Social Learning and Health in Uganda: Relationships among Community and Individual Education and Sexual and Reproductive Health Knowledge

• In Event: Women's issues & rural education

Proposal

Introduction: Access to accurate knowledge about the prevention of unintended pregnancies and HIV is necessary for achieving sexual and reproductive health (SRH) goals in sub-Saharan Africa (SSA) (Bandura, 2001; Paul-Ebhohimhen, 2008; UNICEF, 2015). International development practitioners routinely invest in education interventions as a means of increasing SRH knowledge and women’s overall empowerment, and thereby, improved SRH outcomes (Bloom, 2007; Jukes, 2008; UNICEF 2015). However, few studies explore the association between community education and SRH knowledge (Baxen, 2004; Kravdal, 2002). We use data from Uganda to examine how the distribution of SRH knowledge depends on both individual and community education.

Existing research finds that individual education is associated with greater SRH knowledge (Kenkel, 1991; LeVine, 2001). Educational attainment develops cognitive skills for processing and remembering complex written and verbal information about health risk factors and different methods of pregnancy and HIV prevention (Bloom, 2007; Smith-Greenaway, 2015). Educational attainment also provides access to diverse social environments—classrooms and jobs—that may result in exposure to health information (Jukes, 2008; Grant, 2015). Education is linked to higher rates of employment which, in SSA, often means travel or relocation, particularly from rural to urban areas, with more opportunities to acquire additional health knowledge (Adepoju, 2000; Sahn, 2003). The higher earnings that come with education may garner greater access to health clinics that disseminate information about HIV prevention and contraception (Cutler, 2006; Bloom, 2007).

Education-related skills and knowledge are likely shared within communities in SSA through the process of social learning (Bandura, 2001), whereby individuals garner SRH knowledge from their interactions and observations of others (Benefo, 2006; Kravdal, 2002; Smith-Greenaway, 2017). A key component of social learning is that the knowledge circulating in a community may be more or less accurate, depending on the information available in the social environment. Communities that have higher levels of education may have more accurate knowledge in circulation than communities with lower levels of education.

More educated communities also have features that facilitate the dissemination of health knowledge, including vibrant market economies—marked by more specialized occupational groups and markets where goods and services are exchanged—that foster more frequent contact with wider social networks (Bloom, 2007; Cutler, 2006; Smith-Greenaway, 2017). More educated communities may also have more community resources—health services, social services, schools, and community centers—that provide accurate SRH knowledge (Benefo, 2006; Greenaway, 2012).

We advance prior research by examining whether the association between SRH knowledge and individual educational attainment may depend on community levels of education. Accurate SRH knowledge is an easily shareable resource that is, nevertheless, unlikely to be equitably distributed in the population. Existing research provides little evidence about how community education may modify the association between SRH knowledge and individual education. We hypothesize that high levels of community education will “spillover” and benefit even the least educated in a community. As community education increases, more people will have accurate SRH knowledge, and the institutions that emerge in highly educated communities (e.g., vibrant market economies, social services) may support the dissemination of those ideas. Indeed, those with the highest individual education may have access to accurate SRH knowledge regardless of the community context, but the least educated will benefit the most from living in the most educated communities. Evidence in favor of the spillover perspective could be promising for policy makers in low resource settings, where educating “enough” individuals may be sufficient to result in adequate SRH knowledge within that community.

Methods: We use a nationally representative sample from the Uganda 2011 Demographic Health Survey (DHS) [we plan to update this data using the 2016 DHS data sets]. Individuals are nested within communities (i.e., clusters), allowing us to explore how individual and community education are independently and interactively associated with SRH knowledge. We apply DHS guidance for weighting based on gender and rurality. Our preliminary, unweighted analytic sample includes 2,295 men and 8,674 women from 6,959 households who had complete data on the main independent and dependent variables.

DHS data are ideal for multi-level analyses as they utilize standardized questions within a similar time frame for each geographic area. We assess two measures of SRH knowledge, HIV knowledge and contraceptive awareness, for our dependent variables. Our key independent variable is the continuous measure of years of individual education completed. For community education level, we calculated the mean of all respondent’s individual years of education completed within a cluster. We mean-centered individual and community education levels.
We use a multi-level, multivariate linear regression models to test associations between educational attainment and SRH knowledge, while controlling for other individual and community characteristics. Multi-level models account for the non-independence of individuals within communities. Further, multi-level modelling measures the variability in the education-SRH knowledge relationship between and within different communities, and provides standard errors that accurately account for the different sample size within each cluster.

Preliminary Results: Models that include only main effects show that individual education and average community education are each significantly and substantively associated with increases in contraceptive and HIV knowledge. Models that include cross-level interactions between individual and community education find support for the spillover hypothesis. Specifically, for both types of SRH knowledge, individual education is more strongly and positively associated with SRH knowledge in less educated communities, net of other factors in the model. In more educated communities, both more and less educated individuals had relatively high levels of SRH knowledge, resulting in a weaker association between individual education and SRH knowledge.

Conclusion: We find that community and individual education work in tandem to shape SRH knowledge. Our results are consistent with social learning theory and the spillover hypothesis, which suggests that the least educated individuals will benefit from living in more educated communities. We offer evidence that higher community education facilitates the diffusion of SRH knowledge, and that individual education attainment may matter less for health knowledge within more educated communities. More work on how to harness community education to improve health is needed to create efficient social policies in resource-limited settings like Uganda.

Authors

• Stephanie Chamberlin, University of Colorado Denver

• Patrick Krueger, University of Colorado Denver, Department of Health and Behavioral Sciences