When Does Management Research Produce Cumulative Knowledge? The Diversity–Performance Literature as a Case Study

Research on diversity and organizational performance has generated a large body of empirical work, yet the literature continues to produce mixed and often contradictory findings. This essay argues that the problem is not simply empirical inconsistency, but the structure of the research programme itself. Drawing on classical philosophy of science—particularly Thomas Kuhn’s concept of “normal science,” Imre Lakatos’s methodology of scientific research programmes, and Larry Laudan’s problem-solving model of scientific progress—the essay contends that the diversity–performance literature has not yet achieved the conditions necessary for cumulative knowledge, because scholars routinely investigate different causal relationships using distinct conceptualizations of diversity, varied performance measures, and multiple levels of analysis. A comprehensive review of approximately ninety studies published in leading management journals confirms this structural pattern: findings accumulate episodically rather than cumulatively, and comparable studies remain strikingly rare. Using the diversity–performance literature as a case, the essay maintains that cumulative knowledge in management depends not only on empirical rigor but on coordination at three levels: shared definitions of diversity and performance, a shared ontology specifying the categories and attributes those measures are actually built from, and attention to the mechanisms through which diversity produces its effects rather than merely whether it correlates with them.