In this article, we provide guidance for substantive researchers on the use of structural equation modeling in practice for theory testing and development. We present a comprehensive, two-step modeling approach that employs a series of nested models and sequential chi-square difference tests. We discuss the comparative advantages of this approach over a one-step approach. Considerations in specification, assessment of fit, and respecification of measurement models using confirmatory factor analysis are reviewed. As background to the two-step approach, the distinction between exploratory and confirmatory analysis, the distinction between complementary approaches for theory testing versus predictive application, and some developments in estimation methods also are discussed.

/pdf/structural-equation-modeling-in-practice-a-review-and-1ue2omj492.pdf

Structural equation modeling in practice: a review and recommended two-step approach

Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments

Structural Equations with Latent Variables.

Most methodological fields undertake regular critical reflections to ensure rigorous research and publication practices, and, consequently, acceptance in their domain. Interestingly, relatively little attention has been paid to assessing the use of partial least squares structural equation modeling (PLS-SEM) in marketing research—despite its increasing popularity in recent years. To fill this gap, we conducted an extensive search in the 30 top ranked marketing journals that allowed us to identify 204 PLS-SEM applications published in a 30-year period (1981 to 2010). A critical analysis of these articles addresses, amongst others, the following key methodological issues: reasons for using PLS-SEM, data and model characteristics, outer and inner model evaluations, and reporting. We also give an overview of the interdependencies between researchers’ choices, identify potential problem areas, and discuss their implications. On the basis of our findings, we provide comprehensive guidelines to aid researchers in avoiding common pitfalls in PLS-SEM use. This study is important for researchers and practitioners, as PLS-SEM requires several critical choices that, if not made correctly, can lead to improper findings, interpretations, and conclusions.

An assessment of the use of partial least squares structural equation modeling in marketing research

Confirmatory factor analysis (CFA) is widely used for examining hypothesized relations among ordinal variables (e.g., Likert-type items). A theoretically appropriate method fits the CFA model to polychoric correlations using either weighted least squares (WLS) or robust WLS. Importantly, this approach assumes that a continuous, normal latent process determines each observed variable. The extent to which violations of this assumption undermine CFA estimation is not well-known. In this article, the authors empirically study this issue using a computer simulation study. The results suggest that estimation of polychoric correlations is robust to modest violations of underlying normality. Further, WLS performed adequately only at the largest sample size but led to substantial estimation difficulties with smaller samples. Finally, robust WLS performed well across all conditions.

/pdf/an-empirical-evaluation-of-alternative-methods-of-estimation-xsk0k8dmua.pdf

An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data.

A large-scale simulation design was used to study the sensitivity of maximum likelihood (ML) factor analysis to violations of measurement scale and distributional assumptions in the input data. Pro...

The Sensitivity of Confirmatory Maximum Likelihood Factor Analysis to Violations of Measurement Scale and Distributional Assumptions

Effects of a prison facility on the regional economy

This study was undertaken to examine the influence of wife's employment on major family expenditures. Tobit analysis of data from the 1972–73 Consumer Expenditure Surveys found no relationship between wife's employment and outlays on twelve categories of expenditures. Family income had the greatest overall influence on the expenditure categories examined. Therefore, it appears that the absolute amount of income was more important than its sources in determining expenditures.

Wife's employment—its influence on major family expenditures

C287. Ordinal data and lisrel: a note on significance level

This study illustrates a methodology as a first step toward the development of a benefit/cost model for the evaluation of the Federal Trade Commission regulation of the unavailability of advertised specials in food stores. A Monte Carlo simulation was used to estimate economic losses to shoppers from unavailable advertised specials. Product unavailability in the model occurred as a result of alternative managerial decisions about purchase-order quantities for advertised specials and shelf-stocking policies combined with probabilistic purchase decisions by customers. The model generated probabilistic individual customer and management behavioral responses to unavailability during a one-year time period. As a result, the derived economic losses to customers from unavailability were based on alternative assumptions about managerial behavior in conjunction with empirically derived shopper responses to advertised specials. This study thus provides some idea of the amount of damages being caused by the excessive unavailability of advertised specials. Aggregate customer losses were found to vary primarily with respect to management's purchase-order quantity of advertised specials.

Carl E. Ferguson

Papers

The Sensitivity of Confirmatory Maximum Likelihood Factor Analysis to Violations of Measurement Scale and Distributional Assumptions

Effects of a prison facility on the regional economy

Wife's employment—its influence on major family expenditures

C287. Ordinal data and lisrel: a note on significance level

Simulating food shoppers' economic losses as a result of supermarket unavailability*