The analysis of multitrait-multimethod matrices via constrained components analysis

Multitrait-Multimethod (MTMM) matrices are often analyzed by means of confirmatory factor analysis (CFA). However, fitting MTMM models often leads to improper solutions, or non-convergence. In an attempt to overcome these problems, various alternative CFA models have been proposed, but with none of these the problem of finding improper solutions was solved completely. In the present paper, an approach is proposed where improper solutions are ruled out altogether and convergence is guaranteed. The approach is based on constrained variants of components analysis (CA). Besides the fact that these methods do not give improper solutions, they have the advantage that they provide component scores which can later on be used to relate the components to external variables. The new methods are illustrated by means of simulated data, as well as empirical data sets.This research has been made possible by a fellowship from the Royal Netherlands Academy of Arts and Sciences to the first author. The authors are obliged to three anonymous reviewers and an associate editor for constructive suggestions on the first version of this paper.     

Les manifestations comportementales de l'habilitation au travail : développement d'un cadre conceptuel et d'un instrument de mesure

Despite a strong interest for empowerment in business organizations, specific conceptual framework and measurement instruments do not exist to assess in a behavioural way if employees are really empowered. First, based on a literature review and interviews with twenty managers, seven behavioural indicators associated with empowerment have been identified and integrated in a theoretical framework distinguishing three investment areas (task, group, organization) and two types of contributions (assuring efficacy or improving efficiency). Second, questionnaires based on this framework have been administrated to 374 employees and their immediate supervisors. Factor analyses that have been performed showed a five-factor solution comparable for the self-report version and the version completed by the supervisor. Strengths and limitations of the new measurement instrument as well as the new research possibilities are discussed.     

Robust principal components: A generalized variance perspective

This article compares several methods for performing robust principal component analysis, two of which have not been considered in previous articles. The criterion here, unlike that of extant articles aimed at comparing methods, is how well a method maximizes a robust version of the generalized variance of the projected data. This is in contrast to maximizing some measure of scatter associated with the marginal distributions of the projected scores, which does not take into account the overall structure of the projected data. Included are comparisons in which distributions are not elliptically symmetric. One of the new methods simply removes outliers using a projection-type multivariate outlier detection method that has been found to perform well relative to other outlier detection methods that have been proposed. The other new method belongs to the class of projection pursuit techniques and differs from other projection pursuit methods in terms of the function it tries to maximize. The comparisons include the method derived by Maronna (2005), the spherical method derived by Locantore et al. (1999), as well as a method proposed by Hubert, Rousseeuw, and Vanden Branden (2005). From the perspective used, the method by Hubert et al. (2005), the spherical method, and one of the new methods dominate the method derived by Maronna.     

A Confirmatory Factor Analysis of the Short Form Sex Role Behavior Scale

The Sex Role Behavior Scale (SRBS; Orlofsky & O'Heron, 1987; Orlofsky, Ramsden, & Cohen, 1982) is the only measure of the extent to which people engage in male- and female-valued behaviors, as well as sex-specific behaviors. Because of this, researchers must be assured of its reliability and validity. Although the SRBS has demonstrated good reliability, validity tests have been limited to examinations of scale intercorrelations, correlations with other gender role measures, and tests of gender differences. Tests of the SRBS's construct validity have not been performed. Thus, a scale-based confirmatory factor analysis of the short form SRBS was undertaken to determine the validity of its proposed 12-factor, lower-order and 3-factor, higher-order factor structures. In this sample of undergraduates, both lower-order and higher-order models failed to provide a good fit to the data, which suggests that a new version of the SRBS may be required. Discussion focussed on possible directions for a revision, potential limitations, as well as the need for more measures like the SRBS.     

The Goodman Paradox: Three Different Problems and a Naturalistic Solution to Two of Them

It is now more than 50 years that the Goodman paradox has been discussed, and many different solutions have been proposed. But so far no agreement has been reached about which is the correct solution to the paradox. In this paper, I present the naturalistic solutions to the paradox that were proposed in Quine (1969, 1974), Quine and Ullian (1970/1978), and Stemmer (1971). At the same time, I introduce a number of modifications and improvements that are needed for overcoming shortcomings of the solutions. The discussion of this improved version suggests that the Goodman paradox actually embodies three different problems; yet, one of them is not Goodman’s but Hume’s problem. The discussion also suggests that the naturalistic approach is probably the best for basing on it a theory of confirmation. Finally, I analyze one of Hume’s insights that seems to have been largely ignored. This insight shows a surprising similarity to a central feature of the naturalistic solutions.     

A new OS/8 SKED

A new SKED run-time system and compiler have been designed for use under the OS8 operating system. OS8 is a set of programs designed by DEC for the PDP8 computer with 8K or more core memory locations and a mass-storage device (disk, or DEC-tape). The advantages of OS8 include operator convenience, device independent input-output, standard file formats, and convenient program chaining as well as a set of standard data analysis programs. The new compiler, OSCOMP, differs from the previous version in two ways. The first new feature is the ability to process named input and output files on any OS8 compatible peripheral. The second feature is the utilization of 8K of core, permitting compilation of longer state tables than could be processed with the earlier version. Furthermore, with a disk as the OS8 peripheral, the compilation process is essentially instantaneous, for state tables previously requiring from 3–30 min with paper tape devices. The new run-time system, OSRTS8, contains a variety of new features. The most important improvements are the abilities to record data on the OS8 peripheral as well as to read state tables stored as files on the mass-storage device. Other new features include chaining of state tables, automatic start, automatic output file specification, and capability for as many as 12 simultaneous stations.     

Heuristic and statistical support for protocol analysis with SHAPA Version 2.01

This paper describes SHAPA Version 2.01, an interactive program for performing verbal protocol analysis. Verbal protocol analysis is a time-consuming activity that has hitherto typically been done by hand, whereas SHAPA represents an attempt to build a software environment to aid (but not replace) researchers in this activity. SHAPA allows researchers to develop an encoding vocabulary, to apply it to raw verbal protocol files, and to perform various types of data aggregation and data reduction. When performing verbal protocol analysis, researchers often try out different possible coding schemes before settling on the most effective one. SHAPA has been designed to support quick alterations to an encoding vocabulary and to support the rigorous statistical analysis of content and patterns (sequential data analysis) in protocol data. It is intended as an exploratory, as well as analytical, tool and has been designed to be easy for novices to learn and use, yet fluid and powerful for experts. A prototype version of SHAPA has already been used by a sample of researchers, and their experiences and requests have guided the programming of the present much more powerful version.     

Posttraumatische Belastungsstörung

There are some problems concerning the concepts of trauma-related disorders and especially the category of posttraumatic stress disorder (PTSD) with respect to classification, diagnosis and epidemiological findings, which can only be understood within the historical framework of the development. Even current diagnostic systems, such as the diagnostic and statistical manual of mental disorders version 4 (DSM-IV) and the international classification of diseases version 10 (ICD-10) differ in the classification of these disorders which led to higher prevalence rates being achieved with ICD-10. The diagnostic algorithms have been opened to included subsyndromal PTSD as well as complex PTSD including more severe psychiatric disorders.     

The Development and Testing of a New Version of the Cognitive Reflection Test Applying Item Response Theory (IRT)

The cognitive reflection test (CRT) is a short measure of a person's ability to resist intuitive response tendencies and to produce a normatively correct response, which is based on effortful reasoning. Although the CRT is a very popular measure, its psychometric properties have not been extensively investigated. A major limitation of the CRT is the difficulty of the items, which can lead to floor effects in populations other than highly educated adults. The present study aimed at investigating the psychometric properties of the CRT applying item response theory analyses (a two‐parameter logistic model) and at developing a new version of the scale (the CRT‐long), which is appropriate for participants with both lower and higher levels of cognitive reflection. The results demonstrated the good psychometric properties of the original, as well as the new scale. The validity of the new scale was also assessed by measuring correlations with various indicators of intelligence, numeracy, reasoning and decision‐making skills, and thinking dispositions. Moreover, we present evidence for the suitability of the new scale to be used with developmental samples. Finally, by comparing the performance of adolescents and young adults on the CRT and CRT‐long, we report the first investigation into the development of cognitive reflection. Copyright © 2015 John Wiley & Sons, Ltd.     

Traduction et validation française du questionnaire d’acceptation et d’action (AAQ-II)

Avoidance of painful psychological experience has been chosen as the prime target by several new therapeutic approaches derived from cognitive and behavioral therapies. Attempts to suppress unpleasant emotional events lead to a loss of psychological flexibility strongly correlated with a significant number and variety of psychological disorders. These new therapeutic approaches attempt to increase acceptance in order to decrease this psychological rigidity. Among these approaches, Acceptance and Commitment Therapy (ACT) measures psychological flexibility by mean of Acceptance and Action Questionnaire (AAQ-II), which presents good psychometric properties. A French version of the AAQ-II was evaluated in a group of 210 control subjects and 118 patients with anxiety disorders or depression. The French version of the AAQ-II presents good internal consistency (Cronbach's alpha: 0.76–0.82). Its concurrent validity and construct validity appear strong. The questionnaire is reliable in test–retest evaluation. The one-factor structure of the French version of the AAQ-II reproduces the structure of the original version. These results ensure the use of the French version of the AAQ-II in research as well as in clinical practice.     

Eating disorder NOS (EDNOS): an example of the troublesome "not otherwise specified" (NOS) category in DSM-IV

The “Not Otherwise Specified” (NOS) category within DSM-IV is designed for disorders of clinical severity that are not specified within broad diagnostic classes. “NOS” diagnoses are intended to be residual categories and they tend to be neglected by researchers. This can be inappropriate. The problems associated with certain NOS diagnoses are well illustrated by “Eating Disorder NOS” (sometimes termed EDNOS), which is the most common category of eating disorder encountered in routine clinical practice yet it has barely been studied. Indeed, there has been no research on its treatment. Interim and longer-term conceptual and practical solutions to the anomalous status of eating disorder NOS are proposed including the creation of a new diagnosis termed “mixed eating disorder”. Several of these solutions are of relevance to NOS categories in general. All the solutions should fulfil criteria for clinical utility.     

Moral Responsibility,Alternative Possibilities,and Acting on One’s Own

Frankfurt-style cases (FSCs) have famously served as counterexamples to the Principle of Alternative Possibilities (PAP). The fine-grained version of the flicker defense has become one of the most popular responses to FSCs. Proponents of this defense argue that there is an alternative available to all agents in FSCs such that the cases do not show that PAP is false. Specifically, the agents could have done otherwise than decide on their own, and this available alternative is robust enough to ground moral responsibility. In this paper, I argue that, when relying on definitions of ‘on one’s own’ within the literature on FSCs, a case can be constructed in which the agent could not have done otherwise than make a decision on his own. Insofar as this new case is successful, it will be able to avoid arguments about robustness while showing that moral responsibility does not require alternative possibilities of the type argued for by proponents of the fine-grained version of the flicker defense.

     

Does ‘Ought’ Imply ‘Might’? How (not) to Resolve the Conflict between Act and Motive Utilitarianism

Utilitarianism has often been understood as a theory that concerns itself first and foremost with the rightness of actions; but many other things (e.g., moral rules, motives, laws, etc.) are also properly subject to moral evaluation, and utilitarians have long understood that the theory must be able to provide an account of these as well. In a landmark article from 1976, Robert Adams argues that traditional act utilitarianism faces a particular problem in this regard. He argues that a on a sensible utilitarian account of the rightness of an agent’s motives, right motives will sometimes conflict with right actions, leaving the theory internally incoherent. The puzzle Adams raises has received a good deal of attention but few proposed solutions. Fred Feldman, however, has offered a solution that seems to be gaining adherents. In this paper I argue that Feldman’s approach cannot succeed. At bottom, it relies on a version of the principle that ‘ought’ implies ‘can’—and subsequently an account of an agent’s alternatives—that is far too restrictive to be plausible. Despite the failure of this solution, however, I argue that the conflict Adams develops is not as theoretically troubling as he suggests. While traditional act utilitarianism may fail for other reasons, it will not fail due to the conflict between acts and motives.     

Hypothetico-Deductivism is Still Hopeless

Since Christensen refuted the Bootstrap theory of confirmation in 1990, there have been some trials to improve the Hypothetico-Deductive theory of confirmation. After some trials, Gemes (1998) declared that his revised version completely overcame the difficulties of Hypothetico-Deductivism without generating any new difficulties. In this paper, I will assert that Gemes's revised version encounters some new difficulties, so it cannot be a true alternative to the Bootstrap theory of confirmation and to classical Hypothetico-Deductivism. Also I will assert that, in principle, such new difficulties cannot be overcome by any trials dependent only on formal logic.     

Reasons for Randomness: A Solution to the Axiological Problem for Theists

Abstract

In this paper we bring to light several ways randomness—i.e., undetermined and unintended events—may contribute to our understanding of God's providence and personality. We begin by making clearer a certain problem that randomness has been thought to pose to theism. We then discuss recent criticisms of certain contemporary solutions to this problem that emphasize the value of an autonomous creation. From there, we propose a fresh way of understanding the value of a semi-autonomous creation that does not succumb to these recent critiques. Our end goal is to explore new reasons God might have to value randomness. In particular, we highlight two plausible, interrelated candidate values: (1) There are certain aesthetic properties that a partially random, self-forming creation enjoys; and (2) Such a creation grants God and creatures certain pleasures, such as wonder, anticipation, curiosity, surprise, and appreciation. In articulating our version of the autonomy defense, we position it within two opposing accounts of divine providence, specifically open theism and simple foreknowledge.     

Is a new version of philosophical pragmatism necessary? A reply to Barnes-Holmes

Barnes-Holmes (2000) discussed certain issues regarding philosophy, pragmatism, and behavior analysis, and offered a "behavioral pragmatism" based on or derived from behavior-analytic perspectives. In a comparison of certain philosophical views, Quine's concept of observation sentences was employed for representing pragmatism, but this concept is not sufficiently representative of the literature of philosophical pragmatism to warrant the broad conclusions drawn by Barnes-Holmes. Further, although the extensive and diverse literature of philosophical pragmatism has been shown by a number of writers to have various themes and perspectives in common with Skinner's radical behaviorism, it is unnecessary to extract a limited, generic version of pragmatism because (a) the latter cannot match the range and depth of the various extant versions and (b) the problems raised by Barnes-Holmes in justification for the new version yield readily to the current versions in philosophy. A set of philosophical views may provide additional verbal support for a given system of science, and the science of behavior analysis may eventually contribute to philosophical discourse. The latter, however, will not be achieved by proposing new versions of old philosophy, but rather by approaching established philosophical issues in new ways.     

Bias effects in word fragment completion in young and older adults

Young and older adults were tested on a word fragment completion task in which correct solutions were studied words, words orthographically similar to studied words, or new words. In Experiments 1 and 2, the standard production version of the word fragment completion task was used; older adults had reduced benefits of prior exposure to target words and slightly decreased costs. However, costs and benefits did not differ across age in a forced-choice version of the task (Experiment 3). At a behavioral level, the results are contrary to predictions that age differences in word fragment completion priming effects will be greater when there is a strong competitor for the correct solution and that age differences in both costs and benefits will be smaller for identification than for production tasks. Theoretical implications of these findings are considered.     

A psychometric evaluation of the social skills rating system in children with attention deficit hyperactivity disorder

The social skills rating system (SSRS) was developed to assess social skills of children, as observed by multiple raters (teacher, parent, child). Studies of the SSRS have been conducted with handicapped, mentally retarded and learning disabled children. No studies have reported the psychometric properties of the SSRS in a clinical ADHD sample. This is important, because deficient social functioning is associated with ADHD. The present study assesses the psychometric properties of the teacher, parent and child versions of the SSRS in children with ADHD (n = 123), and normal controls (n = 239). Also, the social skills of children with ADHD, as rated on the SSRS were examined. Results support the factor structure and internal consistency of the original SSRS-teacher version. Moreover, support was found for 3 out of 4 scales of the SSRS-parent version. The factor structure of the SSRS-child version could not be replicated. An explanatory factor analysis on the SSRS-child version yielded two factors. Evidence was found for discriminative ability of the SSRS between normal controls and children with ADHD. Finally, informant agreement between raters was found to be poor.