Data distribution analysis – a preliminary approach to quantitative data in biomedical research

Statistical analysis is an integral part of medical research. It helps transform raw data into meaningful insights, supports hypothesis testing, optimises study design, assesses risk and prognosis, and facilitates evidence-based decision-making. The statistical analysis increases research findings' reliability, validity and generalisability, ultimately advancing medical knowledge and improving patient care. Without it, meaningful analysis of the data collected would be impossible. The conclusions drawn would be unsubstantiated and misleading. Many health professionals are unfamiliar with statistical analysis and its basic concepts. The analysis of clinical data is an integral part of medical research. Identifying the data type (continuous, quasi-continuous or discrete) and detecting outliers are the first and most important steps. When analysing the data distribution for normality, graphical and numerical methods are recommended. Depending on the type of data distribution, appropriate non-parametric or parametric tests can be used for further analysis. Data that are not normally distributed can be normalised using various mathematical methods (e.g., square root or logarithm) and analysed using parametric tests in the next steps. This review provides essential explanations of these concepts without using complex mathematical or statistical equations but with several graphical examples of various statistical terms.