Hi,
I came up with this below hypothesis and don’t know which analysis method I should take on in SPSS.
My hypothesis is “Liberals are more tolerant compared to conservatives and moderates across four groups and further, liberals with group 1 scores the highest tolerance”.
I have political orientations as an ordinal variable (1=right, 4=neutral, 7=left).
I am just not sure if I need to do one-way anova or regression model by making tolerance as DV, turning the four groups into three dummy variables with one reference group and controlling for political orientations.
if political orientation is represented by Likert scale and tolerance too, you can use simple linear regression, assuming, that dependent variable is normally distributed
It would be more reasonable to assume that (Y-1)/6 is beta-distributed.
Why don't you simply force a decision between "left" and "right" political orientation? If a group is really neutral, it would average out, and any average degree of political orientation will be estimable from the binary data (given a sufficiently large sample size). So one could use a simple and easy-to-interpret logistic model, estimating the odds of favoring one of the political orientations.
It sounds like you want to test an interaction effect between your two independent variables, so that the effect of liberal versus conservative varies, according to which of the four groups they are in. If so, the most straightforward way to assess this kind of hypothesis might be ANOVA -- depending on how your variables are scored.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression