Bayesian Latent Class Analysis Tutorial

This article is a how-to guide on Bayesian computation using Gibbs sampling, demonstrated in the context of Latent Class Analysis (LCA). It is written for students in quantitative psychology or related fields who have a working knowledge of Bayes Theorem and conditional probability and have experience in writing computer programs in the statistical language R. The overall goals are to provide an accessible and self-contained tutorial, along with a practical computation tool. We begin with how Bayesian computation is typically described in academic articles. Technical difficulties are addressed by a hypothetical, worked-out example. We show how Bayesian computation can be broken down into a series of simpler calculations, which can then be assembled together to complete a computationally more complex model. The details are described much more explicitly than what is typically available in elementary introductions to Bayesian modeling so that readers are not overwhelmed by the mathematics. Moreover, the provided computer program shows how Bayesian LCA can be implemented with relative ease. The computer program is then applied in a large, real-world data set and explained line-by-line. We outline the general steps in how to extend these considerations to other methodological applications. We conclude with suggestions for further readings.     

Kill or cure? Different types of social class identification amplify and buffer the relation between social class and mental health

The present research investigated different types of social class identification as moderators of the negative relation between social class and mental health problems. Psychology undergraduates (N = 355) completed an online survey that included measures of social class, mental health and well-being, and three aspects of social class identification: importance of identity, salience of identity, and perceived self-class similarity. Perceived self-class similarity buffered the negative association between social class and depressive symptoms. However, importance and salience of social class identity amplified the associations between social class and anxiety and life satisfaction. These findings contribute to a more sophisticated understanding of the way in which social identification may operate as a social cure.     

To be kind or not to be kind: The moderating role of self-compassion in the relationship between general resourcefulness and academic self-regulation

We investigated whether the relationship between students’ general resourcefulness and academic self-regulation changes as a function of self-compassion. A predominantly female sample of 196 undergraduates completed inventories assessing these and other measures. The significant moderating effect of self-compassion revealed that the positive relationship between general resourcefulness and academic self-regulation was stronger for participants scoring low in self-compassion than high in self-compassion. For those low in self-compassion, scoring low in general resourcefulness was associated with the lowest academic self-regulation, whereas scoring high in general resourcefulness was associated with the greatest academic self-regulation. The positive relationship between general and academic self-regulation was attenuated for participants high in self-compassion, with predicted scores for academic self-regulation falling in between the two values described for the low self-compassion function. Implications of the findings are discussed, including the potential value of incorporating self-compassion training alongside programs aimed at increasing general resourcefulness and academic self-regulation.     

Acceptance of Uncommon Information into Group Discussion When That Information Is or Is Not Demonstrable

Research on group problem solving and on social cognition suggests that, during group discussion of a decision-making task, member acceptance of an unknown fact will be affected by the demonstrability of a fact's existence and the familiarity of the decision topic. Four-person groups read and discussed facts about a familiar or unfamiliar issue. Some groups could obtain written fact summaries during discussion, while the others could not. We predicted and found a three-way interaction between demonstrability, familiarity, and number of advocates. Verifiable single-advocate facts were accepted into discussion, but unverifiable facts required two advocates. These effects were more pronounced when the topic was familiar. Regression analysis of the influence of individual items on group choice revealed that, for the unfamiliar topic, the number of advocates of an item was directly related to that item's degree of influence only when the item could not be verified. Verifiable single-advocate facts were as influential as any other verifiable fact.     

Sexual Harassment in the Workplace: Exploring the Effects of Attractiveness on Perception of Harassment

This study was designed to explore the effect of attractiveness on perceptions of sexual harassment. Male and female college students (N = 150) rated four scenarios depicting ambiguous incidents of sexual harassment, each paired with photos of a male boss and a female secretary. Male and female attractiveness was varied in a 2 × 2 factorial design. Participants were asked to rate each photo on a series of traits before making harassment judgments. Overall, females perceived more harassment. The behavior of attractive males was less likely to be seen as harassing. Attractive females were more likely to be seen as harassed, especially when the potential harasser was unattractive. The possible mechanisms underlying the effects of attractiveness were explored with the results supporting a direct effect of stereotypes over a mediating role for implicit personality theories.     

Psychological Correlates of Anorexic and Bulimic Symptomatology

Few studies examining the relationship between eating disorders and personality have been theoretically derived; thus, findings have been equivocal. From a theoretical and empirical perspective (S. Orbach, 1986; R. L. Rogers & T. A. Petrie, 1997; G. J. Williams et al., 1994), this study investigated the connection between eating disorder symptomatology and several psychological correlates—obsessiveness, dependency, overcontrolled hostility, assertiveness, locus of control, and self‐esteem. Regression analyses indicated that obsessiveness and 2 factors of dependency accounted for 21% of the variance in a measure of anorexic attitudes and behaviors. One factor of dependency and obsessiveness accounted for 20% of the variance in a measure of bulimic symptomatology.     

Infant vocal development during the first 6 months: speech quality and melodic complexity

The early development of infant non‐distress vocalizations was investigated in this study. Thirteen infants, from 4 to 24 weeks of age, and their mothers were observed weekly in a face‐to‐face interaction situation. The speech quality (syllabic versus vocalic) and melodic complexity (simple versus complex) of infant vocalizations were coded independently. Based on speech quality and melodic complexity, four types of infant non‐distress vocalizations were categorized: simple and complex syllabic (speech‐like) vocalizations as well as simple and complex vocalic (non‐speech‐like) vocalizations. Results showed that complex syllabic sounds were of longer duration and complex vocalic sounds were less frequent than the other types of vocalizations. Curvilinear developmental trends were found in the rate of simple vocalic sounds and in the mean duration of complex syllabic sounds. Furthermore, before 4 months of age, vocalic sounds were more likely to be associated with simple melodic contours, after which syllabic sounds were more likely to be associated with complex melodic contours. A dynamic systems perspective on the early development of infant vocalization is discussed. Copyright © 2000 John Wiley & Sons, Ltd.     

Sex Differences in Self-reported Infidelity and its Correlates

We examined sex differences in the prevalence, incidence, reasons for, and consequences of infidelity. Participants (Study 1, 543 undergraduates in the Northwestern US; Study 2, 313 undergraduates and 233 community members in the Mid-Atlantic US), reported on infidelity by questionnaire. Using a broad definition of cheating, women reported being as unfaithful or more unfaithful than men. Men were more suspicious about cheating and more likely to discover the cheating than women. Women were more likely to break up with their partners, to begin new relationships after cheating, and to report reasons for cheating that may indicate a desire to switch long-term mates, such as being unhappy in the current relationship. Results are discussed in the context of evolutionary theory.