Fisika & Matematika editorial
Open AccessOA2022
Personalized Assessment: Applying Higher-Order Cognitive Diagnosis Models in Secondary Mathematics
This study applies higher-order cognitive diagnosis models to assess 391 Grade 11 students' mastery of derivative attributes in mathematics. The higher-order GDINA with 1PL best fit the data, showing most students mastered basic derivative concepts but less than half mastered complex transformative contextualization.
Y. Zhang; Yi Jin; Zhen-Rong Xiong; Shing-On Leung; Gao-Wei Chen; Na Li; Bo Liยท Asian Journal for Mathematics Educationยท 2022ยท DOI 10.1177/27527263221136301
The core problem
Personalized assessment is essential in education, yet few studies have applied cognitive diagnosis models (CDMs) in secondary mathematical contexts. This study addresses this gap by applying higher-order CDMs to assess Grade 11 students' mastery of derivative concepts in China. The authors aim to demonstrate how CDMs can provide detailed individual skill profiles and inform personalized feedback. The research questions focus on model fit, Q matrix validation, and attribute mastery patterns at grade and individual levels.
Innovation
A sample of 391 Grade 11 students from a secondary school in China participated. The study employed higher-order generalized deterministic inputs, noisy, 'and' gate (higher-order GDINA) models with different link functions. Model fit was compared using one-parameter logistic (1PL) and other variants. A Q matrix was constructed to define four attributes related to derivatives: B1 (basic concept development: simple equations, zero or extreme points, function range problems), B2 (complex inductive contextualization: induction from known to unknown), B3 (basic routine problem solving: graphs through fixed point or quantitative inequalities), and B4 (complex transformative contextualization: transformation combining numbers and graphs). Q matrix validation was conducted to ensure attribute definitions align with item demands. Data were analyzed using cognitive diagnosis modeling techniques to estimate individual attribute mastery probabilities.
Introduction
Personalized assessment is essential in education, yet few studies have applied cognitive diagnosis models (CDMs) in secondary mathematical contexts. This study addresses this gap by applying higher-order CDMs to assess Grade 11 students' mastery of derivative concepts in China. The authors aim to demonstrate how CDMs can provide detailed individual skill profiles and inform personalized feedback. The research questions focus on model fit, Q matrix validation, and attribute mastery patterns at grade and individual levels.
Why it matters
The findings indicate that while students generally master basic derivative concepts and routine problem solving, they struggle with complex transformative contextualization (B4), which requires combining numerical and graphical representations. This suggests a need for targeted instructional interventions to develop higher-order skills. The study demonstrates the utility of higher-order CDMs in secondary mathematics for personalized assessment. The individual skill profiles enable teachers to tailor feedback and remediation. Implications for personalized assessment are discussed, including the potential for CDMs to support differentiated instruction and formative assessment. Limitations and future research directions are also considered.
Who should read this
Pembaca Research Digest dan praktisi rumpun ini.
Opening member contentโฆ