I am making my first steps in log-linear analysis. I am at a very basic level, so sorry for any error on my part as far as the terminology and/or intepretation is concerned.
I would like to analyze a three ways contingency table, with the aim to explore the relation between three categorical variables:
v63: community size (3 levels: <50, 50-199, >200)
v767: conflicts in local communities (3 levels: Rare, Moderate, High)
v769: conflict management (2 levels: Informal, Formal)
Among the various SPSS outputs, I have focused on (see attached picts):
1) the goodness of fit table (pict 1): which, if I am not mistaken, should inform about the "validity" of the model. Since it is NOT significant, the hypothesis of a significant departure between the observed data and the model can be rejected (=the model fit the data). So, in my case, there should be a good fit between model and data.
2) the tables in the attached pict 2. If I am not wrong, the analysis is indicating (upper table) that there is a significant two-way (K=2) interaction(s) somewhere in the data. These can be located (lower table) in the association between v63 and v769, v63 and v767. (significant=their removal would have a dentrimental effect on the model).
Now, I would like to have some guidance on the interpretation of the attached pict 3 and 4, and on what conclusions can be drawn from the analysis.
I guess that the analsis have eliminated the interactions that does not add a significant improvement to the model (pict 4), and that the analysis is eventually suggesting the most important (?) interaction(s) (pict 3).
But I would have more info from you on my (possibly wrong) understanding.
Besides, is there any follow-up analysis to perform?
Thanks for any suggestion you will kindly provide, and thanks for your attention.
You do not have the required permissions to view the files attached to this post.