Comparison of Type I error rates and statistical power of different propensity score methods

Propensity score analysis (PSA) is a technique to correct for potential confounding in observational studies. Covariate adjustment, matching, stratification, and inverse weighting are the four most commonly used methods involving propensity scores. The main goal of this research is to determine which PSA method performs the best in terms of protecting against spurious association detection, as measured by Type I error rate, while maintaining sufficient power to detect a true association, if one exists. An examination of these PSA methods along with ordinary least squares regression was conducted under two cases: correct PSA model specification and incorrect PSA model specification. PSA covariate adjustment and PSA matching maintain the nominal Type I error rate, when the PSA model is correctly specified, but only PSA covariate adjustment achieves adequate power levels. Other methods produced conservative Type I Errors in some scenarios, while liberal Type I error rates were observed in other scenarios.     

Third-person effect in the context of public relations and corporate communication

Inaccurate perceptions of audience can cause public relations practitioners to overreact or underreact to a situation, which brings about unintended consequences. We draw on the third-person effect research to identify examples and outline the conditions when the overestimation or underestimation of media effects can motivate key players to engage in restrictive, corrective, promotional, and resistance behaviors. Then, we recommend approaches that can help public relations practitioners better reflect and manage corporate communication. In short, the approaches involve: (a) determining the nature of the presumed effect from the audience perspective and going beyond the categorical positive/negative message, (b) assessing the level of the presumed effect and considering also the possibility of the first- as well as the second-person effect, and (c) assessing the influence of the perceived effect on behaviors that have the potential to directly as well as indirectly affect the company.     

Validating protein structure using kernel density estimates

Measuring the quality of determined protein structures is a very important problem in bioinformatics. Kernel density estimation is a well-known nonparametric method which is often used for exploratory data analysis. Recent advances, which have extended previous linear methods to multi-dimensional circular data, give a sound basis for the analysis of conformational angles of protein backbones, which lie on the torus. By using an energy test, which is based on interpoint distances, we initially investigate the dependence of the angles on the amino acid type. Then, by computing tail probabilities which are based on amino-acid conditional density estimates, a method is proposed which permits inference on a test set of data. This can be used, for example, to validate protein structures, choose between possible protein predictions and highlight unusual residue angles.     

The effects of food stamp receipt on weight gained by expectant mothers

With over 66% of Americans overweight, expectant mothers are unusual because they are encouraged to gain weight while pregnant. Food stamp receipt (FSR) may facilitate recommended weight gain by providing resources for food and nutrition. I examine the effects of FSR on the amount of weight gained by low-income expectant mothers using NLSY79 data. Results indicate FSR decreases the probability gaining insufficient weight but does not exacerbate the probability of gaining too much weight. Examining the effects of FSR on pregnancy weight gain is important because low birth weight is more likely when expectant mothers gain insufficient weight.     

Reward‐Modulated Response Inhibition,Cognitive Shifting,and the Orbital Frontal Cortex in Early Adolescence

Immaturities in cognitive shifting are associated with adolescent risk behaviors. The orbital frontal cortex (OFC) regulates reward processing and response inhibition. This study tested the relationship between cognitive shifting, OFC activity, and reward‐modulated response inhibition in young adolescents. An fMRI antisaccade (AS) paradigm examined the effects of reward conditions on inhibitory response and OFC processing. A validated self‐report inventory assessed cognitive shifting. Compared with neutral, reward trials showed better AS performance and increased OFC activation. Cognitive shifting positively associated with AS performance in reward and neutral trials. Poorer cognitive shifting predicted greater OFC activation. Results indicate lower OFC efficiency, as greater activation to achieve correct performance, underlies cognitive shifting problems. These neurocognitive impairments are relevant for understanding adolescent risk behaviors.     

Modeling the mental health service utilization decisions of university undergraduates: A discrete choice conjoint experiment

Objective: We modeled design factors influencing the intent to use a university mental health service. Participants: Between November 2012 and October 2014, 909 undergraduates participated. Method: Using a discrete choice experiment, participants chose between hypothetical campus mental health services. Results: Latent class analysis identified three segments. A Psychological/Psychiatric Service segment (45.5%) was most likely to contact campus health services delivered by psychologists or psychiatrists. An Alternative Service segment (39.3%) preferred to talk to peer-counselors who had experienced mental health problems. A Hesitant segment (15.2%) reported greater distress but seemed less intent on seeking help. They preferred services delivered by psychologists or psychiatrists. Simulations predicted that, rather than waiting for standard counseling, the Alternative Service segment would prefer immediate access to E-Mental health. The Usual Care and Hesitant segments would wait 6 months for standard counseling. Conclusions: E-Mental Health options could engage students who may not wait for standard services.     

Pool Screening: An Example of Independent Nonidentical Bernoulli Trial

Pool screening is a widely used study design to make inference about the probability of a subject being positive for an infection when the probability is extremely low. When the pool sizes are unequal, outcomes of pool screening follow independent nonidentical Bernoulli distributions. This paper presents different calculation methods for the probability distribution of T, the number of positive pools. Marcus and Lopes recursive method is recommended.