Marginal zero-inflated regression models for count data

Data sets with excess zeroes are frequently analyzed in many disciplines. A common framework used to analyze such data is the zero-inflated (ZI) regression model. It mixes a degenerate distribution with point mass at zero with a non-degenerate distribution. The estimates from ZI models quantify the effects of covariates on the means of latent random variables, which are often not the quantities of primary interest. Recently, marginal zero-inflated Poisson (MZIP; Long et al. [A marginalized zero-inflated Poisson regression model with overall exposure effects. Stat. Med. 33 (2014), pp. 5151–5165]) and negative binomial (MZINB; Preisser et al., 2016) models have been introduced that model the mean response directly. These models yield covariate effects that have simple interpretations that are, for many applications, more appealing than those available from ZI regression. This paper outlines a general framework for marginal zero-inflated models where the latent distribution is a member of the exponential dispersion family, focusing on common distributions for count data. In particular, our discussion includes the marginal zero-inflated binomial (MZIB) model, which has not been discussed previously. The details of maximum likelihood estimation via the EM algorithm are presented and the properties of the estimators as well as Wald and likelihood ratio-based inference are examined via simulation. Two examples presented illustrate the advantages of MZIP, MZINB, and MZIB models for practical data analysis.     

Context Matters: Differential Effects of Discrimination by Environmental Context on Depressive Symptoms Among College Students of Color

Clinical Social Work Journal - Experiences with more subtle racism – which have been called microaggressions – have a host of negative effects on health, mental health, educational...     

Conceptualizing reflection in experience-based workplace learning

This paper seeks to understand key dimensions of reflection in experience-based workplace learning for research being collaboratively undertaken by scholars in Dutch and US research institutions. We systematically explore and compare Tara Fenwick’s analysis of five perspectives on cognition to distinguish among constructivist, psychoanalytic, situative, critical-cultural, and enactivist perspectives on reflection. Our aim is to examine how to conceptualize reflection so that one or more perspectives can complement our understanding of learning through experience at work. We position reflection not only from a practical but also from a theoretical perspective that moves beyond the individual focus of the constructivist perspective. Fenwick’s five perspectives are described in terms of whether and how reflection is utilized, key activities in the process of reflection, relative focus on the individual and/or the context, triggers for reflection, role of power and positionality, role of emotions, and reflection outcomes. We conclude that adding other lenses to the dominant constructivist perspective helps expand our understanding of reflection as well as identify and attend to other tools, people, and factors in the work situation that influence reflection processes and learning outcomes. The paper closes with a discussion of promising new approaches that have emerged since Fenwick’s analysis.     

Intimate Partner Violence,Minority Stress,and Sexual Risk-Taking Among U.S. Men Who Have Sex With Men

This article examines the prevalence of intimate partner violence (IPV) among a national sample of Internet-recruited U.S. men who have sex with men (MSM) (n = 1,575), and associations between reporting of IPV, minority stress, and sexual risk-taking. Five outcomes are examined: experiences of physical and sexual violence, perpetration of physical and sexual violence, and unprotected anal intercourse (UAI) at last sexual encounter. MSM who reported experiencing more homophobic discrimination and internalized homophobia were more likely to report experiences of IPV. The results point to the need for prevention messages to address the external and internal stressors that influence both violence and sexual risk among MSM.     

Toward a More Complex View of Career Exploration

This article examines the concept of career exploration in adult populations. In contrast to the prevailing positive view of career exploration, the authors present a more complex and balanced perspective of this process, addressing some of the barriers to career exploration and the applicability of this concept to different populations. They examine differences between voluntary exploration and forced or chance exploration, discuss how relationships may be barriers to exploration, consider various outcomes of career explorations, and call for a more holistic view of the individual in career counseling.     

The Effects of Multitasking on Operations Scheduling

This study considers a typical scheduling environment that is influenced by the behavioral phenomenon of multitasking. Under multitasking, the processing of a selected job suffers from interruption by other jobs that are available but unfinished. This situation arises in a wide variety of applications; for example, administration, manufacturing, and process and project management. Several classical solution methods for scheduling problems no longer apply in the presence of multitasking. The solvability of any scheduling problem under multitasking is no easier than that of the corresponding classical problem. We develop optimal algorithms for some fundamental and practical single machine scheduling problems with multitasking. For other problems, we show that they are computationally intractable, even though in some cases the corresponding problem in classical scheduling is efficiently solvable. We also study the cost increase and value gained due to multitasking. This analysis informs companies about how much it would be worthwhile to invest in measures to reduce or encourage multitasking.     

Importance of interpolation when constructing double-bootstrap confidence intervals

We show that, in the context of double-bootstrap confidence intervals, linear interpolation at the second level of the double bootstrap can reduce the simulation error component of coverage error by an order of magnitude. Intervals that are indistinguishable in terms of coverage error with theoretical, infinite simulation, double-bootstrap confidence intervals may be obtained at substantially less computational expense than by using the standard Monte Carlo approximation method. The intervals retain the simplicity of uniform bootstrap sampling and require no special analysis or computational techniques. Interpolation at the first level of the double bootstrap is shown to have a relatively minor effect on the simulation error.     

Confidence intervals for secondary parameters following a sequential test

In sequential studies, formal interim analyses are usually restricted to a consideration of a single null hypothesis concerning a single parameter of interest. Valid frequentist methods of hypothesis testing and of point and interval estimation for the primary parameter have already been devised for use at the end of such a study. However, the completed data set may warrant a more detailed analysis, involving the estimation of parameters corresponding to effects that were not used to determine when to stop, and yet correlated with those that were. This paper describes methods for setting confidence intervals for secondary parameters in a way which provides the correct coverage probability in repeated frequentist realizations of the sequential design used. The method assumes that information accumulates on the primary and secondary parameters at proportional rates. This requirement will be valid in many potential applications, but only in limited situations in survival analysis.     

Betting Decision Under Break-Streak Pattern: Evidence from Casino Gaming

Cognitive bias is prevalent among gamblers, especially those with gambling problems. Grounded in the heuristics theories, this study contributes to the literature by examining a cognitive bias triggered by the break streak pattern in the casino setting. We postulate that gamblers tend to bet on the latest outcome when there is a break-streak pattern. Moreover, three determinants of the betting decision under break-streak pattern, including the streak length of the alternative outcome, the frequency of the latest outcome, and gender, were identified and examined in this study. A non-participatory observational study was conducted among the Cussec gamblers in a casino in Macao. An analysis of 1229 bets confirms our postulation, particularly when the streak of the alternative outcome is long, the latest outcome is frequent, and the gamblers are females. The findings provide meaningful implications for casino management and public policymakers regarding the minimization of gambling harm.     

Researcher reflexivity: exploring the impacts of CAQDAS use

Computer-assisted qualitative data analysis software (CAQDAS) programs are established tools for qualitative research. Making informed decisions when using them requires researchers to understand how they affect research practices and outcomes. In this article we consider the impact of CAQDAS on researcher reflexivity. Reviewing three decades of literature, we identify specific ‘reflexive moments’ experienced by CAQDAS users, the contexts in which they occur, the issues they raise, and the reflexive awareness they generate. The ways in which CAQDAS can enhance or undermine researcher reflexivity are also reported. By doing so, we aim to help researchers and especially research students (and their supervisors) understand the relationship between CAQDAS and reflexivity and the reflexive moments they may encounter when using such software.