My main interest is to see how well people can identify the reasons behind their decisions. To measure this, I have collected 2 types of data: one set of variables that measures how important people think given factors are, and another set that measures the same factors in real situations when a decision was made. I can analyze the second half by building a mixed-effect logistic regression model using 11 predictor variables and a binary outcome variable in a repeated measures design (because there are 10 measurements per participant per variable).

How should I determine the degree to which people’s insight corresponds with their behavior?

  • $\begingroup$ I share your interest in the topic, but this seems more like a question on research methods in behavioral economics (and a very broad question at that) than about statistics. $\endgroup$ – rolando2 Sep 27 '19 at 10:49
  • $\begingroup$ Well, maybe you are right, but I think my question makes sense from a statistics point of view: what are my options to quantify the relationship between the insights and behavior? I thought about getting separate betas per participant and checking their correlation with the insight measurements. I just thought this would be too specific a question to ask, and maybe people here would come up with something better without me influencing them. $\endgroup$ – hnandi Sep 27 '19 at 12:14
  • $\begingroup$ It’s possible my BE article at yellowbrickstats.com will be a help. $\endgroup$ – rolando2 Sep 28 '19 at 20:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.