Datasets on public knowledge and preference in Kenya

Background: Abortion engenders complex opinions and emotions among diverse groups of people ranging from total pro-life and anti-abortion values to more pro-choice positions. Public opinion plays a critical role and sometimes pressurizes policy and decision-makers to develop policies and laws that respond to people's views. Public opinion about abortion is thus varied, and understanding these variations can be beneficial because abortion is divisive in Africa, and policymakers typically deploy the excuse of negative public sentiments when they fail to enact legal and policy changes. Unfortunately, no rigorous studies have assessed public opinions and attitudes towards abortion in Kenya.

Objective: To assess public opinion of abortion, including perspectives on abortion laws and situations under which abortion should be legal or allowable in Kenya.

Methods: We will conduct a nationwide cross-sectional survey using telephones to collect data from adults 18 years of age or older. We will select the sample from a database of over eight million telephone numbers accumulated by Geopoll through enrollment of potential respondents and in collaboration with mobile telephone providers. The sample will comprise adults of diverse socioeconomic backgrounds from all counties in Kenya. We will use a random digit dialing (RDD) approach to select and recruit participants for the survey. Trained research assistants will collect the data through telephone interviews using structured questionnaires. Data analysis will be conducted using STATA version 15, and continuous variables will be summarized as means with standard deviations, while categorical variables will be presented as proportions. We will conduct univariate, bivariate, and multivariate tests to assess associations between selected variables and the outcomes of interest.

Main Outcomes and Implications: The study's primary outcome will be opinions on abortion, which will be grouped as supportive vs. non-supportive. The outcome variable might be expressed in Likert scale measures, that is, instead of regrouping them, we will use them the way in which we administered the survey. For example, we may not regroup “agree” and “strongly agree” into one category or “disagree” and “strongly disagree” into one. We will report/examine this outcome against the circumstances under which a woman needs or secures an abortion. The evidence we will generate might renew debates and conversations on abortion and generate additional evidence for policymaking. It will also help assess and track shifts in public opinion and attitudes to abortion over time. Advocacy actors might utilize the evidence to facilitate lobbying and effective campaigns toward abortion legal and policy changes.