Statistical methods are widely used for the analysis of data obtained from experimental or observational studies. We examined the statistical methods used for papers published in the Journal of Korean Academy of Rehabilitation Medicine in 1996. The issues included the mean, coefficient of variation, t-test, chi-square test, ANOVA and type IV error. The mean and the standard deviation in descriptive statistics and the t-test in inferential statistics were most frequently used. Statistical methods should incorporate statistical modelling and the needed assumptions. Trials of new statistical models and methods are recommended.