I have a DV ( with three non ordered categories) and one IV ( with two non ordered categories)
Moreover, i have age (with ordered categories), qualification ( with non ordered categories) and gender ( binary category) as moderating variables.
My question is : how should I check moderation?
Thanks in advance
Dear Rafique! The problem is very simple if the dependent variable (DV) is dichotomous then go for simple logistic regression, if the DV is polytomous then go for multinomial logistic regression. If the categories of DV are ordered then go for ordinal logistic regression. However, you need to test characteristics of distributions (esp. DV) lest they may be normal. Usually logistic regression assumes that sample distributions are IID rather than Gaussian. You may use simple spreadsheet application (a.k.a. Excel, Calc, Gnumeric) to study the sample distributions.
Ti use gender as a moderator variable, you can create new variable which multiplies gender with the specific variable (e.g age).
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression