Age Estimation

Age Estimation (GN, 2016, Systems and Models)

Which gender is better at age guessing?

I am conducting an experiment on age estimation and performed it on separate gender and age groups.

According to the Webster dictionary, age is the length of time that a person has lived or a thing has existed. The Census Bureau programs defines age as the length of time in completed years that a person has lived.

When we meet someone for the first time, we form an opinion of that individual in just a few seconds (The impact of visual and auditory cues in age estimation (Gladwell, 2006)). Age estimation is an important part in this process. Age is a dimension on which individuals categorize others rather automatically. Cues to age are perceived from visual cues such as facial morphology, hair color, and also from nonverbal and verbal aspect of individuals’ communication. “Recent studies suggest that people are able to determine a person's age by approximately five years, from looking at a picture of their face,” (The impact of visual and auditory cues in age estimation). Upon presentation of those cues age is readily perceived and can shape our interactions.

Our society is obsessed with youth and anti­aging. We are growing up in a culture glamorizing youth.

Forbes’ 30 under 30, the New Yorker's’ 20 under 40 and, TIME’s 30 under 30 presents youth as some sort of badge of honor. These proliferation of “under 30” lists perpetuate unhealthy views on aging.

They are also problematic because of their negative toll on the way the under 30 may view and set their achievement standards. We are continuously bombarded with messages urging us to join alliance against the war on aging. “​ The market research firm Global Industry Analysts projects

that a boomer­fueled consumer base, "seeking to keep the dreaded signs of aging at bay," will push the U.S. market for anti­aging products from about $80 billion now to more than $114 billion by 2015, ” (​ Shapiro, Lila. "Boomers Will Be Pumping Billions Into Anti­Aging Industry”).

Age shapes different facets of our lives. ​ Just as the racial and ethnic composition of the American

population is shifting, the average age is as well. According to the World Factbook of the Central Intelligence Agency, the national median age is 37.8 years old. “In 2008, the first of the Baby Boomers (the 76.8 million people who were born between 1946 and 1964) reached age sixty­two and qualified for Social Security benefits; in 2011, they reached sixty­five and qualified for Medicare.” (American Government: Roots and Reform, 14).

The changes and issues above hinder relationships between younger and older people. They could potentially pit younger people against older people. This is why it is important to understand the way way in which we perceive age and our relations with it.

My experiment arose from George PA and Hole GJ’s experiment measuring “Factors affecting the accuracy with which adults could assess the age of unfamiliar male faces aged between 5 and 70 years were examined. In the first experiment twenty­five 'young' adult subjects, aged 16­25, and twenty­five 'old' adults, aged 51­60, were used. Each subject saw five versions of three different faces: these consisted of an original version of each face and four manipulated versions of it. The manipulations consisted of mirror reversal, pseudo­cardioidal strain, thresholding, and elimination of all but the internal features of the face. The second experiment as similar except that a between­subjects design was used: each subject saw three faces for each age category of target face, but was exposed to only a

single type of manipulation (plus a set of 'original' faces which were identical for all groups, so that the comparability of the different groups in age estimation could be checked). Results from both experiments were similar. Age estimates for unmanipulated 'original' faces were highly accurate, although subjects were most accurate with target faces that were within their own age range. Results for the manipulated faces implied that the importance of cardioidal strain as a necessary and sufficient cue to age may have been overestimated in previous reports: subjects' age estimates were accurate when cardioidal strain was absent from the stimulus, and poor when cardioidal strain was the only cue available.” (Factors influencing the accuracy of age estimates of unfamiliar faces).

For this observational experiment I was interested in seeing which gender could most accurately age estimate. I was interested in letting people pinpoint how old people were. Many have experienced this awkward encounter where you ask someone how old they are and they respond: How old do you think I am? My many awkward encounters inspired me.

For the purpose of this experiment the following words are defined as follow:

­ Subject: a ​ person or thing that is being discussed, described, or dealt with. In the case of this

­

experiment, the thing being discussed are the faces and the ages of the male and female shown to participants.

Participant: a person who takes part in something. People who took part in this experiment.

Hypothesis:

I think that there will not be a difference observed between the genders’ abilities to accurately age guess/estimate.

Materials:

­ ­ ­ ­ ­

Pencil Time keeping device (Phone) Data sheet Images of female and male being shown for age estimation Confidence (in order to get participants to partake in your experiment)

Images of female and male participants were asked to age estimate!

For this experiment, I needed a location with plenty of people. I conduct my experiment at Harold Washington College due to its convenience. Harold Washington College is a part of the City Colleges of Chicago. It is located at 30 East Lake street. It has approximately 110,000 students each year.

­ Students demographics: 7% Asians, 36% Black, 38% Hispanic, and 16% White.

This experiment could be replicated at any other location with similar subjects. I created a data sheet for making observations that had the following categories:

1. Group/alone: This category notes whether or not the subject was alone or in group of friends or colleagues.

2. Time taken to look at the image 3. Gender 4. Race 5. Age

In this experiment there was no age or nationality restrictions for participants. All participation was voluntary. Participants received auditory instructions from me. At the start of the experiment, all participants provided information about their age, gender, and race/ethnicity. The picture shown were approximately 8 by 10 inches. Participants wrote the female/male estimated age, and rated their judgement on a scale from 1 to 5, which indicated whether they felt very confident (5) about their answer or not (1). At the end of the experiment, participants were asked 2 questions: Did you find any of the two person harder to age estimate than the other? What information did you use while age estimating? The answers from this questionnaire were used during my result evaluation. The experiment took approximately 5 minutes or less for each participant. This experiment was

Procedure:

1. Took the red line train heading in the south bound direction towards 95th/Dan Ryan. I got off at State/Lake station. It was about 3:45 when I arrived.

2. With copies of the data sheets and the images of the subjects being age estimated, I took place at the entrance of Harold Washington College.

3. I introduced myself to students and staff members of the college. “Hello, my name is GN. I am currently a senior at Global Citizenship Experience High School. I am conducting an experiment on people’s ability to age estimate. Do you have a minute or two to partake in my experiment?”

4. Ask participants to age estimate both the female and male pictures I showed them.

5. Asked a participant to rate themselves on a scale from 1 to 5 on how confident they felt about their estimation.

6. Asked participant: Did you find any of the two person harder to age estimate than the other?

What information did you use while age estimating? I recorded their responses.

7. Thanked participants for their participation. Repeated steps 3 to 6. I conducted my experiment on 47 participants.

Data:

Results:

● ● ● ● ●

Both genders are equally good at age estimating in general.

Both gender are equally good at guessing the age of male.

Women are better than men at guessing the age of females.

It is easier to guess a man’s age.

The highest accuracy of guesses occurred when the photo was looked at for 30 seconds and above.

Analysis:

Female amounted to 48.9% of my participants. Of the 47 participants, 23 identified as females. In this chart, blue represents the accuracy rates for the female participants to the female subject’s age. The female subject was 28 years old. The Pink represents the accuracy rate for the female participants to the male subject’s age. The male subject was 24 years old.

Male amounted to 51.1% of the experiment. There were 24 males. In this graph, the blue represents 8.33%. I used the same images/participants). Pink represents 47.6% accuracy rate within male subjects.

❖ Gender composition of participants: 51.1% male and 48.9% female. ❖ Racial breakdown: ­ Black: 31.9% ­ Hispanic: 36.2% ­ Asian: 12.8% ­ White: 19.1%

❖ 59.6% of participants were alone when asked to age estimate ❖ 40.4% of participants were in a group when asked to age estimate. ❖ Average time taken to look at the images 35.8 seconds. ❖ Average age guessed for the male subject was 23 years old. The male subject was 24 years old. ❖ Average age estimated/guessed for the female subject was 26 years old. ❖ Average subjects’ age was 25 years old. ❖ 18­26 wa the age group with the highest accuracy rate

3 ways to verify that my data is valid:

­

­ ­

The ​ observer​ ­expectancy effect is a form of reactivity in which a researcher's cognitive ​ bias causes them to unconsciously influence the participants of an experiment. I tried to account for this while conducting my experiment by diversifying my participant pool through age and race.

Documented my process. Documentation allows me to also communicate my thoughts to others who could validate my logic.

I surveyed my participants at the end of the experiments, questioning their reasoning and how comfortable they felt about their decisions. This prevented me from having any assumptions about their reasoning/logic, thus reducing that observer­expectancy effect previously mentioned.

Conclusion:

My hypothesis was incorrect. Females had higher rates of accuracy than males. The results show that there are factors that drastically affect the accuracy of the genders to estimate/guess age. The data showed a correlation between the time spent looking at the images and the accuracy of your estimation. One thing I would do differently is have more subjects.

Works cited:

­ Census. "Age and Sex." ​ - People and Households​ . The United States Census Bureau, 1 Apr. 2010. Web. 28 Feb. 2016. <​ http://www.census.gov/population/age/​ >.

-

C.I. A. "World Factbook - Median Age." ​ Central Intelligence Agency​ . Central Intelligence Agency, n.d. Web. 28 Feb. 2016.

<​ https://www.cia.gov/library/publications/the-world-factbook/fields/2177.html​ >.

-

George, P. & Hole G. (1995). Factors influencing the accuracy of age estimations of unfamiliar faces. Perception.

-

Gladwell, M. (2006). Blink: The Power of Thinking without Thinking. New York: Time Warner Book Group.

­

Gregoire, Carolyn. "Here's Everything That's Wrong With Our 'Under 30' Obsession." ​ The Huffington Post​ . TheHuffingtonPost.com, 20 Feb. 2014. Web. 01 Mar. 2016.

<​ http://www.huffingtonpost.com/2014/02/20/why-30-under-30-lists-mis_n_4791178.html​ >.

-

Han, Hu, Charles Otto, and Anil K. Jain. "Age Estimation from Face Images: Human vs.

Machine Performance." ​ 2013 International Conference on Biometrics (ICB)​ (2013): 1-8. 4-7 June 2013. Web. 28 Feb. 2016.

<​ http://www.cse.msu.edu/biometrics/Publications/Face/HanOttoJain_AgeEstimationFaceIma ges_HumanvsMachinePerformance_ICB13.pdf​ >.

-

"Result Filters." ​ National Center for Biotechnology Information​ . U.S. National Library of Medicine, n.d. Web. 01 Mar. 2016. <​ http://www.ncbi.nlm.nih.gov/pubmed/8552458​ >.

-

Richeson, Jennifer A. "A Social Psychological Perspective on the Stigmatization of Older Adults." ​ A Social Psychological Perspective on the Stigmatization of Older Adults​ . U.S. National Library of Medicine, 2006. Web. 28 Feb. 2016. <​ http://www.ncbi.nlm.nih.gov/books/NBK83758/​ >.

-

SABATO, O'CONNOR &. ​ American Government: Roots and Reform​ . Twelfth Edition ed. N.p.: Pearson Education, 2014. Print.

­

Shapiro, Lila. "Boomers Will Be Pumping Billions Into Anti-Aging Industry."​ The Huffington Post​ . TheHuffingtonPost.com, 20 Aug. 2011. Web. 01 Mar. 2016.

<​ http://www.huffingtonpost.com/2011/08/20/boomers-anti-aging-industry_n_932109.html​ >.

-

"Take This Quiz And See If You Can Guess Random People's Ages (It's For Science)."

Co.Exist​ . N.p., 28 May 2014. Web. 28 Feb. 2016.

<​ http://www.fastcoexist.com/3030752/take-this-quiz-and-see-if-you-can-guess-random-peo ples-ages-its-for-science​ >.