Data analytics and business statistics: From data quality to statistical evidence

Data analytics turns observations into evidence, but statistical technique cannot rescue data whose meaning, quality or provenance is misunderstood. This chapter develops the quantitative foundation needed for responsible business and trade analysis. It follows the data lifecycle from business question, source and governance through cleaning, exploratory analysis and visualization, then builds probability, sampling distributions, confidence intervals, hypothesis testing, t-tests, chi-square tests, analysis of variance, correlation, simple and multiple regression, and introductory time-series forecasting. Worked examples connect formulas to managerial interpretation and repeatedly distinguish statistical significance from practical importance, association from causation, and model fit from truth. Rwanda's 2026 consumer-price data and World Bank Enterprise Survey metadata provide African applications, while large-scale online experimentation illustrates global statistical practice and the risks of multiple testing. Christian moral reasoning treats numerical analysis as accountable truth-telling: analysts must not manufacture certainty, hide inconvenient results, violate privacy or use technically valid summaries to mislead decision makers.

Keywords: business statistics; data analytics; data quality; descriptive statistics; probability; statistical inference; hypothesis testing; regression; time series; statistical ethics