The common problem?: Examining the impact of common-method variance on the measurement of job attitudes

Emotional intelligence (EI) is the ability of an individual to use emotions intelligently for positive outcomes. This ability is said to have an impact on many aspects of our personal and professional lives. Well-being (WB) is a person�s state of mind regarding his/her life. Theory suggests that highly emotionally intelligent individuals are likely to experience psychological well-being at a higher level than individuals who are low in emotional intelligence. Based on this concept, this paper examines the empirical relation between EI and WB, both of which are attitudinal variables. A sample of 104 respondents from the manufacturing sector of South India was considered for the study and their emotional intelligence and well-being levels were assessed using appropriate tools. When two attitude-based constructs are obtained from the same respondent, using the same instrument, at the same time, then the relationship between them is said to be influenced by Common Method Variance (CMV). It is the variance that is attributable to the measurement method rather than to the constructs the measures represent and is said to introduce a spurious correlation between the variables of interest. In this study, both EI and WB were obtained from the same respondent, through a single instrument. Hence CMV was considered and acted upon. While controlling the effect of CMV on the empirical relation between EI and WB, this paper aims at the following sub-objectives: To assess the EI and WB levels of the respondents To implement procedural remedies to control CMV To implement statistical remedies to control CMV To prove that EI is a predictor of WB, over and above the effect of CMV. Procedural remedies were used with respect to questionnaire design and statistical remedies were used to partial out the effect of CMV. This paper should serve as a framework for researchers, who wish to evaluate the potential biasing effects of method variance in their research, pertaining to behavioural aspects.

Download Full-text

Impact of Social Desirability and Common Method Variance on Two Measures of Entitlement

Psychological Reports ◽

10.1177/0033294120937439 ◽

2020 ◽

pp. 003329412093743

Author(s):

Brian K. Miller

Keyword(s):

Social Desirability ◽

Discriminant Validity ◽

Common Method ◽

Inventory Data ◽

Common Method Variance ◽

Method Variance ◽

Psychological Entitlement ◽

Two Measures ◽

The Impact ◽

Variance Explained

Entitlement is the predisposition toward expecting or demanding more rewards than someone else regardless of one’s own contribution, effort, or performance. This study examines the impact of social desirability and common method variance on the discriminant validity of the two most popular measures of entitlement: The Psychological Entitlement Scale and the Entitlement subscale of the Narcissistic Personality Inventory. Data were gathered using three independent samples. In all three studies, forcing items to load on their intended and separate constructs measuring entitlement fit the data well thus providing some evidence of discriminant validity. The correlation between the two entitlement constructs was larger than in previous studies but short of that which is necessary for collinearity providing further evidence of discriminant validity in each study. In each study the amount of shared variance between the two latent entitlement constructs was more than the variance explained in either construct thus failing to provide evidence of discriminant validity. These two measures are similarly affected by socially desirable responding but not likely contaminated by common method variance. Because of their consistently and moderately strong correlation, the PES and the entitlement subscale of the NPI likely measure different but strongly related constructs.

Download Full-text