Assessment of the dimensionality of the Wijma delivery expectancy/experience questionnaire using factor analysis and Rasch analysis

BMC Pregnancy and Childbirth

Research Article

H. M. Haines¹  J. F. Pallant¹  P. Green²  J. Toohill³  J. Gamble³  D. K. Creedy³  J. Fenwick⁴ 

[1] Department of Rural Health, The University of Melbourne, Graham St, Shepparton, Australia;Faculty of Health Sciences and Medicine, Bond University, Queensland, Australia;Menzies Health Institute, Griffith University, Meadowbrook Queensland, Australia;Menzies Health Institute, Griffith University, Meadowbrook Queensland, Australia;Gold Coast University Hospital Southport, 4215, Queensland, Australia;

关键词: Fear of childbirth;    Measurement;    Factor analysis;    Rasch analysis;   

DOI  :  10.1186/s12884-016-1157-8

 received in 2016-04-21, accepted in 2016-11-14,  发布年份 2016

来源: Springer

【 摘 要 】

BackgroundFear of childbirth has negative consequences for a woman's physical and emotional wellbeing. The most commonly used measurement tool for childbirth fear is the Wijma Delivery Expectancy Questionnaire (WDEQ-A). Although originally conceptualized as unidimensional, subsequent investigations have suggested it is multidimensional. This study aimed to undertake a detailed psychometric assessment of the WDEQ-A; exploring the dimensionality and identifying possible subscales that may have clinical and research utility.MethodsWDEQ-A was administered to a sample of 1410 Australian women in mid-pregnancy. The dimensionality of WDEQ-A was explored using exploratory (EFA) and confirmatory factor analysis (CFA), and Rasch analysis.ResultsEFA identified a four factor solution. CFA failed to support the unidimensional structure of the original WDEQ-A, but confirmed the four factor solution identified by EFA. Rasch analysis was used to refine the four subscales (Negative emotions: five items; Lack of positive emotions: five items; Social isolation: four items; Moment of birth: three items). Each WDEQ-A Revised subscale showed good fit to the Rasch model and adequate internal consistency reliability. The correlation between Negative emotions and Lack of positive emotions was strong, however Moment of birth and Social isolation showed much lower intercorrelations, suggesting they should not be added to create a total score.ConclusionThis study supports the findings of other investigations that suggest the WDEQ-A is multidimensional and should not be used in its original form. The WDEQ-A Revised may provide researchers with a more refined, psychometrically sound tool to explore the differential impact of aspects of childbirth fear.

【 授权许可】

CC BY   
© The Author(s). 2016