This dissertation investigates the consequences of fractional dynamics for political modeling. Using Monte Carlo analyses, Chapters II and III investigate the threats to statistical inference posed by including fractionally integrated variables in bivariate and multivariate regressions. Fractional differencing is the most appropriate tool to guard against spurious regressions and other threats to inference. Using fractional differencing, multivariate models of British politics are developed in Chapter IV to compare competing theories regarding which subjective measure of economic evaluations best predicts support levels for the governing party; egocentric measures outperform sociotropic measures. The concept of fractional cointegration is discussed and the value of fractionally integrated error correction mechanisms are both discussed and demonstrated in models of Conservative party support. In Chapter V models of presidential approval in the United States are reconfigured in light of the possibilities of fractionally integrated variables. In both the British and American case accounting for the fractional character of all variables allows the development of more accurate multivariate models.

This study develops and tests a group conflict model as an explanation for international immigration beliefs in the United States and Canada. Group conflict is structured by evaluations concerning group relationships and group members. At a conceptual level group conflict explains a broad range of policy beliefs among a large number of actors in multiple settings. Group conflict embodies attitudes relating to objective-based conditions and subjective-based beliefs.