Statistical Analysis: Data Presentation and Statistical Tests
Data analysis should be planned in line with objectives of the study. Quality assured data entry, data cleaning and construction of dummy tables are essential steps before data analysis.
Descriptive statistics summarize the data obtained from the study subjects. The measures could be rate, proportion, odds ratio, relative risk, mean, median, mean difference, etc.
Inferential statistics make inference about the population using the data from the study sample. In inferential statistics, 95% confidence interval and p-value are calculated.
In comparison with p-value, 95% confidence interval gives information on the magnitude of effect in addition to statistical significance. Clinical significance is important than statistical significance.
Choice of statistical test of significance depends on the type of data (numerical, categorical) and its distribution (normal or non-normal).
The authors thank Dr Tanveer Rehman, Junior Resident and Dr Gunjan Kumar, Senior Resident, Department of PSM for critically reviewing the manuscript
- 2.Sivia DS. Data analysis- a dialogue with the data. 2005. http://bayes.wustl.edu/sivia/lisbon05.pdf. Accessed 28 July 2017.
- 3.Kirkwood BR, Sterne JAC, Kirkwood BR. Essential medical statistics. Victoria, Australia: Blackwell Science; 2003.Google Scholar
- 7.Paul Vogt W, editor. Dummy table. In: Dictionary of statistics & methodology. Thousand Oaks, CA: SAGE Publications, Inc. https://doi.org/10.4135/9781412983907.n596.
- 9.Developing a quantitative data analysis plan for observational studies developing a quantitative data analysis plan 2013. http://nceph.anu.edu.au/files/Data_Analysis_Plan_Guide_20131125.pdf. Accessed 13 Aug 2017.
- 12.Gravetter FJ, Wallnau LB. Essentials of statistics for the behavioral sciences. Baltimore: Wadsworth Cengage Learning; 2014.Google Scholar
- 13.Gordis L. Epidemiology. Philadelphia: Elsevier; 2013.Google Scholar
- 15.Epidemiology kept simple. http://www.sjsu.edu/faculty/gerstman/eks/formula_sheet.pdf. Accessed 13 August 2017.
- 17.The UNION. Course material of UNION SORT-IT operational research workshop.Google Scholar
- 20.Thompson ML, Myers JE, Kriebel D. Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data: what is to be done? https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1757577/pdf/v055p00272.pdf. Accessed 31 Aug 2017.
- 23.Coe R. It’s the effect size, stupid: what effect size is and why it is important. September 2002. https://www.leeds.ac.uk/educol/documents/00002182.htm. Accessed 25 Aug 2017.
- 25.Government of Canada SC. Statistics: power from data! Using graphs. http://www.statcan.gc.ca/edu/power-pouvoir/ch9/using-utilisation/5214829-eng.htm. Accessed 15 Aug 2017.