Evaluating expectations about negative emotional states of aggressive boys using Bayesian model selection.

Researchers often have expectations about the research outcomes in regard to inequality constraints between, e.g., group means. Consider the example of researchers who investigated the effects of inducing a negative emotional state in aggressive boys. It was expected that highly aggressive boys would, on average, score higher on aggressive responses toward other peers than moderately aggressive boys, who would in turn score higher than nonaggressive boys. In most cases, null hypothesis testing is used to evaluate such hypotheses. We show, however, that hypotheses formulated using inequality constraints between the group means are generally not evaluated properly. The wrong hypotheses are tested, i.e.. the null hypothesis that group means are equal. In this article, we propose an innovative solution to these above-mentioned issues using Bayesian model selection, which we illustrate using a case study. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Model inference or model selection: Discussion of Klugkist, Laudy, and Hoijtink (2005).

I. Klugkist, O. Laudy, and H. Hoijtink (2005) presented a Bayesian approach to analysis of variance models with inequality constraints. Constraints may play 2 distinct roles in data analysis. They may represent prior information that allows more precise inferences regarding parameter values, or they may describe a theory to be judged against the data. In the latter case, the authors emphasized the use of Bayes factors and posterior model probabilities to select the best theory. One difficulty is that interpretation of the posterior model probabilities depends on which other theories are included in the comparison. The posterior distribution of the parameters under an unconstrained model allows one to quantify the support provided by the data for inequality constraints without requiring the model selection framework. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Empirical evaluation and justification of methodologies in psychological science.

The purpose of this article is to describe a relatively new movement in the history and philosophy of science, naturalism, a form of pragmatism emphasizing that methodological principles are empirical statements. Thus, methodological principles must be evaluated and justified on the same basis as other empirical statements. On this view, methodological statements may be less secure than the specific scientific theories to which they give rise. The authors examined the feasibility of a naturalistic approach to methodology using logical and historical analysis and by contrasting theories that predict new facts versus theories that explain already known facts. They provide examples of how differences over methodological issues in psychology and in science generally may be resolved using a naturalistic, or empirical, approach. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Postscript: Theory development should not end (but always begins) with good empirical fits: Response to Roberts and Pashler's (2002) reply.

Presents a response to Roberts and Pashler's reply (see record 2002-13781-009) to the current author's original article (see record 2002-13781-008). Roberts and Pashler (2002) have shifted their original criterion to make it appear that we "have failed to come up with even one clear counterexample" (Roberts & Pashler, 2002, p. 607). However, no one ever argued that goodness of fit stands alone (note the title of our original criticism). Their original challenge was, exactly, to find "theories originally supported mainly or entirely by good fits to data that eventually found support from other sources" (Roberts & Pashler, 2000, p. 362). Roberts and Pashler (2002) rejected our analogy between criticism of null hypothesis significance testing (NHST) and their criticism of using goodness of fit as a starting point in the development of theories. We believe that Roberts and Pashler (2000, 2002) made important statements about theory development. Although those statements are embedded in an extreme, unreasonable, and unacceptable position as regards goodness of fit, we can still extract great value from their principles for theory development. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Analysis of correlation matrices using covariance structure models.

It is often assumed that covariance structure models can be arbitrarily applied to sample correlation matrices as readily as to sample covariance matrices. Although this is true in many cases and leads to an analysis that is mostly correct, it is not permissible for all structures. This article reviews three interrelated problems associated with the analysis of structural models using a matrix of sample correlations. Depending upon the model, applying a covariance structure to a matrix of correlations may (a) modify the model being studied, (b) produce incorrect values of the omnibus test statistic, or (c) yield incorrect standard errors. An important class of models are those that are scale invariant (Browne, 1982), for then Errors a and b cannot occur when a correlation matrix is analyzed. A number of examples based on restricted factor analysis are presented to illustrate the concepts described in the article. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Testing the race model inequality in redundant stimuli with variable onset asynchrony.

In speeded response tasks with redundant signals, parallel processing of the signals is tested by the race model inequality. This inequality states that given a race of two signals, the cumulative distribution of response times for redundant stimuli never exceeds the sum of the cumulative distributions of response times for the single-modality stimuli. It has been derived for synchronous stimuli and for stimuli with stimulus onset asynchrony (SOA). In most experiments with asynchronous stimuli, discrete SOA values are chosen and the race model inequality is separately tested for each SOA. Due to the high number of statistical tests, Type I and II errors are increased. Here a straightforward method is demonstrated to collapse these multiple tests into one test by summing the inequalities for the different SOAs. The power of the procedure is substantially increased by assigning specific weights to SOAs at which the violation of the race model prediction is expected to be large. In addition, the method enables data analysis for experiments in which stimuli are presented with SOA from a continuous distribution rather than in discrete steps. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Theory and philosophy in social psychology.

Social psychologists are generally very empirical, and their research is usually theoretically oriented. However, what passes for theory is often no more than a couple of vague statements. Nevertheless, coming from the tradition of Kurt Lewin, we tend to reject papers that do not test a "theory". On the other hand, we pay no attention to theories lacking experimental support. Clinical observations are not enough. There is very little concern with philosophical issues. Ken Gergen of Swarthmore College is a notable exception and is often on convention programs. His comments about "historicism" in social psychology have caused some published rebuttals, but our members are largely not very concerned with this or with Gergen's more recent forays in "social re-constructionism". (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Analyzing longitudinal data with multilevel models: An example with individuals living with lower extremity intra-articular fractures.

Objective: The use and quality of longitudinal research designs has increased over the past 2 decades, and new approaches for analyzing longitudinal data, including multilevel modeling (MLM) and latent growth modeling (LGM), have been developed. The purpose of this article is to demonstrate the use of MLM and its advantages in analyzing longitudinal data. Research Method: Data from a sample of individuals with intra-articular fractures of the lower extremity from the University of Alabama at Birmingham's Injury Control Research Center are analyzed using both SAS PROC MIXED and SPSS MIXED. Results: The authors begin their presentation with a discussion of data preparation for MLM analyses. The authors then provide example analyses of different growth models, including a simple linear growth model and a model with a time-invariant covariate, with interpretation for all the parameters in the models. Implications: More complicated growth models with different between- and within-individual covariance structures and nonlinear models are discussed. Finally, information related to MLM analysis, such as online resources, is provided at the end of the article. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The Barnum effect and validity of computer-based test interpretations: The Human Resource Development Report.

This article addresses 2 conceptual problems surrounding the validation of computer-based test interpretations (CBTIs): assessing the discriminating power of a real CBTI compared with a generic bogus report and assessing the overall level of accuracy or validity of the report. The full Barnum research design, based on split-plot analysis of covariance, was developed and used to evaluate the Human Resources Development Report (HRDR), a CBTI from the Sixteen Personality Factor Questionnaire (16PF). Real HRDR CBTIs were rated by 54 college student examinees as 76.3% accurate, compared to 71.1% perceived accuracy for bogus reports. In the CBTI statement analysis, statements appearing only in real reports were rated as 76.2% accurate, compared to 68.6% for statements in bogus reports only. Procedures are also described for assessing the influence of context on the perceived accuracy of a statement contained in a report. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Detecting differential item functioning with confirmatory factor analysis and item response theory: Toward a unified strategy.

In this article, the authors developed a common strategy for identifying differential item functioning (DIF) items that can be implemented in both the mean and covariance structures method (MACS) and item response theory (IRT). They proposed examining the loadings (discrimination) and the intercept (location) parameters simultaneously using the likelihood ratio test with a free-baseline model and Bonferroni corrected critical p values. They compared the relative efficacy of this approach with alternative implementations for various types and amounts of DIF, sample sizes, numbers of response categories, and amounts of impact (latent mean differences). Results indicated that the proposed strategy was considerably more effective than an alternative approach involving a constrained-baseline model. Both MACS and IRT performed similarly well in the majority of experimental conditions. As expected, MACS performed slightly worse in dichotomous conditions but better than IRT in polytomous cases where sample sizes were small. Also, contrary to popular belief, MACS performed well in conditions where DIF was simulated on item thresholds (item means), and its accuracy was not affected by impact. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comparisons of means using exploratory and confirmatory approaches.

This article discusses comparisons of means using exploratory and confirmatory approaches. Three methods are discussed: hypothesis testing, model selection based on information criteria, and Bayesian model selection. Throughout the article, an example is used to illustrate and evaluate the two approaches and the three methods. We demonstrate that confirmatory hypothesis testing techniques have more power—that is, have a higher probability of rejecting a false null hypothesis—and confirmatory model selection techniques have a higher probability of choosing the correct or the best hypothesis than their exploratory counterparts. Furthermore, we show that if more than one hypothesis has to be evaluated, model selection has advantages over hypothesis testing. Another, more elaborate example is used to further illustrate confirmatory model selection. The article concludes with recommendations: When a researcher is able to specify reasonable expectations and hypotheses, confirmatory model selection should be used; otherwise, exploratory model selection should be used. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of erotica upon men's loving and liking responses for women they love.

In 2 studies, the construct validity of the Rubin Love scale and its discriminant validity in relation to the Rubin Liking scale were examined. In Study 1, 51 male undergraduates were asked to describe their loved ones on a series of measures after having been exposed to either erotica or control materials. Analysis of the within-condition correlations revealed convergent and discriminant patterns indicating construct validity. Analysis of covariance also revealed convergent and discriminant patterns: Only scores on the Love measures reliably increased from the control to the erotica condition. In presenting a behavioral analysis of the differential impact of erotica on Love vs Liking scale responses, it was assumed that (a) women are more likely to reinforce suitors for emitting statements more similar to the Love than to the Liking items and (b) men are more likely when sexually aroused than when not aroused to express statements more similar to the Love than to the Liking items to their loved ones. In Study 2, a series of surveys of 77 female and 64 male undergraduates offer support for these assumptions. Overall, results are interpreted as corroborating the construct validity of the Love scale. (17 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Processing attitude statements from in-group and out-group members: Effects of within-group and within-person inconsistencies on reading times.

The authors examined reading times of attitude statements made by group members as a function of consistency of statements with stereotypic expectancies (between-member) and consistency of statements with other statements from the same member (within-member). Stereotype-inconsistent statements were studied longer than consistent statements only when the target group was an outgroup or when subjects were instructed to focus on the group as a whole. Results suggested that the out-group was perceived as a single homogeneous whole regardless of experimental instructions. Inconsistencies within individual group members instigated the longest reading times. This effect was stronger for inconsistencies within out-group members than within in-group members, suggesting that subjects not only expected more within-group variability in in-groups than in out-groups, but they also expected more within-person variability. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

On the likelihood ratio test in structural equation modeling when parameters are subject to boundary constraints.

The authors show how the use of inequality constraints on parameters in structural equation models may affect the distribution of the likelihood ratio test. Inequality constraints are implicitly used in the testing of commonly applied structural equation models, such as the common factor model, the autoregressive model, and the latent growth curve model, although this is not commonly acknowledged. Such constraints are the result of the null hypothesis in which the parameter value or values are placed on the boundary of the parameter space. For instance, this occurs in testing whether the variance of a growth parameter is significantly different from 0. It is shown that in these cases, the asymptotic distribution of the chi-square difference cannot be treated as that of a central chi-square-distributed random variable with degrees of freedom equal to the number of constraints. The correct distribution for testing 1 or a few parameters at a time is inferred for the 3 structural equation models mentioned above. Subsequently, the authors describe and illustrate the steps that one should take to obtain this distribution. An important message is that using the correct distribution may lead to appreciably greater statistical power. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Les nuages sont un troupeau de moutons: à quel age les enfants comprennent-ils qu'il n'y pas de moutons dans le ciel?

This study examined the evolution of metaphor understanding in 4- to 7-year-old children. In Experiment 1, children listened to metaphoric and literal statements that were preceded by a context, and then had to choose among three proposed interpretations (literal, contextual, and correct for metaphors; one correct and two erroneous for literal statements), which one corresponded to the statement. Results showed that, between 4 and 7, children improved at choosing the correct interpretation only for metaphors, and that they frequently chose contextual interpretations. The metaphor-specific improvement might be explained in terms of a gain in cognitive flexibility, since for these statements, three acceptable interpretations were proposed compared to only one for literal statements. In Experiment 2, for all statements, children had to choose between three proposed interpretations, one correct and two erroneous. The difference between metaphors and literal statements was replicated. Metaphor understanding thus evolves between 4 and 7; depending on their age, children use the context or perform a semantic analysis of the words composing the statements. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Sociogenèse et psychobiologie: Les sources de la continuité et de la discontinuité du développement.

The aim of this article is to reflect conceptually upon the various stakes in development from both psychobiological and sociogenetical perspectives, two prominent contemporary theories of social sciences. Where does developmental continuity belong?...Does it belong within the individual or within his or her environment? Despite explicitly recognizing the influence of social experience in human development, the traditionally old nature-nurture debate remains at the center of theoretical analyses and leads to a futile integration of the two positions. The authors propose a theoretical model, which suggests not to integrate these two antagonistic positions but rather to recognize and utilize their interdependency as a means of viewing continuity and discontinuity as complementary facets of development. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Introducing the grandmother test into psychological science.

It is a commonly held belief among many psychological scientists that the more plausible a hypothesis, the better it is scientifically. It is also often believed that if there are two alternative theories which are equally strong in terms of explanatory power, one should choose the plausible theory over the implausible one. The goal of this article is to challenge these assumptions. Based on Popper's philosophical views, it will be argued that psychological scientists should be encouraged to formulate implausible (yet testable) hypotheses and theories. Examples from other scientific disciplines will be presented to illustrate this argument. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Thoughts of agoraphobic people during scary tasks.

The authors examined the occurrence of theoretically derived patterns of thinking in 74 agoraphobic participants as they drove alone or tolerated an enclosed place. During the increasingly scary tasks in a behavioral test hierarchy, participants responded to a periodic beep by stating aloud what they were thinking at that moment, yielding more than 1,800 tape-recorded statements. Content analyses revealed that participants were mainly preoccupied with their current anxiety (expressed in 29% of the statements) and with their self-efficacy (15%). Despite participants' mounting feelings of anxiety, fewer than 1% of their statements expressed a thought of danger or an anticipation of future anxiety or panic. The rarity of danger thoughts poses an explanatory challenge for all cognitive theories of phobia and especially for the perceived danger theory of A. T. Beck (1976) and A. T. Beck, G. Emery, and R L. Greenberg (1985). (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Selective recall of positive and negative feedback, self-control behaviors, and depression.

Tested 2 sets of hypotheses, derived from cognitive–behavioral theories of depression, that (a) compared to a sample of nondepressed controls, depressed Ss would underestimate the frequency of reinforcement and overestimate the frequency of punishment received during an ambiguous laboratory task; and (b) when given the opportunity to self-reinforce or self-punish, depressed Ss would self-reinforce less often and self-punish more often than controls. Three of these predictions were supported. In an experiment with 24 depressed and 21 nondepressed undergraduates (Beck Depression Inventory), depressed Ss recalled less positive and more negative feedback than controls. As expected, these differences were significant only at a high rate of reinforcement and at a low rate of punishment. In the latter condition, however, depressed Ss were accurate in their recall, while nondepressed Ss underestimated the frequency of negative feedback. Depressed Ss self-reinforced less often than controls, but there were no differences in rates of self-punishment. Implications for cognitive and behavioral theories of depression are discussed. (22 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The analytic environment in times of bodily dysfunction: The effect of psychoanalytic psychosomatic theories.

This article examines the analytic environment in which psychoanalytic work occurs when patients struggle with complex somatic experiences, such as disease or physiological dysfunction. Patients express fantasies and beliefs about the etiology of their somatic experiences; they elaborate theories about why infertility, irritable bowel disease, or some other disease, syndrome, or crisis is happening to them. I consider these to be patients’ multiply determined, fantasy-saturated psychosomatic theories, and suggest that the analyst’s understanding of patients’ ideas about their somatic experiences is organized by the analyst’s both articulated and not articulated psychosomatic theories. Using brief clinical vignettes, I highlight the potentially constricting effect of clinicians’ theories on their analyses of patients’ psychosomatic theories. I examine psychoanalytic theories of psychosomatics, suggesting that adherence to the theories we have, which often posit linear psychogenesis of somatic phenomena, can result in collusion with, rather than analysis of, the psychic stances reflected by patients’ theories. To address the problematic adherence to our theories, and characterize a less constricting clinical and theoretical stance, I critique aspects of the theories and suggest directions for replenishing them. (PsycINFO Database Record (c) 2010 APA, all rights reserved)