I have a bit of rudimentary question and am very new to SPSS and statistics in general so I apologise in advance if anything I say is unclear or doesn't make sense - I'll try to clarify as much as I can!
I have created and run an experiment with 3 Independent variables - Appeal Type (as expressive or instrumental); Financial Information (as present or excluded); and Pivot probability (as high or low) - and two dependent variables - voter behaviour (with 3 scales as expressively, instrumentally or financially motivated) and affective system usage (with 2 scales describing the use of a voter's disposition or surveillance system).
My understanding thus far is that I am required to treat my two DVs, with 3 and 2 dimensions respectively as FIVE DVs in order to run 3-way ANOVAs to describe the difference in means between groups and determine if my IVs have significant effects on each of the five scales. The trouble is, this would result in a boat load of hypotheses testing for main and interaction effects between my IVs and DVs which a) I'm not interested in exploring all of and b) will simply take up far too much space and time in the write up of the results and hypotheses for effects I am not concerned with.
The main objective of the study is to observe the differences in the mean scores across the three scales (for the DV of voter behaviour) and across the two scales (for the DV of affective system) when exposed to each and combinations of the IVs.
It is my understanding that the 3-way ANOVA tells me if the IVs have a significant impact on each scale individually, but I'm not terribly interested in this. What I would like to observe is, for example, if the financial information IV will result in significantly higher mean scores for financially motivated behaviours than for the scales describing expressively and instrumentally motivated behaviours.
I am also interested in the interaction effects between the IVs, for example, I would like to observe the effects of financial information and high pivot probability on financially motivated behaviours as compared to expressively and instrumentally motivated behaviours (which are all measured with independent scales). What sort of test (or tests) does this involve?
I hope I have explained my intentions clearly but please let me know if you would like me to explain anything further at all and I will be happy to do so! Again, I apologise for my basic understanding of the statistical tests involved and am looking forward to your input!
Thank you!!
