2015
DOI: 10.16986/huje.2015014663
|View full text |Cite
|
Sign up to set email alerts
|

Bilişsel Tanı Modellerinde Parametre Kestirimini ve Sınıflama Tutarlılığını Etkileyen Faktörlerin İncelenmesi

Abstract: ÖZ: Bu çalışmanın amacı, Bilişsel Tanı Modelleri'nde madde parametre kestirimini, madde uyumunu ve sınıflama tutarlılığını etkileyen faktörlerin neler olduğunun incelenmesidir. Bu amaç doğrultusunda, tamamlayıcı olmayan model (DINA) kullanılarak çeşitli faktörlere (örneklem büyüklüğü, özelikler arası korelasyon, özelik sayısı, madde sayısı, s ve g parametre düzeyleri) göre veri üretilmiştir. Üretilen veri, DINA analiz modeline göre analiz edilmiş, R 3.0 programı ve CDM paketi kullanılmış ve her bir durum için … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1

Citation Types

0
1
0

Year Published

2021
2021
2021
2021

Publication Types

Select...
1

Relationship

0
1

Authors

Journals

citations
Cited by 1 publication
(1 citation statement)
references
References 5 publications
0
1
0
Order By: Relevance
“…For instance, Akbay ( 2016 ) showed that the non-parametric cognitive diagnosis approach (Chiu and Douglas, 2013 ) performs as well as the CDM based empirical Bayes estimation method for attribute classification in the presence of small sample sizes such as 250, 500, and 1,000. Sünbül and Kan ( 2016 ) investigated the effect of several factors including number of attributes and sample size (i.e., 200, 500, 1,000, and 5,000) on model fit, item recovery, and classification accuracy of the DINA model. The number of attributes and sample size had positive effects on the model estimates.…”
Section: Introductionmentioning
confidence: 99%
“…For instance, Akbay ( 2016 ) showed that the non-parametric cognitive diagnosis approach (Chiu and Douglas, 2013 ) performs as well as the CDM based empirical Bayes estimation method for attribute classification in the presence of small sample sizes such as 250, 500, and 1,000. Sünbül and Kan ( 2016 ) investigated the effect of several factors including number of attributes and sample size (i.e., 200, 500, 1,000, and 5,000) on model fit, item recovery, and classification accuracy of the DINA model. The number of attributes and sample size had positive effects on the model estimates.…”
Section: Introductionmentioning
confidence: 99%