Statitical Techniques Reported in Korean Dermatology Journals During 1990-1994

Lee, Sung Yul; Lee, Jong Suk; Hwang, Jae Young; Kim, Chong Hyeok

Abstract

BACKGROUND
Effective evaluation of an article's scientific merit requires familiarity with the methodology described, especially when quantitative techniques, such as statistical procedures, are invoked to clarify research findings or to summarize data.
OBJECTIVE
The purpose of this study was to describe the frequency with which various statistical concepts were reported in journals important to dermatology. From these results, dermatologists can identify the major statistical skills needed to critically evaluate their literature.
METHODS
All 376 original articles of the Korean Journal of Dermatology(KJD) and the Annals of Dermatology(KAD) during 1990-1994 were chosen for review. Each article was re-viewed to determine the statistical content.
RESULTS
The two most commonly used statistical techniques in the two journals were mean and standard deviation. The t-test was the next most frequently used statistical technique in the reviewed journals, followed by non-parametric, chi-square test, orphan p, ANOVA, and correlation/regression.
CONCLUSION
These results indicate the need for wider education about the use of descriptive and basic comparative statistics. It is impossible to evaluate the dermatological literature critically without these skills.