Öğrenci başarılarının belirlenmesi sınavından klasik test kuramı, tek ve çok boyutlu madde tepki kuramı modelleri ile kestirilen başarı puanlarının karşılaştırılması

Özkan, Yeşim Özer

doi:10.14687/ijhs.v11i1.2739

Cited by 6 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hambleton, Swaminathan and Rogers (1991) indicate that difficulty parameters usually vary between -2.0 and 2.0 and item discrimination parameters usually vary between .00 and 2.00 and when related studies (Can, 2003;Çelen, 2008;Özer Özkan, 2012) are investigated it can be seen that these parameters mostly vary between a narrower range. The sample sizes are 500, 1000 and 2000; data sets with item difficulties [-1, +1] and discriminations [0.5 -1.5] were produced with WinGen 2 (Han, 2007) software.…”

Section: Data Setsmentioning

confidence: 98%

Examination of the Effects of Different Missing Data Techniques on Item Parameters Obtained by CTT and IRT

Akbaş¹

2017

IOJES

View full text Add to dashboard Cite

The purpose of this study is to examine the effect of different missing data techniques on the item parameters estimated for Classical Test Theory (CTT) and Item Response Theory (IRT) comparatively through simulated and real data sets. For this purpose, data sets with missing completely at random pattern with different sample sizes and missing data rates have been generated. Item parameters based on CTT and IRT are estimated after applying different missing data techniques (listwise deletion, regression imputation and expectation -maximization). Estimated parameters were compared with the parameters of complete data sets. The same procedure was performed on the data of the PM3 subtest in the PISA 2012 application. It is found that higher difficulty and lower discrimination values are obtained with listwise deletion while accurate item difficulties can be obtained with regression imputation and expectation -maximization algorithm, but the same situation is not valid for item discriminations. It is also seen that missing data at a ratio of %2 can lead to serious problems even if they have missing completely at random pattern.

show abstract

Section: Data Setsmentioning

confidence: 98%

Examination of the Effects of Different Missing Data Techniques on Item Parameters Obtained by CTT and IRT

Akbaş¹

2017

IOJES

View full text Add to dashboard Cite

show abstract

“…Although they are generally similar in item parameters, it is concluded that IRT provides more detailed results than CTT; CTT is useful in pass-fail decisions; IRT is superior in item invariance or individualized test. Although there is not much research based on reliability comparison, the a and b parameters have been examined on the basis of the item, and it has been seen that reliability interpretations are made only on the item and test functions (Çelen & Aybek, 2013;Doğan & Tezbaşaran, 2003;Gelbal, 1994;İlhan, 2016;Kan, 2006;Kelecioğlu, 2001;Kim & Feldt, 2010;Koch, 1983;Köse, 2015;Lee, Torre & Park, 2012;Morales, 2009;Nartgün, 2002;Özdemir, 2004;Özer-Özkan, 2012;Sebille et al, 2010;Sünbül, 2011).…”

Section: Purpose Of the Studymentioning

confidence: 99%

Examination of the Reliability of the Measurements Regarding the Written Expression Skills According to Different Test Theories

Seheryeli

Tan

2019

Eğitimde Ve Psikolojide Ölçme Ve Değerlendirme Dergisi

View full text Add to dashboard Cite

show abstract

“…Bunun için KTK'dan elde edilen iç tutarlık katsayısı, MTK'da ölçülen özelliğin farklı düzeyleri için kestirilen güvenirlik katsayılarının ortalaması olarak kabul edilen marjinal güvenirlik katsayısıyla karşılaştırılmıştır. KTK ile MTK'nın karşılaştırılmasına yönelik çalışmaların önemli bir kısmında ise (Courville, 2004;Çelen & Aybek, 2013;Fan, 1998;Hwang, 2002;İlhan, 2016;Macdonald ve Paunonen, 2002;Ndalichako ve Todd Rogers, 1997, Özer Özkan, 2014Zaman, Kashmiri, Mubarak ve Ali, 2008)…”

Section: Introductionunclassified

Likert Tipi Ölçeklerde Klasik Test Kuramı ile Rasch Analizinden Elde Edilen Yetenek Kestirimleri Arasındaki Uyumun Test Edilmesi

İlhan¹,

Güler²

2017

Ege Eğitim Dergisi

View full text Add to dashboard Cite

The ability estimations calculated in Classical Test Theory (CTT) with the ones calculated in Rasch analysis was compared in the present study in terms of i) relative agreement, ii) absolute agreement, and iii) distribution properties. The research data were collected from a study group of 200 teachers by means of Teacher Efficacy Scale for Classroom Diversity (TESCD) containing 10 items with one-factor structure. Correlation coefficient between ability measures in CTT and in Rasch model was found to be .95, and it was concluded that there was a high relative agreement between ability estimations made in these two theories. The findings obtained in paired samples t-test suggested that the ability estimations made in CTT were significantly higher than those made in Rasch analysis, and consequently, the result was considered as evidence that there were no absolute agreements between the ability scores estimated in these two theories. Finally, it was found that the ability scores estimated in Rasch analysis had more symmetrical and sharper distribution than the ones in CTT.

show abstract

Öğrenci başarılarının belirlenmesi sınavından klasik test kuramı, tek ve çok boyutlu madde tepki kuramı modelleri ile kestirilen başarı puanlarının karşılaştırılması

Cited by 6 publications

References 24 publications

Examination of the Effects of Different Missing Data Techniques on Item Parameters Obtained by CTT and IRT

Examination of the Effects of Different Missing Data Techniques on Item Parameters Obtained by CTT and IRT

Examination of the Reliability of the Measurements Regarding the Written Expression Skills According to Different Test Theories

Likert Tipi Ölçeklerde Klasik Test Kuramı ile Rasch Analizinden Elde Edilen Yetenek Kestirimleri Arasındaki Uyumun Test Edilmesi

Contact Info

Product

Resources

About