Some recommended statistical analytic practices when reliability generalization studies are conducted

Precursors of the reliability generalization (RG) meta‐analytic approach have not established a single preferred analytic method. By means of five real RG examples, we examine how using different statistical methods to integrate coefficients alpha can influence results in RG studies. Specifically, we compare thirteen different statistical models for averaging reliability coefficients and searching for moderator variables that differ in terms of: (a) whether to transform or not the coefficients alpha, and (b) the statistical model assumed, distinguishing between ordinary least squares methods, the fixed‐effect (FE) model, the varying coefficient (VC) model, and several versions of the random‐effects (RE) model. The results obtained with the different methods exhibited important discrepancies, especially regarding moderator analyses. The main criterion for the model choice should be the extent to which the meta‐analyst intends to generalize the results. RE models are the most appropriate when the meta‐analyst aims to generalize to a hypothetical population of past or future studies, while FE and VC models are the most appropriate when the interest focuses on generalizing the results to a population of studies identical to those included in the meta‐analysis. Finally, some guidelines are proposed for selecting the statistical model when conducting an RG study.     

The impact of familiarization strategies on the missing-letter effect

When reading a text and searching for a target letter, readers make more omissions of the target letter if it is embedded in frequent function words than if it is in rare content words. While word frequency effects are consistently found, few studies have examined the impacts of passage familiarity on the missing-letter effect and studies that have present conflicting evidence. The present study examines the effects of passage familiarity, as well as the impacts of passage familiarization strategy promoting surface or deep encoding, on the missing-letter effect. Participants were familiarized with a passage by retyping a text, replacing all common nouns with synonyms, or generating a text on the same topic as that of the original text, and then completed a letter search task on the familiar passage as well as an unfamiliar passage. In Experiment 1, when both familiar and unfamiliar passages use the same words, results revealed fewer omissions for the retyping and synonyms conditions. However, in Experiment 2, when different words are used in both types of texts, no effect of familiarization strategy was observed. Furthermore, the missing-letter effect is maintained in all conditions, adding support to the robustness of the effect regardless of familiarity with the text.     

How representationalism can account for the phenomenal significance of illumination

In this paper, I defend a representationalist account of the phenomenal character of color experiences. Representationalism, the thesis that phenomenal character supervenes on a certain kind of representational content, so-called phenomenal content, has been developed primarily in two different ways, as Russellian and Fregean representationalism. While the proponents of Russellian and Fregean representationalism differ with respect to what they take the contents of color experiences to be, they typically agree that colors are exhaustively characterized by the three dimensions of the color solid: hue, saturation, and lightness. I argue that a viable version of representationalism needs to renounce this restriction to three dimensions and consider illumination to be a genuine phenomenal dimension of color. My argument for this thesis falls into two parts. I first consider the phenomenon of color constancy in order to show that neither Russellian nor Fregean representationalism can do justice to the phenomenal significance of local illumination. I subsequently formulate a version of representationalism that accounts for illumination by taking it as a phenomenal dimension of color.     

Examining the validity of the family investment and stress models and relationship to children's school readiness across five cultural groups

Using the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B) dataset, this study examined whether the family investment and the family stress models generalized to non-European American (EA) families. Specifically, we examined whether parenting processes mediated the association between family demographics and children's school readiness, and whether the pathways vary across cultural groups. Both models were most salient for EAs followed by African Americans (AAs) and Spanish-speaking Hispanics, but less so for English-speaking Hispanics (EHs) and Asian Americans. Findings indicated that sensitive parenting was a salient mediator between family demographics and children's school readiness for all groups except EHs; negative parenting and parent-child activities were salient mediators for EAs only.     

Self-concept and psychosocial adjustment in adolescence

This study analyses the relationship between a multidimensional measure of self-concept, Self-concept Form-5 Questionnaire (AF5), and a broad set of adolescents' psychosocial adjustment indicators. From the responses of 1,281 participants (53.7% females) aged 12 to 17 years ( M = 14.98 years, SD = 1.74 years), results indicated that higher self-concept scores corresponded to better psychological adjustment, good personal skills and fewer behavioral problems. Although a positive relationship between social self-concept and drug use was found, this significant relationship disappeared once the adolescent's age and sex was controlled for. These results support the idea that the self-concept is a basic theoretical construct closely related to the psychosocial adjustment in adolescence. Also this study helps explain some contradictory results reported in the literature (i.e., a positive relationship between social self-concept and drug use), by showing how the statistical control of a third variable effect (i.e., age) avoids reaching conclusions based on spurious relationships.     

The PANAS structure revisited: on the validity of a bifactor model in community and forensic samples

The Positive and Negative Affect Schedule (PANAS) is a widely used inventory for the assessment of affect in psychology and other applied sciences. Despite its popularity, the structure of the PANAS is still under debate. On the one hand, there is evidence of the traditional 2-factor model with Positive Affect (PA) and Negative Affect (NA) as uncorrelated factors. On the other hand, a more complex structure of the PANAS has been discussed. To shed further light on the core dimensions of the PANAS, 2 studies investigated the structure of the PANAS in 2 German samples (N = 354 and N = 364, respectively) by means of confirmatory factor analysis. The factor analysis results of Study 1 for a traitlike time frame instruction suggested a suboptimal model fit for the uncorrelated 2-factor model and the 3-factor model with PA, Afraid, and Upset as factors, whereas a superior model fit occurred for a bifactor model with traitlike PA, NA, and a general 3rd factor named Affective Polarity. In Study 2, the bifactor model was replicated for a statelike PANAS time frame instruction and evidence of criterion validity was provided for PA, NA, and Affective Polarity factors in 2 sex offender subgroups and in a community sample. With Affective Polarity, we introduce an affect dimension that captures additional variance beyond PA and NA. Because of the adjectives with relevant loadings on Affective Polarity, this general factor represents an individual's orientation toward approach and withdrawal, respectively.     

Are cyberbullies less empathic? Adolescents' cyberbullying behavior and empathic responsiveness

Meta-analyses confirm a negative relationship between aggressive behavior and empathy, that is, the ability to understand and share the feelings of others. Based on theoretical considerations, it was, therefore, hypothesized that a lack of empathic responsiveness may be characteristic for cyberbullies in particular. In the present study, 2.070 students of Luxembourg secondary schools completed an online survey that included a cyberbullying questionnaire(4) and a novel empathy short scale. According to the main hypothesis, analyses of variances indicated that cyberbullies demonstrated less empathic responsiveness than non-cyberbullies. In addition, cyberbullies were also more afraid of becoming victims of cyberbullying. The findings confirm and substantially extend the research on the relationship between empathy and aggressive behavior. From an educational point of view, the present findings suggest that training of empathy skills might be an important tool to decrease cyberbullying.     

Talking about visually perceived events: Communication effects on eyewitness memory

Communicators' tuning of a message about a social target to their audience's evaluation can shape their representation of the target. This audience‐tuning effect has been demonstrated with ambiguous text passages as input material. We examined whether the effect also occurs when communicators learn about the target's behaviours from visual (nonverbal) input material. In Experiment 1, participants watched a soundless video depicting ambiguous behaviours of a target, described the video to an audience who liked (vs. disliked) the target, and subsequently recalled the video. Both message and recall were biased towards the audience's judgement. In Experiment 2, the video depicted a forensically relevant event, specifically ambiguous behaviours of two persons involved in a bar brawl. Participants tuned their event retellings to their audience's responsibility judgement and remembered the event accordingly. In both experiments, the effect of the audience's judgement on recall was statistically mediated by the extent to which the message was tuned to the audience. The more participants experienced a shared reality with their audience the stronger was the message‐recall correlation (Experiment 2). We conclude that the audience‐tuning effect for visually perceived information depends on the communicators' creation of a shared reality with their audience. Copyright © 2011 John Wiley & Sons, Ltd.