Conceptual Complexity and Construct Validity

Despite a century of methodological and conceptual advances in the technology of psychosocial measurement, poor correspondence between indicators and the constructs they are intended to represent remains a limiting factor to the accumulation of scientific knowledge. Longstanding conventions in measurement may contribute to the failure to develop optimal criteria. These conventions include the focus on complex over simple constructs and the use of multi-item measures of disparate content to represent those constructs. Several arguments suggest that such a measurement model compromises the potential for developing measures that accurately reflect psychosocial phenomena. The article concludes with some preliminary suggestions concerning an alternative model that may address this construct validity problem more effectively.     

Assessing the validity of the achievement motive in the presence of random measurement error

The presence of random measurement error in indicators of theoretical constructs biases observed estimates of relations among those constructs. Correcting for this bias is particularly important when random measurement error is substantial, or is substantially different for indicators of distinct constructs included in a theoretical model. Validity assessment in the case of thematic apperceptive measures of the achievement motive (TAT n Achievement) has been vulnerable to interpretive errors because these indicators of the achievement motive are typically much less reliable than indicators of other constructs to which the motive may be related, and no correction has been made for the bias introduced by such differential measurement error. Here we illustrate a causal modeling approach to validity assessment for TAT n Achievement that incorporates explicit true-score measurement models of theoretical constructs. We confirm the hypothesis that the achievement motive construct is positively related to work satisfaction in a representative sample of adult males in the United States, taken in 1976 (N = 413). Evidence for the discriminant validity of story content versus story length, an issue raised in the literature on the TAT, is also presented in this nomological network.     

Positive Indicators of Child Well-Being: A Conceptual Framework, Measures, and Methodological Issues

This article suggests a comprehensive conceptual framework which identifies constructs for positive well-being as well as potential indicators and extant measures that fit with those constructs. In addition, the article reviews existing data sources for examples of positive measures that are found in the proposed framework as well as research studies that have been successful in measuring these indicators. The article then notes the data and measurement gaps that exist in comprehensively measuring the positive in children and youth. Finally, it identifies a number of conceptual and methodological issues that need consideration as efforts to define and measure positive indicators of well-being and well-becoming go forward. We argue that positive indicators represent good science, reflect and promote a positive vision of children, and establish positive goals for nations.     

The effect of Method Variance on Relationships Between the Work Ethic and Individual Difference Variables

A great deal of research relating the work ethic to other individual difference variables has been conducted using questionnaires to measure all the variables. Many researchers believe that using a single method of measurement for multiple constructs results in artificially inflated correlations. A Confirmatory Factor Analysis of the work ethic, locus of control, and conscientiousness reveals that approximately 16% of the variance in measures of these constructs is a function of the method of measurement. This suggests that correlations in work ethic research based on only one method of measurement may be overstated.     

Validation of scales measuring attitudes and norms related to mammography screening in women veterans.

Validation of psychosocial measures for use in mammography screening research has been given inadequate attention in the literature. The authors report on the validation of 5 measures examining 4 attitudinal constructs (i.e., pros, cons, outcome expectations, and cancer worries) and 1 social influence construct (i.e., subjective norms) in a 22-item inventory. The study participants consisted of a national, randomly sampled population of women veterans (n = 2,910). After minor revision of scales, the authors found independent measures for 4 constructs: pros, cons, cancer worries, and subjective norms. The authors concluded that these scales have acceptable psychometric properties; support construct validity; and provide brief, reliable, and valid measures of attitudes toward and norms regarding mammography screening. These scales may be useful for intervention research.     

Shared, not unique, components of personality and psychosocial functioning predict depression severity after acute-phase cognitive therapy

In a sample of 100 patients with recurrent major depression, we collected depression severity data early and late in acute-phase cognitive therapy, plus a wide range of psychosocial variables that have been studied extensively in depression research, including measures of interpersonal, cognitive, and social functioning, and personality traits using an inventory that is linked with the Big-Three tradition in personality assessment theory. By assessing this broad range of variables in a single study, we could examine the extent to which relations of these variables with depression were due to (a) a common factor shared across this diverse set of constructs, (b) factors shared among each type of construct (personality vs. psychosocial measures), or (c) specific aspects of the individual measures. Only the most general factor shared across the personality and psychosocial variables predicted later depression.     

The problem of measurement model misspecification in behavioral and organizational research and some recommended solutions

The purpose of this study was to review the distinction between formative- and reflective-indicator measurement models, articulate a set of criteria for deciding whether measures are formative or reflective, illustrate some commonly researched constructs that have formative indicators, empirically test the effects of measurement model misspecification using a Monte Carlo simulation, and recommend new scale development procedures for latent constructs with formative indicators. Results of the Monte Carlo simulation indicated that measurement model misspecification can inflate unstandardized structural parameter estimates by as much as 400% or deflate them by as much as 80% and lead to Type I or Type II errors of inference, depending on whether the exogenous or the endogenous latent construct is misspecified. Implications of this research are discussed.     

The theoretical status of latent variables

This article examines the theoretical status of latent variables as used in modern test theory models. First, it is argued that a consistent interpretation of such models requires a realist ontology for latent variables. Second, the relation between latent variables and their indicators is discussed. It is maintained that this relation can be interpreted as a causal one but that in measurement models for interindividual differences the relation does not apply to the level of the individual person. To substantiate intraindividual causal conclusions, one must explicitly represent individual level processes in the measurement model. Several research strategies that may be useful in this respect are discussed, and a typology of constructs is proposed on the basis of this analysis. The need to link individual processes to latent variable models for interindividual differences is emphasized.     

Reconsidering formative measurement

The relationship between observable responses and the latent constructs they are purported to measure has received considerable attention recently, with particular focus on what has become known as formative measurement. This alternative to reflective measurement in the area of theory-testing research is examined in the context of the potential for interpretational confounding and a construct's ability to function as a point variable within a larger model. Although these issues have been addressed in the traditional reflective measurement context, the authors suggest that they are particularly relevant in evaluating formative measurement models. On the basis of this analysis, the authors conclude that formative measurement is not an equally attractive alternative to reflective measurement and that whenever possible, in developing new measures or choosing among alternative existing measures, researchers should opt for reflective measurement. In addition, the authors provide guidelines for researchers dealing with existing formative measures.     

Buyers’ Trust of the Salesperson: An Item-Level Meta-Analysis

The issue of trust is an important one in the marketing literature. To assess the nomological validity of trust-related measures, this research uses an item-level measurement meta-analysis based on correlations from 32 studies that measure trust-related constructs in the context of a buyer’s assessments of sellers. Analysis reveals that 16-item measures converge into three constructs that are indicative of the seller’s credibility, expertise, and compatibility and three items each that are indicative of trust and trustworthiness. In a structural equation model based on the meta-analysis, a seller’s characteristics are predictive of trust and trustworthiness.     

An Investigation of the Structural Validity of Loevinger's Model and Measure of Ego Development

This study demonstrates the use of structural equation modeling for investigating a hierarchical model of personality based on Loevinger's (1976) conceptualization. This model was constructed to have a multifactorial and hierarchical structure in which the ego development construct served as the source of covariation for specific personality constructs of impulse control, interpersonal style, conscious preoccupations, and cognitive style. This model and three other conceivable models were investigated to determine which structure best represented the relations among specific personality constructs and ego development. The specific personality constructs were identified by measures chosen from the available pool of measures that were compatible with Loevinger's conceptualization. The four conceivable models were fitted to sample data from 267 adult subjects using confirmatory factor analysis and were compared with indices of model fit and methods for comparison of nested models. Results suggested that the four specific personality constructs and ego development are part of the same process. Although a hierarchical structure appeared tenable for this process, the role of ego development perhaps is not as dominant as envisioned by Loevinger. These results must be considered tentative due to measurement limitations.     

A checklist for testing measurement invariance

The analysis of measurement invariance of latent constructs is important in research across groups, or across time. By establishing whether factor loadings, intercepts and residual variances are equivalent in a factor model that measures a latent concept, we can assure that comparisons that are made on the latent variable are valid across groups or time. Establishing measurement invariance involves running a set of increasingly constrained structural equation models, and testing whether differences between these models are significant. This paper provides a step-by-step guide to analysing measurement invariance.     

Reflections On The Intensive Measurement Of Stress,Coping, And Mood,With An Emphasis On Daily Measures

Abstract

This paper discusses reasons for the shift from longer-term measurement of stress, coping, and mood to more intensive measurement strategies. Daily assessments are common in intensive studies of these constructs. We review measures available to the researcher, focusing on various aspects of the constructs. While daily assessments are an improvement over less intensive methods, we also discuss shortcomings of these measures. Furthermore, current assessments may not capture many important dimensions of stressful occurrences, and we suggest that such occurrences be thought of in terms of episodes. Modifications of current assessments are discussed, as are even more intensive assessments of stress, coping, and mood.     

A comparison of Holland's model using constructs measured by two different methods

The relationship was studied between the set of constructs specified by Holland's 1973 theory (viz, Realistic, Intellectual, Artistic, Social, Enterprising, and Conventional) measured by an Adjective List (AL) derived from Holland's theory and the same set of constructs as measured by the Vocational Preference Inventory (VPI). The correspondence of the two measures of Holland's model compared the distances between the constructs as measured by the AL and the constructs measured by the VPI in factor space using an adaptation of a model testing procedure reported by Wakefield and Doughtie 1973. The subjects were 174 undergraduates at a state university. The correspondence between the two separate measures of Holland's model was significant (p < .01). These results provide further support of the construct validity of Holland's theory and offer evidence that the relationships hypothesized by the theory are not tied to any one particular method of measurement.     

The Theory of Planned Behavior Without Compatibility? Beyond Method Bias and Past Trivial Associations1

Overreliance on one measurement approach can challenge accurate statements about reality, as findings can represent by‐products of the compulsory measurement paradigm. Within the theory of planned behavior (TPB), the compatibility principle represents one such strictly imposed paradigm. Using 2 cross‐sectional surveys of 1,394 volunteers and involving structural equation models, we demonstrate that the widely employed practice of measuring TPB constructs is confounded with method‐implied bias. This means the theory cannot conclusively reveal origins of a behavior. Our results also suggest that on an aggregated level, when method bias is eliminated, its constructs are linked in hypothesized ways. Adopting a more general model—thus, adopting a more traitlike conceptualization of attitudes—has interesting implications for social psychology and its current trends.     

Validating an older adult driving behaviour model with structural equation modelling and confirmatory factor analysis

The Multilevel Older Persons Transportation and Road Safety (MOTRS) model postulates a role for demographic, functional, and psychosocial factors in driving self-regulation. This study formally tested the constructs of the MOTRS model, and its ability to account for variation in older adult’s driving self-regulation. Four exogenous constructs dealing with sociodemographic and driving-related variables (driving space, dependency on other drivers, health, and driving performance) and three endogenous constructs of psychosocial appraisals (driving confidence, attitudes and beliefs towards driving) were used to predict driving self-regulation. A sample of 277 Australian drivers, aged 65 years or over, completed standardised questionnaires to assess these variables. Structural equation modelling was used to evaluate one of the model’s assumptions. Findings established that sociodemographic and driving-related factors influence older adults’ driving self-regulation by impacting psychosocial variables. A re-specified model accounted for 75% of the variance in self-reported driving self-regulation. When other variables were considered, attitude towards driving was the strongest single predictor of driving self-regulation. These findings provide an initial partial validation of the MOTRS model. The results indicate that psychosocial variables are important determinants of driving self-regulation by older adults. This model could guide future research into older adults’ driving self-regulation.     

The reliability and validity of the Perceived Stigmatization Questionnaire (PSQ) and the Social Comfort Questionnaire (SCQ) among an adult burn survivor sample

In this study, 361 adult burn survivors completed the Perceived Stigmatization Questionnaire (PSQ), the Social Comfort Questionnaire (SCQ), and other measures. Both the PSQ and SCQ had good internal consistency indices. Factor analysis of the PSQ yielded 3 factors (absence of friendly behavior, confused/staring behavior, and hostile behavior). The SCQ had 1 factor. Conjoint factor analysis with measures of related constructs (body esteem, body-esteem importance, depression, social support) suggested that PSQ and SCQ measure distinct constructs. Correlations with the related psychosocial constructs and burn characteristics suggested the PSQ and SCQ have good convergent and discriminant validity. Limitations of the study are discussed.     

The Relation Between Self-Continuity and Measures of Identity

Conceptualizations of Erikson's (1968) fifth psychosocial stage, identity formation, have been criticized for failing to be inclusive of an individual's sense of self-continuity. However, recent research has shown that identity commitment is significantly correlated to measures of continuity. This study further examined the relation among measures of identity, including identity commitment and continuity. The results suggest that many of the current measures of identity do indeed share a significant amount of variance with continuity. The specific results concerning identity commitment and continuity add to the growing evidence that these constructs are reliably related.