The direct initiative process, often referred to as a gun behind the door, provides an incentive for legislators to pass legislation more in line with voters' wishes. Concomitantly, legislative procedures such as the filibuster and executive veto often impede the ability of the legislature to pass policies. We explore the tension between these two forces by incorporating legislative procedures and initiative proposal into a spatial model of the policymaking process. We find that the ability to propose initiatives sometimes breaks legislative gridlock, but that other times pivotal players may prefer the initiative outcome and therefore prevent the legislature from preempting a ballot measure. In particular, we show that initiative use increases with the distance between pivotal actors and the median voter. An empirical analysis of initiative use in the American states provides support for this prediction.

In early work on women in Congress, scholars consistently identified a tendency among women legislators to be more liberal roll‐call voters than male copartisans. Recent changes in Congress point to the polarization of women, where Democratic women remain more liberal than Democratic men but Republican women are no different from, or more conservative than, Republican men. We use newly available state legislative roll‐call data to determine whether women state legislators are more liberal or polarized than male copartisans. We find that while Democratic women state legislators remain consistently more liberal than male copartisans in most state chambers, Republican women legislators are growing more conservative. Thus, women state legislators are increasingly polarized in most U.S. states. Legislator replacement and increasing polarization among state legislators in office contribute to this effect. We argue that polarization among women legislators has implications for the representation of women in the states.

No abstract

Duration data are often subject to various forms of censoring that require adaptations of the likelihood function to properly capture the data generating process, but existing spatial duration models do not yet account for these potential issues. Here we develop a method to estimate spatial duration models when the outcome suffers from right censoring, the most common form of censoring in this area. In order to address this issue, we adapt Wei and Tanner’s (1991) imputation algorithm for censored (nonspatial) regression data to models of spatially interdependent durations. The algorithm treats the unobserved duration outcomes as censored data and iterates between multiple imputation of the incomplete, i.e., right censored, values and estimation of the spatial duration model using these imputed values. We explore performance of estimators for Weibull and log-normal durations in the face of varying degrees of right censoring via Monte Carlo and provide empirical examples of its estimation by analyzing spatial dependence in states’ entry dates into World War I.

No abstract