Detecting intervention effects with a cognitive diagnostic model for learning trajectories

As students are exposed to more learning materials or resources, educators and researchers become more interested in selecting effective tools in assisting students' learning process. Specifically, students' mastery and growth of skills in a longitudinal manner depend on the various instructional impacts. However, most current studies tracking transitions of students' mastery of skills rarely addressed instructional effects. For those studies involving intervention covariates, instructional effects are not differentiated considering their categories, the time points they are assigned, and their interactions. This study focuses on the common educational setting when students are assessed after certain instructions. We proposed a cognitive diagnostic model framework in detecting instructional intervention effects while assessing students' learning trajectories. Two specific probit regression models are introduced under the framework. A Bayesian modeling formulation is presented, and Gibbs sampling algorithm is proposed for parameter estimation. Simulation study results show that the proposed model provides accurate estimation of intervention effects and reliable recovery of students' latent attributes.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Li, Anqi

Contributors dc:contributor

• Culpepper, Steven A.
• Chang, Hua-Hua

• NA

Statement dc:rights

• Copyright 2019 Anqi Li

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106500

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/106500