International Comparative Study of Statistics Learning Trajectories Based on PISA Data on Cognitive Diagnostic Models

Abstract

The cognitive diagnosis model is an emerging evaluation theory. The mastery of fine-grained knowledge points of students can be obtained via the cognitive diagnostic model (CDM), which can subsequently describe the learning trajectory. The latter is a description of the learning progress of students in a specific area, through which teaching and learning can be linked. This research is based on nine statistical items in the Program for International Student Assessment (PISA) 2012 and an analysis of the response data of 30,092 students from 14 countries from four attributes based on CDM. Then, it obtains the learning trajectory of students in statistical knowledge. The study found that Bulgaria, Costa Rica, Peru, Mexico, and Serbia have the same learning trajectories. The learning trajectories of almost 14 countries are as follows: (1) uncertainty, (2) data handling, (3) statistical chart, and (4) average.

International comparative study of learning trajectories based on TIMSS 2019 G4 data on cognitive diagnostic models

“…Learning trajectory can be obtained through continuous observation of individual learning processes or through quantitative research on many students’ tests ( Jia et al, 2021 ). The research in this field first started in the field of science and then gradually expanded to various disciplines.…”

“…The result showed that students from Australia, Canada, the United Kingdom, and Russia shared similar main learning trajectories, whereas Finland and Japan were consistent with their main learning trajectories ( Wu et al, 2020 ). Subsequently, a study used the same method with PISA data to study learning trajectories in statistics ( Jia et al, 2021 ). These studies follow a similar line of thought but with improved methods.…”

“…These studies either focus on the factors that affect TIMSS scores or the impact of TIMSS scores on society and education, but few studies analyze TIMSS projects, especially exploring the deep-level mining of information presented by Mathematics Content Domains in TIMSS. Relevant research has been conducted under the Program for International Student Assessment (PISA) to explore students’ learning trajectories (see Wu et al, 2020 ; Jia et al, 2021 ).…”

A new person‐fit method based on machine learning in CDM in education

“…Unlike traditional measurement models, which assume underlying low‐dimensional continuous latent traits, CDMs model students’ performance as a function of the multidimensional binary attribute mastery profile (α). There have been many successful CDM applications in educational assessment (e.g., Chang, 2012, 2015; Jia, Zhu, & Gao, 2021; Wu, Wu, Chang, Kong, & Zhang, 2020; Zhang & Wang, 2020). One of the simplest and most popular CDM models is the deterministic input, noisy ‘and’ gate (DINA) model (Henson, Templin, & Willse, 2009; Junker & Sijtsma, 2001).…”

Using country-specific Q-matrices for cognitive diagnostic assessments with international large-scale data

“…The G-DINA aims to measure to what extent students master a set of cognitive attributes (e.g., fractions, proportions, and decimals as a fine-grained cognitive mathematic attribute) to improve educational policy and practice. Some studies have already employed the G-DINA to analyze data from international comparative tests, such as PISA (Jia et al, 2021;Wu et al, 2020). However, before the G-DINA can be used on international assessments to perform the CDA, a content analysis of the test must occur (von Davier & Lee, 2019).…”