Separating trait effects from trait-specific method effects in multitrait-multimethod models: a multiple-indicator CT-C(M-1) model

An overview of several models of confirmatory factor analysis for analyzing multitrait-multimethod (MTMM) data and a discussion of their advantages and limitations are provided. A new class of multi-indicator MTMM models combines several strengths and avoids a number of serious shortcomings inherent in previously developed MTMM models. The new models enable researchers to specify and to test trait-specific-method effects. The trait and method concepts composing these models are explained in detail and are contrasted with those of previously developed MTMM models for multiple indicators. The definitions of the models are explained step by step, and a practical empirical application of the models to the measurement of 3 traits x 3 methods is used to demonstrate their advantages and limitations.     

The double-anchoring theory of lightness perception: a comment on Bressan (2006)

Recently, a double-anchoring theory (DAT) of lightness perception was proposed (P. Bressan, 2006), which offers explanations for all the data explained by the original anchoring theory (A. Gilchrist et al., 1999), as well as a number of additional lightness phenomena. Consequently, DAT can account for an unprecedented range of empirical results, potentially explaining everything from the basic simultaneous contrast display to subtle variations of the Gelb effect. In this comment, the authors raised 4 concerns that demonstrate serious theoretical and empirical difficulties for DAT.     

Predicting group-level outcome variables from variables measured at the individual level: a latent variable multilevel model

In multilevel modeling, one often distinguishes between macro-micro and micro-macro situations. In a macro-micro multilevel situation, a dependent variable measured at the lower level is predicted or explained by variables measured at that lower or a higher level. In a micro-macro multilevel situation, a dependent variable defined at the higher group level is predicted or explained on the basis of independent variables measured at the lower individual level. Up until now, multilevel methodology has mainly focused on macro-micro multilevel situations. In this article, a latent variable model is proposed for analyzing data from micro-macro situations. It is shown that regression analyses carried out at the aggregated level result in biased parameter estimates. A method that uses the best linear unbiased predictors of the group means is shown to yield unbiased estimates of the parameters.     

A comment on Birnbaum's three-parameter logistic model in the latent trait theory

Birnbaum's three-parameter logistic model for the multiple-choice item in the latent trait theory is considered with respect to the item response information function and the unique maximum condition. It is clarified that with models of knowledge or random guessing nature, which include the three-parameter logistic model, the unique maximum condition is not satisfied for the correct answer, and the item response information function is negative for the interval (− ∞,θ _g ). It is suggested that we should useθ _g as a criterion in selecting optimal items for a specified group of examinees, so that we can practically avoid the possibility of non-unique maxima of the likelihood function on the response pattern given by an examinee in the group. The work described in this paper was partially done while the author was at University of New Brunswick, Canada, in 1968–1970, supported by NRC Grant APA-345 from National Research Council of Canada.     

Consequential validity of the implicit association test: comment on Blanton and Jaccard (2006)

Numeric values of psychological measures often have an arbitrary character before research has grounded their meanings, thereby providing what S. J. Messick (1995) called consequential validity (part of which H. Blanton and J. Jaccard now identify as metric meaningfulness). Some measures are predisposed by their design to acquire meanings easily, an example being the sensitivity measure of signal detection theory. Others are less well prepared, illustrated by most self-report measures of self-esteem. Counter to Blanton and Jaccard's characterization, the Implicit Association Test (IAT) has properties that predispose it to acquire consequential validity rapidly. With the IAT as the subject of over 250 publications since 1998, there is now much evidence for its consequential validity. The IAT has attracted more scholarly criticism than have other measures designed for similar purposes. The authors speculate as to why the IAT is an attractive target.     

What can be measured in dating data: comment on Vaillant (2006)

Vaillant's 2006 paper presents data from a marriage bureau to give evidence on partner-seeking behavior. Three questions may be asked about the nature and usefulness of such work, whether it can (i) measure the efficiency of such bureaux, (ii) indicate the volume of resources devoted to partner search from commercially sourced dating data, and (iii) be used to test hypotheses about preference formation.     

Toward a process-based framework for classifying personality tests: comment on Meyer and Kurtz (2006)

Meyer and Kurtz (2006) argued that the longstanding psychological test labels "objective" and "projective" have outlived their usefulness, and invited further work focusing on alternative terms for these measures. This Comment describes a framework for classifying personality tests based on the psychological processes that occur as people respond to test stimuli. Because an attribution process is involved in responding to both types of measures, those instruments formerly called "objective" tests are labeled self-attribution tests, and those formerly classified as "projective" tests are labeled stimulus-attribution tests. The possibility of extending the process-based framework beyond personality, to psychological tests in general, is also discussed. Clinical and empirical implications of a process-based framework are considered.     

The meaning and computation of causal power: comment on Cheng (1997) and Novick and Cheng (2004)

D. Hume (1739/1987) argued that causality is not observable. P. W. Cheng (1997) claimed to present "a theoretical solution to the problem of causal induction first posed by Hume more than two and a half centuries ago" (p. 398) in the form of the power PC theory (L. R. Novick & P. W. Cheng, 2004). This theory claims that people's goal in causal induction is to estimate causal powers from observable covariation and outlines how this can be done in specific conditions. The authors first demonstrate that if the necessary assumptions were ever met, causal powers would be self-evident to a reasoner--they are either 0 or 1--making the theory unnecessary. The authors further argue that the assumptions the power PC theory requires to compute causal power are unobtainable in the real world and, furthermore, people are aware that requisite assumptions are violated. Therefore, the authors argue that people do not attempt to compute causal power.     

Factor analysis with (mixed) observed and latent variables in the exponential family

We develop a general approach to factor analysis that involves observed and latent variables that are assumed to be distributed in the exponential family. This gives rise to a number of factor models not considered previously and enables the study of latent variables in an integrated methodological framework, rather than as a collection of seemingly unrelated special cases. The framework accommodates a great variety of different measurement scales and accommodates cases where different latent variables have different distributions. The models are estimated with the method of simulated likelihood, which allows for higher dimensional factor solutions to be estimated than heretofore. The models are illustrated on synthetic data. We investigate their performance when the distribution of the latent variables is mis-specified and when part of the observations are missing. We study the properties of the simulation estimators relative to maximum likelihood estimation with numerical integration. We provide an empirical application to the analysis of attitudes.     

On the meaning of formative measurement and how it differs from reflective measurement: comment on Howell, Breivik, and Wilcox (2007)

D. Howell, E. Breivik, and J. B. Wilcox (2007) have presented an important and interesting analysis of formative measurement and have recommended that researchers abandon such an approach in favor of reflective measurement. The author agrees with their recommendations but disagrees with some of the bases for their conclusions. He suggests that although latent variables refer to mental states or mental events that have objective reality, to gain knowledge of the existence of these states or events requires that emphasis be placed on the nature and interpretation of the relationship between latent and manifest variables. This relationship is not a causal one but rather a kind of correspondence rule that contains theoretical, empirical, operational, and logical meanings as part of its content and structure. Implications of the above views are discussed for formative and reflective measurement.     

Partial support from a non-replication: comment on Roma, Silberberg, Ruggiero, and Suomi (2006)

A purported replication by P. G. Roma, A. Silberberg, A. M. Ruggiero, and S. J. Suomi of the authors' previous study claims to contradict their finding that capuchin monkeys (Cebus apella) refuse to exchange with an experimenter if their partner receives a superior reward. Roma et al. used no exchange task, however, or any other task. Roma et al. offered frustration as explanation of their findings, yet failed to statistically prove that the effect of frustration is stronger than that of inequity. They also misrepresented the dependent measure of the authors' study. Reanalysis of the authors' own data indicated no role of frustration, that is, no effect of previous experience with a superior reward. The authors conclude that Roma et al.'s study is not a replication and does not disprove the authors' findings.     

Latent mnemonic strengths are latent: A comment on Mickes, Wixted, and Wais (2007)

Mickes, Wixted, and Wais (2007) proposed a simple test of latent strength variability in recognition memory. They asked participants to rate their confidence using either a 20-point or a 99-point strength scale and plotted distributions of the resulting ratings. They found 25% more variability in ratings for studied than for new items, which they interpreted as providing evidence that latent mnemonic strength distributions are 25% more variable for studied than for new items. We show here that this conclusion is critically dependent on assumptions—so much so that these assumptions determine the conclusions. In fact, opposite conclusions, such that study does not affect the variability of latent strength, may be reached by making different but equally plausible assumptions. Because all measurements of mnemonic strength variability are critically dependent on untestable assumptions, all are arbitrary. Hence, there is no principled method for assessing the relative variability of latent mnemonic strength distributions.     

Observed and latent variables: Discussion of a basic LISREL paradigm

The aim of the present paper is to discuss a general paradigm of LISREL: that observed psychologocal variables are causally determined by non-observed, latent variables. An example from the LISREL manual is used to show that this basic assumption is not generally meaningful. The possibility of understanding the theoretical variables of LISREL in an instrumentalist way is also considered. But instrumentalism, it is concluded, does not provide a tenable way of defending the LISREL paradigm. For LISREL conceives latent variables as causes of observed variables, and this is not compatible with an instrumentalist understanding of theoretical terms.     

On the status of self in social prediction: comment on Karniol (2003)

A challenge to mainstream notions on the status of the self in social prediction is welcome. The self-as-distinct model (R. Karniol, 2003) is thoughtful, provocative, and parsimonious, but it is also underspecified, undertested, and selective in its treatment of the evidence. More important, the model does not provide compelling answers to issues pertaining to the origins of prototypic social knowledge, the status of self-knowledge, the content of the self-representation, whether the use of self in social prediction is a logical contradiction, and whether the self's role in social prediction is amotivated.     

Anxiety,aspirations, and self-concept in the achievement process: A longitudinal model with latent variables

Recent research suggests that anxiety is not a single, unified reaction to perceived threat, but rather a cluster of interrelated factors whose relationships to performance change as the individual progresses from one test event to another. This study investigated the presumed linkages between traitlike predispositions to perceive threat and achievement performance, as mediated by statelike anxiety arousal on a longitudinal basis (Perceived Threat Anxiety Arousal Impaired Performance). College students were administered self-report questionnaire measures during a preenrollment period, after the first two midterms, and following the last two midterms in a general psychology course. Four performance measures and 26 motivational indicators were fitted to a 10-factor latent model using LISREL model-fitting techniques. Path-analytic interpretations of this structural model provided little evidence for the commonly held view that traitlike threat perceptions mediate performance via statelike anxiety reactions. Far more promising, theoretically, are those influences on test performance stemming from the self-attributional, cognitive domain. Overall, the findings support a recent reinterpretation of achievement anxiety as stemming from the disruptive effects of diminished ability perceptions (and hence, impaired personal worth), rather than from the interfering influence of diffused emotional arousalper se.     

Faulty assumptions: A comment on Blanton, Jaccard, Gonzales, and Christie (2006)

Blanton, Jaccard, Gonzales, and Christie [Blanton, H., Jaccard, J., Gonzales, P., Christie, C. (2006). Decoding the Implicit Association Test: implications for criterion prediction. Journal of Experimental Social Psychology, 42, 192-212.] assert that the Implicit Association Test (IAT) imposes a model that portrays relative preferences as the additive difference between single attitudes. This assertion is misplaced because relative preferences do not necessarily reduce to component attitudes. BJGC also assume that the IAT conditions represent two indicators of the same construct. This assumption is incorrect, and is the cause of their poor-fitting models. The IAT, like other experimental paradigms, contrasts performance between interdependent conditions, and cannot be reduced to component parts. This is true whether calculating a simple difference between conditions, or using the IAT D score. D—an individual effect size that is monotonically related to Cohen’s d—codifies the interdependency between IAT conditions. When their unjustified psychometric assumptions are replaced with plausible assumptions, the models fit their data very well, and basis for their poor-fitting models becomes clear.     

Associating versus proposing or associating what we propose: comment on Gawronski and Bodenhausen (2006)

This commentary highlights the strengths of the associative-propositional evaluation model. It then describes problems in proposing a qualitative separation between propositional and associative processes. Propositional processes are instead described as associative. Propositions are ordered associations, whereas many other associations do not depend on the order of the involved elements. Implications of this alternative definition for the phenomenology of thought and for social psychology are discussed.