Validation of a new questionnaire assessing the health impact of divorce on women: An exploratory factor analysis

• Main Authors: Abd. Aziz, Karimah Hanim, Abd. Aziz, Aniza, Salleh, Halim, Ambak, Nurul Jannah, Ab Rahman, Nor Hakimah, Oo, San San, Abed, Naghem Farouk, Husain, Rohayah
• Format: Article
• Language: English; English
• Published: Malaysian Public Health Physicians Association 2016
• Subjects: RA0421 Public health. Hygiene. Preventive Medicine
• Online Access: http://irep.iium.edu.my/51244/1/Dr.%20Karimah%20-%20VALIDATION%20OF%20A%20NEW%20QUESTIONNAIRE%20ASSESSING%20THE%20HEALTH%20IMPACT%20OF%20DIVORCE%20ON%20WOMEN%20AN%20EXPLORATORY%20FACTOR%20ANALYSIS.pdf; http://irep.iium.edu.my/51244/7/51244-Validation%20of%20a%20new%20questionnaire%20assessing%20the%20health%20impact%20of%20divorce%20on%20women_SCOPUS.pdf; http://irep.iium.edu.my/51244/; http://www.mjphm.org.my/mjphm/index.php?option=com_content&view=article&id=722:validation-of-a-new-questionnaire-assessing-the-health-impact-of-divorce-on-women-an-exploratory-factor-analysis&catid=108:2016-volume-16-2&Itemid=122

The aim of this study was to develop a new validated questionnaire to assess the health impact of divorce on women. A cross sectional study was undertaken in a district in Terengganu, using a newly developed self-administered Malay language questionnaire comprising of six domains and 82 items. The questionnaire was constructed based on the literature review and discussions with experts. Exploratory factor analysis was applied in construct validity and internal consistency was used for reliability analysis. A total of 51 respondents were involved in this study. All the consented divorcees were Malays. The mean (SD) age for the divorced women was 35.8 (10.00). Majority (82.4%) of them had secondary education. All the final three domains demonstrated Keiser-Meyer-Olkin (KMO) of more than 0.6, significant Bartlett’s test of Sphericity and Cronbach’s alpha of more than 0.80. The overall Cronbach’s alpha was excellent (0.920). The final questionnaire consisted of two domains. Factor loading for all items in each subdomains were satisfactory. The final functional somatic symptoms domain consisted of 6 subdomains and 19 items with factor loading from 0.402 to 0.914. For emotional instability domain, the final items were 13 items with factor loading from 0.548 to 0.878. The questionnaire was valid and reliable to be used for functional somatic symptoms and emotional instability domains. It is useful to assess the health impact and related intervention of divorced women. A confirmatory factor analysis may further confirm the final model.