Analyzing the IDMS data Make sense of the data. Look for causal relationships. Always keep in mind that analysis is precise – each finding tells you one thing – so you need to be careful not to misuse data. In the same way that wrong diagnosis by a doctor to a
patient’s illness could lead to illness, impairment, or worse, death; wrong analysis of the community’s current situation could result to plans that do not cater to the needs or the community nor address their real and most urgent concerns. Here are some of the common techniques in analysis that you can use:
Techniques in Data Analysis Trends Trends can be up and down, linear or exponential, steady or fluctuating, seasonal or random and there can be changes at a defined rate. Check out your data. Do you see apparent direction or trends when looking at the data on: • Level of knowledge about disaster? • Disability type and access to services? • Involvement in community activities by age? • Household income?
Up and down vs. Flat Linear vs. Exponential Steady vs. Fluctuating Seasonal vs. Random Rate of Change vs. Steepness
Comparison Comparisons can be based on ranking, measurements, range, context, relationships. Do you see a difference or similarity in data such as: • Gender and access to services? • Gender and knowledge about disaster? • Family costs for disability according to types of disability? Patterns A pattern is a series of data that repeats in a recognizable way. Does your data show: • Clear connection or relationship between or among the indicators? You may look at the disaster knowledge level according to the level of educational attainment. • Gaps such as lack of service access and knowledge about their rights? • Outliers? There might be a specific segment of the population that has zero knowledge about disasters, for example. 45
IDMS GUIDEBOOK
Categorical comparison and Proportions RANKING: Big, small, medium MEASUREMENTS/VALUES: Absolutes Range and Distribution CONTEXT: Targets, forecasts, averages Hierarchical Relationships
Exceptions/outliers Intersections Correlations Connections Clusters Associations Gaps