Exploring Some Analytical Characteristics of Finite Mixture Models

Finite mixture models have become increasingly prevalent in criminology over the past two decades. Yet there is no consensus about the appropriate criterion for model selection with finite mixture specifications. In this paper, we use simulation evidence to examine model selection criteria. Our focus is on mixture models for event count data like those often encountered in criminology. We use two indices to measure model selection performance. First, we examine how often each criterion chooses the correct specification. Then, we investigate how closely the finite mixture models selected by these criteria approximate the true mixing distribution used to simulate the event count data. We consider three sets of simulations. In the first set, the underlying model is itself a three component Poisson-based finite mixture model. In the two other sets of simulations, the underlying distribution of the Poisson rate parameter follows a continuous distribution. The analysis shows that both AIC and BIC perform well under certain sets of circumstances likely to be encountered by criminologists.

Going, going, gone: the effects of aid policies on graduation at three large public institutions

This paper exploits uniquely detailed data and cross-institution variation in aid for three large public universities to identify the effects of aid on the probability of college graduation. The results indicate that need-based and merit-based aid both increase graduation rates at large public institutions, but primarily through the types of students that ‘select’ these institutions. Merit-based aid facilitates an institution attracting students who have higher observed academic ability that raises the probability of graduation. Need-based aid enables an institution to attract students with non-academic attributes such as social and cultural networks that, while often unobserved, improve graduation success. Broadly, our results suggest that recent aid policy that has moved away from need-based aid for low-income students (reducing their ability to find the best institutional match) and toward merit-based aid (that alters the distribution of high ability students across colleges) could foster stagnant graduation rates even with rising enrollment rates that have been observed over the last three decades.

Ability, Effort, and Performance Among First-Year Law Students at Brigham Young University

This study considers the relationships between first-year law students' admission credentials, the amount of time they spend in study, and the grades they receive on examination. Findings include that there is a significant drop in effort during the first year, that while effort invested in study pays off in improved grades this effort is much less significant in explaining grades than is student ability as measured by LSAT and undergraduate grades, that students in the middle and bottom of the class are helped more by substantial study than are those in the top, that class attendance is much more valuable in raising grades than is equivalent time in other study, and that none of the various study techniques examined could be linked with major differences in results.