February 2008 Poll - Discussion
Type and Party Membership
S-N and T-F preferences seem related to a person's political self-concept as liberal, moderate, or conservative. But, are type preferences related to one's choice of a political party?
Consulting Pscyhologists Press' construction of the Form M collected information about type and political party preference that provides some interesting insights into this question. In particular, the following hypotheses seem reasonable based on this analysis:
One of the objectives of the Type & Politics Poll is to try and replicate these results. Thus, in the November 2008 Poll we examined statistically whether or not the following variables would be associated with party identification (Republican, Democratic, Independent): E-I, S-N, T-F, J-P, and political orientation (liberal, moderate, conservative). We used a technique called Nominal Regression using the sample of APTi only members (with password) as well as the total set of responses (for exploratory purposes and to enhance statistical analysis). Results for the November Poll indicated that poll members with preferences for S, T, and J were more likely to identify as Republican than Democrat. Results also indicated those with preferences for T and J were more more likely to identify with Independent than Democrat. The results showing T & J associated with Republican moreso than Democrat seem consistent with CPP's findings.
These results, however, were produced without considering political orientation in the analysis. When political orientation was included along with the type preferences, only political orientation (liberal, moderate, conservative) was a significant predictor of party identification. As discussed in the analysis of the November 2007 Poll, these results are similar to results found in a sample of Italian voters using the Five-Factor Model measures as well as a measure of values. That is, the personality variables were found to relate to political party identification, but not as strongly as the measures of personal values. Thus, it seems possible that type preferences may influence political orientations which then are more closely associated with party identification.