Dept Menu

Audience Menu

 
Measuring abortion stigma for program design and evaluation

Anu Kumar, Ipas, Inc., 2010
See also executive summary.

Project abstract

Abortion stigma can have profound social and health consequences for women. When faced with unplanned or unhealthy pregnancies, fear of abortion stigma often prevents women from seeking early safe abortions, even in contexts where the procedure is legal. To improve women's health, interventions are needed that can reduce abortion-related stigma.

Our research is aimed at understanding what abortion stigma is, how it is created and perpetuated, and how this stigma can affect women's access to safe abortion care. To more fully understand both the causes and contexts of abortion stigma, we intend to develop a scale tool for both measuring abortion stigma and evaluating relevant interventions.

The research comprises a sequential, mixed-methods design, using both qualitative (focus group discussions) and quantitative (survey) data. We will conduct this research in two countries, Zambia and Ghana, and in two communities in each country. We will generate an item pool for the abortion stigma scale from focus group discussion data collected from groups of married and unmarried women and men in the study communities. Once these scale items have been identified by a study team, which includes local language speakers, a set of items will be included in a questionnaire that will be administered in each country.

In Zambia this study will build on ongoing community engagement work that centers on service-provision sites, and in Ghana we will work in communities in which Ipas has previously trained service providers. We will use a factor analysis technique to develop a statistically valid scale, with the possibility of subscales, which can then be further refined as part of our ongoing work in these two countries. We will also examine correlations between abortion stigma, personal characteristics, and geographic locations to help develop scales and interventions that are broadly adaptable.


Back to top