Yükseköğretimde Program Yetkinliği: Bir Ölçek Geliştirme Çalışması


Derin Kılıç A., Küçükoğlu A.

JOURNAL OF HIGHER EDUCATION AND SCIENCE (YUKSEKOGRETIM VE BILIM DERGISI), cilt.14, sa.3, ss.462-484, 2024 (Hakemli Dergi)

Özet

Yükseköğretimde kalite hedefleri her geçen gün biraz daha ön plana çıkarken gerek ulusal gerekse de uluslararası ölçekte etkinliği yüksek, yenilikçi program(lar)ın tasarlanabilmesi noktasında kritik bir öneme sahip olduğu düşünülen öğretim elemanlarının programa ilişkin (öğretim, araştırma vb.) yetkinlik düzeylerinin incelenebilmesi amacıyla beşli likert türünde derecelendirilen Program Yetkinliği Ölçeği’nin 105 madde ve tek boyuttan oluştuğu saptanmıştır. Bu araştırma kapsamında araştırma grubu belirlenirken, bilgi açısından zengin durumların seçilerek derinlemesine bir araştırmanın yapılmasına olanak sağlayan amaçsal örnekleme yöntemine başvurulmuş olup öğretim elemanlarından oluşan bir araştırma grubundan veri toplanmıştır. Araştırma grubundan elde edilen veriler rastgele iki alt gruba bölünmüş olup ilk grup üzerinde (336 kişi) AFA ve diğer grup üzerinde (336 kişi) ise DFA gerçekleştirilmiştir. Bunun yanı sıra ilgili alan yazında yer alan benzer araştırmalar göz önünde bulundurularak Klasik Test Kuramı’na dayalı analizlerin gerçekleştirildiği araştırma grubu üzerinden Madde Tepki Kuramı çatısı altında yer alan “Sıralama Ölçekli Modele Dayalı Rasch Analizi” gerçekleştirilmiştir. Veri analizi sonucunda AFA bulguları incelendiğinde, öz değeri birden büyük olan 26 faktörlü bir yapının ortaya çıktığı gözlenmiştir. Öte yandan ilgili alanyazında program yetkinliğine ilişkin çeşitli boyutların olduğunu öne süren araştırmalar bulunsa da bu boyutların iç içe geçmesi nedeniyle boyutlar arasında tam bir ayrımın yapılamadığı göz önünde bulundurularak bu araştırma kapsamında birinci ve ikinci öz değer oranı da incelenmiştir. Nitekim birinci ve ikinci öz değer oranının üçten büyük olması (18.820/6.273) nedeniyle tek boyutlu bir yapının ortaya çıktığı sonucuna varılmıştır. Tek faktörlü bu yapının toplam varyansın %46.475’ini açıkladığı bulgulanmıştır. Ayrıca tek boyutluluğun temel ve katı olmak üzere ikiye ayrıldığı göz önünde bulundurularak bu ölçekte öz değeri birden büyük olan ikincil minör boyutların/faktörlerin bulunması nedeniyle ölçek yapısının temel tek boyutluluğa uygun olduğu düşünülmüştür. Başka bir ifadeyle, söz konusu faktörlerin yorumlanması ve adlandırılması noktasında yaşanan problemler nedeniyle ölçeğin tek faktörlü/boyutlu olarak kullanılmasının uygun olacağı sonucuna ulaşılmıştır. Bununla birlikte ölçeğin tek faktörlü yapısını test etmek amacıyla doğrulayıcı faktör analizi gerçekleştirilmiş olup veri analizi sonucunda uyum indekslerinin (GFI=.86, AGFI=.87, CFI=.96 ve RMSEA=.07) kabul edilebilir aralıkta olduğu saptanmıştır. Ayrıca Rasch analizinden elde edilen sonuçlar incelendiğinde, (0.54 ile 1.39 değerleri arasında yer alan) uyum-içi ve (0.51 ile 1.38 değerleri arasında yer alan) uyum-dışı istatistiklerinin bu istatistiklere ilişkin kabul edilebilir sınır olarak kabul edilen 0.5 ile 1.5 aralığında yer aldığı saptanmış olup model-veri uyumunun sağlandığı gözlenmiştir. Bu bulgulara dayanarak araştırma verilerinin ölçek geliştirme sürecinde esas alınan tek boyutlu yapıyı doğruladığı şeklinde bir sonuca varılabilmektedir. Öte yandan bu araştırma kapsamında geliştirilen ölçeğin güvenirlik düzeyi, Cronbach alfa güvenirlik yöntemi kullanılarak hesaplanmış olup Cronbach alfa güvenirlik katsayısı a=0.962 olarak tespit edilmiştir. Bu değerin 0.80’den yüksek olması ise ölçeği oluşturan maddeler arasında yüksek düzeyde iç tutarlılık olduğuna işaret etmektedir. Sonuç olarak Program Yetkinliği Ölçeği’nin geçerlik ve güvenirlik değerlerinin yüksek düzeyde olduğu saptanmış olup elde edilen bulgular, öğretim elemanlarının program yetkinliği düzeylerini belirlemeye yönelik ilk ölçme aracı olma özelliği taşıyan bu ölçeğin öğretim elemanlarının program yetkinliği düzeylerini belirleme noktasında geçerli ve güvenilir bir araç olduğunu ortaya koymuştur.

Anahtar Sözcükler: Yükseköğretim, Öğretim elemanı, Program yetkinliği, Ölçek geliştirme 

While the quality goals in higher education are coming to the fore day by day, it has been determined that the Curriculum Competence Scale, which was developed in the five-point likert type in order to examine the curriculum competence levels of the lecturers which is considered to be of critical importance in designing innovative curriculum(s) with high efficiency both nationally and internationally, consists of 105 items and a single dimension. Within the scope of this research, while determining the research group, the purposive sampling method, which allows an in-depth research to be carried out by selecting information-rich situations, was used and data were collected from a research group consisting of lecturers. The data obtained from the research group were randomly divided into two subgroups, and EFA was performed on the first group (336 people) and CFA was performed on the other group (336 people). In addition, considering the similar studies in the relevant literature, the “Ranking-Scale Model-Based Rasch Analysis” under the umbrella of Item Response Theory was performed on the research group that Classical Test Theory was carried out. As a result of the data analysis, when the AFA findings were examined, it was seen that a 26-factor structure with an eigenvalue greater than one emerged. On the other hand, although there are studies suggesting that there are various dimensions related to curriculum competence in the relevant literature, considering that a complete distinction cannot be made between the dimensions due to the intertwining of these dimensions, the ratio of the first and second selfvalue to each other has also been examined within the scope of this research. It was concluded that a one-dimensional structure emerged due to the fact that the first and second eigenvalue ratio was greater than three (18.820/6.273). It was found that this one-factor structure explained 46.475% of the total variance. In addition, considering that one-dimensionality is divided into two as essential and strict, it is thought that the scale structure is suitable for essential one-dimensionality due to the presence of secondary minor dimensions/factors with eigenvalues greater than one on this scale. In other words, it has been concluded that it would be appropriate to use the scale as a single factor/dimensional due to the problems experienced in the interpretation and naming of these factors. However, in order to test the single-factor structure of the scale, confirmatory factor analysis was performed and as a result of the data analysis, it was determined that the fit indices (GFI=.86, AGFI=.87, CFI=.96 and RMSEA=.07) were within the acceptable range. In addition, when the results obtained from the Rasch analysis were examined, it was determined that the in-fit and out-fit statistics (between 0.54 and 1.39) were in the range of 0.5 and 1.5, which is accepted as the acceptable limit for these statistics, and it was understood that model-data fit was achieved. Based on these findings, it can be concluded that the research data confirm the one-dimensional structure in the scale development process. On the other hand, the reliability level of the scale developed within the scope of this study was calculated using the Cronbach alpha reliability method. Also the Cronbach alpha reliability coefficient was determined as a=0.962. If this value is higher than 0.80, it indicates that there is a high level of internal consistency among the items. As a result, it was determined that the validity and reliability values of the Curriculum Competence Scale were high, and the findings revealed that this scale, which is the first measurement tool to determine the curriculum competence levels of the lecturers, is a valid and reliable tool for determining the curriculum competence levels of the lecturers.

Keywords: Higher education, Lecturer, Curriculum competence, Scale development