An Evaluation of the Statistical Techniques used in the 1994~1998 Editions of The Journal of the Korean Rheumatism Association

Park, Ile Kyu; Kim, Think You

Abstract

OBJECTIVE
Quantitative experiment and analysis of the result with statistical techniques are an essential part of the medical article for acquiring objective confidence. But errors on application, calculation, and interpretation of statistics and insufficient explanation of the used statistical technique deprive the reader of reliance on the article. We identified the statistical techniques and errors that were commonly encountered, and both of researchers and readers should recognize, in the Journal of the Korean Rheumatism Association (JKRA). We identified the frequency of the statistical techniques and errors from the 136 articles in the 1994~1998 editions of the JKRA. Each different kind of statistical techniques and errors in the same article were counted separately, but two or more of the same kind of statistical techniques and errors in an article were counted as one. One hundred and seventy-seven statistical techniques were encountered and 75 statistical errors were identified. The most frequently used techniques were descriptive statistics, and common errors and the observed numbers of the errors were as follow : 1. performing statistical techniques for two groups on the comparison of three or more groups(14), 2. mentioning the statistical technique without using it(14), 3. using mean and standard deviation instead of median and range on the ordinal data(12), 4. using standard error instead of standard deviation for the description of data spread(11), 5. using p value without the statistical name(11).
CONCLUSION
The most frequently encountered statistical technique and error are descriptive technique and performing statistical techniques for 2 groups on the comparison of 3 or more groups, respectively. The authors? careful application of the basic statistics would be the real solution to reduce the statistical errors.