Supplementary Materials for Jeffrey A. Karp and Susan A. Banducci. "Political Efficacy and Participation in Twenty Seven Democracies: How Electoral Systems Shape Political Behavior"

To verify that our indicator of efficacy from the CSES is indeed tapping into external efficacy and not internal efficacy, we can compare the question with other items that are intended to measure either internal or external efficacy. The deposited CSES data set does not include any of the traditional internal efficacy measures. However, if we use the original election studies that contain items tapping internal efficacy and other external efficacy items, we can compare the CSES item to these other questions. Of those election studies that participated in the CSES, Australia, the Netherlands, New Zealand, and the United States each included additional efficacy items necessary for this supplementary analysis. We have used the available items tapping internal and external efficacy in combination with the CSES item to test whether the CSES items fits with external or internal efficacy dimension. The results of a factor analysis performed on external and internal efficacy items from four national election studies that were part of the CSES project can be viewed here. In all four election studies, the CSES item loaded most strongly with the other items measuring external efficacy. In New Zealand, the factor loading is not strong for the CSES item but it does load more strongly on the external efficacy dimension. Therefore, there is clear evidence that the CSES item measures an aspect of external efficacy.

The following tables were estimated when using the question, "Are you close to a political party?". The results are similar to those reported in the paper which relies instead on the party likes and dislikes scale as a measure of party preference. The advantage of the party likes and dislikes measure is that three quarters of the respondents evaluate one party more highly than another. In contrast, only a minority of the sample claim to be close to a political party.

For further inquires, contact Jeffrey Karp at