Learning through the Distribution of Failures within an Organization: Evidence from Heart Bypass Surgery Performance

Vinit Desai
Academy of Management Journal, Volume 58 Issue 4, p1032-1050

While research has suggested that organizations can improve by investigating and learning from failures, some work has found that they may generate incorrect lessons or fail to learn. This study addresses the debate by turning attention to the processes that underlie learning, using attribution theory to highlight the way in which decision makers interpret information about where failures occurred or who was involved. This approach is notable because it suggests that different organizations with similar experiences may have quite distinct reactions based on where that experience originates. Specifically, I predict that organizations learn less effectively when their failures are relatively concentrated in origin, meaning that failures typically involve a particular unit or even a specific individual, compared to when failures are more broadly dispersed. I also examine factors that intensify or ameliorate this effect, including an organization’s size or its performance relative to aspirations. I test related hypotheses on a panel of hospitals that offered a specific surgical procedure within California from 2003 through 2010.