How Effective Are Different Approaches to Higher Education Provision in Increasing Access, Quality and Completion for Students in Developing Countries?

BACKGROUND: How to effectively increase access to and quality in higher education in developing countries is a highly debated topic. There is no consensus as to what policies or provisions best increase access to higher education, nor is there a firm understanding of how each form or provision or policy impacts the quality of higher education. Empirical evidence on these issues is also lacking. Moreover, while a large body of literature on the implementation of such policies and provisions exists for the developed world, little evidence exists for the developing world, resulting in a comparative dearth of literature that analyses the impact of such policies or provisions for developing nations. METHODS: We systematically reviewed research available in the English language on the impact of higher education policies and methods of provision on access, quality and gender issues in developing countries. We also examined the potential differences of these outcomes in terms of gender. We discussed the types of outcomes for which there is evidence and addressed gaps in the evidence base. We selected studies for inclusion based on the relevance of the study method and context, as well as study quality. Given the small number of quantitatively rigorous studies addressing the review question, we also included a number of qualitative and descriptive studies in our review. We synthesised the studies using thematic summaries, using quantitative studies as the basis for our analysis and qualitative studies as supplemental evidence. RESULTS: We identified 175 studies in total. Twenty-four of these studies were regression-based studies, and the remaining 151 were qualitative or descriptive studies. The majority of the studies fell into six main categories: affirmative action; cross-border, distance and open education; financial policies/programmes (including cost-sharing policies and student loan programmes); technical and vocational education and training (TVET); and institutional type. Other studies did address topics not covered by these categories such as part-time education; quality assurance; and academic advising. Eight of the included studies, moreover, did not address any specific policy (or addressed multiple policies), but focused on increasing access for certain subgroups such as females or disabled students. CONCLUSIONS: Although the identified studies failed to reach a firm consensus on the policies most effective for increasing access to and the quality of higher education, several of them nonetheless demonstrated positive effects for the interventions studied. Moreover, while many studies lacked evidence on how these outcomes varied by gender, several did address this issue and highlighted key areas of concern for improving not only access to and the quality of higher education, but for promoting equity as well. We find positive effects for affirmative action in increasing access for targeted subgroups but also noted that these policies may have unintended negative consequences. Financial programmes and policies such as fee-sharing, dual-track tuition policies and different types of student loans may also positively increase access to higher education while shifting some portion of the costs of higher education from the government to the student. Careful consideration, however, must be taken to formulate the right mix of policies to ensure access to lower-income students. The cost of such programmes and their long-term sustainability must also be taken into account. We find little evidence for the impact of cross-border and transnational provision and TVET in increasing access to and the quality of higher education. A few randomised trials of vocational education programmes, however, did show significant gains to lower-income women who participated. FUTURE RESEARCH: The scarcity of robust evidence on this topic for developing countries demonstrates the need for improved data. Studies using larger datasets that span multiple institutions are needed to yield more robust and generalisable findings for some types of interventions. More studies that use randomised trials or natural experiments to measure the impact of a particular method of provision or policy for treated versus control groups would also be valuable. In cases where this is not possible, comparative studies could offer some evidence on the impact of policy interventions. Finally, additional evidence on outcomes of interest, such as enrollment, retention, graduation and employment, is needed. Because context matters, however, it is not always possible to identify 'one size fits all' solutions.