Assessing Arabic couples: An evidence-based approach

Abstract

Assessing couple relationships across diverse languages and cultures has important implications for both clinical intervention and prevention. However, empirically validated measures of relationship functioning for couples from non-Western cultures have generally been lacking. The current study examined the psychometric properties and measurement invariance of an Arabic translation of the Marital Satisfaction Inventory – Revised (MSI-R), a 150-item, 13-scale measure of couple relationship functioning, and a brief 10-item screening scale (MSI-B) derived from this measure. Data were collected from 84 community and 32 clinic Middle Eastern couples living in Qatar. The Arabic MSI-R demonstrated moderate-to-strong internal consistency, discriminative validity, and factorial invariance across Qatari and US standardization samples. Similarly, findings regarding the 10-item Arabic MSI-B supported its internal consistency and discriminative validity. Overall, psychometric characteristics for both the full-length MSI-R and the MSI-B screener suggest the clinical utility of these measures for identifying Arabic-speaking couples for whom secondary prevention or intervention protocols may be appropriate.

Keywords:

• marital discord
• marital satisfaction
• assessment
• Arabic
• Middle Eastern
• Muslim

Acknowledgements

The preparation of this article was supported, in part, by a grant from the Doha International Institute for Family Studies and Development, Doha, Qatar.

Notes

1. Both the complete 150-item Marital Satisfaction Inventory – Revised (MSI-R) and the 10-item Marital Satisfaction Inventory – Brief form (MSI-B) can be obtained in either English or Arabic versions from the publisher: Western Psychological Services, 625 Alaska Avenue, Torrance, California, 90503, U.S.A., www.wpspublish.com

2. A sixth test examining invariance in factor covariances across groups, applicable only to models with two or more factors, was not relevant to the evaluation of this one-factor model and hence was omitted.