Basic principles of descriptive statistics in medical research

Descriptive statistics provides tools to explore, summarize and illustrate the research data. In this tutorial we discuss two main types of data - qualitative and quantitative variables, and the most common approaches to characterize data distribution numerically and graphically. This article presents two important sets of parameters - measures of the central tendency (mean, median and mode) and variation (standard deviation, quantiles) and suggests the most suitable conditions for their application. We explain the difference between the general population and random samples, that are usually analyzed in studies. The parameters which characterize the sample (for example, measures of the central tendency) are point estimates, that can differ from the respective parameters of the general population. We introduce the concept of confidence interval - the range of values, which likely includes the true value of the parameter for the general population. All concepts and definitions are illustrated with examples, which simulate the research data.

1. Fisher R.A. On the mathematical foundations of theoretical statistics. Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character. 1922. Vol. 222. P. 309-368. https://doi.org/10.1098/rsta.1922.0009

2. Bulanov N.M., Blyuss O.B., Munblit D.B., et al. Venn diagrams and probability in clinical research. Sechenov Med J. 2020; 11(4): 5-14. https://doi.org/10.47093/2218-7332.2020.nA5-14

3. Bulanov N.M., Blyuss O.B., Munblit D.B., et al. Studies and research design in medicine. Sechenov Med J. 2021; 12(1): 4-17. https://doi.org/10.47093/2218-7332.2021.12.1A17

4. Kirkwood B., Stern J. Essential Medical Statistics. 2nd ed. Blackwell Publishing; 2003; 512 p. ISBN: 978-0-865-42871-3.

5. RothmanK. Random error and the role of statistics. Epidemiology: An Introduction. 2nd ed. Oxford University Press; 2012: 148-163. ISBN: 9780199754557.

6. MotulskyH. Intuitive Biostatistics. 4th ed. Oxford University Press; 2018; 568 p. ISBN-13: 978-0190643560. ISBN-10: 0190643560ъ.

7. Ghasemi A., Zahediasl S. Normality tests for statistical analysis: a guide for non-statisticians. Int J Endocrinol Metab. 2012 Spring; 10(2): 486-489. https://doi.org/10.5812/ijem.3505. Epub 2012 Apr 20. PMID: 23843808. PMCID: PMC3693611

8. Mohd Razali N.M., Wah Y.B. Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests. J Stat Model Anal. 2011; 2: 21-33.

9. Nuzzo R.L. Histograms: A useful data analysis visualization. PM R. 2019 Mar; 11(3): 309-312. https://doi.org/10.1002/pmrj.12145. Epub 2019 Mar 7. PMID: 30761760

10. Spriestersbach A., Rohrig B., Prel J.B., et al. Descriptive Statistics: The specification of statistical measures and their presentation in tables and graphs - Part 7 of a series on evaluation of scientific publications. Dtsch Arztebl. 2009; 106(36): 578-583. https://doi.org/10.3238/arztebl.2009.0578

11. Langford E. Quartiles in elementary statistics. Journal of Statistics Education. 2017; 14(3). https://doi.org/10.1080/10691898.2006.!1910589