Structural Equation Methodology Research Articles

Redundancy in constructs: Problem, assessment, and an illustrative example

This article argues that attention to redundancy in constructs is important for theoretical, empirical, and pragmatic reasons. In particular, the article advocates that redundancy between two or more constructs be evaluated from conceptual and empirical perspectives. To facilitate such assessment, this article outlines a systematic, multipronged approach. This approach is based on a directed review of the literature and well-defined methods and principles, such as the Latent Variable Structural Equations (LVSE) methodology, and discriminant, convergent, and nomological validity. The author illustrates the suggested approach by utilizing two constructs often used to assess consumers' attitudes toward business practices— consumer alienation and discontent. The illustration underscores the usefulness of the suggested approach in revealing redundancy gaps that heretofore have remained unidentified. Specifically, consumer discontent and alienation are found to be conceptually nonredundant, but their current operationalizations are not. Implications for future research into redundancy issues are discussed.

Japanese migration in contemporary Japan: Economic segmentation and interprefectural migration

This paper examines the economic segmentation model in explaining 1985-86 Japanese interregional migration. The analysis takes advantage of statistical graphic techniques to illustrate the following substantive issues of interregional migration: (1) to examine whether economic segmentation significantly influences Japanese regional migration and (2) to explain socioeconomic characteristics of prefectures for both in- and out-migration. Analytic techniques include a latent structural equation (LISREL) methodology and statistical residual mapping. The residual dispersion patterns, for instance, suggest the extent to which socioeconomic and geopolitical variables explain migration differences by showing unique clusters of unexplained residuals. The analysis further points out that extraneous factors such as high residential land values, significant commuting populations, and regional-specific cultures and traditions need to be incorporated in the economic segmentation model in order to assess the extent of the model's reliability in explaining the pattern of interprefectural migration.

Factors Affecting Alcohol Consumption in Black Women. Part II

An eight-variable model for understanding and predicting alcohol consumption in a sample of 289 African American women is evaluated using a structural equation methodology. We found that life events, physical health problems, and internalized racialism played important roles in accounting for variance in alcohol consumption. Marital status did not have the predicted inverse effect on alcohol consumption. While religious orientation did not have the expected inverse effect on alcohol consumption, it had an unexpected direct effect on internalized racialism, which had a direct effect on alcohol consumption. We found that the effects of socioeconomic status and developmental status on alcohol consumption were mediated through other variables specified in the model. Overall the model, which provided partial to complete support for five of eight hypotheses, provided a statistically adequate fit.

The matching model: An examination of the processes underlying realistic job previews.

Using information systems and data processing personnel in 9 major insurance companies, this research assessed the validity of Wanous's (1980)matching model as a framework of the processes linking realistic job previews (RJPs) to employee adjustment and employment stability. Because the matching model is an adaptation of the Theory of Work Adjustment (TWA), this study also assessed the TWA's applicability to RJP and turnover research. Structural equation methodology tested the fit of the matching model to the data of newcomers and of more tenured employees. Also, two-group structural analysis was performed to determine differences between groups. Support for the matching model was obtained with newcomer data but not with data from tenured employees. Results also supported the applicability of the TWA to RJP and turnover research. Implications of the findings to work adjustment and RJP research are discussed.

Evaluating the Long Range Implications of Compliance and Internalization on Turnover Attitudes and Behavior.

The long term impact of two forms of psychological attachment, compliance and internalization, on employee turnover was supported with structural equation methodology. Compliance positively influen...

A Longitudinal Study of Mobley's (1977) Turnover Model.

This study evaluated the validity of Mobley's (1977) theory. Using a longitudinal design, 126 nurses were surveyed on three occasions which assessed all model constructs. Structural equation methodology failed to confirm Mobley's model. Instead, another model specifying alternative causal pathways better accounted for the data.

Narcotics use, property crime, and dealing: Structural dynamics across the addiction career

Structural-equation modeling is used to test causal relationships between narcotics addiction and the associated cost-support activities of property crime and drug dealing across four critical periods of the addiction career. It is argued that structural-equation methodology yields greater insight into the causal dynamics of such activities than the typical methodologies of comparing means and proportions. Using structural coefficients of longitudinal models to infer causal relationships and stability, it is found that (1) dealing is often a predictor of future narcotics involvement; (2) narcotics use, property crime, and dealing are mutually interrelated during periods of elevated narcotics use and are not related during periods of reduced narcotics use; (3) dealing appears to be the most stable of the three variables, although narcotics use and property crime show appreciable stability as well; and (4) property crime and dealing appear to be inversely related to a moderate extent. The relevance of these findings to the understanding of the economic behavior of addicts during the initiation and cessation of addiction, relapse, and “maturing out” is discussed.

Labor force participation and fertility: a social analysis of their antecedents and simultaneity.

The nature of the relationship between labor force participation and fertility is examined for 172 families comprising a marriage cohort. The results show that labor force participation and fertility are spuriously related in that they are under the common influence of sex-role norms, attitudes toward the wife working, and the value or importance of children for the spouses. Comparisons are made among social psychological models, wherein either the husband or wife provides information as they perceive their relationship, and a sociological model wherein group constructs are formed with the husband and wife acting as informants on the pattern of norms guiding their relation. A structural equation methodology is employed to better model measurement error and errors in equations simultaneously.

Re-Analysis of Morse Code Learning Data Using Structural Equation Models

The correlation matrix of 14 tests and the time to reach four levels of Morse code proficiency reported in Fleishman, Roberts and Friedman (1958) and Fleishman and Fruchter (1960) was analyzed using structural equation methodology. This methodology was not available at the time of the original research and is especially suited for the data in these studies. The method is based on the standard maximum-likelihood approach for parameter estimation and goodness of fit. Five factors were extracted from the correlation matrix of 14 tests. These factors were significantly correlated and differed only slightly from the factors obtained in the original factor analysis. The factors were auditory perceptual speed, auditory rhythm perception, speed of closure, visualization and verbal ability. The time to reach a level of proficiency on Morse code was significantly predicted by the time to reach the previous two proficiency levels. The Auditory Perceptual Speed factor was a significant predictor of Morse code learning to each of the four proficiency levels. A speed of closure factor significantly predicted learning to the second proficiency level. The visualization factor was a significant negative predictor of learning to the 2nd and 3rd proficiency levels. This result suggests that individuals who are high in visualization ability may adopt a visual strategy (such as visualizing dots and dashes) that is detrimental to performance at the middle learning levels. In contrast to the earlier findings, the auditory rhythm factor was unrelated to reaching any of the proficiency levels. Overall, the results demonstrate the usefulness of correlation structure models in the area of selection and training.

A Model of the Distributor's Perspective of Distributor-Manufacturer Working Relationships

Building upon work from social exchange theory and channels of distribution, a model of distributor-manufacturer working relationships from the distributor's perspective is presented. An initial empirical test, using a structural equation methodology, provided acceptable support of the model, given some measurement limitations. Further work on modeling both perspectives of the exchange relationship is discussed.

Fertility, labor force participation, and tastes: An economic psychology perspective

Economic and economic psychology models of wife's labor force participation and family size decisions are examined within the context of a marriage cohort. Among other results, it was found that tastes influence the demand for work and number of children, along with traditional economic determinants such as opportunity costs and income. Tastes have an especially strong impact when modeled as unobservables, in order to correct for measurement error. Further, the often observed empirical association between labor force participation and fertility is found to be spurious and not necessarily causal in nature. That is, labor force participation and fertility negatively covary largely as a consequence of common economic and taste antecedents and not due to reciprocal interactions. Hypotheses were tested using a new structural equation methodology.

An Examination Of The Validity Of Two Models Of Attitude.

A structural equation methodology is used to assess the reliability and validity of expectancy-value and semantic differential measures of attitude toward the act of donating blood. Hypotheses as to the internal consistency and the convergent, concurrent, discriminant, predictive, and nomological validities of responses are tested in the context of a quasi-experiment performed on 284 students, faculty, and staff. Semantic differential attitudes are found to exist as unidimensional responses while expectancy-value attitudes are found to occur as multidimensional reactions. However, although both operationalizations achieve convergent, concurrent, discriminant, and predictive validity, the nomological validity of attitude is brought into question. The extent of past behavior and the elapsed time since performance of that behavior appear to offer the greatest explanatory content for subsequent behavioral intentions. Finally, the role of personal and social normative beliefs as copredictors of intentions along with attitude and past behavior is also investigated. The findings are interpreted from attitude, learning, and attribution theory arguments.

The Nature and Causes of Self-Esteem, Performance, and Satisfaction in the Sales Force: A Structural Equation Approach

Research into the sales force suggests that the behavior of salespeople will be a function of the person; the interactions the person has with customers, managers, and significant others; and the situation or context in which these interactions take place (Churchill, Ford, and Walker 1976; Bagozzi 1978). This article addresses the impact of two classes of these determinantsthat is, the person and role relationships-as they influence particular outcomes on the job. Specifically, motivation and felt role strain and role ambiguity are investigated as they affect certain emotional, cognitive, and behavioral (i.e., performance) outcomes of salespeople. The discussion begins with a conceptualization of three broad classes of job outcomes experienced by salespeople. Next, motivation, role strain, and role ambiguity are defined, and the mechanisms relating these independent variables to job outcomes are specified. Third, the methodology and structural equation models used to test the hypotheses are presented. Included in the treatment is a discussion of the sample and Salesmen experience three broad outcomes on the job. The first deals with feelings about the self and is termed self-esteem. The second refers to actual performance, such as dollar volume of sales achieved, new business generated, or expenses incurred. The third represents evaluations about specific dimensions of the work situation and is termed here job satisfaction. This article presents the results of a study investigating the determinants of each type of job outcome for a sample of industrial salesmen. Using a structural equation methodology, the research shows job outcomes are a function of role ambiguity and motivation but not necessarily job tension.

Relation between loneliness and depression: a structural equation analysis.

Research on loneliness has been hampered by its strong association with depression. The two states frequently co-occur, and measures of the two states are substantially correlated. Inability to manipulate experimentally loneliness or depression makes it difficult to untangle the causal influence of one on the other. The combination of longitudinal design and structural equation methodology is proposed as a solution to this general problem. Measures of loneliness and depression were administered to undergraduates at two points 5 weeks apart. Data from 333 subjects were correlated and analyzed under a succession of structural equation models. Results indicated that loneliness and depression were correlated but clearly different constructs; neither was a direct cause of the other, though both probably share some common origins; both were highly stable over to 5-week-period.

The Construct Validity of the Tripartite Classification of Attitudes

The convergent, discriminant, and predictive validity of the tripartite model of attitudes is investigated. On the basis of a structural equation methodology, evidence is obtained for convergent va...

The Construct Validity of the Tripartite Classification of Attitudes

The convergent, discriminant, and predictive validity of the tripartite model of attitudes is investigated. On the basis of a structural equation methodology, evidence is obtained for convergent validity. Moreover, a factor analysis of the attitude measures and measures of perceived extraneous events and personal and social normative beliefs provides evidence for discriminant validity. Finally, evidence for predictive validity based on actual and intended behaviors is found to be mixed. The implications of these findings for research pertaining to the prediction of consumer behavior are discussed.

Toward a general theory of fertility: A causal modeling approach

A general theory of fertility is derived hypothesizing that the demand for children is primarily an outcome of social psychological processes within the family, subject to certain socioeconomic constraints. Two broad social psychological processes are posited as determinants of fertility. The first suggests that the attitudes or tastes of family members influence the demand for children. The second maintains that the nature of the husband-wife interaction (in terms of power, conflict, decision making, and marital satisfaction) determines family size. Socioeconomic variables, in the form of the normative social structure and social stratification, and economic constraints, such as income and price, are hypothesized to influence fertility through their impact on social psychological processes within the family. The overall theory is tested on two independent samples--one in Ankara, Turkey, the second in Mexico City, Mexico--using a structural equation methodology.