Common Method Variance and Specification Errors: A Practical Approach to Detection

Abstract

The purpose of this study was to demonstrate how examining the bivariate correlations between items in self-report measures can assist in differentiating between possible common method variance vs. model specification errors. Specifically, social desirability was viewed as either a possible source of common method variance or as a theoretically meaningful construct that should be included in the model of interest (i.e., a specification error). In the first instance, LISREL was used, and the level of correlation between measures of social desirability and measures of the five constructs of interest was manipulated. These results provided some insight as to when one needs to be concerned about the possible "common variance effects" on the structural model. In the second instance, the correlations between measures of social desirability and the measures of only two constructs of interest were again manipulated. These analyses illustrated the point at which the omission of social desirability as a theoretically relevant variable began to result in a poor fit of the structural model.

Predictors of Threat and Error Management: Identification of Core Nontechnical Skills and Implications for Training Systems Design

“…One possible implication of the study design was the potential for common method variance (CMV) to result in a circularity of the relations between observers' perceptions of threat and error management and their ratings of crews' nontechnical skills. As Kline, Sulsky, and Rever-Moriyama (2000) suggested, CMV is a common problem facing self-report measures in which relations between a range of variables are detected in data collected using the same instrument. In these situations, the relations that emerge might be the result of a common relation with a third spurious and unmeasured variable rather than an independent relation between the two or more variables as measured.…”

A Special Kind of Downsizing

“…As we discussed previously, surveys rely on participants to provide as reliable responses as possible to a number of questions pertaining to different issues (Kline et al, 2000). Arguably, the significant correlation between union and organizational commitment could suggest that common method variance is a problem in this study.…”

“…The survey method relies on participants to provide as reliable responses as possible to a number of questions pertaining to a number of different issues (Kline, Sulsky and Rever-Moriyama, 2000). There is often concern that relationships, which are found through conducting correlations, may be partly due to a "spurious" relationship.…”

“…That being said, correlations between these measures are not uncommon (see Snape et al, 2000) and the correlation matrix revealed a number of insignificant correlations, suggesting that common method variance may not be an issue in this study. Note that we also followed the advice of Kline et al (2000) to reduce common method variance through the use of reverse coded items. Nevertheless, this limitation cannot be ruled out in this study, suggesting that the time may be ripe for longitudinal surveys in this field of exploration.…”

"Entrepreneurship, Legitimacy and Online Social Communities: An Empirical Analysis"

“…A web based questionnaire was given to the raters to perform the categorization process. Measurement error due to context effects like grouping of items (Kline, Sulsky, and Rever-Moriyama 2000) and item priming effects (Salancik 1984) was avoided by randomizing the order of items that form the various constructs (Tourangeau and Rasinski 1988). The questionnaire was peer reviewed to ensure principles of good item writing such as avoiding double barreled questions, jargon, leading items, and negatively worded items.…”