Average Effects Based on Regressions with a Logarithmic Link Function: A New Approach with Stochastic Covariates

Psychometrika - Researchers often use regressions with a logarithmic link function to evaluate the effects of a treatment on a count variable. In order to judge the average effectiveness of the...     

Analysis of Variance Models with Stochastic Group Weights

The analysis of variance (ANOVA) is still one of the most widely used statistical methods in the social sciences. This article is about stochastic group weights in ANOVA models – a neglected aspect in the literature. Stochastic group weights are present whenever the experimenter does not determine the exact group sizes before conducting the experiment. We show that classic ANOVA tests based on estimated marginal means can have an inflated type I error rate when stochastic group weights are not taken into account, even in randomized experiments. We propose two new ways to incorporate stochastic group weights in the tests of average effects – one based on the general linear model and one based on multigroup structural equation models (SEMs). We show in simulation studies that our methods have nominal type I error rates in experiments with stochastic group weights while classic approaches show an inflated type I error rate. The SEM approach can additionally deal with heteroscedastic residual variances and latent variables. An easy-to-use software package with graphical user interface is provided.     

Emotional intelligence: new ability or eclectic traits?

Some individuals have a greater capacity than others to carry out sophisticated information processing about emotions and emotion-relevant stimuli and to use this information as a guide to thinking and behavior. The authors have termed this set of abilities emotional intelligence (EI). Since the introduction of the concept, however, a schism has developed in which some researchers focus on EI as a distinct group of mental abilities, and other researchers instead study an eclectic mix of positive traits such as happiness, self-esteem, and optimism. Clarifying what EI is and is not can help the field by better distinguishing research that is truly pertinent to EI from research that is not. EI--conceptualized as an ability--is an important variable both conceptually and empirically, and it shows incremental validity for predicting socially relevant outcomes. (PsycINFO Database Record (c) 2008 APA, all rights reserved).     

Testing different types of genotype-environment correlation: an extended children-of-twins model

This study presents an extended children-of-twins model, which allowed the authors to test the direction of the association between parenting and child adjustment. Three mechanisms were examined: direct phenotypic influence of parenting on child behavior (controlling for both parental and child genotype), passive genotype-environment correlation, and evocative genotype-environment correlation. This model was tested with Monte Carlo simulations. The authors generated data sets consisting of 1,000 twin parent pairs together with their children and 1,000 twin children pairs together with their parents. These simulated data sets were then used to estimate the model, and the procedure was repeated 1,000 times. The simulation results showed that this model recovered the true values of parameters with high precision. The model was also applied to an observed data set to analyze, as a first example, the association between maternal emotional overinvolvement and child internalizing problems. The results showed that this association was best explained by evocative genotype-environment correlation.     

On What it Takes for There to Be No Fact of the Matter

Philosophers are very fond of making non‐factualist claims—claims to the effect that there is no fact of the matter as to whether something is the case. But can these claims be coherently stated in the context of classical logic? Some care is needed here, we argue, otherwise one ends up denying a tautology or embracing a contradiction. In the end, we think there are only two strategies available to someone who wants to be a non‐factualist about something, and remain within the province of classical logic. But one of these strategies is rather controversial, and the other requires substantially more work than is often supposed. Being a non‐factualist is no easy business, and it may not be the most philosophically perspicuous way to go.     

Persuasive Appeals and the Reduction of Skin Cancer Risk: The Roles of Appearance Concern, Perceived Benefits of a Tan, and Efficacy Information

This experiment compared the effectiveness of appearance-based messages focusing on the benefits of tanning and the efficacy of preventive measures on students' intentions to protect themselves against sun-induced skin damage. Subjects who were high or low on a dispositional measure of appearance concern responded to 1 of 4 essays providing either high or low benefits information, and high or low efficacy information. Overall, subjects low in appearance concern expressed greater intentions to take precautionary measures than did those high in appearance concern. Additionally, the low benefits of tanning message produced greater intentions to take precautions than did the high benefits message. The benefits results suggest 1 mechanism through which sun protection behaviors may be encouraged.     

Comparing the results of laboratory and World-Wide Web samples on the determinants of female attractiveness

The Internet provides a new method to obtain subjects. Arguments supporting the use of the Internet as an experimental medium include the ease of data collection, limited resources needed, and ability to gain a sample more representative of the population. However, it has not been demonstrated that the Internet provides the conditions necessary to yield valid data. This paper compares results from experiments conducted over the Internet and in a laboratory. Correlating the means from the Internet samples to the laboratory samples on important interactions yielded values near 1.0, indicating that the two data sets are driven by the same psychological variables. Further evidence of the validity of these Web studies comes from regression analyses. However, more studies of this sort are needed to further identify what types of experiments may be validly conducted over the Internet.     

The effect of transgressor's and victim's sex on children's punish-help judgments

The present investigation, expanding upon a study by Miller and McCann, sought to determine whether children's reactions to the perpetrators and victims of antisocial behavior are influenced by the sex of the individuals involved. Third- and fifth-grade children were asked their opinions of a series of stories in which one child (boy or girl) was described as having intentionally harmed another child (boy or girl). The results indicate that the relative importance that children attribute to punishing transgressors and helping victims is mediated by the sex composition of the transgressor-victim dyad.