The development and application of a scale to measure the extent and forms of work-family conflict in collectivist cultures

Purpose - The aim of this paper is to advance the conceptualisation and measurement of work-family conflict (WFC) by developing and validating a scale that is relevant in a collectivist culture setting. Design/methodology/approach - First, qualitative interviews with 15 bank employees were conducted to establish whether WFC was an issue, its meaning and form and the relevance of the Carlson, Kacmar and Williams' (2000) scale. Second, drawing on role theory and work-family border theory, an additional psychological dimension was developed, and the new scale was tested with data from a self-report survey of bank employees (n = 569). Third, the validity, reliability and measurement invariance of the scale were confirmed with data from a sample of secondary school teachers (n = 223). Findings - The characteristics of collective societies pertinent to WFC were relevant to these middle-class employees, and they experience high levels of WFC. A model with a six-factor structure (time-based, strain-based and psychological-based WFC and FWC) represents the most theoretically and statistically sound measure of WFC for these samples. Practical implications - WFC has many negative social and economic consequences. However, there is inadequate evidence on which to base human resource policies to address the issue in collective societies. This study developed and applied a more reliable measure to assess its extent and form to assist in the design of appropriate WFC management practice. It will be of interest to scholars researching and teaching international management, management consultants, policy makers and managers seeking to understand the problem of WFC in collective societies. Originality/value - This is the first study to establish the validity of a psychological dimension of WFC in a collectivist culture. It confirms the relevance of the strain and time dimensions of the most commonly used multi-dimensional measure, but found no evidence of behavioural WFC.