Bayesian credible intervals for response surface optima

In response surface methodology, one is usually interested in estimating the optimal conditions based on a small number of experimental runs which are designed to optimally sample the experimental space. Typically, regression models are constructed from the experimental data and interrogated in order to provide a point estimate of the independent variable settings predicted to optimize the response. Unfortunately, these point estimates are rarely accompanied with uncertainty intervals. Though classical frequentist confidence intervals can be constructed for unconstrained quadratic models, higher order, constrained or nonlinear models are often encountered in practice. Existing techniques for constructing uncertainty estimates in such situations have not been implemented widely, due in part to the need to set adjustable parameters or because of limited or difficult applicability to constrained or nonlinear problems. To address these limitations a Bayesian method of determining credible intervals for response surface optima was developed. The approach shows good coverage probabilities on two test problems, is straightforward to implement and is readily applicable to the kind of constrained and/or nonlinear problems that frequently appear in practice.     

Digital identities – overcoming visual bias through virtual embodiment

ABSTRACT

This study sheds new light on the role of identity in virtual environments when virtual representation of self is in support of disabled individuals and the potential impact of their virtual identity on work opportunities. It is widely understood that individuals who live with disability often experience a lifetime of bullying, exclusion, marginalization and rejection. They often experience workplace bias and discrimination. Yet, when they can create an identity and experience embodiment in virtual environments, the results can be extraordinarily powerful – even life-changing. This research builds on nearly a decade of ethnographic research in 3D online immersive social virtual worlds; seven of those years working with disability communities to answer the following: RQ1: In a virtual world where one can choose any avatar form, how does that visual sense of self-representation influence one’s ability to gain access to a social network, to be a leader in that network, and to find work? RQ2: How does realism in representation influence work experiences in these digital worlds? The results reveal the importance of choice in online representation of avatars in creating work and online social engagement. Implications contribute to our understanding of visual bias in the workplace and how emerging virtual reality technologies may open new avenues for meaningful work and social interactions for people with disabilities.     

Manipulating motivating operations to facilitate the emergence of mands for a child with autism

Research on the functional independence of verbal operants (Skinner, 1957) has demonstrated inconsistent findings. One explanation may be that these studies have not manipulated the motivating operation (MO) to facilitate the emergence of mands (Hall & Sundberg, 1987; Lamarre & Holland, 1985). In the current study, 1 participant, diagnosed with autism, was taught to tact high-preference and low-preference leisure items, and emergence of mands was tested under varying MO conditions. Results showed the emergence of mands following periods of arranged deprivation, and greater maintenance for a highly preferred relative to a less preferred stimulus. However, mands only emerged when presession tact trials were conducted. These results suggest that in a state of deprivation, transfer of stimulus control from discriminative to motivational conditions may occur without direct training.     

Me,Myself, and Money II: Relative Deprivation Predicts Disordered Gambling Severity via Delay Discounting,Especially Among Gamblers Who Have a Financially Focused Self-Concept

In the current research, we examined whether the known link between relative deprivation and disordered gambling (via delay discounting; i.e., preferences for immediate smaller rewards relative to delayed larger rewards) is moderated by the extent to which gamblers have a financially focused self-concept. Specifically, we hypothesized that delay discounting would be a strong predictor of disordered gambling among those who base their self-worth on their financial success. To test this moderated-mediation model, a community sample of gamblers (N = 239) completed measures that assessed relative deprivation, delay discounting, financially focused self-concept, and disordered gambling severity. As predicted, people who felt more relative deprivation reported more severe symptoms of disordered gambling and this association was mediated by delay discounting. Importantly, this mediated relationship was moderated by the extent to which participants’ self-concept was focused on financial success. Among participants whose self-concept was high in financial focus, greater delay discounting (stemming from relative deprivation) was a strong predictor of disordered gambling. Among people whose self-concept was low in financial focus, delay discounting (stemming from relative deprivation) was a weak predictor of disordered gambling. Thus, the magnitude of the indirect effect of relative deprivation on disordered gambling severity was larger among people with a more financially focused self-concept—an effect mediated by delay discounting. These findings suggest that targeting gamblers’ financial focus in prevention and treatment interventions may be instrumental in curtailing the development and maintenance of disordered gambling.     

Do conditional cash transfers influence migration? A study using experimental data from the Mexican progresa program

Prior research on Mexican migration has shown that social networks and economic incentives play an important role in determining migration outcomes. We use experimental data from PROGRESA, Mexico's primary poverty-reduction program, to evaluate the effects of conditional cash transfers on migration both domestically and to the United States. Our study complements a growing body of literature aimed at overcoming longstanding hurdles to the establishment of causal validity in empirical studies of migration. Analysis based on the data collected before and after the program's onset shows that conditional transfers reduce U.S. migration but not domestic migration. The data also enable us to explore the role of existing family and community migration networks. The results show that migration networks strongly influence migration, but that the effect of conditional transfers on migration is apparently not mediated by existing migration network structures. Our results suggest that conditional transfers may be helpful in managing rural out-migration, particularly to the United States.     

Alcohol use among college students: responses to raising the purchase age

On December 1, 1985, New York State raised its alcohol purchase age from 19 to 21. We used a quasi-experimental research design to explore the changes in alcohol use behaviors and attitudes of undergraduates at a large central New York university before and after this legislation was enacted. The overwhelming majority of this undergraduate population is under 21 years old and is thus affected by the new legislation. A comparison of data from the two survey times revealed that 90% of the undergraduates sampled continued to drink at least occasionally. Our analysis of drinking quantity showed a slight moderation in alcohol consumption overall, with the greatest changes occurring for the heaviest drinkers--men and members of Greek organizations. Even with apparent moderation in student drinking, reported negative consequences such as physical injuries were more common. A change in drinking location to less-controlled environments, such as private rooms and unmonitored parties, is offered as one possible explanation.     

Studying categorical data over time

A review of recently developed simple techniques for analyzing data in the form of proportions from replicated, cross-sectional, sample surveys. (A) One variable over time—tests for homogeneity, pooling homogeneous proportions, tests for linear trends with as few as three time points, tests for departures from linearity; (B) Two variables with a constant percentage difference—four fold tables as rudimentary causal models, decomposing change with linear flow graphs; (C) Extension of the flow graph techniques to—three or more variables and changing coefficients; (D) Comments implying panel designs are somewhat over-rated and successive cross-sections somewhat under-rated.