Data analysis is the systematic process of inspecting, cleaning, transforming, and modelling data to extract useful information, support decision-making, and test hypotheses. It encompasses descriptive summarisation, exploratory analysis to surface structure and anomalies, inferential statistics to generalise from samples, and predictive modelling. Data analysis spans manual statistical work through to automated analytics pipelines, and it is the disciplinary core from which data science, business intelligence, and machine-learning workflows draw. Rigorous analysis attends to data quality, sampling bias, and the validity of inferential assumptions.
Content
- The analysis workflow typically proceeds through problem framing, data acquisition and cleaning, exploratory analysis, modelling, validation, and communication of findings. Each stage shapes the credibility of the conclusions; in particular, data cleaning and assumption checking often consume the majority of analytical effort.
- The distinction between confirmatory and exploratory analysis is methodologically important: exploratory work generates hypotheses and must not be presented as if it tested them, while confirmatory analysis pre-specifies hypotheses and procedures to preserve the validity of inferential statistics.