The Effectiveness of Electronic Learning on Academic Performance with Meta-Analysis

Over the last few decades, electronic learning has been the focus of many studies and many researchers have examined the effect of electronic learning on academic performance, which yielded different results that highlighted a demand for a combination of research in this area. The aim of this research is a meta-analysis of the effect of electronic learning on academic performance.

The present study was conducted using meta-analysis method. The corpus of the present study included all valid research published which was available in computer databases that examined the effectiveness of electronic learning on academic performance. The intended data were available in related studies between 1996-2022. For gathering data, Mesrabadi research proposal checklist (2011) was used. Using the purposive sampling method and extensive search based on the specified keywords and applying entry and exit criteria, 243 studies were selected as a sample and were entered into the meta-analysis process. Sensitivity analysis, fixed and random effects models, and heterogeneity analysis with CMA software were used for data analysis. From selected studies as a sample, in total, 311 effect sizes were extracted which 17 extreme effect sizes were eliminated in sensitivity analysis and the next analyzes were performed on 294 effect sizes. The value of “fail-safe N” statistic was obtained in 1911, which indicated that after entering this number of non-significant studies, the combined effect size would be non-significant.

The findings indicated that the mean of the total effect of the studies was 0.726 for the fixed effects model and 1.009 for the random effects model; both of them were statistically significant (p<0.001) which indicated a positive and significant relationship between electronic learning and academic performance. As the effect sizes were heterogeneous, gender, type of sample, type of electronic learning, type of course, age, and type of research method were examined as the mediating variables. The results indicated that the effect size of electronic learning on academic performance is higher among the females compared to that of the males, and in preschool students more than others, in mobile-based instruction more than other types, in teaching methods courses more than other courses, and in experimental research more than correlation research. Also, the results of the research showed that the age of the subjects had a very small contribution to explaining the heterogeneity. The results showed that electronic learning had a significant effect on academic performance and based on Cohen's criterion, the effect size of this relationship was found to be high (ES=1/009).

Supplying programs to educate the effect of electronic learning on academic performance is a necessity that needs to be considered more than before. Considering the obtained results and the quantity and quality of existing research studies, it is suggested that the studies of these two fields be carried out at a higher level in observance of the rules of academic paper writing and conducting research. Also, to achieve a reliable result in the first stage, parametric statistics in their design and strong statistical methods with appropriate sample and appropriate sampling methods in accordance with the population should be used.