Spatial point pattern data describes locations of events observed over a given domain, with the number of and locations of these events being random. Historically, data analysis for spatial point patterns has focused on rejecting complete spatial randomness and then on fitting a richer model specification. From a Bayesian standpoint, the literature is growing but primarily considers versions of Poisson processes, focusing on specifications for the intensity. However, the Bayesian literature on, e.g., clustering or inhibition processes is limited, primarily attending to model fitting. There is little attention given to full inference and scant with regard to model adequacy or model comparison.The contribution here is full Bayesian analysis, implemented through generation of posterior point patterns using composition. Model features, hence broad inference, can be explored through functions of these samples. The approach is general, applicable to any generative model for spatial point patterns.The approach is also useful in considering model criticism and model selection both in-sample and, when possible, out-of-sample. Here, we adapt or extend familiar tools. In particular, for model criticism, we consider Bayesian residuals, realized and predictive, along with empirical coverage and prior predictive checks through Monte Carlo tests. For model choice, we propose strategies using predictive mean square error, empirical coverage, and ranked probability scores. For simplicity, we illustrate these methods with standard models such as Poisson processes, log-Gaussian Cox processes, and Gibbs processes. The utility of our approach is demonstrated using a simulation study and two real datasets.

Group-based interventions have been developed for treating patients across a range of health conditions. Enrollment into such groups often occurs on an open (or rolling) basis. Conditional autoregression modeling of random session effects has been proposed to account for the expected correlation in session effects associated with the overlap in patient participation session to session. However, when the analytic objective is to examine the relationship between a fixed-effect session feature and a patient outcome using conditional autoregression, confounding might arise if the fixed session feature of interest and the random session effects vary across sessions in similar ways, resulting in bias and inflated standard errors of a fixed-effect session feature of interest. Motivated by the goal of examining the relationships between outcomes and the session features of leader and session module theme, we applied restricted spatial regression to the analysis of patient outcomes collected from 132 participants in an open-enrollment group for treating depression among patients of a residential alcohol and other drug treatment program, adapting the approach to the multilevel data structure of open-enrollment group data. As compared with standard conditional autoregression, the restricted regression approach resulted in more precise estimates of regression coefficients of the module theme and leader predictor variables. The restricted regression approach provides an important analytic tool for group therapy researchers who are investigating the relationship between key components of open-enrollment group therapy interventions and patient outcomes.

Background: Driving under the influence (DUI) is a significant problem, and there is a pressing need to develop interventions that reduce future risk.Methods: We pilot-tested the acceptance and efficacy of web-motivational interviewing (MI) and in-person MI interventions among a diverse sample of individuals with a first-time DUI offense. Participants (N = 159) were 65 percent male, 40 percent Hispanic, and an average age of 30 (SD = 9.8). They were enrolled at one of three participating 3-month DUI programs in Los Angeles County and randomized to usual care (UC)-only (36-h program), in-person MI plus UC, or a web-based intervention using MI (web-MI) plus UC. Participants were assessed at intake and program completion. We examined intervention acceptance and preliminary efficacy of the interventions on alcohol consumption, DUI, and alcohol-related consequences.Results: Web-MI and in-person MI participants rated the quality of and satisfaction with their sessions significantly higher than participants in the UC-only condition. However, there were no significant group differences between the MI conditions and the UC-only condition in alcohol consumption, DUI, and alcohol-related consequences. Further, 67 percent of our sample met criteria for alcohol dependence, and the majority of participants in all three study conditions continued to report alcohol-related consequences at follow-up.Conclusions: Participants receiving MI plus UC and UC-only had similar improvements, and a large proportion had symptoms of alcohol dependence. Receiving a DUI and having to deal with the numerous consequences related to this type of event may be significant enough to reduce short-term behaviors, but future research should explore whether more intensive interventions are needed to sustain long-term changes.

Little is known about the effect of group therapy treatment modules on symptom change during treatment and on outcomes post-treatment. Secondary analyses of depressive symptoms collected from two group therapy studies conducted in substance use treatment settings were examined (n = 132 and n = 44). Change in PHQ-9 scores was modeled using longitudinal growth modeling combined with random effects modeling of session effects, with time-in-treatment interacted with module theme to test moderation. In both studies, depressive symptoms significantly decreased during the active treatment phase. Symptom reductions were not significantly moderated by module theme in the larger study. However, the smaller pilot study's results suggest that future examination of module effects is warranted, given the data are compatible with differential reductions in reported symptoms being associated with attending people-themed module sessions versus thoughts-themed sessions.

Woody invasive plants are an increasing component of the New England flora. Their success and geographic spread are mediated in part by landscape characteristics. We tested whether woody invasive plant richness was higher in landscapes with many forest edges relative to other forest types and explained land use/land cover and forest fragmentation patterns using socioeconomic and physical variables. Our models demonstrated that woody invasive plant richness was higher in landscapes with more edge forest relative to patch, perforated, and especially core forest types. Using spatially-explicit, hierarchical Bayesian, compositional data models we showed that infrastructure and physical factors, including road length and elevation range, and time-lagged socioeconomic factors, primarily population, help to explain development and forest fragmentation patterns. Our social–ecological approach identified landscape patterns driven by human development and linked them to increased woody plant invasions. Identifying these landscape patterns will aid ongoing efforts to use current distribution patterns to better predict where invasive species may occur in unsampled regions under current and future conditions.

Compositional data analysis considers vectors of nonnegative-valued variables subject to a unit-sum constraint. Our interest lies in spatial compositional data, in particular, land use/land cover (LULC) data in the northeastern United States. Here, the observations are vectors providing the proportions of LULC types observed in each 3 km×3 km grid cell, yielding order 104 cells. On the same grid cells, we have an additional compositional dataset supplying forest fragmentation proportions. Potentially useful and available covariates include elevation range, road length, population, median household income, and housing levels.

This article describes a collaborative learning experience in experimental design that closely approximates what practicing statisticians and researchers in applied science experience during consulting. Statistics majors worked with a teaching assistant from the chemistry department to conduct a series of experiments characterizing the variation in measured voltage output of Smestad and Grätzel's nanocrystaline titanium dioxide (TiO2) solar cells. These solar cells can be constructed easily in a laboratory, and they are reported to produce an open circuit voltage in direct sunlight of 0.3 to 0.5V. Statistics students planned a series of experiments as part of an experimental design class, and the chemistry TA performed the experiments in the lab where the statistics students could observe. The students wrote a description of what they did and the results. From the students' comments about what they learned from this experience, it appears that this type of exercise could be very beneficial in training future consulting statisticians and scientists or technologists who will use experimentation in their work.