Journals / Eğitim ve Bilim / 2012 / Cilt: 37 - Sayı: 165

An ınvestigation of ıtem bias in Turkish sub tests in level determination exam

Seviye belirleme sınavındaki Türkçe alt testlerinin madde yanlılığının incelenmesi

Pages
348–362
Publication Date
2022-10-18
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

In the light of the experts’ views, it was found that only one of the itemsthat include DIF caused item bias and other items did not cause any item bias according to thegender and school types; and the difference between the groups was caused by the item effect.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The purpose of this study is to determine whether 2009 Level Determination Exam (LDE) 6th, 7thand 8th year Turkish sub tests indicate item bias in terms of the gender of the students and school types.In the light of this purpose, Mantel-Haenszel (MH) and Logistic Regression (LR) methods were usedin order to determine the differential item functioning of the items in the tests.The study was carried out with 22656 students in total, 6922 of whom were 6th year students(3620 males and 3302 females), 6360 of whom were 7th year students (3290 males and 3070 females)and 9374 of whom were 8th year students (4290 males and 5084 females).
📊
Key Findings & Data:In the study, in terms ofthe gender variable, a) DIF was not found in any of the items at B and C level in the 6th year, (b) DIFwas found in 2 items in the 7th year according to the MH technique and in 1 item according to LRtechnique, (c) DIF was also found in 5 items in the 8th year according to MH technique and in 3 itemsaccording to LR technique.
💡
Conclusion & Impact:In the light of the experts’ views, it was found that only one of the itemsthat include DIF caused item bias and other items did not cause any item bias according to thegender and school types; and the difference between the groups was caused by the item effect.
🔬Design:Research Article
📍Scope:2009 Verileri
⚙️Variables:Domain Parameters

Abstract

The purpose of this study is to determine whether 2009 Level Determination Exam (LDE) 6th, 7thand 8th year Turkish sub tests indicate item bias in terms of the gender of the students and school types.In the light of this purpose, Mantel-Haenszel (MH) and Logistic Regression (LR) methods were usedin order to determine the differential item functioning of the items in the tests.The study was carried out with 22656 students in total, 6922 of whom were 6th year students(3620 males and 3302 females), 6360 of whom were 7th year students (3290 males and 3070 females)and 9374 of whom were 8th year students (4290 males and 5084 females). In the study, in terms ofthe gender variable, a) DIF was not found in any of the items at B and C level in the 6th year, (b) DIFwas found in 2 items in the 7th year according to the MH technique and in 1 item according to LRtechnique, (c) DIF was also found in 5 items in the 8th year according to MH technique and in 3 itemsaccording to LR technique. In terms of school type variable, 7 items in the 6th year had DIF accordingto MH technique and 5 items had DIF according to LR technique; 2 items in the 7th year accordingto MH technique and 1 item according to LR technique had DIF; and 2 items in the 8th year had DIFaccording to MH technique. In the light of the experts’ views, it was found that only one of the itemsthat include DIF caused item bias and other items did not cause any item bias according to thegender and school types; and the difference between the groups was caused by the item effect.

Özet

Bu çalışmanın amacı, 2009 yılı Seviye Belirleme Sınavı (SBS) içerisindeki Türkçe alt testlerininöğrencilerin cinsiyetine ve okul türlerine göre madde yanlılığı gösterip göstermediğini belirlemektir.Bu amaç doğrultusunda testler içerisindeki maddelerin değişen madde fonksiyonlarını belirlemekiçin Mantel-Haenszel (MH) ve Lojistik Regresyon (LR) yöntemleri kullanılmıştır.Araştırma; 6922’si (3620 erkek, 3302 kız) 6. sınıf, 6360’ı (3290 erkek, 3070 kız) 7. sınıf ve9374’ü (4290 erkek ve 5084 kız) 8. sınıf olmak üzere toplam 22656 öğrenci üzerinde yürütülmüştür.Araştırmada cinsiyet değişkenine göre; a) 6. sınıfta B ve C düzeyinde hiçbir maddede DMF’yerastlanmamıştır, b) 7. sınıfta MH tekniğine göre 2 madde, LR tekniğine göre 1 madde, c) 8. sınıfta MHtekniğine göre 5 madde, LR tekniğine göre ise 3 maddede DMF’ye rastlanmıştır. Okul değişkeninegöre; 6. sınıfta MH tekniğine göre 7 maddenin, LR tekniğine göre 5 madde, 7. sınıfta MH tekniğine göre2, LR tekniğine göre 1 maddenin ve 8. sınıfta MH tekniğine göre 2 maddenin DMF’ye sahip olduğugörülmüştür. Uzman görüşleri doğrultusunda DMF içeren maddeler arasından yalnızca bir tanesininmadde yanlılığı oluşturduğu, diğer maddelerin ise cinsiyete ve okul türüne göre madde yanlılığıoluşturmadığı, gruplar arasındaki farkın madde etkisinden kaynaklandığı sonucuna ulaşılmıştır.

🔖

Cite This Article

KUTLU, Ö., KARAKAYA, İ. (2022). An ınvestigation of ıtem bias in Turkish sub tests in level determination exam. Eğitim ve Bilim, Cilt: 37 - Sayı: 165, 348–362. https://ajindex.com/articles/497614

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (9)

  1. Allalouf, A., Hambleton, R., ve Sireci, S. (1999). Identifying the Causes of DIF in Translated Verbal Items. Journal of Educational Measurement, 36, 185-198.
  2. Amrein, A. L. ve Berliner, D. C. (2002). High-stakes testing, uncertainty, and student learning. Education Policy Analysis Archives, 10(18). http://epaa.asu.edu/epaa/v10n18/.
  3. Anıl, D. ve Güzeller, C. O (2010). Seviye Belirleme Sınavı Yedinci Sınıf Fen ve Teknoloji Alt Testi ile Diğer Alt Testler Arasındaki İlişkinin Yol Analizi ile İncelenmesi. Eğitimde ve Psikolojide Ölçme ve Değerlendirme II. Ulusal Kongresi. Mersin Üniversitesi,
  4. Mersin. Anıl, D., Güzeller,C.O, Çokluk,Ö. ve Şekercioğlu, G. (2010) “Level Determination Exam (SBS- 2008) the Determination of the Validity and Reliability of 7th Grade Mathematics Sub- Test”, Procedia Social and Behavioral Sciences, 2(2) 5292–5298.
  5. Bekçi, B. (2007). Ortaöğretim Kurumları Öğrenci Seçme Sınavının Değişen Madde Fonksiyonlarının Cinsiyete ve Okul Türüne Göre İncelenmesi, Hacettepe Üniversitesi Sosyal Bilimler Enstitüsü. Berberoğlu, G., Demirtaşlı, N., Güzel Ç. İ., Arıkan, S. ve Tuncer, Ç. Ö. (2010). Okul Dışı Etmenlerin Öğrenci Başarısı ile İlişkisi. Cito Eğitim: Kuram ve Uygulama, Sayı 7- 27-36.
  6. Camilli, G. ve Shepard, L. A. (1994). Methods for identifying biased test items. London: Sage Publications. Clauser, E. B. ve Mazor, M. K. (1998). Using statistical procedures to identify differentially functioning test items. Educational Measurement: Issues And Practice. 17 (1), 31–44.
  7. Cochran, W. G. (1977),
  8. Sampling Techniques, Third Edition, New York: John Wiley & Sons, Inc. Doğan, N. ve Sevindik, H. (2011). İlköğretim 6. Sınıflar İçin Uygulanan Seviye Belirleme Sınavının Uygunluk Geçerliği. Eğitim ve Bilim Dergisi, 36 (110), 309-319.
  9. Geske, A. ve Ozola, A. (2010). Differential item functioning in the aspect of gender differences in reading literacy. Paper Presented at the The 4th IEA International Research Conference. www.iea.nl/fileadmin/user_upload/IRC/IRC./IRC2010_Geske_Ozola.pdf Gierl, M., Khaliq, N. S ve Boughton, K. (1999). Gender differential item functioning in mathematics and science: Prevalence and policy implications. Paper Presented at the Symposium entitled “Improving Large-Scale Assessment in Education” at the Annual Meeting of the Canadian Society for the Study of Education, Canada. http://www.education.ualberta.ca/educ/ psych/ crame/files/dif_csse99.pdf Gök, B., Kelecioğlu, H. ve Doğan, N. (2010). Değişen Madde Fonksiyonunu Belirlemede Mantel– Ha