Arab Educational Information Network

The current study was designed to examine and evaluate the extent to which t-test assumptions (normality, homogeneity of variances and independence) were violated in the psychological and educational research at the local level and golf state level to accomplish the aforementioned goal the actual data of 37 ma theses and 37 paper published at the golf state academic journal in which t-test was used as the major statistical analysis tool were reanalyzed and examined. After completing the collection of the needed data, several statistical analyses technique was employed to analyze the data. The major statistical techniques were Kolmogorov-Smirnov test for normality distribution, levene’s-test for homogeneity of variances in ma theses and chi square test for comparative study. The result of the statistical analysis revealed the following finding local and golf state researcher are found to be weak with respect to their knowledge of the t-test assumptions and are lacking of its awareness, among the examined master degree theses, considerable percentage were found to be violating the t-test assumptions and graduate students tend to use larger sample than the researchers who published their paper in the golf state academic. (Published Abstract)