BUSINESS RESEARCH, STATISTICS, AND EVIDENCE-BASED DECISION-MAKING

Business decisions routinely require action before uncertainty can be eliminated. The relevant scholarly problem is therefore not merely how to collect more data, but how to decide which evidence can support which inference, under what assumptions, and at what cost. This chapter develops a critical, practice-oriented account of business research, statistics, and evidence-based decision-making for enterprises operating across formal and informal institutions, with particular attention to Rwanda and comparable African economies. It integrates evidence-based management, research philosophy, measurement theory, statistical inference, causal identification, forecasting, decision theory, and responsible analytics. The secondary-source design combines seminal research, professional reporting guidance, official institutional methodology, and the National Institute of Statistics of Rwanda's 2024 Integrated Business Enterprise Survey. A reproducible analysis of published enterprise indicators illustrates percentage-point contrasts, denominator discipline, and the limits of descriptive comparisons. Constructed datasets provide independently checkable examples of dispersion, interval estimation, regression reasoning, and decision trees; constructed values are not passed off as field observations. An original evidence-to-decision workflow and provenance protocol show how questions, data, uncertainty, stakeholder interests, implementation, and evaluation must connect. The chapter argues that defensible enterprise intelligence is a governed inferential process rather than a dashboard, algorithm, or significant p-value. Its educational conclusion is that evidence quality depends jointly on relevance, measurement, identification, interpretability, and ethical consequences, while managerial responsibility remains irreducible to statistical calculation.