Hello everyone!
I am conducting a study on the intention to pay a service after applying three different treatments.
Respondents will first be asked whether or not they pay for the service. Then they will be shown a stimulus. After the stimulus they will be asked again about their intention to pay for the same service in the future.
Therefore, I will have three variables:
-Pretest: whether they pay for the service before (dichotomous variable).
-Treatment: which of the three treatments was applied (categorical variable: 1,2,3).
-Posttest: whether they pay for the service afterwards (dichotomous variable).
I read that I can use McNemar's test, but I would like to know if with a logistic regression I could use all three variables, and know if the treatments produce a change in payment intention and which treatment is better.
I would also like to know if other factors (age, ethical level, etc.) can be included to determine if they influence the change in intention to pay.
Best regards.
I think you could use generalized mixed effects models (GLMM) with a binomial distribution, logit link function, and participant ID as a random effect in order to account for the repeated measures. Then timepoint (pre or post test)*treatment, age, ethical level, etc. could all be added as fixed effects.
Thanks Lisa K. Chinn
Do you know any source (book, article...) I can consult about GLMM? and how to use it in spss?
Thsnaks again for your help.
Diana Serrano I have only ever done GLMM in R using the lme4 package (https://www.rdocumentation.org/packages/lme4/versions/1.1-27.1/topics/glmer), so I can't help with specific details regarding GLMM in SPSS. I know GEE can be done in SPSS and can also handle a repeated-measures design with a binary dependent variable.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing