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.     

Social validation in group decision-making: Differential effects on the decisional impact of preference-consistent and preference-inconsistent information

Shared information has a stronger impact on group decisions than unshared information. A prominent explanation for this phenomenon is that shared information can be socially validated during group discussion and, hence, is perceived as more accurate and relevant than unshared information. In the present study we argue that this explanation only holds for preference-inconsistent information (i.e., information contradicting the group members’ initial preferences) but not for preference-consistent information. In Experiments 1 and 2 participants studied the protocol of a fictitious group discussion. In this protocol, we manipulated which types of information were socially validated. As predicted, social validation increased the decisional impact of preference-inconsistent but not preference-consistent information. In both experiments the effect of social validation was mediated by the perceived quality of information. Experiment 3 replicated the results of the first two experiments in an interactive setting in which two confederates discussed a decision case face-to-face with one participant.     

Coping with loneliness in old age: a cross-cultural comparison

The present study is an examination of the influence of cultural background on the ways in which the aged cope with loneliness. Thirty-six participants from Canada and 105 from Portugal (60-83 years old) answered an 86-item questionnaire which examined thebeneficial strategies which they used to cope with loneliness. The strategies which were examined included Acceptance and Reflection, Self-Development and Understanding, Social Support Network, Distancing and Denial, Religion and Faith, and Increased Activity. Results indicated that the elderly of the two cultures differ on three subscales. Gender differences between and within cultures were also examined.     

Young children's selective learning of rule games from reliable and unreliable models

We investigated preschoolers’ selective learning from models that had previously appeared to be reliable or unreliable. Replicating previous research, children from 4 years selectively learned novel words from reliable over unreliable speakers. Extending previous research, children also selectively learned other kinds of acts – novel games – from reliable actors. More important, – and novel to this study, this selective learning was not just based on a preference for one model or one kind of act, but had a normative dimension to it. Children understood the way a reliable actor demonstrated an act not only as the better one, but as the normatively appropriate or correct one, as indicated in both their explicit verbal comments and their spontaneous normative interventions (e.g., protest, critique) in response to third-party acts deviating from the one demonstrated. These findings are discussed in the broader context of the development of children's social cognition and cultural learning.     

The nonverbal manifestations of cognitive processes in clinical listening

Nonverbal behaviors were observed during communicative sequences in videotaped clinical interviews. The interviews were conducted by two groups of physicians-in-training with contrasting cognitive profiles as determined by well-known laboratory tasks prior to the interviews. In each phase of the communicative sequences, significant differences emerged between the two groups for various types of nonverbal behaviors. Differences also emerged between the two groups of interviewers in their questioning patterns and in the nature of the narrative elicited from their interlocutors. A correlational analysis revealed consistent patterns of behaviors across phases of the communicative sequences. The findings provide additional evidence for the intimate connection between nonverbal behavior and cognitive activity and point to nonverbal behaviors as objective measures of the processes underlying clinical listening.     

An experimental evaluation of tokens as conditioned reinforcers in retarded children

Since the publication of experimental reports by Wolfe (1936) and Cowles (1937), that tokens could be established as conditioned reinforcers, many accounts have appeared on the implementation of token systems in human subjects. Tokens have been successfully used as prosthetic motivational devices in a great variety of situations and subjects, such as retarded children (Birnbrauer, Wolf, Kidder and Tague, 1965), adult psychotics (Ayllon and Azrin, 1969), delinquent children and adolescents (Tyler, 1967; Phillips and Wolf, 1968), low achievers in schools (Clark, Lachowicz and Wolf, 1968), emotionally disturbed children (Hewett, Taylor and Artuso, 1969) and in normal classroom settings (O'Leary, Becker, Evans and Saudargas, 1969). In these reports, administration of tokens usually produced an increase in the frequency of desirable social and productive behavior. However, none of these studies showed that the effect of the tokens depended on their role as conditioned reinforcers based on the standard exchange systems being used. No assessment was provided about the separate contribution of the social reinforcement operations involved in dispensing tokens in a social setting.A suggestion about the possible confounding effects of social reinforcement in token administration comes from the failure to establish token systems in subjects who are partially unresponsive to social reinforcement (Hamblin et al., 1970; Ribes and Souza e Silva, in preparation). The present research was conducted with the aim of isolating the effects of conditioned reinforcement as such—apart from those produced by the social reinforcement provided in the delivery of tokens.     

On objectivity

The following definition of objective is proposed: A statement S is objective if and only if in S all parameters that are relevant to its truth value are made explicit. The objectivity of predicates and relations can be defined in a similar manner. This simple conception of objectivity-which could be called explicitness conception of objectivity-can be found in Hermann Weyl and plays a central part in the natural sciences. There are grades of objectivity depending on the quality and the number of parameters our predicates are relativized to A relativistic Ockham principle has to be recognized: Relativization parameters are not to be multiplied beyond necessity. The explicitness conception of objectivity is accessible to mathematical specifications, is the core of the idea of invariance, has a lot of philosophical applications and leads to precise notions of subjectivity and a precise formulation of the problem of the limits of objectivity.I am grateful to Andreas Bartels, Ulrich Gähde, Mark Helme, Andreas Kamlah, Andreas Kemmerling, Winfred Klink, Toni Koch, Hilary Putnam and Matthias Varga von Kibéd who have, in one way or another, contributed to the present content and form of this paper.     

The effects of metabolic stress and vagotomy on emotional learning in an animal model of anxiety

The aim of the present study is to evaluate the role of the blood glucose (BG) level in emotional learning in the elevated plus maze (EPM), an animal model of anxiety. In Experiment 1, male Wistar rats were submitted to different EPM trial lengths (1- or 5-min). Blood samples were withdrawn before and after the maze exploration, through a polyethylene cannula chronically implanted into the jugular vein. In Experiment 2, the animals received either saline or 2-deoxy-D-glucose, a glucoprivic drug (2-DG, 250 or 500 mg kg(-1)) by i.p. route, 30 min before a 5-min EPM exposure and were retested in the maze (Trial1/Trial2 EPM procedure) 24 h later. In an independent group of rats, blood samples were withdrawn 0, 5, 15, and 30 min after 2-DG administration, through the jugular vein, to determine BG. In Experiment 3, the animals underwent a vagotomy and were tested in a Trial1/Trial2 EPM procedure four weeks later. The results showed that rats exploring the EPM for 5 min displayed increased fear and higher hyperglycemia than those exploring the EPM for 1 min. In addition, rats submitted to 5-min EPM Trial1 length displayed higher level of fear on Trial2, as well as higher percentage of shortening of the %Open arm entries and %Open arm time from Trial1 to Trial2, which characterizes the occurrence of emotional learning. In contrast, rats previously vagotomized or treated with 2-DG (500 mg kg(-1)) showed the same level of fear on both EPM trials and a low percentage of shortening, from Trial1 to Trial2, of the %Open arm entries and %Open arm time, indicating poor emotional learning. The data is discussed regarding the role of glycaemia in emotional learning in the EPM.