The study was conducted by Mahadih Kyambade of the Department of Leadership and Governance and Afulah Namatovu of the Department of Applied Computing and IT, both at Makerere University Business School in Kampala. Their starting point was a gap in the organizational learning literature: while the mechanics of learning from failure have been studied extensively in hospitals, hotels, and technology firms, higher education in developing countries has received comparatively little empirical attention. Universities, the authors argue, are precisely the kind of institutions where learning from failure should flourish, since inquiry and reflection are their core business, yet hierarchical cultures and competitive pressures can make failure something to conceal rather than examine.

To test their ideas, the researchers used a quantitative design, collecting structured survey data from 214 academic staff across three public universities in Uganda. The survey measured three constructs: the quality of workplace relationships, the degree of psychological safety respondents felt, and the extent to which staff reported openly analyzing and learning from failures. The analysis followed the mediation procedure introduced by Baron and Kenny in 1986, a classic statistical approach for testing whether a third variable explains the pathway between a cause and an effect. Regression diagnostics were used to check the data for problems such as collinearity before the models were estimated.

The concept of high-quality relationships at the heart of the study comes from positive organizational scholarship, particularly the work of Jane Dutton and colleagues. High-quality connections are characterized by mutual trust, respect, and energizing engagement; people in such relationships feel valued and are willing to invest in one another. Earlier research by Abraham Carmeli and Jody Gittell, published in the Journal of Organizational Behavior in 2009, showed that these relational capacities promote learning from failure in work organizations, partly by building psychological safety. The Ugandan study extends that framework into a new institutional and cultural setting, asking whether the same mechanism holds in resource-constrained public universities.

Psychological safety, the mediating variable in the study, is a construct popularized by Harvard’s Amy Edmondson in landmark papers from 1996 and 1999. It describes a shared belief that a team or workplace is safe for interpersonal risk-taking, that members can ask questions, admit mistakes, and raise concerns without fear of humiliation or punishment. Edmondson’s original fieldwork in hospital units found that teams with better climates actually reported more errors, not because they made more, but because they felt able to discuss them. The construct later gained global fame when Google’s Project Aristotle identified psychological safety as the leading characteristic of its most effective teams, and it has since been studied in settings ranging from supply chains to software engineering.

The results of the Ugandan survey were clear on the first question. High-quality relationships significantly and positively predicted learning from failures among academic staff. In other words, respondents who reported trusting, respectful, and supportive connections with colleagues were more likely to say that they reflected on mistakes, discussed what went wrong, and adjusted their practice accordingly. This is consistent with social exchange theory, rooted in Peter Blau’s 1964 work, which holds that relationships built on reciprocity and goodwill generate obligations of openness and mutual support. When colleagues treat each other well, the theory predicts, they exchange not just favors but also the candid information that learning requires.

The second finding added nuance. Psychological safety partially mediated the relationship between high-quality connections and learning from failure. That means supportive relationships do their work partly by creating an environment in which staff feel safe to be vulnerable about their errors, but the relationships also appear to contribute through other channels, such as the practical help and diverse perspectives that good colleagues provide when dissecting a failure. Partial mediation is a common and informative pattern in this literature; it suggests that psychological safety is an important mechanism but not the only one, and that interventions targeting relationships and interventions targeting safety climate can each add value.

The practical implications for university leaders are significant. If learning from failure depends on trust, respect, and open communication, then leaders who want resilient, continuously improving institutions should invest in the relational fabric of their organizations, not merely in training programs or evaluation systems. That could mean modeling openness about their own missteps, responding to reported problems with curiosity rather than blame, and building inclusive climates where staff of all statuses feel their voices matter. Research on leader inclusiveness in health care by Ingrid Nembhard and Amy Edmondson has shown that such behaviors disproportionately help lower-status members speak up, suggesting that deliberate leadership attention can widen who participates in learning conversations.

The study also contributes to a growing body of work on psychological safety in African institutional contexts, including prior surveys in Ghana and South Africa and the authors’ own earlier research on socially responsible leadership, job engagement, and innovative work behavior in Ugandan universities. Higher education across much of the continent faces intense pressures, from surging enrollment to constrained budgets, and Mushemeza’s 2016 review documented the challenges academic staff confront. In such environments, the capacity to learn from failure may be a decisive ingredient of institutional adaptation, and the finding that everyday relationship quality matters offers a comparatively low-cost lever for change.

Like all survey-based research, the study has limits that the authors and readers should keep in view. The data are cross-sectional and self-reported, so the findings capture associations and a theoretically grounded mediation pathway rather than definitive proof of causation, and common-method bias cannot be entirely excluded. The sample of 214 staff across three public universities is respectable but specific to one national context. Still, the convergence of these results with a broad international literature, from hospital units to flight decks, strengthens the case that the pattern is general: organizations learn from failure when their people are connected well enough, and feel safe enough, to talk honestly about what went wrong. For universities hoping to turn setbacks into improvement, the message is that the soft stuff of relationships and psychological safety is, in fact, the hard currency of learning.

Subject of Research: The influence of high-quality workplace relationships on learning from failure among university academic staff, mediated by psychological safety.

Article Title: Learning from failure is essential in university context: how high-quality relationships influence learning from failures via the lens of psychological safety

Article References: Kyambade, M., & Namatovu, A. (2026). Learning from failure is essential in university context: how high-quality relationships influence learning from failures via the lens of psychological safety. SN Social Sciences, 6(10), Article 482. https://doi.org/10.1007/s43545-026-01778-x

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01778-x

Keywords: psychological safety, learning from failure, high-quality relationships, organizational learning, higher education, Uganda, public universities, academic staff, mediation analysis, social exchange theory, workplace trust, Learning