Suchergebnisse

AbstractMultinomial logit (MNL) differs from many other econometric methods because it estimates the effects of variables upon nominal, not ordered outcomes. One consequence of this is that the estimated coefficients vary depending upon a researcher's decision about the choice of a reference, or "baseline," outcome. Most researchers realize this in principle, but many focus upon the statistical significance of MNL coefficients for inference in the same way that they use the coefficients from models with ordered dependent variables. In some instances, this leads researchers to report statistics that do not reflect the correct quantities of interest and reach flawed conclusions. In this note, I argue that researchers need to orient their approach to analyzing both the substantive and statistical significance of predicted probabilities of interest that match their research questions.

The academic debate concerning public opinion about war focuses upon two explanations: cost/benefits and partisan cues. Both sides of this debate use laboratory experiments to estimate the influence of events and cues, but Gelpi is notable for using a well-designed experiment to compare the theories simultaneously. He argues that his results support the cost/benefits explanation as "surprising events" that counter individuals' prior attitudes have significantly more effect than "surprising opinions" upon people's attitudes toward the Iraq War. His analysis, however, considers only the direction, but not the strength, of people's attitudes toward the war. Additionally, the measure of source credibility for determining the influence of cues is not optimal. When the analysis accounts for attitude strength and uses a better measure of source credibility, the results show little support for the effect of surprising events and markedly greater support for the influence of partisan cues.

The Institutional Revolutionary Party (PRI) dominated Mexican politics for over 70 years, but its fortunes have been sliding in the past two decades, capped by its loss of the presidency in 2000. The results for the PRI were even worse in 2006, with a third place finish in the presidential race and significant losses in both houses of the legislature. This paper examines whether or not the PRI risks going from being a dominant party to a third party in a Mexican two-party system. The results suggest that the PRI has fallen behind, but that it may have a limited window to regroup based upon comparative regional advantage in the northern and southern states. Adapted from the source document.

The 2001 Legislative Yuan elections in Taiwan were the first to be held following the change in control of the presidency that occurred after the 2000 elections. The transition from a one-party dominant system to one where the opposition had control of the government provides a good opportunity to examine the prevalence of democratic attitudes in Taiwan. Democracy can be frustrating when control over public policy is divided between two opposing groups. Under these circumstances, it could be reasonable for voters to desire a return to authoritarianism government. In a survey conducted following the 2001 Legislative Yuan elections, Taiwanese citizens were asked a series of questions about democratic norms. In this article, I analyze these attitudes as a means of exploring how Taiwan's voters feel about democracy as Taiwan moves away from a one-party dominant system and faces potential difficulties operating within a multi-party semi-presidential system.

Research in political science is often concerned with modeling dependent variables that are proportions. Proportions are relevant in a wide variety of substantive areas, including elections, the bureaucracy, and interest groups. Yet because most researchers rely upon an approach, OLS, that does not recognize key aspects of proportions, the conclusions we reach from normal models may not provide the best understanding of phenomena of interest in these areas. In this paper, I use Monte Carlo simulations to show that maximum likelihood estimation of these data using the beta distribution may provide more accurate and more precise results. I then present empirical analyses illustrating some of these differences.