Contrary to media reports, California's 2003 recall election was anything but a circus. Despite the presence of 135 candidates, just 3 managed to split 94% of the vote, and the winner came close to achieving a majority. In this article, the author uses elite interviews and a social network analysis of campaign donations to study how the Republican Party sought to impose order on the potentially chaotic political environment. The author finds that a network of Republican donors, activists, and officeholders coordinated their efforts to advantage Arnold Schwarzenegger and pressure other Republicans out of the race.

Since the Civil War, the president's party has lost seats in the U.S. House of Representatives in all but three midterm election cycles. Many attribute this pattern to “balancing” by moderate voters who prefer a Democratic Congress when Republicans control the White House, and vice versa. Although a number of scholars have tested the balancing hypothesis, the debate remains unsettled. We argue that the U.S. states provide an excellent way to analyze the issue further. Similar to the national government, states feature an executive and (in most cases) a bicameral legislature whose members come up for election in both gubernatorial and state midterm years. If voters balance, one should observe such behavior in state elections when an executive's partisanship is known and a legislative choice is necessary. We examine state legislative elections from 1978 to 2009 and find evidence consistent with the balancing hypothesis.

What led elites in some U.S. states to surrender policy-making power to voters between 1898 and 1918, while leaders elsewhere retained only representative democracy? The authors argue that progressives behaved as strategic politicians by supporting direct democracy when they were stymied at achieving their goals in the legislature and were confident that the voters who would be empowered by initiatives that agreed with progressive policies. They made their delegation of power conditional on who would receive it. The presence of these underlying conditions made adoption of the citizen initiative likely, the authors posit, while the timing of reforms came when insurgent reformers had a strong presence in state government, when the results of a galvanizing election sent a clear signal, or when the adoption of the initiative in one state diffused to its neighbors. Exploring these hypotheses by analyzing a new data set, the authors find strong support for their expectations about the conditions that created fertile ground for direct democracy.

One common critique of direct democracy posits that minority rights are endangered by institutions like ballot initiatives and referenda. Empirical research testing this claim, however, has produced conflicting results that leave the question of direct democracy's effect on minority rights open to debate. This study extends previous research by providing a more direct test of this criticism—it compares anti-minority policy proposals from direct democracy states to similar proposals from states without direct democracy institutions. The author examines both ballot proposals and traditional legislative bills to account for both the direct and indirect effects of direct democracy. Analyzing anti-minority proposals from all 50 states from 1995 to 2004 shows that direct democracy states are more likely to pass these proposals than states without direct democracy institutions.

U.S. state politics researchers often analyze data with observations grouped into clusters. This structure commonly produces unmodeled correlation within clusters, leading to downward bias in the standard errors of regression coefficients. Estimating robust cluster standard errors (RCSE) is a common approach to correcting this bias. However, despite their frequent use, recent work indicates that RCSE can also be biased downward. Here the author provides evidence of that bias and offers a potential solution. Through Monte Carlo simulation of an ordinary least squares (OLS) regression model, the author compares conventional standard error (OLS-SE) and RCSE performance to that of a bootstrap method that resamples clusters of observations (BCSE). The author shows that both OLS-SE and RCSE are biased downward, with OLS-SE being the most biased. In contrast, BCSE are not biased and consistently outperform the other two methods. The author concludes with three replications from recent work and offers recommendations to researchers.