Evaluation of the psychometric properties of the Reproductive Autonomy Scale for use in the UK

Abstract

BackgroundReproductive autonomy—control over outcomes including contraceptive use and childbearing—is a human right and vital to women’s empowerment. Those whose reproductive autonomy is threatened by the structures and relationships in their lives are at risk of coercion and unplanned pregnancy and could benefit from additional services. The Reproductive Autonomy Scale (RAS) was developed in the USA to assess women’s reproductive autonomy; this study evaluates the RAS for use in the UK.MethodsAfter testing, the RAS was incorporated into an online survey of women of reproductive age. Those who were sexually active were asked to complete the RAS, which was evaluated according to classical test theory. Reliability was assessed via internal consistency and a 3-month test-retest. Construct validity was assessed using hypothesis testing and confirmatory factor analysis.ResultsFor 826 women the RAS was highly acceptable, with a response rate of >97.7%. Almost the whole range of reproductive autonomy scores were captured. Internal consistency was good, with a Cronbach’s α of 0.75. Test-retest reliability was fair-good with an intraclass correlation coefficient of 0.67. Construct validity analysis found the scale to be valid based on our hypothesis that among women who want to avoid pregnancy, those with higher reproductive autonomy will be more likely to use contraception. The three-factor structure of the scale was confirmed on confirmatory factor analysis.ConclusionThe RAS is valid and reliable for use in the UK. This tool holds potential utility across research, clinical practice, health interventions and policy development.

Turkish adaptation of the reproductive autonomy scale: Validity and reliability study

“…The scale was found to have Cronbach's alpha coefficients between 0.61–0.82 and 0.68–0.81 in adaptation studies for women in the United Kingdom and Brazil 17,18 . Besides, Cronbach's alpha coefficient values in the original scale are between 0.65 and 0.82, and our results parallel other and the original scale study 2,15,16 …”

“…However, in both studies, the lowest mean score was taken from the decision sub‐dimension, while the lowest was from the decision sub‐dimension. In this respect, it also contains findings parallel to our study 15,16 …”

“…In this respect, it also contains findings parallel to our study. 15,16 This study evaluated the Turkish validity and reliability of a measurement tool needed to evaluate reproductive autonomy in women. In this context, first of all, the stages of translation, expert opinion, and pilot application, which are the steps for the cultural adaptation of the scale, were carried out.…”

“…17,18 Besides, Cronbach's alpha coefficient values in the original scale are between 0.65 and 0.82, and our results parallel other and the original scale study. 2,15,16 In assessing internal consistency for reliability, the item-total correlation coefficient is evaluated. Item-total correlation levels higher than 0.25 are considered sufficient, and the items measure the same structure.…”