نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه آمار، دانشکده علوم پایه، دانشکاه قم، قم، ایران
2 گروه علم اطلاعات و دانش شناسی، قم، دانشگاه قم
3 گروه علم اطلاعات و دانش شناسی، دانشکده ادبیات، دانشگاه قم، قم، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Objective: Statistical methods have an important role in reaching results and analysis. The current study aims to analyze the status of statistical methods in information and science knowledge journals, while reviewing the statistical methods used, it also discovers the frequently of their methodological problems.
Methods: The nature of the research is a descriptive survey. The research community consists of descriptive survey articles in 13 active journals of information and science knowledge until the end of year 1400. Based on Cochran's formula, 141 articles were selected through stratified sampling. In this study, the research methodology was investigated in the statistical methods, the sampling methods, validity and reliability of the selected articles. In addition, the variables affected on the characteristics of the articles and their statistical errors were inspected. Data analysis was carried out in Excel and SPSS 26 software with the methods of descriptive statistics.
Results: The statistical methods have been used a total of 599 times, the most frequent being related to descriptive statistics. The T, chi-square and the Kolmogorov-Smirnov tests are the most frequent in parametric and non-parametric inferential statistical methods, respectively. In the section about predictive methods and post hoc tests, correlation coefficient and Tukey's test are the most frequent. The results of the research show that the most statistical error observed in the articles is the use of parametric tests without checking the normality of the data.
Conclusions: The findings of the research show that although the authors of the of the articles use the descriptive statistics well in their articles, they pay less attention to the inferential statistical methods. The important point in using parametric methods is satisfying the underlying assumption of tests, since the violation of them will invalidate the research results
کلیدواژهها [English]