I have a doubt about hierarchical regression analysis.

My dependent variable is Presenteeism (the act of attending work while sick). Originally that variable has 4 levels (Has it happened to you that you had gone to work despite thinking that you should stay at home because of illness? categories: never, 1 time, 2-5 times, more than 5 times. The first two categories indicate 'no presenteeism', last two categories indicate 'There is presenteeism').

Then, I recoded that variable into 0 and 1 value (0 - no presenteeism, 1 - presenteeism). Maybe I shouldn't have done that?

My main doubt is: can I use hierarchical linear regression when my dependent variable is dichotomous and should it stay as 4-categorical variable or should it be dichotomous?

Or I MUST use logistic regression?

  • $\begingroup$ Something like ordinal logit or its relatives makes sense for your 4 level variable and logit and its relatives make sense for your 2 level response. EIther or both could fit your research goals. But why regard 1 time as fundamentally different from 2-5? What makes this hierarchical? $\endgroup$ – Nick Cox Jan 24 '16 at 11:01
  • $\begingroup$ Hierarchical - because I have two blocks of predictors of Presenteeism. Thanks for reply. So, linear regression can not be used for such type of data? $\endgroup$ – Hayley Marshall Jan 24 '16 at 11:07
  • $\begingroup$ Why would you want to apply linear regression to such responses? Neither is measured or counted. Both are categorical. Some people would defend the so-called linear probability model on a binary (dichotomous) response, but I think that's a minority view. Modern texts on categorical data analysis discuss this. $\endgroup$ – Nick Cox Jan 24 '16 at 11:13
  • $\begingroup$ For best learning and to draw the best answers, avoid broad "should I?" and "must I?" questions. Instead, consider what will be the consequences of certain choices, then ask for more info about those consequences if you have specific questions. $\endgroup$ – rolando2 Jan 24 '16 at 15:52

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.