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.

A spatial analysis of green teams: a tactical response to marijuana production in British Columbia

Over a period of 4 years (1997–2000), British Columbia (BC) experienced tremendous growth in the illicit production and distribution of domestically grown marijuana. By the close of 2000, each policing jurisdiction in BC had adopted a particular policy in response to grow operation proliferation. In summary, four policy responses were noted. First, some maintained the status quo wherein enforcement of police initiated investigations and citizens’ tips continued, but with no additional resources specifically dedicated to grow operations. Second, some jurisdictions suspended the majority of investigation and enforcement of grow operations. Third, some agencies implemented or reinforced existing resource intensive drug squads, which focused on trafficking, sales and production of all types of drugs. Finally, some of the jurisdictions formed specialized tactical units known as “green teams” that focused solely on the enforcement of marijuana production. In this paper, we evaluate the effectiveness of green teams using a Geographic Information System (GIS) and difference-in-difference estimates. The results indicate that green teams decrease grow operations within their target area without significant displacement to surrounding areas.

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.

Detection of visible and latent fingerprints using micro-X-ray fluorescence elemental imaging

Using micro-X-ray fluorescence (MXRF), a novel means of detecting fingerprints was examined in which the prints were imaged based on their elemental composition. MXRF is a nondestructive technique. Although this method requires a priori knowledge about the approximate location of a print, it offers a new and complementary means for detecting fingerprints that are also left pristine for further analysis (including potential DNA extraction) or archiving purposes. Sebaceous fingerprints and those made after perspiring were detected based on elements such as potassium and chlorine present in the print residue. Unique prints were also detected including those containing lotion, saliva, banana, or sunscreen. This proof-of-concept study demonstrates the potential for visualizing fingerprints by MXRF on surfaces that can be problematic using current methods.