I want to analyze how the difference between men and women (dichotomous independent variable) in a dichotomous dependent variable (daily consumption vs. lesser consumption) has changed over the long-term (39 annually repeated cross-sectional surveys).
one approach would be calculate the difference in the proportion of daily consumption between men and women for each year and do a regression of these differences on time (year)
James Leigh definitely describes a possible approach. From what you describe, it is not clear to me whether that would be the best method.
To be able to advise on this, please provide more information:
Why are you interested in this question?
Why does it matter how the difference between men and women changed over time?
Do you have one or more testable hypotheses?
If not:
- Is this intended as exploratory, and possibly hypotheses generating, research?
- Or do you aim for conclusions? If so: what do you see as potential conclusions?
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS