A Point to Consider in the Statistical Analysis of Scientific Studies
International Journal of Medical Biochemistry
a non-parametric test, was used in intergroup comparisons  . Data analyzed using non-parametric statistics should be reported in tables or shown as median figures with an appropriate range (e.g., median, minimum and maximum values). Presenting variables that do not show normal distribution as mean±standard deviation can lead to the misinterpretation of the data . In the Results section of the study, the ferritin and folate levels in the TE group are given in Table 1 as 17.35±18.54 ng/mL
... d 7.94±8.98 ng/mL, respectively (as mean±SD). Presenting the results in this way can lead to the misinterpretation that some of the results were below the minimum detection limit (e.g., folate: -1.04-16.92 ng/mL). Mean and standard deviation (or standard error) values are not suitable for reporting data analyzed using non-parametric statistics. If the distribution of the data is not sufficiently normal and requires a non-parametric analysis, mean and standard deviation will not accurately represent the location and range of the data. When data are analyzed using nonparametric methods, such as the Mann-Whitney U test, it is generally more appropriate to report median values with the minimum and maximum values, the 25 th and 75 th percentile values, or other central intervals [median (min-max)] . Recognizing and understanding these errors will help researchers select the appropriate statistical methods and use them correctly, thereby improving the quality of their published study. References 1. Ertug EY, Yilmaz RA. Reduced ferritin, folate, and vitamin B12 levels in female patients diagnosed with telogen effluvium. Int J Med Biochem 2018;1:111-4. 2. Ozdemir O. Medikal istatistik. 1st ed. İstanbul: İstanbul Medikal Yayıncılık; 2006. 3. Olsen CH. Review of the use of statistics in infection and immunity. Infect Immun 2003;71:6689-92.