Abstract

DU PLESSIS, M  and  MARTINS, N. Testing for measurement invariance for employee engagement across different demographic groups in South Africa. JCMAN [online]. 2017, vol.14, n.1, pp.24-59. ISSN 1815-7440.

The purpose of this study was to assess the measurement invariance of the employee engagement instrument (EEI), as a factor of race and gender. Previous research has advised that existing engagement instruments cannot be used with confidence to make comparisons between different demographic groups in South Africa. Given the demographic differences between individuais, the focus of this study was to determine whether the instrument could be used with confidence among the mentioned demographic groups. A quantitative research approach was followed in this study. The EEI was administered electronically to a sample of business people, who were representative of the South African working population. 4 099 completed questionnaires were received. Differential item functioning (DIF) was used to test for measurement invariance within an item response theory (IRT) framework. The results revealed that the assumption for metric invariance does not hold across ali six sub-scales. It was further found that the DIF for race is less problematic than for gender. Eight items demonstrated DIF across gender. The Organisational Commitment, Immediate Manager and Strategy and Implementation sub-scales demonstrated the highest degree of invariance for race.

Keywords : differential item analysis (DIF); employee engagement; gender; measurement invariance testing; race.

        · text in English     · English ( pdf )