3 x 2 ANOVA to examine differences in perceived target political ideology by condition. Although there was no significant main effect of trait, F(2, 254) = 1.65, p = .19, ηp2 = .01, nor a significant interaction between trait and target race, F(2, 254) = 2.05, p = .13, ηp2 = .02, on perceived political ideology, there was a significant main effect of target race on perceived political ideology, F(1, 254) = 23.58, p < .001, ηp2 = .09, such that the White therapist (M = 3.40, SD = 0.99) was perceived as significantly

more conservative than the Black therapist (M = 2.81, SD = 1.02). Therefore, we also controlled for perceived political ideology in subsequent analyses examining differences in the dependent variables by condition.

Additionally, we conducted a 3 (trait: high agree vs. low agree vs. control) x 2 (target race: Black woman vs. White woman) x 2 (participant race: Asian American woman vs. Hispanic/Latinx woman) ANOVA to examine differences in the outcome variables by participant race (see supplemental file for analyses). We did not control for participant race in subsequent analyses.

Participants’ level of agreeableness did not vary significantly across trait conditions, F(2, 254) = 0.60, p = .55, ηp2 = .01, nor target race conditions, F(1, 254) = 0.32, p = .58, ηp2 = .00. Additionally, participants’ therapy experiences did not vary significantly across conditions, Χ2(5) = 4.25, p = .51. Perceived sexual orientation of the therapist did not vary significantly across conditions either, Χ2(20) = 23.62, p = .26. Thus, we did not control for participant agreeableness, participant therapy experiences, nor perceived sexual orientation in subsequent analyses.

Open Response Impressions

The first author first coded the open responses based on whether participants indicated a generally positive, neutral, or negative first impression of the therapist depicted in the profile (see Table 2). More than 85% of participants in the control condition reported positive first impressions of the therapist. Additionally, more than 65% of participants had a positive impression of the agreeable therapist, whereas 50–59% of participants reported a positive impression of the disagreeable therapist. Among participants exposed to the control or agreeable therapist, 89–96% of those who learned about a Black therapist reported a positive first impression, while 68–86% of those who learned about a White therapist reported a positive first impression.

The first author identified different themes in the open responses and coded for the presence of each theme in participants’ responses. Table 3 depicts the breakdown of these themes by condition. Responses that stated something positive or negative about the specific trait of the condition (i.e., agreeable for the agreeable condition, attentive or energetic for the control condition, disagreeable for the disagreeable condition) were coded as either “trait positive” or “trait negative.” Positive and negative statements about the race of the therapist were coded as “race positive” and “race negative,” respectively. If participants indicated that the client reviews were what led the participants to view the therapist positively, the responses were coded as “reviews,” whereas positive perceptions due to the gender of the therapist were coded as “female positive.”

Responses that indicated that a participant would be uncomfortable or comfortable talking to the therapist and at the first therapy session were coded as “disclosure discomfort” and “disclosure comfort,” respectively. Responses where the participants indicated that they believed that the therapist would be warm and friendly were coded as “warm/friendly.”

Less than 5% of participants in the agreeable and control conditions reported trait negative responses, whereas 25–31% of participants in the disagreeable condition had trait negative responses. Interestingly, approximately 11% of participants who viewed a disagreeable therapist gave a trait positive response. Participants only had race positive responses when the therapist was Black, although these responses were given infrequently.

Interestingly, even though we did not find significant differences in disclosure comfort between the different conditions through the quantitative analyses, the open­ended responses seem to suggest that clients have

First General Impressions of the Therapist by Condition

Themes of First Impressions of the Therapist and Their Frequencies by Condition

Identity, Personality, and Prejudice in Therapy | Philip and Maimon

varying levels of disclosure comfort with the different therapists. More participants indicated being comfortable with disclosures in the control (13.95–16.28%) and agreeable conditions (10.64–21.62%) than in the disagreeable condition (0–6.52%). For the control and agreeable therapists, a greater percentage of participants indicated that they would be comfortable with disclosing personal and sensitive information to the Black therapist (16.28–21.62%) than to the White therapist (10.64–13.95%).

Perceived Prejudice and Discrimination

We conducted 2 x 3 ANCOVAs while controlling for likeability and perceived target political ideology to examine differences in perceived racism, perceived sexism, perceived heterosexism, and perceived SDO by condition (see Table 4). There were no significant main effects of trait nor significant interactions between trait and target race for any of these outcomes. There were also no significant main effects of target race on perceived sexism, perceived heterosexism, and

perceived SDO. However, there was a significant main effect of target race on perceived racism, F(1, 252) = 12.68, p < .001, η p 2 = .05, such that the White therapist (Mmarg = 2.81, SE = 0.09) was perceived as more likely to be racist than the Black therapist (Mmarg = 2.36, SE = 0.09). We also conducted 2 x 3 ANCOVAs while controlling for likeability and perceived target political ideology to examine differences in perceived clientele demographics by condition (see Table 5). There were no significant main effects of trait, but there were significant main effects of target race, such that participants perceived the White therapist to have more clients with the same racial background as herself (Mmarg = 66.82%, SE = 1.64) than the Black therapist (Mmarg = 44.66%, SE = 1.34).

Meta-SDO and Perceived Cultural Competence as Mediators

Because we did not find any significant effects of trait or target race on meta­SDO and some of the perceived prejudice measures, we did not examine meta­SDO as a mediator between the therapist’s level of agreeableness and target outcomes and between the therapist’s ascribed race and target outcomes. We conducted mediation analyses using Hayes’ (2013)

PROCESS

Macro for SPSS 26 Model 4 with bias­corrected confidence intervals (CIs) and 10,000 resamples to examine whether perceived cultural competence mediates the relationship between therapist race and the other anticipated therapeutic relationship outcomes. We controlled for likeability and perceived political ideology of the therapist and compared impressions of the White therapist (1) to impressions of the Black therapist (2). There were significant indirect effects of therapist race on therapy interest, B = 0.10, 95% CI = [0.04, 0.17], relationship genuineness, B = 0.13, 95% CI = [0.06, 0.22], and disclosure comfort, B = 0.12, 95% CI = [0.04, 0.20], through perceived cultural competence.

Therapeutic Relationship

We conducted 2 x 3 ANCOVAs while controlling for likeability and perceived target political ideology to examine differences in therapy interest, relationship genuineness, disclosure comfort, and perceived cultural competence by condition (see Table 6). There were no significant main effects of trait on any of the therapeutic relationship outcomes. Additionally, there were no significant main effects of target race on therapy interest and disclosure comfort. There was a significant main effect of target race on relationship genuineness, F(1, 252) = 10.36, p = .001, ηp 2 = .04, such that participants anticipated a more genuine relationship with the Black therapist (Mmarg = 5.08, SE = 0.07) than the

4

White therapist (Mmarg = 4.76, SE = 0.07). We also found a significant main effect of target race on perceived cultural competence, F(1, 252) = 20.04, p < .001, ηp2 = .07, such that participants perceived the Black therapist (Mmarg = 5.13, SE = 0.09) as more culturally competent than the White therapist (Mmarg = 4.57, SE = 0.09).

Discussion

In the present study, we predicted that participants would perceive a disagreeable therapist as more likely to be racist, sexist, heterosexist, and high in SDO than an agreeable therapist. We predicted that this effect would be stronger when the therapist was a White woman than when she was a Black woman. We also hypothesized that participants would anticipate more positive therapy dynamics (i.e., cultural competence, disclosure comfort, relationship genuineness, and therapy interest) with a Black therapist than with a White therapist, and with an agreeable therapist than with a disagreeable therapist.

Importantly, we found significant differences in perceived racism, relationship genuineness, perceived clientele demographics, and perceived cultural competence between the Black therapist and White therapist conditions. Specifically, we found that Asian American and Latinx female clients anticipated a more genuine therapeutic relationship with the Black therapist and expected her to be more culturally competent and less likely to be racist than the White therapist. These findings provide preliminary support that the race of a therapist can serve as an identity safety cue for racial/ ethnic minority clients, such that racial/ethnic minority clients anticipate better therapeutic relationship outcomes with a Black therapist than with a White therapist, even when they do not share the same ascribed race with either therapist. This is similar to what was found about identity safety cues in the medical field, as minority representation cues cause individuals with marginalized identities to anticipate better interactions with and treatment from medical providers (Cipollina & Sanchez, 2019, 2022).

Additionally, we found that the greater perceived cultural competence of the Black therapist compared to the White therapist related to greater therapy interest, disclosure comfort, and expectations of a genuine relationship with the therapist. Consistent with past research on cultural competence in healthcare settings (Cipollina & Sanchez, 2022), these findings demonstrate the importance of a therapist’s cultural competence for racial/ethnic minority clients.

Diversifying the field of psychology is crucial to ensuring that people with diverse identities receive the support they need in a therapeutic context. White psychologists make up a staggering 88% of the health

service psychology workforce, despite the United States population being only 62% White (Lin et al., 2018). The underrepresentation of psychologists with racial/ethnic minority identities could contribute to existing racial disparities in mental health and mental healthcare, as racial/ethnic minority individuals anticipate more racism, less cultural competence, and a less genuine relationship from White therapists than from Black therapists, and these therapeutic relationship factors impact treatment outcomes for clients (Anderson et al., 2019; Orlinsky & Howard, 1987). Therefore, increasing the presence of practitioners with racial/ethnic minority identities in the field could help to improve treatment outcomes and increase comfort in therapy for individuals with racial/ethnic minority identities.

Counter to our predictions, we found no significant differences in the outcomes of interest between the disagreeable therapist and the agreeable therapist. Because female clients tend to prefer to have a female therapist (Ilagan & Heatherington, 2022), and there are more female than male therapists (Lin et al., 2015), participants were asked to evaluate a female therapist in the present study. However, because common ingroup identities have been shown to reduce intergroup threat (Gaertner et al., 1993; Riek et al., 2010), it is possible that female participants’ perceptions of prejudice of the female therapist were reduced in this study and that participants were less attuned to possible cues of threat (i.e., personality) than in previous studies where participants did not share a common identity with the target (i.e., Asian American and Latinx female participants formed impressions of White or Black men; Philip et al., 2022). Additionally, the manipulation of the therapist’s personality was subtle. The therapist’s personality was only mentioned and manipulated in one client review, so participants’ opinions might not have been heavily

Identity, Personality, and Prejudice in Therapy | Philip and Maimon

influenced by one client’s perspective. Participants might have focused more on other aspects of the profile and reviews, such as the identity, experience, and specializations of the therapist, as these topics were frequently brought up in the open­ended responses.

Although it may be undesirable to interact with some people with a disagreeable personality (e.g., a boss, friend, family member), there could be some benefits to having a therapist who is not highly agreeable. In the open responses for the disagreeable therapist condition, some participants noted that they were glad that the therapist was disagreeable, as they believed this would encourage them to be the best version of themselves that they could be. Some participants indicated that having a therapist say what they think clients want to hear and agree with everything clients say would not help clients to improve and better themselves.

Strengths, Limitations, and Future Directions

The present work added to the literature on identity cues, perceptions of prejudice, and the effect of therapist race on anticipated therapeutic relationship outcomes with racial/ethnic minority clients. The current study expanded the literature of perceptions of prejudice to include female targets of different races and suggests that the race of a therapist can potentially serve as an identity safety cue for racial/ethnic minority clients, thus improving psychological health outcomes for these individuals.

One limitation of the current work is that we did not account for social desirability, so there is a possibility that participants responded to the questionnaires in such a way as to not be perceived as being racist, which could have impacted the results. Another possible limitation is that participants might have misunderstood the term “disagreeable” to mean “someone who does not agree with you” instead of the intended meaning of stubborn, demanding, and unsympathetic, as evidenced by their responses to the open ­ ended questions. This could potentially explain why the disagreeable therapist was rated so highly in the open­ended responses and why we did not find the expected differences from the personality manipulation. Therefore, future work should include terms such as “stubborn” or “unsympathetic” instead of “disagreeable” to reduce the risk of participants misunderstanding the trait.

Finally, we focused broadly on participants who self­identified as Asian American or Hispanic/Latinx. It is important to acknowledge that these racial/ ethnic groups are not homogenous and that there are differences in the experiences, treatment of, and even social status of individuals within these racial/ethnic identity groups as well as between them (Teranishi,

2002; Weinick et al., 2004). Therefore, it is possible that perceptions of the therapist differ within the racial/ ethnic groups included in the present study, and these differences may limit the generalizability of these findings to other populations.

Future studies should ask participants directly if they have preferences for the personality of their therapists to get a better sense of whether therapist personality influences the client’s desire to see the therapist among other therapeutic relationship outcomes. Future work could also examine whether preferences for a therapist’s personality differs from preferences for the personality of a friend, boss, coworker, or romantic partner. To extend the present work, future research could also examine whether manipulating other aspects of a therapist’s identity (e.g., gender identity, sexual orientation, ethnicity) impacts prospective clients’ perceptions of prejudice and anticipated therapeutic relationship outcomes, as individuals with dual stigmatized identities may benefit more from having therapists who are similarly stigmatized on multiple identity dimensions (Pietri et al., 2018). Subsequent research should also examine the perceptions of clients with different identities (i.e., Black, multiracial, men, LGBTQIA+), as this work may not generalize to these other identity groups. Additionally, future longitudinal research should examine the retention of racial/ethnic minority clients in therapy with therapists of different identities, as racial/ethnic minority individuals are more likely to drop out of treatment prematurely than White individuals (Cooper & Conklin, 2015).

Although we did not find expected differences in perceptions of prejudice and expectations for the therapeutic relationship based on personality, we did find that the race of a therapist impacted perceptions of racism, anticipated relationship genuineness, and perceived cultural competence among Asian American and Hispanic/Latinx women. This work provided support for diversifying the psychological workforce, which is currently predominantly White, as it demonstrated that representation of racial/ethnic minority group members within the mental healthcare field has important implications for improving therapeutic dynamics (e.g., relationship genuineness and cultural competence) and treatment outcomes for individuals with racial/ethnic minority identities. This could reduce racial disparities in mental healthcare by increasing satisfaction with treatment and treatment retention for racial/ethnic minorities. It is also important to consider how the identities of practitioners can impact the interactions they have with their clients, so future work should explore how likely people would be to seek therapy depending on the identity of the practitioners that are accessible to them.

References

Anderson, K. N., Bautista, C. L., & Hope, D. A. (2019). Therapeutic alliance, cultural competence and minority status in premature termination of psychotherapy. American Journal of Orthopsychiatry, 89(1), 104–114. https://doi.org/10.1037/ort0000342

Babbitt, L. G., Gaither, S. E., Toosi, N. R., & Sommers, S. R. (2018). The role of gender in racial meta-stereotypes and stereotypes. Social Cognition, 36(5), 589–601. https://doi.org/10.1521/soco.2018.36.5.589

Beede, D. N., Julian, T. A., Langdon, D., McKittrick, G., Khan, B., & Doms, M. E. (2011). Women in STEM: A gender gap to innovation. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.1964782

Benuto, L. T., Casas, J., & O’Donohue, W. T. (2018). Training culturally competent psychologists: A systematic review of the training outcome literature. Training and Education in Professional Psychology, 12(3), 125–134. https://doi.org/10.1037/tep0000190

Breslau, J., Aguilar-Gaxiola, S., Kendler, K., Su, M., Williams, D., & Kessler, R. (2006). Specifying race-ethnic differences in risk for psychiatric disorder in a USA national sample. Psychological Medicine, 36(1), 57–68. https://doi.org/10.1017/S0033291705006161

Cabral, R. R., & Smith, T. B. (2011). Racial/ethnic matching of clients and therapists in mental health services: A meta-analytic review of preferences, perceptions, and outcomes. Journal of Counseling Psychology, 58(4), 537–554. https://doi.org/10.1037/a0025266

Carkhuff, R. R., & Truax, C. B. (1965). Training in counseling and psychotherapy: An evaluation of an integrated didactic and experiential approach. Journal of Consulting Psychology, 29(4), 333–336. https://doi.org/10.1037/h0022187

Chaney, K. E., & Sanchez, D. T. (2018). Gender-inclusive bathrooms signal fairness across identity dimensions. Social Psychological and Personality Science, 9(2), 245–253. https://doi.org/10.1177/1948550617737601

Chaney, K. E., Sanchez, D. T., & Remedios, J. D. (2018). We are in this together: How the presence of similarly stereotyped allies buffer against identity threat. Journal of Experimental Social Psychology, 79, 410–422. https://doi.org/10.1016/j.jesp.2018.09.005

Chang, D. F., & Yoon, P. (2011). Ethnic minority clients' perceptions of the significance of race in cross-racial therapy relationships. Psychotherapy Research, 21(5), 567–582. https://doi.org/10.1080/10503307.2011.592549

Chaudoir, S. R., & Quinn, D. M. (2010). Revealing concealable stigmatized identities: The impact of disclosure motivations and positive firstdisclosure experiences on fear of disclosure and well-being. Journal of Social Issues, 66, 570–584. https://doi.org/10.1111/j.1540-4560.2010.01663.x

Cicero, D. C. (2016). Measurement invariance of the Schizotypal Personality Questionnaire in Asian, Pacific Islander, White, and multiethnic populations. Psychological Assessment, 28(4), 351–361. https://doi.org/10.1037/pas0000180

Cipollina, R., & Sanchez, D. T. (2019). Reducing health care disparities through improving trust: An identity safety cues intervention for stigmatized groups. Translational Issues in Psychological Science, 5(4), 315–325. https://doi.org/10.1037/tps0000207

Cipollina, R., & Sanchez, D. T. (2020, September 10). Racial identity safety cues and healthcare provider expectations. Stigma and Health. Advance online publication. http://dx.doi.org/10.1037/sah0000265

Cipollina, R., & Sanchez, D. T. (2021). Identity cues influence sexual minorities’ anticipated treatment and disclosure intentions in healthcare settings: Exploring a multiple pathway model. Journal of Health Psychology, 27(7), 1569–1582.https://doi.org/10.1177/1359105321995984

Cook, B. L., Trinh, N.-H., Li, Z., Hou, S. S.-Y., & Progovac, A. M. (2017). Trends in racial-ethnic disparities in access to mental health care, 2004–2012. Psychiatric Services, 68(1), 9–16. https://doi.org/10.1176/appi.ps.201500453

Cooper, A. A., & Conklin, L. R. (2015). Dropout from individual psychotherapy for major depression: A meta-analysis of randomized clinical trials. Clinical Psychology Review, 40, 57–65. https://doi.org/10.1016/j.cpr.2015.05.001

Drinane, J. M., Owen, J., & Tao, K. W. (2018). Cultural concealment and therapy outcomes. Journal of Counseling Psychology, 65(2), 239–246. https://doi.org/10.1037/cou0000246

Farber, B. A., Berano, K. C., & Capobianco, J. A. (2004). Clients' perceptions of the process and consequences of self-disclosure in psychotherapy. Journal of Counseling Psychology, 51(3), 340–346. https://doi.org/10.1037/0022-0167.51.3.340

Farsimadan, F., Draghi-Lorenz, R., & Ellis, J. (2007). Process and outcome of therapy in ethnically similar and dissimilar therapeutic dyads.

Psychotherapy Research, 17(5), 567–575. https://doi.org/10.1080/10503300601139996

Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

Flückiger, C., Del Re, A. C., Wampold, B. E., Symonds, D., & Horvath, A. O. (2012). How central is the alliance in psychotherapy? A multilevel longitudinal meta-analysis. Journal of Counseling Psychology, 59(1), 10–17. https://doi.org/10.1037/a0025749

Gaertner, S. L., Dovidio, J. F., Anastasio, P. A., Bachman, B. A., & Rust, M. C. (1993). The common ingroup identity model: Recategorization and the reduction of intergroup bias. European Review of Social Psychology, 4(1), 1–26. https://doi.org/10.1080/14792779343000004

Garb, H. N. (2021). Race bias and gender bias in the diagnosis of psychological disorders. Clinical Psychology Review, 90, 102087. https://doi.org/10.1016/j.cpr.2021.102087

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. The Guilford Press.

Ho, A. K., Sidanius, J., Kteily, N., Sheehy-Skeffington, J., Pratto, F., Henkel, K. E., Foels, R., & Stewart, A. L. (2015). The nature of social dominance orientation: Theorizing and measuring preferences for intergroup inequality using the new SDO7 scale. Journal of Personality and Social Psychology, 109(6), 1003–1028. https://doi.org/10.1037/pspi0000033

Horvath, A. O., Re, A. C. D., Flückiger, C., & Symonds, D. (2011). Alliance in individual psychotherapy. In J. C. Norcross (Ed.), Psychotherapy relationships that work: Evidence-based responsiveness (pp. 25–69). Oxford University Press. https://doi.org/10.1093/acprof :oso/9780199737208.003.0002

Ilagan, G. S., & Heatherington, L. (2022). Advancing the understanding of factors that influence client preferences for race and gender matching in psychotherapy. Counselling Psychology Quarterly, 35(3), 694–717. https://doi.org/10.1080/09515070.2021.1960274

Imel, Z. E., Baldwin, S., Atkins, D. C., Owen, J., Baardseth, T., & Wampold, B. E. (2011). Racial/ethnic disparities in therapist effectiveness: a conceptualization and initial study of cultural competence. Journal of Counseling Psychology, 58(3), 290–298. https://doi.org/10.1037/a0023284

Inman, M. L., & Baron, R. S. (1996). Influence of prototypes on perceptions of prejudice. Journal of Personality and Social Psychology, 70(4), 727–739. https://doi.org/10.1037/0022-3514.70.4.727

Johnson, I. R., Pietri, E. S., Fullilove, F., & Mowrer, S. (2019). Exploring identitysafety cues and allyship among Black women students in STEM environments. Psychology of Women Quarterly, 43(2), 131–150. https://doi.org/10.1177/0361684319830926

Jung, E., Wiesjahn, M., Rief, W., & Lincoln, T. M. (2015). Perceived therapist genuineness predicts therapeutic alliance in cognitive behavioural therapy for psychosis. British Journal of Clinical Psychology, 54(1), 34–48. https://doi.org/10.1111/bjc.12059

Kelley, F. A., Gelso, C. J., Fuertes, J. N., Marmarosh, C., & Lanier, S. H. (2010). The real relationship inventory: Development and psychometric investigation of the client form. Psychotherapy: Theory, Research, Practice, Training, 47(4), 540–553. https://doi.org/10.1037/a0022082

Kelly, A. E., & Yuan, K.-H. (2009). Clients' secret keeping and the working alliance in adult outpatient therapy. Psychotherapy: Theory, Research, Practice, Training, 46(2), 193–202. https://doi.org/10.1037/a0016084

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593–602. https://doi.org/10.1001/archpsyc.62.6.593

LaFromboise, T. D., Coleman, H. L. K., & Hernandez, A. (1991). Development and factor structure of the Cross-Cultural Counseling Inventory—Revised. Professional Psychology: Research and Practice, 22(5), 380–388. https://doi.org/10.1037/0735-7028.22.5.380

Landes, S. J., Burton, J. R., King, K. M., & Sullivan, B. F. (2013). Women’s preference of therapist based on sex of therapist and presenting problem: An analog study. Counselling Psychology Quarterly, 26(3–4), 330–342. https://doi.org/10.1080/09515070.2013.819795

Liang, J., Matheson, B. E., & Douglas, J. M. (2015). Mental health diagnostic considerations in racial/ethnic minority youth. Journal of Child and Family Studies, 25(6), 1926–1940. https://doi.org/10.1007/s10826-015-0351-z

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

SPRING 2023

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH

Identity, Personality, and Prejudice in Therapy | Philip and Maimon

Lin, L., Stamm, K., & Christidis, P. (2018, February). How diverse is the psychology workforce? Monitor on Psychology, 49(2), 19.

https://www.apa.org/monitor/2018/02/datapoint#:~:text=In%202015%2C%20 86%20percent%20of,from%20other%20racial%2Fethnic%20groups

Love, M., & Farber, B. A. (2019). Honesty in psychotherapy: Results of an online survey comparing high vs. low self-concealers. Psychotherapy Research, 29(5), 607–620. https://doi.org/10.1080/10503307.201

Maimon, M. R., Sanchez, D. T., & Rodriguez, S. (2021, February 9-14). Personality as an identity cue: Lay perceptions of the link between personality and prejudice [Poster presentation]. Annual Convention of the Society for Personality and Social Psychology, online.

https://meeting.spsp.org/2021/programming/full-schedule

Marshall, G. N., Schell, T. L., & Miles, J. N. V. (2009). Ethnic differences in posttraumatic distress: Hispanics’ symptoms differ in kind and degree. Journal of Consulting and Clinical Psychology, 77(6), 1169–1178. https://doi.org/10.1037/a0017721

Meyer, O. L., & Zane, N. (2013). The influence of race and ethnicity in client’s experiences of mental health treatment. Journal of Community Psychology, 41(7), 884–901. https://doi.org/10.1002/jcop.21580

McKnight-Eily, L. R., Okoro, C. A., Strine, T. W., Verlenden, J., Hollis, N. D., Njai, R., Mitchell, E. W., Board, A., Puddy, R., & Thomas, C. (2021). Racial and ethnic disparities in the prevalence of stress and worry, mental health conditions, and increased substance use among adults during the COVID-19 pandemic - United States, April and May 2020. MMWR. Morbidity and Mortality Weekly Report, 70(5), 162–166. https://doi.org/10.15585/mmwr.mm7005a3

Mills, D., & Gaia, A. C. (2012). Prototype use in perceptions of prejudice in interactions between Whites and Latinos. Psi Chi Journal of Psychological Research, 17(1), 10–18. https://doi.org/10.24839/2164-8204.JN17.1.10

Orlinsky, D. E., & Howard, K. I. (1987). A generic model of psychotherapy. Journal of Integrative & Eclectic Psychotherapy, 6(1), 6–27. https://psycnet.apa.org/record/1988-30326-001

Ortega, A. N., & Rosenheck, R. (2000). Posttraumatic stress disorder among Hispanic Vietnam veterans. American journal of psychiatry, 157(4), 615–619. https://ajp.psychiatryonline.org/doi/full/10.1176/appi.ajp.157.4.615

Pietri, E. S., Johnson, I. R., & Ozgumus, E. (2018). One size may not fit all: Exploring how the intersection of race and gender and stigma consciousness predict effective identity-safe cues for Black women. Journal of Experimental Social Psychology, 74, 291–306. https://doi.org/10.1016/j.jesp.2017.06.021

Philip, J. S., Maimon, M. R., & Sanchez, D. T. (2022, February 16-19). Interpersonal identity cues: The perceived link between identity, personality, and prejudice [Poster presentation]. Annual Convention of the Society for Personality and Social Psychology, San Francisco, CA, United States. https://meeting.spsp.org/programming/full-schedule

Rasheed, M. (2011). South Asian women's experiences in counseling: An exploration of working alliance, multicultural competence, acculturation, and cultural value conflicts (Publication No. 3529251) [Doctoral dissertation, University of North Texas]. ProQuest Dissertations Publishing. https://www.proquest.com/dissertations-theses/south-asianwomens-experiences-counseling/docview/1041250765/se-2

Riek, B. M., Mania, E. W., & Gaertner, S. L. (2008). Does a common ingroup identity reduce intergroup threat? PsycEXTRA Dataset. https://doi.org/10.1037/e506102012-212

Russell, G. L., Fujino, D. C., Sue, S., Cheung, M. K., & Snowden, L. R. (1996). The effects of therapist-client ethnic match in the assessment of mental health functioning. Journal of Cross-Cultural Psychology, 27(5), 598–615. https://doi.org/10.1177/0022022196275007

Sanchez, D. T., Chaney, K. E., Manuel, S. K., Wilton, L. S., & Remedios, J. D. (2017). Stigma by prejudice transfer: Racism threatens White women and sexism threatens men of color. Psychological Science, 28(4), 445–461. https://doi.org/10.1177/0956797616686218

Sibrava, N. J., Bjornsson, A. S., Pérez Benítez, A. C. I., Moitra, E., Weisberg, R. B., & Keller, M. B. (2019). Posttraumatic stress disorder in African American and Latinx adults: Clinical course and the role of racial and ethnic discrimination. American Psychologist, 74(1), 101–116. https://doi.org/10.1037/amp0000339

Soto, C. J., & John, O. P. (2017). Short and extra-short forms of the Big Five Inventory–2: The BFI-2-S and BFI-2-XS. Journal of Research in Personality, 68, 69–81. https://doi.org/10.1016/j.jrp.2017.02.004

Stout, J. G., Dasgupta, N., Hunsinger, M., & McManus, M. A. (2011). Steming the tide: Using ingroup experts to inoculate women's self-concept in science, technology, engineering, and Mathematics (STEM). Journal of Personality and Social Psychology, 100(2), 255–270. https://doi.org/10.1037/a0021385

Schwartz, R. C., & Blankenship, D. M. (2014). Racial disparities in psychotic disorder diagnosis: A review of empirical literature. World Journal of Psychiatry, 4(4), 133–140. https://doi.org/10.5498/wjp.v4.i4.133

Sue, S., Fujino, D. C., Hu, L.-tze, Takeuchi, D. T., & Zane, N. W. (1991). Community mental health services for ethnic minority groups: A test of the cultural responsiveness hypothesis. Journal of Consulting and Clinical Psychology, 59(4), 533–540. https://doi.org/10.1037/0022-006x.59.4.533

Teranishi, R. T. (2002). The myth of the super minority: Misconceptions about Asian Americans. College Board Review, 195(2), 17–21.

Turner, R., Lloyd, D., & Taylor, J. (2006). Physical disability and mental health: An epidemiology of psychiatric and substance disorders. Rehabilitation Psychology, 51(3), 214–223. https://doi.org/10.1037/0090-5550.51.3.214

Weinick, R. M., Jacobs, E. A., Stone, L. C., Ortega, A. N., & Burstin, H. (2004). Hispanic healthcare disparities: Challenging the myth of a monolithic Hispanic population. Medical Care, 42(4), 313–320. https://doi.org/10.1097/01.mlr.0000118705.27241.7c

Author Note

Jessica S. Philip https://orcid.org/0000­0003­4535­5473

Melanie R. Maimon https://orcid.org/0000­0003­3355­2218

Jessica S. Philip is now at the Department of Psychology at Northern Illinois University. Materials, data, and syntax for this study can be accessed at https://osf.io/sjecw/. We have no known conflicts of interest to disclose. Special thanks to Dr. Diana Sanchez for her feedback and support.

Positionality Statement: Jessica identifies as a heterosexual, cisgender Asian woman. Melanie identifies as a queer, cisgender White woman. Both authors acknowledge that their perspectives are influenced by their positions within all these dimensions of identity.

Correspondence concerning this article should be addressed to Jessica S. Philip. Email: jessicasphilip1@gmail.com

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

Replication of the Interpersonal Sunk Cost Effect

Tristan C. Bolinger, Mason A. Ostermiller, and Scott D. Martin* Department of Psychology, Brigham Young University-Idaho

ABSTRACT. Olivola (2018) postulated that the sunk cost fallacy, choosing a less preferred option simply because an original investment was already committed, would be facilitated by interpersonal factors. The prior work had compared sunk cost decisions made individually (intrapersonally) and those made on behalf of someone else (interpersonally) and found the relationship between type of investment and commitment of the sunk cost fallacy to be impacted by both intrapersonal and interpersonal influences. Our study sought to replicate Olivola’s original experimental methods and analyses but advanced his original research question with additional analytic investigation using factorial logistic regression. We found similar results to Olivola, however, we also discovered that committing the sunk cost fallacy was significantly more impacted by the price of the sunk cost rather than the person investing, χ2 (420, N = 423) = 209.41, p < .001, R² = .53, OR = 33.11, p < .001. This study expanded prior research findings on how individual investment decisions are influenced by others’ investments by contrasting this outcome with how the amount that is invested or given might be more important than the giver or the situation. This is an important distinction as future research seeks to find what influences people’s decisions for personal investments of money, time, etc.

Keywords: interpersonal sunk cost, behavioral decisions, sunk cost effect, sunk cost fallacy

Maximizing utility (i.e., profit or benefit) in any given domain through decisive action is crucial to making investment decisions, especially in relation to time and money. Utility in this case refers to whether or not the decision being made would be considered beneficial and profitable. When decisions regarding investments are made (of time, money, etc.), understanding the past, present, and future influences on those decisions is critical. For example, cognitions toward and expectation of maximizing one’s utility significantly influences investment decisions and is often a critical goal (Dawes & Hastie, 2001; Samuelson, 1958). The way time and money are spent is important when making initial investment decisions as well as deciding whether to continue investing substantial resources in the future. A particular fallacy to avoid, yet common to initial investment decisions, is committing

additional resources to a sunk cost (unrecoverable resources already spent). Many factors are crucial in their influence on decisions about investments and when investments should be foregone.

Sunk Cost Fallacy and Its Predictors

The sunk cost fallacy is a specific type of cognitive error linked to investments of sunk costs (either of time or money) and is important to understand and recognize. Imagine a person who has paid a substantial monthly fee for a gym membership but has only attended the gym once in the six months since signing up. Now suppose that, rather than cancelling the membership, this individual reasons that they will use the gym in the future. If instead of recognizing a bad investment (purchasing a membership when unlikely to use it), this individual continues to invest in the membership, reasoning that

Open Materials, Open Data, and Replications badges earned for transparent research practices. Data and materials are available at https://osf.io/qt26b/

Interpersonal Sunk Cost Effect | Bolinger, Ostermiller, and Martin

the cost already sunk into monthly premiums obliges them to continue paying in order to access some imagined future benefit from the investment, they would have committed the sunk cost fallacy. The sunk cost effect does not necessarily lead to fallacious thinking, however. Some gym users may be spurred to use the gym more often, given that they have already paid for access. The fallacy applies when the influence of the sunk cost leads a person to ignore indicators of a bad investment decision and a reasonable outcome to forego further investment. In the case of the gym membership, if the behavior does not change (i.e., improved attendance), it would be a better fiscal decision not to renew the membership. Such decisions are manifest in reality. Many people do in fact continue to pay for gym memberships (or other types of memberships) without utilizing the benefits of such memberships, thereby committing the sunk cost fallacy. Such an error may occur because the context for the decision being made, although critical, is often ignored. As a result, some may fall victim to cognitive fallacies like believing that a decision is independent of other influences, such as the number of resources invested (Arkes & Blummer, 1985; Haita­Falah, 2017). In the case of financial investments that compound over time, it may be perceived that a loss in one investment decision can be made up by additional investments. Wanting to maximize this utility influences one of two decisions: either continue an investment or forgo and choose something different. Regardless of the decision, one may first believe the premise of maximizing utility to be “common sense,” but personal biases in decision­making are not often recognized. For instance, after making a poor investment of time or money and considering whether to continue such investments, people may justify a similar poor choice, believing that “this time things will be different.” Another perspective explaining why one continues to invest unrecoverable resources in a poor decision is to avoid possible future regret over what “may happen” (regret aversion, Wong & Kwong, 2007). In reality, cognitive fallacies are ever present and bear significant influence on decisions, even if unrecognized.

The moment the effect of a sunk cost becomes fallacious is when there are better alternatives, and yet people choose not to stop current investment approaches. The fallacy is not due to the mere presence of alternatives, but only by the presence of a better alternative, such as not renewing a membership that would not be utilized. Tykocinski and Ortmann (2011) believed that the sunk cost fallacy is a form of self­justification and a misapplication of a social norm. They purported that continued investment is one’s attempt to convince oneself or others that the prior resources were not wasted. Ironically, by continuing to invest time and money in

a poor investment, more resources are being wasted in pursuit of progress while there is a low probability of success. Similar to Tykocinski and Ortmann, Arkes (1996) hypothesized that being seen as wasteful is so aversive that decisions contrary to economic interest will be made in order to maintain the appearance of taking full advantage of the benefits from previous investments, even when this is not the case and when the prior investment should be ignored. Here decisions may not be as much about the actual investment, but the social comparative components intertwined within the decision itself (see Garland & Newport, 1991, as well as Tykocinski & Ortmann, 2011).

Additionally, negative emotional reactions following a sunk cost influence one's proclivity to engage in the sunk cost fallacy (Dijkstra & Hong, 2019). According to Dijkstra and Hong, some people may be more likely to keep engaging in resource expenditure because they may feel frustrated or upset. Tykocinski and Pittman (2004) studied how feelings about an investment influence the commitment of the sunk cost fallacy and found that, when feelings of a failed investment suggest something has been lost, the likelihood of the sunk cost fallacy will increase. Through counterfactual thinking, or what­if scenarios, the decision to keep investing is made in hopes of a return on the sunk cost even when it is impossible (Tykocinski & Pittman, 2004).

The sunk cost fallacy has unique individual manifestations, which make studying the application of this effect challenging. Those who engage in this fallacy have been shown to be influenced more by the proportion of resources expended, rather than the absolute amount (Garland & Newport, 1991). Paradoxically, the probability of committing the sunk cost fallacy increases in a situation where costs (financial, for this example) are equal to a higher proportion of total monthly salary, say 15%, when compared to only 5% (Newport, 1991). The implication is that a lower ratio of investment is correlated with a lower likelihood of committing the sunk cost fallacy. One therefore cannot assume that all people react similarly to the same number of resources invested. This makes it difficult to generalize individual sunk cost investment decisions to other people and infer how others would react. Just as the worth of a million dollars to one person may not be the same for another, so is the relationship between the resources expended and associated thoughts and feelings connected to the expense of those resources.

Cunha and Caldieraro (2009) found that the more time one takes in making an initial decision, the more likely that person is to pull out of a sunk cost. The time spent making a decision might illustrate the importance one feels about the investment decision. It is interesting

to consider whether a factor like time taken to make an initial investment would impact the outcome of making investment decisions on behalf of others in relation to committing the sunk cost fallacy.

Interpersonal Sunk Cost Effect

Pronin et al. (2007) studied how theoretical decisions are made on behalf of others or in other people’s situations. Their study investigated how people make investment decisions in the present and in the future for other people. People tend to make the same future decisions for others as they would for present decisions due to a perceived helpful, futuristic mindset. When tasked with making decisions for themselves, however, they more often made different present decisions than decisions about the future. Many current decisions are rewarded in the future and research suggests that people have a better ability to evaluate future outcomes when making choices for others compared to when making future­oriented decisions for themselves (Pronin et al., 2007). More often, personal decisions are made in the short term based on emotion and often lead to lesser, and even poor, decisions influencing engagement in the sunk cost fallacy (Pronin et al., 2007).

One would expect that, even if people have a higher proclivity to make better decisions for others, the sunk cost fallacy would still be a factor in those decisions. Research completed by Rego et al. (2016) found that, even when decisions were made on behalf of another, those whose decision was chosen for them still found it difficult to leave an unwise investment. Even though sunk cost fallacy situations may be less likely when deciding for others instead of oneself; when those decisions turn out to be poor investments, a similar connection keeps others tied to it as is the case when investing in oneself (Rego et al., 2016). Rego et al. (2016) also determined that those who previously invested money and effort into a relationship were more likely to continue in a bad relationship after knowing there was a better alternative, due to the money and effort previously invested in it. These findings suggest a need to differentiate between the sunk cost effect not only being intrapersonal (individual investment), but interpersonal (investment on another’s behalf). Does a sunk cost that someone invests on a person’s behalf influence that person’s decision making, and if it does, are the person’s decisions similar to having personally invested those resources?

Much research has been conducted on the prevalence of the sunk cost fallacy being primarily intrapersonal, but little research has been conducted on interpersonal investments. Olivola (2018) investigated sunk cost effects with interpersonal factors. He suspected

that the sunk cost effect would generalize beyond the intrapersonal domain, hypothesizing that interpersonal sunk costs would increase the proclivity of committing the sunk cost fallacy. He looked at whether investment on another's behalf significantly influenced continuance in a failed investment. He tested his hypothesis by presenting his sample with a scenario where a sunk cost had been incurred either by participants themselves or by someone personally close to them. The participants were then asked whether to continue with the investment, even though it wasn’t positive, or whether they would choose another alternative, thereby avoiding a sunk cost fallacy. Olivola’s results found that people did commit the sunk cost fallacy when considering intrapersonal and interpersonal investments, concluding that the sunk cost effect does generalize beyond the intrapersonal domain and is present in sunk cost situations involving interpersonal relationships (Olivola, 2018).

The Present Study

The American Psychological Association (APA) and Open Science Collaboration have been encouraging the replication of psychological research (Woodell, 2020). The APA has purported that replication is critical for social sciences due to the complexity of understanding human behavior. The Open Science Collaboration completed a replication of 100 experimental and correlational research in 2015 and found that 97% of the novel research had statistically significant results, while the replicated research only found 36% were significant and 47% of replication effect sizes in the 95% confidence interval of the original (Open Science Collaboration, 2015). The Open Science Collaboration concluded that authors or publishers should not gain fame or credibility because of their authority, but by the reproducibility of their methods, procedures, and their results (Open Science Collaboration, 2015).

In the literature review above, we have discussed the importance of identifying strategies implemented for making important investment decisions and discussed predictors related to engaging in the sunk cost fallacy, pointing out how much of the focus in prior research has emphasized the influence of intrapersonal effects. As a way to further an understanding of the antecedents to committing the sunk cost fallacy, we have argued for the need to investigate interpersonal effects on this outcome. Olivola’s work (2018) was novel in this regard and presented interesting findings from interpersonal effects. Given the initial results from Olivola and the aforementioned need for replication of novel findings, the present study investigated the interpersonal effects on committing the sunk cost fallacy and addressed this inquiry following a direct replication of Olivola’s

Bolinger, Ostermiller, and Martin | Interpersonal Sunk Cost Effect

methods and procedure from Study 1c to further understand and articulate the interpersonal sunk cost effect. We hypothesized that the sunk cost fallacy would extend beyond the intrapersonal domain and would be present in sunk cost situations when the cost is incurred or paid by another (interpersonal domain). We hoped that the replication of Olivola’s findings would further elucidate avenues to recognize and forgo negative sunk cost situations.

Method

Participants

Participants included 424 undergraduate psychology students from a private college in the Northwest. One respondent was excluded for noncompletion, for a final sample of 423 participants. More than two thirds of the participants reported their biological sex as women (66.75%), a third reported as men (32.08%), with 1.17% indicating “prefer not to answer.” The average age of participants was 24.84 years old (SD = 8.53). With the primary purpose of this study being to replicate Olivola’s original methodology directly, race and ethnicity classification of participants was not assessed in the current study as this demographic assessment was not included in the measures of Olivola’s study. This omission, while an error, was not motivated by a theory of “color blind research” (Rouse, 2021), supposing invariability of responses among or between individuals of varying racial or ethnic identification, yet the impacts of this omission on Constraints of Generality (COG; see Rouse, 2021) are noteworthy and addressed in the discussion section below. No financial incentive was offered for study completion, although many in our sample had a psychology course with a class assignment for students to participate in at least one research project at some point in the semester.

Procedure

All procedures and measures used were approved by our institution’s institutional review board. Participants completed our study online via Qualtrics (Qualtrics, 2020). Each person presented with our study was given a brief explanation of the purpose of the project, consent documentation, followed by the survey. If participants chose not to consent to participate in the study, their survey ended, and no further information was collected. If they accepted, they continued to the body of the survey. We utilized a two­person condition (person incurring the sunk cost: self vs other) x 2 sunk cost condition (present vs absent) fully between ­ subjects’ factorial design. In the body of the survey, participants were asked to imagine a version of the hotel­movie scenario (Frisch, 1993; Strough et al., 2008), in which participants were to

imagine that, while being on a trip with some friends, they had fallen ill part way through their vacation. As a result, they decided to stay at the hotel and watch a movie, but encouraged their friends to go to a museum, something previously planned for the whole group. In the scenario, before the friends leave, a set condition was presented; a movie is found for the participant to watch that is either free (absent sunk cost) or $19.95 (present sunk cost), and the friend (other condition) or the sick individual (self­condition) decided on the movie and paid the cost. Participants were randomly assigned to one of the four study conditions (self/other incurring the cost) and the nature of the sunk cost (present/absent); see Table 1 for display of study conditions.

Five minutes after the friends left for the museum, participants found the movie to be boring. Participants were then asked to decide whether they would continue watching the boring movie or find something else to watch on TV. After indicating their decision following the situational presentation, participants were directed to the final portion of the survey where they completed demographic questions.

Measures

Person Condition (Predictor 1)

Participants were instructed to imagine that either they would pay the cost, or the friend would pay the cost. Response options included either self or other. In the data, “self” was scored as 1 and “other” was scored as 0. Participants were randomly assigned to a condition.

Sunk Cost Condition (Predictor 2)

Two groups, “present” or “absent,” characterized the sunk cost. Present was characterized as a monetary cost, whereas absent was characterized by no monetary cost (no sunk cost). Our variable was measured “present,” with a movie price of $19.95 USD and a free movie indicated “absent.” Each group was scored as 1 or 0, respectively. As with the person condition above, the sunk cost condition presented was randomized for study participants.

Outcome Decision

Participants indicated whether they would either continue watching the boring film (yes = 1), or find something else on TV (no = 0). A “yes” decision indicates the sunk cost fallacy as participants were explicitly told that the movie was boring. In all study conditions, this was fixed, so regardless of the person condition or cost condition, participants were told the movie was boring.

Analysis Approach

As mentioned above, we replicated Olivola’s (2018) methods from Study 1c. Olivola found evidence to support his hypothesis that the sunk cost effect generalized beyond the interpersonal domain, meaning it was present in both interpersonal and intrapersonal settings. In his approach, he ran a 2 (individual vs. other) x 2 (present vs. absent) x 2 (continue watching vs. change movie) analysis where he combined each study group to determine independence between groups, using a chi­square test of independence. Olivola concluded that the variables were not independent, meaning that the results of the test were in fact dependent on one of the predictors. Although using a chi­square test can determine whether responses are independent of one another, it does not offer further investigation between the responses by cells. Olivola could not explicitly identify which variables were critical to the outcome. Given the expanded contingency table analyzing a 2 x 2 x 2 factorial used in Olivola’s chi­square test, it is challenging to elucidate which specific cells are different from one another, particularly because no post­hoc cell comparison was presented. The analysis approach to do this for a chi­square table is of little utility because its complexity lacks clear interpretability following chisquare tests and lacks widespread use in the literature. In our replication of Olivola’s study and hope to support his findings, we wanted to further understand which predictors had the greatest influence on engaging in the sunk cost fallacy. The chi­square test of independence does not explicitly provide this nor do we feel Olivola could draw specific conclusions regarding this from only running a chi­square analysis. As we used Olivola’s data

and ran a chi­square test, it appeared that the self/other conditions were remarkably similar in their responses, but that the explanation of the result’s variability came from the level of the sunk cost (see Table 2; Figures 1–2). As previously mentioned, there is difficulty with differentiating a chi­square framework with specific factors of influence when there are multiple levels to consider with multiple predictors. Additionally, we believe a more parsimonious model can adequately fit the data from this study. Running a logistic regression analysis would accomplish these goals and allow greater confidence to specify whether the person incurring the cost matters, or if the price of the cost plays the critical role in our commitment of the sunk cost fallacy.

To articulate our analysis approach clearly, we analyzed Olivola’s original data, and the data collected from our study in the same way, by computing and reporting on a logistic regression analysis. We then compared results from Olivola’s findings with results from the present study.

Results

Figures 1 and 2 illustrate the proportion of people who commit the sunk cost fallacy (i.e., refuse to stop watching the boring movie) when controlled for the person investing and for the amount invested. Figure 1 shows a small visual difference between interpersonal and intrapersonal factors. Figure 2 shows an apparently large difference when the variable cost is compared (present vs. absent; see Figure 2; Table 2). Evaluation of the overall model fit statistics for regression analyses on Olivola’s original data, and the present study’s data resulted in strong fit indices, χ2 (602, N = 605) = 332.30, p < .001, R2 = .57 and χ2 (420, N = 423) = 209.41, p < .001, R2 = .53, respectively; see Table 4 for comparison of the model. Related to Figure 2, Olivola found a significant

FIGURE 1

Estimate Plot of Person Predictor Variable

odds ratio for the costs influence on the sunk cost fallacy (OR = 39.7) as did the current study (OR = 33.11). This is a much larger and more significant difference than what was found when the person was isolated. Olivola found a significant odds ratio of 1.6, meaning the intrapersonal investment had higher odds of influencing the sunk cost fallacy than the interpersonal, however, our study did not find a significant odds ratio (OR= 1.3; see Table 3).

Results from our data collection revealed that the predominant factor influencing the likelihood of engaging in the sunk cost fallacy was a cost being present, not who incurred the cost. This finding was slightly different than the findings and conclusions reported from Olivola’s original report and our hypotheses, even though clarifying the specific influence of interpersonal effects was difficult from the analysis utilized in that study (Olivola, 2018). These results were also different from our hypothesis and emphasize the sunk cost fallacy being significantly influenced by interpersonal effects. Results from a factorial logistic regression helped clarify the specific influence from incurring an initial cost on committing the sunk cost fallacy, showing that the presence of a cost was more influential on the probability of committing the sunk cost fallacy and that interpersonal effects were not significant. Potential explanations and further implications are discussed below.

Discussion

As previously discussed, prior research findings suggested that being seen as wasteful is so aversive that investments will be made to “save face” (Arkes, 1996; Tykocinski & Ortmann, 2011). After controlling for the person who invested, study results show this variable made little to no difference in the willingness to commit the sunk cost fallacy. When there was a more substantial price involved (perhaps a situation that sets one up not to be seen as wasteful) participants were more likely to commit the sunk cost fallacy.

Cunha and Caldieraro (2009) concluded that greater time taken prior to making an investment decision decreased the likelihood of committing the sunk cost fallacy. In our research, a decision had already been made for participants, precluding them from evaluating options to choose from and making an informed decision. Understanding the connection between length of time and proclivity to commit the sunk cost fallacy may help clarify why some participants committed the sunk cost fallacy. Furthermore, with the survey being rather short and providing little context, the participants likely did not spend very much time contemplating all the options. Future studies could control for time to help understand whether an effect from this factor exists.

We do not know the financial situations of our participants. However, in line with research by Garland and Newport (1991), the proportion of resources expended is more important than the absolute amount. We saw very little commitment of the sunk cost fallacy when there was no monetary value, and varying responses by participants to continue with the initial decision when there was a cost present. This may suggest that a deeper dive into financial history of participants, and adjusting

FIGURE 2

Estimate Plot of Cost Predictor Variable Olivola Replication

Note. Illustrates sunk cost fallacy committers when price and person are compared separately.

TABLE 3

Comparative Inferential Summary Statistics

the amount of money committed, might help us to better understand whether proportion of wealth, or perceived proportion, might have a meaningful influence on committing the sunk cost fallacy.

It is apparent that individuals are more susceptible to social influence when they are not aware of or lack the knowledge to make a decision related to a specific investment or commitment (Hoffmann & Broekhuizen, 2009). This suggests that some people may be more susceptible to commit the sunk cost fallacy based on their knowledge of the given activity. To use the gym membership example again, one who understands fitness by knowing how often to exercise, what good exercises are, how to balance good nutrition, and etc. might feel tied to their initial investment differently than a more novice gym goer. However, Hoffman and Broekhuizen (2009) might suggest (though their findings were made following a different context) that the individual with more experience and knowledge would back out of investments first as opposed to the novel gym­goer who may simply be attending because of a New Year’s resolution, influence from friends and family, or to maintain a particular image of commitment and/ or activity. Investigating which scenario is more likely to result in committing the sunk cost fallacy would provide an interesting direction for future studies to consider. Social psychology, as a discipline, would argue that behaviors often predict attitudes more often than attitudes predict behaviors (Myers & Twenge, 2015). In application to this study, this suggests that the participants saying how they would act (i.e., stop watching the movie) does not necessarily explain what they would actually do in a real situation. Further research connecting sunk cost decisions in real life by having participants make investment decisions and choose outcomes following these decisions as opposed to simply surveying participants about what they would do in a hypothetical context would be very interesting and something we are interested in following up with.

Conclusion

Generalizability of these results to a broader population will likely be untenable for serval reasons. For example, Olivola’s original design did not control for sex, gender, or race and ethnicity effects on the outcome of interest. In accordance with Rouse (2021), it may not be adequate to assume study context and design to be race/ethnicity (or sex/gender)­neutral. Our analytic plan did not include sex/gender predictors nor race and ethnicity as covariates. Neither did Olivola’s. Our primary intent was to replicate as directly as we could Olivola’s study design and then apply a new analytic model to the investigation of the stated predictors’ effects on the sunk cost fallacy.

Given the alteration to the statistical model we used, it would not have been difficult to include demographic characteristic as predictors, but unfortunately, as mentioned in the Procedure section, we did not gather this data from participants. This creates an opportunity for future research to investigate interpersonal effects on committing the sunk cost fallacy by including various sociodemographic variability. Such research would be interesting and may help respond to constraints on generality (see Rouse, 2021).

Another possible limitation includes the analytic frame we chose to use. Although we argued that using a logistic regression would better identify specific influences on committing the sunk cost fallacy, it was a very parsimonious model, leaving out other potentially important predictors, such as sociodemographics and other considerations as detailed below. Much of this limitation is subject to correction with design and measurement, and we hope future research will address these limitations by including more contextual influences. Future research should focus on future sunk cost investments, meaning any sunk cost decision made in the present that is rewarded or realized in the future. Urminsky (2017) found that the more into the future the sunk cost is the less value is appraised toward that cost. Our replication focused on a decision where the result (finding out the movie was poor, or boring) was immediate. Our study’s procedures may also lack substantial pressure or importance of decision due to the fairly negligible impact associated with choosing to watch or not watch a movie. Weightier consequences for decisions in a controlled research environment, with results being experienced in the future, may cast greater light on the impact future interpersonal sunk costs could have.

One element of these study procedures important to note (Olivola’s and this study’s) is that the individual did not ask the friend to pay for the movie. At least study protocol did not specify this, so it may lead to misinterpretation by participants, but the question is worded to suggest that the friend decided to pay for the movie, not that they were asked to pay for the movie. Would asking someone to pay a sunk cost on one’s behalf, influence one’s willingness to further commit the fallacy after they had agreed to pay? Such questions would add further clarity on interpersonal effects.

Although Olivola included an assessment of the relationship between investor and decision maker (the one committing the sunk cost fallacy), he did not find a significant result (Study 4). Further research needs to be conducted to determine what, if any, influence the relationship between investors has on committing a sunk cost fallacy when someone else has incurred the sunk cost

Bolinger,

SPRING

Interpersonal Sunk Cost Effect | Bolinger, Ostermiller, and Martin

on one’ behalf. For example, would strength of relationship or how close one is to the person investing in their behalf alter a sunk cost fallacy decision? Or, what would results show when investigating the sunk cost fallacy in connection to important relational dyads, such as a father–son, mother–daughter, grandfather–grandson, or friend–friend, compared to stranger–stranger, and etc.? There is still much to investigate and to be said of investment behavior, specifically those related to sunk costs. What we can conclude from our replication and adjustment of the analysis is that many factors influence decisions, especially decisions that follow an initial investment. With continued research needed to determine more causal relationships and findings, it is important to be aware of and alert to the decisions that one makes. Although the specific influence of money, time, and presence/influence of other people may not be fully elucidated, it seems they all play some factor in decision­making. Knowing how these factors influence investment decisions, and knowing that initial cost significantly influences someone, as our study demonstrates, helps people become more aware to ultimately make smarter and wiser investment decisions.

References

Arkes, H. R. (1996). The psychology of waste. Journal of Behavioral Decision Making, 9(3), 213–224.

https://doi.org/10.1002/(SICI)10990771(199609)9:3%3C213::AID-BDM230%3E3.0.CO;2-1

Arkes, H. R., Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140.

https://doi.org/10.1016/0749-5978(85)90049-4

Cunha, J. M., & Caldieraro, F. (2009). Sunk-cost effects on purely behavioral investments. Cognitive Science, 33(1), 105–113.

https://doi.org/10.1111/j.1551-6709.2008.01005.x

Dawes, R., & Hastie R. (2001). Rational choice in an uncertain world: The psychology of judgment and decision making. Sage.

Dijkstra, K. A., & Hong, Y. Y. (2019). The feeling of throwing good money after bad: The role of affective reaction in the sunk-cost fallacy. Plos One, 14(1).

https://doi.org/10.1371/journal.pone.0209900

Frisch, D. (1993). Reasons for framing effects. Organizational Behavior and Human Decision Processes, 54(3), 399–429.

https://doi.org/10.1006/obhd.1993.1017

Garland, H., & Newport, S. (1991). Effects of absolute and relative sunk costs on the decision to persist with a course of action. Organizational Behavior and Human Decision Processes, 48(1), 55–69.

https://doi.org/10.1016/0749-5978(91)90005-e

Haita-Falah, C. (2017). Sunk-cost fallacy and cognitive ability in individual decision-making. Journal of Economic Psychology, 58, 44–59.

https://doi.org/10.1016/j.joep.2016.12.001

Hoffmann, A. O., & Broekhuizen, T. L. (2009). Susceptibility to and impact of interpersonal influence in an investment context. Journal of the Academy of Marketing Science, 37(4), 488–503.

https://doi.org/10.1007/s11747-008-0128-7

Myers, D. G., & Twenge, J. M. (2015). Social Psychology. 12th edition. McGraw-Hill.

Olivola, C. Y. (2018). The interpersonal sunk-cost effect. Psychological Science, 29(7), 1072–1083. https://doi.org/10.1177/0956797617752641

Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349(6251). https://doi.org/10.1126/science.aac4716

Pittman, T. S., & Tykocinski, O. E. (2004). The psychology of economic decisions (Vol. 2). Oxford University Press.

Pronin, E., Olivola, C. Y., & Kennedy, K. A. (2007). Doing unto future selves as you would do unto others: Psychological distance and decision making. Personality and Social Psychology Bulletin, 34(2), 224–236. https://doi.org/10.1177/0146167207310023

Qualtrics XM - experience management software. (2021, February 23). Retrieved March 14, 2021, from https://www.qualtrics.com/

Rego, S., Arantes, J., & Magalhães, P. (2016). Is there a sunk cost effect in committed relationships? Current Psychology, 37(3), 508–519. https://doi.org/10.1007/s12144-016-9529-9

Rouse, S. V. (2021). Increasing the representation of diversity in the Psi Chi Journal. Psi Chi Journal of Psychological Research, 26(4), 360–362. https://doi.org/10.24839/2325-7342.JN26.4.360

Samuelson, P. (1958). Economics: An introductory analysis (4th ed.). McGraw-Hill. Strough, J., Mehta, C. M., Mcfall, J. P., & Schuller, K. L. (2008). Are older adults less subject to the sunk-cost fallacy than younger Adults? Psychological Science, 19(7), 650–652. https://doi.org/10.1111/j.1467-9280.2008.02138.x

Tykocinski, O. E., & Ortmann, A. (2011). The lingering effects of our past experiences: The sunk-cost fallacy and the inaction-inertia effect. Social and Personality Psychology Compass, 5(9), 653–664. https://doi.org/10.1111/j.1751-9004.2011.00369.x

Urminsky, O. (2017). The role of psychological connectedness to the future self in decisions over time. Current Directions in Psychological Science, 26(1), 34–39. https://doi.org/10.1177/0963721416668810

Wong, K. F. E., & Kwong, J. Y. Y. (2007). The role of anticipated regret in escalation of commitment. Journal of Applied Psychology, 92(2), 545–554. https://doi.org/10.1037/0021-9010.92.2.545

Woodell, A. (2020). Leaning into the replication crisis: Why you should consider conducting replication research. American Psychological Association. https://www.apa.org/ed/precollege/psn/2020/03/replication-crisis

Author Note

Tristan C. Bolinger https://orcid.org/0000­0002­2181­1748 Mason A. Ostermiller https://orcid.org/0000­0002­4193­6935 Mason A. Ostermiller is now at the College of Health Sciences at Midwestern University, Glendale, AZ

This replication research project of Olivola (2018) was approved by the BYU­Idaho Institutional Review Board in June 2020. Materials and data for this research can be accessed at https://osf.io/qt26b/. We have no conflicts of interest to disclose. We would like to acknowledge Bradford J. Wiggins for his critique of our manuscript.

Correspondence concerning this article should be addressed to Tristan Bolinger, 737 Michael St. Iowa City, IA 52246. Email: bol16007@byui.edu

COPYRIGHT 2023 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 28, NO. 1/ISSN 2325-7342)

Psi Chi Journal of Psychological Research 2022 Reviewers

We sincerely appreciate the hard work on the part of the following individuals who each completed at least one review in 2022. Without the assistance of such dedicated professionals, Psi Chi Journal would not be able to function.—Steve Rouse (Editor)

Leslie D. Alvarez

University of Denver

Glena Lynne Andrews RAF Lakenheath USAF Medical Center

Lara K. Ault

Saint Leo University

Karl Bailey Andrews University

Michelle Beddow

Saginaw Valley State University

Chris Bjornsen Longwood University

Stefanie S. Boswell

University of the Incarnate Word

Debi Brannan

Western Oregon University

MaryAnne Britt

Northern Illinois University

Sheila Brownlow

Catawba College

Candice Burkett

University of Indianapolis

Alissa A. Call Black Hills State University

Mary Jo Carnot Chadron State College

EB Caron

Fitchburg State University

Kevin R. Carriere

Washington and Jefferson College

Bradley James Caskey Birmingham–Southern College

Shawn Charlton

University of Central Arkansas

Sarah Cronin Bemidji State University

Kristen A. DilibertoMacaluso Berry College

Urska Dobersek

University of Southern Indiana

Martin J. Downing

Lehman College

Ryan C. Ebersole Albany Medical Center

Deb Esty

Baldwin Wallace University

Warren Fass

University of Pittsburgh, Bradford

Keith Feigenson

Albright College

Frank Ferraro

Midland University

Shlomit Flaisher-Grinberg

Saint Francis University

Tifani Fletcher

West Liberty University

Andrew Sheldon Franklin

Norfolk State University

Azenett A. Garza Caballero

Weber State University

James Gigliello St Joseph’s University

Rebecca Gilbertson

University of Minnesota Duluth

Gary W. Giumetti

Quinnipiac University

Heather Haas

University of Montana

Western

Ralph Hale

University of North Georgia

Jonathan J. Hammersley

Western Illinois University

Thomas F. Harlow

University of Maryland

Global Campus

Deborah Harris O’Brien

Trinity Washington University

Marya T. Howell-Carter

Farmingdale State College

Shadab F. Hussain

National Institutes of Health

Robert Hymes

University of Michigan–Dearborn

Lori E. James

University of Colorado

Colorado Springs

Tammy J. Jechura

Albion College

Nabil J. Jondi

Hebron University

Marla Jorgensen

Farmingdale State College

Damian Kelty-Stephen State University of New York at New Paltz

Allen H. Keniston

University of Wisconsin—

Eau Claire

Lisa M. Korenman

United States Military Academy

David S. Kreiner

University of Central Missouri

Antonio V. Laverghetta

Saint Leo University

Timothy J. Legg

University of North Dakota

Lisa K. Lockhart

University of the

Incarnate Word

Richard Lopez

Bard College

Gina Mariano

Troy University

Diego Martinez

University of Arkansas

Tammy McClain

West Liberty University

Tara Mitchell

Lock Haven University

Betsy L. Morgan

University of Wisconsin–La Crosse

Beverly J. Myers

Stillman College

Lydia Njoroge

Morehead State University

Susan L. O’Donnell

George Fox University

Janelle S. Peifer

University of Richmond

Marilyn Petro

Nebraska Wesleyan University

Amy Posey

Benedictine College

Dee Posey

Washington State University

Patricia Prunty

Blackburn College

Chelsea Robertson

West Liberty University

Lisa Rosen

Texas Woman’s University

Raylene Ross

University of South Carolina

Kristen A. Rost

Troy University

Donald Saucier

Kansas State University

Bryan Saville

James Madison University

Christina S. Sinisi

Charleston Southern University

Paul C. Smith

Alverno College

Kathryn Sperry

Weber State University

Crystal N. Steltenpohl

Dartmouth Institute for Health Policy and Clinical Practice

David B. Strohmetz

University of West Florida

Jill Swirsky

Holy Family University

Leonell Torres-Pagan

Pace University

Dunja Trunk

Bloomfield College

Mary Utley

Drury University

Scott VanderStoep

Hope College

Taylor Wadian

University of Cincinnati

Blue Ash College

J. Austin Williamson

Augustana College

Caitlin Wong

Wantagh UFSD

William D. Woody

University of Northern

Colorado

Jill Yamashita

California State University, Monterey Bay

Evan L. Zucker

Loyola University

New Orleans

Publish Your Research in Psi Chi Journal

Undergraduate, graduate, and faculty submissions are welcome year round. Only one author (either first author or coauthor) is required to be a Psi Chi member. All submissions are free. Reasons to submit include

• a unique, doctoral­level, peer­review process

• indexing in PsycINFO, EBSCO, and Crossref databases

• free access of all articles at psichi.org

• our efficient online submissions portal

View Submission Guidelines and submit your research at www.psichi.org/?page=JN_Submissions

Become a Journal Reviewer

Doctoral­level faculty in psychology and related fields who are passionate about educating others on conducting and reporting quality empirical research are invited become reviewers for Psi Chi Journal. Our editorial team is uniquely dedicated to mentorship and promoting professional development of our authors—Please join us!

To become a reviewer, visit www.psichi.org/page/JN_BecomeAReviewer

Resources for Student Research

Looking for solid examples of student manuscripts and educational editorials about conducting psychological research? Download as many free articles to share in your classrooms as you would like.

Search past issues, or articles by subject area or author at www.psichi.org/journal_past Psi Chi is the International Honor Society in Psychology. Membership is primarily open to undergraduates, graduate students, transfer students, full­time and part­time faculty members, and alumni.

Learn About Psi Chi

See membership benefits and a link to apply at www.psichi.org/page/member_benefits

Register an account: http://pcj.msubmit.net/cgi-bin/main.plex