Türkiye'de 5e öğrenme modelinin matematik başarısına etkisi üzerine bir meta-analiz çalışması
[ X ]
Tarih
2022
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Siirt Üniversitesi
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Yapılandırmacı yaklaşıma dayalı 5E öğrenme modeli, bir dizi yerleşik aşama yoluyla öğrencilere 21. Yüzyıl becerilerini kazandırmak, matematikle ilgili kavram yanılgılarını gidermek ve akademik başarılarını arttırabilmek için kullanılan bir öğrenme yöntemidir. Bu doğrultuda araştırmanın amacı, 5E öğrenme modelinin geleneksel öğrenme yöntemlerine kıyasla öğrencilerin matematik başarısı üzerindeki etkisini deneysel yöntemlerle inceleyen çalışmaları meta-analiz yöntemiyle birleştirmek, elde edilen bulguların değerlendirilmesiyle kapsamlı bir sonuca ulaşmaktır. Bu doğrultuda, 2007-2022 yılları arasında Türkiye'de yapılmış 5E öğrenme modeline yönelik lisansüstü tezler ve makaleler taranmıştır. Tarama sonucunda 5E öğrenme modelinin matematik başarısı üzerindeki etkisini inceleyen, meta-analize dâhil edilme ölçütlerini karşılayan toplam 20 çalışma (13 yüksek lisans tezi, 4 doktora tezi ve 3 makale) meta-analize dâhil edilmiştir. Verilerin analizleri Comprehensive Meta-Analysis programında yapılmıştır. Meta-analize dâhil edilen çalışmaların etki büyüklükleri sabit etkiler ve rastgele etkiler modeline göre ayrı ayrı hesaplanmıştır. Bu modellerden hangisinin kullanılacağını belirlemek için yapılan heterojenlik testi sonucunda çalışmaların heterojen yapıda olduğu belirlenmiş ve analizler rastgele etkiler modeline göre yürütülmüştür. Bulunan genel etki büyüklüğünün geçerli sayılabilmesi için Huni Saçılım Grafiği, Rosenthal'ın Güvenli N ve Orwin'in Güvenli N yayın yanlılığı değerlendirme yöntemleriyle incelenmiştir. Çalışmalarda yayın yanlılığının olmadığı belirlenmiştir. Ayrıca çalışmaların yayın türü, yayın yılı, öğretim kademesi, uygulama süresi, örneklem büyüklüğü, yapıldıkları bölgeler, desen türü ve öğrenme alanına göre aralarında anlamlı farklılığın olup olmadığı moderatör analizi yardımıyla gerçekleştirilmiştir. Çalışmaların rastgele etkiler modeline göre genel etki büyüklüğü 0,718-1,102 güven aralığında Hedges's G=0,910 (%95 CI, SE=0,098) olarak hesaplanmıştır. Genel etki büyüklüğünün Thalheimer ve Cook tarafından yapılan sınıflandırmaya göre "geniş düzeyde" olduğu tespit edilmiştir. Çalışmaların tamamı pozitif etki büyüklüğüne (deney grubunun lehine) sahiptir. Moderatör analizi sonucunda 5E öğrenme modelinin öğrencilerin matematik başarısı üzerindeki etkisinde çalışmaların yayın türüne, öğretim kademesine, uygulama süresine, yapıldıkları bölgelere, desen türüne ve öğrenme alanına göre anlamlı bir farklılık bulunmamıştır. Ancak yayın yılına ve örneklem büyüklüğüne göre anlamlı bir farklılık göstermemiştir. Araştırma sonucuna göre 5E öğrenme modeline dayalı öğrenmenin, öğrencilerin matematik başarısını geliştirmede geleneksel öğrenme yöntemlerine kıyasla daha etkili olduğu sonucuna varılmıştır. Bu bağlamda 5E öğrenme modelinin matematik derslerinde kullanılması için öğretmenlerin teşvik edilmesi önerilebilir. Anahtar Kelimeler: Etki büyüklüğü, Matematik başarısı, Meta-analiz, Öğrenme alanları, 5E öğrenme modeli.
The 5E learning model, based on the constructivist approach, is a learning method used to provide students with 21st century skills, eliminate mathematical misconceptions and increase their academic success through a series of built-in stages. In this direction, the aim of the research is to combine the studies examining the effect of the 5E learning model on students' mathematics achievement with experimental methods compared to traditional learning methods with the meta-analysis method, and to reach a comprehensive conclusion by evaluating the findings. In this direction, postgraduate theses and articles on the 5E learning model made in Turkey between 2007-2022 were scanned. As a result of the screening, a total of 20 studies (13 master's theses, 4 doctoral dissertations and 3 articles) examining the effect of the 5E learning model on mathematics achievement and meeting the inclusion criteria in the meta-analysis were included in the meta-analysis. Analyzes of the data were made in the Comprehensive Meta-Analysis program. The effect sizes of the studies included in the meta-analysis were calculated separately according to the fixed effects and random effects models. As a result of the heterogeneity test performed to determine which of these models to use, it was determined that the studies were heterogeneous and the analyzes were carried out according to the random effects model. In order for the overall effect size found to be valid, the Funnel Scatterplot was examined using Rosenthal's Safe N and Orwin's Safe N publication bias evaluation methods. It was determined that there was no publication bias in the studies. In addition, it was carried out with the help of moderator analysis whether there was a significant difference between the studies according to the publication type, publication year, teaching level, application period, sample size, regions, design type and learning area. According to the random effects model of the studies, the overall effect size was calculated as Hedges's G=0.910 (95% CI, SE=0.098) in the confidence interval of 0.718-1.102. It has been determined that the overall effect size is "wide" according to the classification made by Thalheimer and Cook. All of the studies have positive effect size (in favor of the experimental group). As a result of the moderator analysis, no significant difference was found in the effect of the 5E learning model on students' mathematics achievement according to the publication type, teaching level, application period, the regions where they were made, the type of design and the learning area. However, it did not show a significant difference according to publication year and sample size. According to the results of the research, it was concluded that learning based on the 5E learning model is more effective in improving students' mathematics achievement compared to traditional learning methods. In this context, it can be suggested to encourage teachers to use the 5E learning model in mathematics lessons. Keywords: Effect Size, Meta-analysis, 5E Learning model, Math success, Learning spaces.
The 5E learning model, based on the constructivist approach, is a learning method used to provide students with 21st century skills, eliminate mathematical misconceptions and increase their academic success through a series of built-in stages. In this direction, the aim of the research is to combine the studies examining the effect of the 5E learning model on students' mathematics achievement with experimental methods compared to traditional learning methods with the meta-analysis method, and to reach a comprehensive conclusion by evaluating the findings. In this direction, postgraduate theses and articles on the 5E learning model made in Turkey between 2007-2022 were scanned. As a result of the screening, a total of 20 studies (13 master's theses, 4 doctoral dissertations and 3 articles) examining the effect of the 5E learning model on mathematics achievement and meeting the inclusion criteria in the meta-analysis were included in the meta-analysis. Analyzes of the data were made in the Comprehensive Meta-Analysis program. The effect sizes of the studies included in the meta-analysis were calculated separately according to the fixed effects and random effects models. As a result of the heterogeneity test performed to determine which of these models to use, it was determined that the studies were heterogeneous and the analyzes were carried out according to the random effects model. In order for the overall effect size found to be valid, the Funnel Scatterplot was examined using Rosenthal's Safe N and Orwin's Safe N publication bias evaluation methods. It was determined that there was no publication bias in the studies. In addition, it was carried out with the help of moderator analysis whether there was a significant difference between the studies according to the publication type, publication year, teaching level, application period, sample size, regions, design type and learning area. According to the random effects model of the studies, the overall effect size was calculated as Hedges's G=0.910 (95% CI, SE=0.098) in the confidence interval of 0.718-1.102. It has been determined that the overall effect size is "wide" according to the classification made by Thalheimer and Cook. All of the studies have positive effect size (in favor of the experimental group). As a result of the moderator analysis, no significant difference was found in the effect of the 5E learning model on students' mathematics achievement according to the publication type, teaching level, application period, the regions where they were made, the type of design and the learning area. However, it did not show a significant difference according to publication year and sample size. According to the results of the research, it was concluded that learning based on the 5E learning model is more effective in improving students' mathematics achievement compared to traditional learning methods. In this context, it can be suggested to encourage teachers to use the 5E learning model in mathematics lessons. Keywords: Effect Size, Meta-analysis, 5E Learning model, Math success, Learning spaces.
Açıklama
Fen Bilimleri Enstitüsü, Matematik ve Fen Bilimleri Eğitimi Ana Bilim Dalı, Matematik Eğitimi Bilim Dalı
Anahtar Kelimeler
Matematik, Mathematics
