This is my first research assignment and I feel overwhelmed and confused.

I have been informed by my supervisor to do a hierarchical multiple regression using SPSS. I noticed that my dependent variable (DV) is ordinal. Is a hierarchical multiple regression correct or should I perform an ordinal (logistics) regression.

Research question: What proportion of variance in health-promoting behaviour is explained by the linear combination of predictors (independent variables specified below):

The (DV) = Health-promoting behaviour (Measured by HPLP 4-point Likert scale; 52 items) Independent variables (IV) 1) Worldview: Measured by OMPI 2-point - 26 items 2) Self-efficacy: Questionnaire 4-point Likert - 13 items 3) Psychological distress: HOP-25 4 point Likert - 25 items 4) Gender 5) Study year (First year, Second year, third year) 6) Race (African, Coloured, White, Indian, Other)


1 Answer 1


If the DV is ordinal, as yours is, you should do ordinal logistic regression.

However, ordinal logistic regression can also be hierarchical and multiple: Those terms refer to the number of independent variables and how they are entered into the regression. "Multiple" means there are more than one IV and "hierarchical" means they are entered into the equation in a "hierarchy" - that is, in a specific order.

  • 2
    $\begingroup$ I would add that in some fields "hierachicical"=random effects (or mixed effects) models. Again you can have mixed effects ordinal regression. But from the description not clear where the random effect is. $\endgroup$
    – charles
    Commented Nov 18, 2013 at 1:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.