Bu çalışmanın amacı “peer learning”in (PL) öğrenci çıktılarına etkisini incelemektir. Bu doğrultuda 11 first order meta analiz (FOM) araştırması incelenmiştir. FOM’lara Web of Science, Scopus, Medline, Eric, Akademik Search Ultimate, PsycNet and Google Academic veri tabanlarından erişilmiştir. Aramalar 2000-2024 yılları arasını kapsamaktadır. FOM araştırmaları K12 ve yükseköğretim kademesini kapsamaktadır. FOM araştırmalarının beş çalışma Cohen's d etkisini (k=5), on bir çalışma ise Hedge's g etkisini (k=11) raporladı.FOM araştırmalarından elde edilen 16 etki büyüklüğü istatistiksek olarak analiz edilmiştir.Araştırma yöntemi olarak second order meta analiz yöntemi kullanılmıştır. İstatistiksel analizler random model altında yapılmıştır. Etki büyüklüklerinin yayın yanlılığı incelemiştir. Çalışma sonucunda PL’nin öğrenci çıktılarına etkisinin zayıf düzeyde olduğu sonucuna varılmıştır (ES=.40 [CI=.33-.47] p
The purpose of this study is to examine the impact of peer learning (PL) on student outcomes. To this end, 11 first-order meta-analysis (FOM) studies were reviewed. FOMs were accessed from the Web of Science, Scopus, Medline, Eric, Academic Search Ultimate, PsycNet, and Google Academic databases. The searches cover the years between 2000 and 2024. FOM (Factoring-Out Misconceptions) research encompasses both K-12 and higher education levels. While k=5 studies reported Cohen’s d effect size indexes (d), k=11 studies reported Hedge’s g effect size indexes (g).İn this analyses totals sixteen effect size (k=16) values obtained from FOM studies have been statistically analyzed. The research method used was second-order meta-analysis. Statistical analyses were conducted under the random model. Publication bias of effect sizes was examined. The study concluded that the impact of PL on student outcomes is weak (ES= .40 [CI= .33-0.47], p < .01). It was found that the effect sizes between PL and student outcomes have low levels of publication bias. It has been observed that PL has a higher impact on academic achievement compared to affective, professional skills, and social skills. It was also found that the effect sizes vary according to the languages of primary studies and the publication bias status of FOM studies. The group of FOMs exhibiting publication bias has a high-level effect size (while other groups have produced low-level effect sizes.
