Explain Individual Differences
Lesson
Goal During this Stage
We want to try to characterize why some Subjects make different Decisions than others. Thus, we are going to use demographic factors and/or dispositional psychological factors to try to predict these differences.
How to Achieve this Goal
Common Dispositional Psychological Factors to Consider
Note
I’m not an expert on the psychometrics of any of these psychological factors: I’ve just provided these to give you something to start with and think about in terms of how you want to predict behavior.
Personality - Big Five Personality Inventory or HEXACO Personality Inventory
Morality - Moral Foundations Questionnaire
Individualism/Collectivism - Auckland Individualism-Collectivism Scale (AICS)
Dark Triad - Short Dark Triad (SD3)
Social Dominance - Social Dominance Orienation Scale
Emotional Intelligence - Rotterdam Emotional Intelligence Scale
Approach/Avoidance Tendencies - BIS/BAS Scale
Our answer to this question depends entirely on what we want to focus on: this table should help you determine what you may want to use. We’re not going to bother using conceptual examples, all implemented examples are shown below.
1 Free Parameter |
2+ Free Parameters |
Cluster or Bin |
|
|---|---|---|---|
1 Binary Variable |
2 Sample t-test (Ex. 1) |
Cluster Strength Analysis (Ex. 2) |
Chi-Square (Ex. 3) |
1 Categorical Variable with 3+ Levels |
One-Way ANOVA (Ex. 4) |
Cluster Strength Analysis (Ex. 5) |
Chi-Square (Ex. 6) |
1 Continuous Variable |
Correlation (Ex. 7) |
Matrix Correlation (Ex. 8) |
Logistic Regression (Ex. 9) |
2+ Categorical Variables |
Multiple Regression (Ex. 10) |
Cluster Strength Analysis (Ex. 11) |
Logistic Regression (Ex. 12) |
Any 2+ Variables |
Multiple Regression (Ex. 13) |
Matrix Correlation (Ex. 14) |
Logistic Regression (Ex. 15) |
Any 2+ Continuous Variables |
Matrix Correlation (Ex. 16) |
Matrix Correlation (Ex. 17) |
Cluster Strength Analysis (Ex. 18) |
Miscellaneous Examples
Example 19: Using Weighted Averages of Free Parameters over Conditions as Predictors of Preference-Relevant Atittudes