Ogunbiyi Bolatito O, Bingenheimer Jeffrey B, Baird Sarah, Vyas Amita
Department of Prevention and Community Health, Milken Institute School of Public Health, George Washington University, Washington, DC, USA.
Department of Global Health, Milken Institute School of Public Health, George Washington University, Washington, DC, USA.
J Adolesc. 2025 Jan;97(1):219-232. doi: 10.1002/jad.12414. Epub 2024 Oct 5.
The last decade has experienced a surge of interventions focused on improving adolescent girls' agency. Yet measuring adolescent girls' agency continues to be a challenge, limiting the ability to track impact. This study addresses this evidence gap by constructing and validating a multidimensional measure of agency among adolescent girls in Ethiopia.
This study utilized cross-sectional data from 3033 in school adolescent girls aged 10-12 years and 15-17 years and their adult female caregivers collected as part of the Gender and Adolescent: Global Evidence study in 2017-2018 in Ethiopia. This study constructed a measure of agency among the sample and evaluated both known group and convergent validity of the scale. Twenty-two indicators across three domains (decision-making, voice, and mobility) were used to characterize adolescent girls' agency. The data was randomly divided into two halves for exploratory and confirmatory factor analyses separately.
While six factors of agency emerged from the initial exploratory analysis, two factors (decision-making and mobility factors) defined adolescent girls' agency from the confirmatory factor analysis. Known-groups validity of the agency scale was confirmed-the scores on the two domains (decision-making and mobility) and the overall agency scale was higher for older girls compared to younger girls and for girls in urban households compared to those in rural households. Convergent validity of the scale was not confirmed.
This study advances adolescent girls' agency measurement by providing a validated multidimensional measure that can be used to support future policy and research.
在过去十年中，旨在增强少女自主权的干预措施激增。然而，衡量少女的自主权仍然是一项挑战，限制了追踪影响的能力。本研究通过构建并验证埃塞俄比亚少女自主权的多维测量方法，填补了这一证据空白。
本研究利用了2017 - 2018年在埃塞俄比亚进行的“性别与青少年：全球证据”研究中收集的3033名10 - 12岁和15 - 17岁在校少女及其成年女性照料者的横断面数据。本研究构建了样本中自主权的测量方法，并评估了该量表的已知群体效度和收敛效度。使用三个领域（决策、发言权和行动能力）的22个指标来描述少女的自主权。数据被随机分成两半，分别用于探索性因素分析和验证性因素分析。
虽然在初始探索性分析中出现了六个自主权因素，但在验证性因素分析中，两个因素（决策和行动能力因素）定义了少女的自主权。自主权量表的已知群体效度得到了证实——与年龄较小的女孩相比，年龄较大的女孩在两个领域（决策和行动能力）以及总体自主权量表上的得分更高；与农村家庭的女孩相比，城市家庭的女孩得分更高。量表的收敛效度未得到证实。
本研究通过提供一种经过验证的多维测量方法，推进了少女自主权的测量，该方法可用于支持未来的政策和研究。
