Treatment Choice With Trial Data: Statistical Decision Theory Should Supplant Hypothesis Testing

Abstract

A central objective of empirical research on treatment response is to inform treatment choice. Unfortunately, researchers commonly use concepts of statistical inference whose foundations are distant from the problem of treatment choice. It has been particularly common to use hypothesis tests to compare treatments. Wald’s development of statistical decision theory provides a coherent frequentist framework for use of sample data on treatment response to make treatment decisions. A body of recent research applies statistical decision theory to characterize uniformly satisfactory treatment choices, in the sense of maximum loss relative to optimal decisions (also known as maximum regret). This article describes the basic ideas and findings, which provide an appealing practical alternative to use of hypothesis tests. For simplicity, the article focuses on medical treatment with evidence from classical randomized clinical trials. The ideas apply generally, encompassing use of observational data and treatment choice in nonmedical contexts.     

Realistic Free-Spins Features Increase Preference for Slot Machines

Despite increasing research into how the structural characteristics of slot machines influence gambling behaviour there have been no experimental investigations into the effect of free-spins bonus features—a structural characteristic that is commonly central to the design of slot machines. This series of three experiments investigated the free-spins feature using slot machine simulations to determine whether participants allocate more wagers to a machine with free spins, and, which components of free-spins features drive this preference. In each experiment, participants were exposed to two computer-simulated slot machines—one with a free-spins feature or similar bonus feature and one without. Participants then completed a testing phase where they could freely switch between the two machines. In Experiment 1, participants did not prefer the machine with a simple free-spins feature. In Experiment 2 the free-spins feature incorporated additional elements such as sounds, animations, and an increased win frequency; participants preferred to gamble on this machine. The Experiment 3 “bonus feature” machine resembled the free spins machine in Experiment 2 except spins were not free; participants showed a clear preference for this machine also. These findings indicate that (1) free-spins features have a major influence over machine choice and (2) the “freeness” of the free-spins bonus features is not an important driver of preference, contrary to self-report and interview research with gamblers.     

Public Multichannel Management – an Integrated Framework of Off- and Online Multichannel Government Services

Government institutions have introduced multichannel environments over time following trends and technological innovations. These public multichannel environments often neglect specific channel characteristics and potentials and thus miss the opportunity to make full use of a strategic channel integration and combination with a focus on customer value creation. This study offers a systematic approach to establish an integrated public multichannel system not only taking into account user preferences, habits and their environment, but also the respective channel issuer and the characteristics of channels and public interfaces. Against this background, the study presents a systematic approach to expand services and channels according to their capacities, and provides academics and practitioners with a framework on public multichannel strategies.     

The Development of Object Construction From Infancy Through Toddlerhood

Early in development, children explore and combine objects in increasingly complex ways. One manual skill, object construction, represents a major shift in how objects are explored relative to other objects. Despite recent connections with cognitive functioning such as spatial skills, the development of object construction ability has rarely been studied and its trajectory has not been adequately described. The purpose of this study was to describe the development of three types of object construction (stacking, nesting, and affixing) across 5 monthly infant visits and 7 monthly toddler visits using a longitudinal design and a large sample size. Infants (10–14 months, n = 131) and toddlers drawn from the infant sample (18–24 months, n = 65) were given sets of age‐appropriate toys each of which elicited at least one type of constructive play. We described success at different construction tasks and identified trends for construction skill for infant and toddler development using multilevel modeling. We found that each of the three construction actions developed in unique ways across the 10‐ to 24‐month period. Infant construction skill predicted the development of toddler skill, but toddler construction skill diverged from infant trajectories. We discuss the role of combination strategies in the development of object construction skill and how object construction could be related to other domains of development.     

A latent variable model for analyzing mixed longitudinal (k,l)-inflated count and ordinal responses

A random effects model for analyzing mixed longitudinal count and ordinal data is presented where the count response is inflated in two points (k and l) and an (k,l)-Inflated Power series distribution is used as its distribution. A full likelihood-based approach is used to obtain maximum likelihood estimates of parameters of the model. For data with non-ignorable missing values models with probit model for missing mechanism are used.The dependence between longitudinal sequences of responses and inflation parameters are investigated using a random effects approach. Also, to investigate the correlation between mixed ordinal and count responses of each individuals at each time, a shared random effect is used. In order to assess the performance of the model, a simulation study is performed for a case that the count response has (k,l)-Inflated Binomial distribution. Performance comparisons of count-ordinal random effect model, Zero-Inflated ordinal random effects model and (k,l)-Inflated ordinal random effects model are also given. The model is applied to a real social data set from the first two waves of the national longitudinal study of adolescent to adult health (Add Health study). In this data set, the joint responses are the number of days in a month that each individual smoked as the count response and the general health condition of each individual as the ordinal response. For the count response there is incidence of excess values of 0 and 30.     

Social Cognitive Factors,Support, and Engagement: Early Adolescents’ Math Interests as Precursors to Choice of Career

The authors examined the central hypothesis that students’ early perceptions of support and sense of engagement in math classes and math activities strongly influence the broadening or narrowing of their math interest. The focus was on the first wave of qualitative data collected from 5th‐, 7th‐, and 9th‐grade students during the 2007–2008 academic year as part of a longitudinal study. Findings indicate the importance of using group work and extrinsic motivation in middle school math classes to broaden interest; peer classroom behavior was often a detractor of math interest.     

Bayesian averaging of classical estimates in asymmetric vector autoregressive models

The estimated vector autoregressive (VAR) model is sensitive to model misspecifications, resulting to biased and inconsistent parameter estimates. This article extends the Bayesian averaging of classical estimates, a robustness procedure in cross-section data, to a vector time-series that is estimated using a large number of asymmetric VAR models. The proposed procedure was applied to simulated data from various forms of model misspecifications. The results of the simulation suggest that, under misspecification problems, particularly if an important variable and moving average (MA) terms were omitted, the proposed procedure gives robust results and better forecasts than the automatically selected equal lag-length VAR model.