Heterogeneous heterogeneity by default: Testing categorical moderators in mixed-effects meta-analysis

Categorical moderators are often included in mixed-effects meta-analysis to explain heterogeneity in effect sizes. An assumption in tests of categorical moderator effects is that of a constant between-study variance across all levels of the moderator. Although it rarely receives serious thought, there can be statistical ramifications to upholding this assumption. We propose that researchers should instead default to assuming unequal between-study variances when analysing categorical moderators. To achieve this, we suggest using a mixed-effects location-scale model (MELSM) to allow group-specific estimates for the between-study variance. In two extensive simulation studies, we show that in terms of Type I error and statistical power, little is lost by using the MELSM for moderator tests, but there can be serious costs when an equal variance mixed-effects model (MEM) is used. Most notably, in scenarios with balanced sample sizes or equal between-study variance, the Type I error and power rates are nearly identical between the MEM and the MELSM. On the other hand, with imbalanced sample sizes and unequal variances, the Type I error rate under the MEM can be grossly inflated or overly conservative, whereas the MELSM does comparatively well in controlling the Type I error across the majority of cases. A notable exception where the MELSM did not clearly outperform the MEM was in the case of few studies (e.g., 5). With respect to power, the MELSM had similar or higher power than the MEM in conditions where the latter produced non-inflated Type 1 error rates. Together, our results support the idea that assuming unequal between-study variances is preferred as a default strategy when testing categorical moderators.     

Macro- and microlevel factors related to sense of community: The case of temporary neighborhoods in Israel

This study examines macro and micro factors influencing the development of sense of community (Davidson & Cotter, 1980) in two different populations (immigrants and veterans) in new temporary neighborhoods in Israel. At the macrolevel, the major factors examined were population size, population density, number of dwelling units in the site, urbanity of the area, ethnic heterogeneity, and peripheriality of the region. Three kinds of variables were examined on the microlevel: (a) personal attitudes: evaluation of the dwelling unit and satisfaction with public services; (b) social networks; and (c) sociodemographic characteristics. Data were collected from 242 immigrants from the former USSR and from 60 Israeli veterans, residing in 5 different sites. Two different patterns of predictors of sense of community emerged in the two different samples. In the veteran sample, only one macrolevel variable entered the stepwise analysis equation: the number of dwelling units in the site. In the immigrant sample, three significant microlevel factors entered the equation: evaluation of the dwelling unit, external network, and age. The data thus suggest that the major determinant factors underlying sense of community vary for different groups of people. The paper is based on a study initiated and financed by the Ministry of Housing and Construction, the Unit of Social Policy, Israel.     

Moral Values Reveal the Causality Implicit in Verb Meaning

Prior work has found that moral values that build and bind groups—that is, the binding values of ingroup loyalty, respect for authority, and preservation of purity—are linked to blaming people who have been harmed. The present research investigated whether people's endorsement of binding values predicts their assignment of the causal locus of harmful events to the victims of the events. We used an implicit causality task from psycholinguistics in which participants read a sentence in the form “SUBJECT verbed OBJECT because…” where male and female proper names occupy the SUBJECT and OBJECT position. The participants were asked to predict the pronoun that follows “because”—the referent to the subject or object—which indicates their intuition about the likely cause of the event. We also collected explicit judgments of causal contributions and measured participants' moral values to investigate the relationship between moral values and the causal interpretation of events. Using two verb sets and two independent replications (N = 459, 249, 788), we found that greater endorsement of binding values was associated with a higher likelihood of selecting the object as the cause for harmful events in the implicit causality task, a result consistent with, and supportive of, previous moral psychological work on victim blaming. Endorsement of binding values also predicted explicit causal attributions to victims. Overall, these findings indicate that moral values that support the group rather than the individual reliably predict that people shift the causal locus of harmful events to those affected by the harms.     

A Process Model of Causal Reasoning

How do we make causal judgments? Many studies have demonstrated that people are capable causal reasoners, achieving success on tasks from reasoning to categorization to interventions. However, less is known about the mental processes used to achieve such sophisticated judgments. We propose a new process model—the mutation sampler—that models causal judgments as based on a sample of possible states of the causal system generated using the Metropolis–Hastings sampling algorithm. Across a diverse array of tasks and conditions encompassing over 1,700 participants, we found that our model provided a consistently closer fit to participant judgments than standard causal graphical models. In particular, we found that the biases introduced by mutation sampling accounted for people's consistent, predictable errors that the normative model by definition could not. Moreover, using a novel experimental methodology, we found that those biases appeared in the samples that participants explicitly judged to be representative of a causal system. We conclude by advocating sampling methods as plausible process-level accounts of the computations specified by the causal graphical model framework and highlight opportunities for future research to identify not just what reasoners compute when drawing causal inferences, but also how they compute it.     

Knowing When Help Is Needed: A Developing Sense of Causal Complexity

Research on the division of cognitive labor has found that adults and children as young as age 5 are able to find appropriate experts for different causal systems. However, little work has explored how children and adults decide when to seek out expert knowledge in the first place. We propose that children and adults rely (in part) on “mechanism metadata,” information about mechanism information. We argue that mechanism metadata is relatively consistent across individuals exposed to similar amounts of mechanism information, and it is applicable to a wide range of causal systems. In three experiments, we show that adults and children as young as 5 years of age have a consistent sense of the causal complexity of different causal systems, and that this sense of complexity is related to decisions about when to seek expert knowledge, but over development there is a shift in focus from procedural information to internal mechanism information.     

Perceived Disagreement and Heterogeneity in Social Networks: Distinct Effects on Political Participation

ABSTRACT

Although the coexistence of conflicting opinions in society is the very core of democracy, people’s tendency to avoid conflict could keep them away from political discussion and participation. On the other hand, being exposed to diverse political views could motivate citizens to participate. We conducted secondary analyses on two 2013 ITANES (Italian National Election Studies) probability samples in order to test the hypotheses that perceived network disagreement (between an individual and her/his discussion partners) and heterogeneity (among discussants holding different political opinions) exert independent and opposite effects on political participation through motivation and knowledge. Results converged in showing that disagreement dampened, while heterogeneity encouraged, political participation (voting, propensity to abstain in future, offline and online activism, and timing of vote decision) by decreasing or increasing, respectively, political interest and, in turn, knowledge.     

The development of causal categorization

Two experiments examined the impact of causal relations between features on categorization in 5- to 6-year-old children and adults. Participants learned artificial categories containing instances with causally related features and noncausal features. They then selected the most likely category member from a series of novel test pairs. Classification patterns and logistic regression were used to diagnose the presence of independent effects of causal coherence, causal status, and relational centrality. Adult classification was driven primarily by coherence when causal links were deterministic (Experiment 1) but showed additional influences of causal status when links were probabilistic (Experiment 2). Children's classification was based primarily on causal coherence in both cases. There was no effect of relational centrality in either age group. These results suggest that the generative model (Rehder, 2003a) provides a good account of causal categorization in children as well as adults.