# Expectations, Tensions, and Brokerage: A Discourse Analysis of Community Engagement with Health Research in South Africa

## TL;DR

This study explores how different groups in South Africa view community involvement in health research, highlighting mismatched expectations and the heavy reliance on community representatives.

## Contribution

The paper introduces a discourse analysis framework revealing three key themes: expectations, tensions, and brokerage in community-academic relationships.

## Key findings

- Community members expect tangible benefits from research, which often clash with academic priorities.
- Tensions arise from power imbalances and structural inequalities in research partnerships.
- Community representatives act as brokers but face ethical and practical burdens due to lack of institutional support.

## Abstract

Research is increasingly claimed to be done in collaboration with communities, but community members may have entirely different expectations of engagement and research participation than what typically follows the logic of academia. In South Africa, intersecting inequalities further complicate relationships with communities and stakeholders. To understand how different actors view and construct the relationships between academic institutions and communities, we undertook a multiperspective discourse analysis. We conducted 11 in‐depth interviews with 12 participants categorized as researchers, community representatives, and community members. These interviews reflect three interconnected discourses: expectations, tensions, and brokerage. Expectations pattern intergroup dynamics, such as community members’ expectations of research benefits, while tensions primarily capture challenging relationships between different research actors. Our analysis also illustrates how, in the absence of comprehensive institutional support, community engagement relies on brokerage by community representatives, and how this reliance disproportionately burdens them. There is a need to lessen this ethical burden and invite community input without also burdening community members and representatives with the challenges of academia. Our findings also have wider relevance for community‐based health research because engagement practices are often hindered by institutional and structural factors.