A Diagnostic Comparison of Turkish and Korean Students’ Mathematics Performances on the TIMSS 2011 Assessment

Abstract

The purpose of the present study was to analyze an international large-scale data set using a cognitive assessment approach. Although some researchers question the usefulness of international large-scale assessments (e.g., TIMSS), participating countries have continued to use the results from these large-scale assessments to improve their curricula and teaching methods. Despite the common reporting practice-single-score-in these large scale assessments gives useful insights about students' overall performances, they still lack diagnostic information. Cognitive diagnosis models (CDMs) were developed to provide more feedback on students' cognitive strengths and weaknesses. This study retrofitted the TIMSS 2011 eighth grade mathematics assessment by applying a specific CDM called the DINA (the deterministic, inputs, noisy, "and" gate) model to data from South Korea and Turkey. Results of the DINA model were used to make a detailed comparison between students of these two countries.

Growth Modeling in a Diagnostic Classification Model (DCM) Framework–A Multivariate Longitudinal Diagnostic Classification Model

“…DCMs have been increasingly used for empirical data analysis in recent years. For example, DCMs have been retrofitted to existing large-scale assessments to identify examinees' mastery status of tested skills (e.g., Lee and Sawaki, 2009;George and Robitzsch, 2014;Sedat and Arican, 2015;Ravand, 2016). In addition, some researchers successfully demonstrated the practical uses of DCMs in test development .…”

Investigating Preservice Middle School Mathematics Teachers’ Competencies in Statistics and Probability in Terms of Various Variables

“…Bilişsel tanı modellerinin güçlü olduğu yönlerden birisi, az sayıda test maddesi kullanılsa bile KTK'ya kıyasla daha güvenilir tahminler sunmalarıdır (Templin & Bradshaw, 2013). Son yıllarda araştırmacılar bilişsel tanı modellerini kullanarak öğrencilerin (örn., Choi, Lee & Park, 2015;Dogan & Tatsuoka, 2008;Im & Park, 2010;Lee, Park & Taylan, 2011;Sen & Arican, 2015), öğretmenlerin (örn., Bradshaw, Izsak, Templin & Jacobson, 2014) ve öğretmen adaylarının (örn., Arican & Kuzu, 2019) çeşitli matematik konularından elde ettikleri sonuçlar üzerine tanılayıcı değerlendirmede bulunmuşlardır.…”

A Nondiagnostic Assessment for Diagnostic Purposes: Q-Matrix Validation and Item-Based Model Fit Evaluation for the TIMSS 2011 Assessment

“…The attributes of Q-matrix in this study were adapted from the Common Core State Standards for Mathematics (Common Core State Standards Initiative, 2010). Table 2 presents attribute description for each content domain reported by TIMSS researchers (Mullis et al, 2012), and subattributes were determined by four experts in mathematics education (Şen & Arıcan, 2015). Note that as Items M052503A and M052503B were the same in the original 32-item list, one of them (Item M052503A) was dropped from the Q-matrix (Sen & Arıcan, 2015), thus, a total of 31 items was used in this study.…”

“…The first purpose is to validate attribute specifications in the Q-matrix under the G-DINA model in that any reduced model needs not to be known. Rather than constructing the Q-matrix of the test, it was adapted from Şen and Arıcan (2015). The validation of attribute specifications was implemented by the G-DINA model discrimination index (GDI; de la Torre & Chiu, 2016).…”