واقع مؤشرات الدلالات الإحصائية والعملية في الدراسات التربوية بالجامعات العمانية : دراسة تحليلية لرسائل الماجستير

Abstract

The present study aimed to investigate the reality of the use of indicators of statistical significance and the practical significance indicators of Master's theses at Omani universities. A random sample of (628) Master's theses was selected which is 33.14% from the total population 1895. The researcher designed a content analysis tool to collect data related to the statistical tests included in this study, which are t- tests, analysis of variance, Pearson correlation coefficient, and regression coefficient. The results determined that there was a deficiency in the Cohen standard to explain the values of the practical significance indicators and suggested that a criterion for the practical significance as follows: small when the values of the index (d) are less than 0.631 ,medium when the values are between 0.631-1.50, and large when the values are equal to or greater than 1.51. The study showed the justifications that distinguish this criterion from the Cohen criterion in educational and psychological studies. The results of the applied side of the study showed that the number of statistical tests used in master’s theses reached 11926 tests, and that the practical significance indicators 7% of the total statistical tests. Most of the results of the (t) tests and analysis of variance tests were not statistically significant at a level of significance less than 0.05, and the opposite in the results of Pearson correlation coefficient and regression coefficient. The practical significance was calculated for all tests. It was found medium in (t) tests and Pearson correlation coefficient tests, small in analysis of variance tests, and large in regression coefficient tests. The results also indicated that there were statistically significant positive relationships between statistical significance and levels of practical significance and sample size in all statistical tests. Most of the statistically significant results were associated with a medium or large practical significance, and the most of not statistically significant results were associated with a small practical significance and there was a statistically significant negative relationship between the sample size and the practical significance levels for all statistical tests except for Pearson correlation coefficients. The results also indicated that 26.5% of the decisions reached by the results of master’s theses in Omani universities may be inaccurate and unreliable. The practical significance of the t- tests and analysis of variance with master’s theses at Sultan Qaboos University was better than those of private universities whereas the practical significance of the Pearson correlation coefficients at private universities was better than those of Sultan Qaboos University. The results showed that students of psychology were better than their colleagues in other disciplines in average practical significance.