I am analysing a set of data which have both within-subject and between-subject variables. However, since my assigned task is just simply focusing on part of this data set, I am not interested in examining the within-subject variable.
To make you understand it easier, here is an example:
The original study is to examine how the branding of the product (independent variable) affects participants' happiness (dependent variable), and all participants are required to try both brand A and brand B product. However, extending from this original study, I was asked to examine how participants' demographic may actually affect their happiness instead.
My questions are:
How can I solve these problems using SPSS? Can anyone help? Thank you
I'm not sure I fully understand your design and analytic goals but it sounds like multilevel (hierarchical linear) modeling could be used to address the issue(s). Multilevel modeling takes into account the nested data structure (non-independence of observations) and allows you to focus on either within or between level independent variables (or to include both at the same time).
If your research question(s) is/are completely addressed by the between-Ss factors, then I would say YES in response to your first question. But I imagine that even in that case, reviewers and editors will probably want you to report all of the usual results from your mixed design ANOVA. And I think that's fair enough, given that readers may have other questions beyond your research question(s).
I agree with @Bruce Weaver because a complete combined between and within ANOVA allows you to test the generality (or lack thereof) of between group effects across values of your within group variable. Such ANOVA is manageable within SPSS.
- Badges
- Science topic
- Similar topics
- Psychology