Assessing Change in Latent Skills Across Time With Longitudinal Cognitive Diagnosis Modeling: An Evaluation of Model Performance

Abstract

Cognitive diagnosis models are diagnostic models used to classify respondents into homogenous groups based on multiple categorical latent variables representing the measured cognitive attributes. This study aims to present longitudinal models for cognitive diagnosis modeling, which can be applied to repeated measurements in order to monitor attribute stability of individuals and to account for respondent dependence. Models based on combining latent transition analysis modeling and the DINA and DINO cognitive diagnosis models were developed and then evaluated through a Monte Carlo simulation study. The study results indicate that the proposed models provide adequate convergence and correct classification rates.

Reliably Assessing Growth with Longitudinal Diagnostic Classification Models

“…For example, to obtain the deterministic‐inputs, noisy‐and‐gate model (DINA; e.g., Haertel, ; Junker & Sijtsma, ), both main effects, and , would be constrained to be zero, and only the intercept and interaction terms would be estimated. Previous longitudinal DCMs have used constrained DCMs like the DINA model (Kaya & Leite, ; Li et al., ; Wang et al., ). By specifying the LCDM measurement model, the TDCM serves as a generalized version of other longitudinal DCMs.…”

“…Most often, classical test theory (CTT) and IRT models have been used to provide measures of student growth on a continuous scale in the form of gain scores or ability gain scores, respectively. Recently, however, several studies have developed and applied longitudinal DCMs to study how examinees transition between different attribute mastery statuses over time (Hansen, ; Kaya & Leite, ; Li, Cohen, Bottge, & Templin, ; Madison & Bradshaw, , ; Wang, Yang, Culpepper, & Douglas, ). Wang et al.…”

A reinforcement learning approach to personalized learning recommendation systems

“…The current work focuses on developing a general method for optimizing the recommendation system for adaptive learning and showcases a concrete example in which learners are modelled by a dynamic cognitive diagnosis model (Kaya & Leite, 2017;Li, Cohen, Bottge, & Templin, 2016;Wang, Yang, Culpepper, & Douglas, 2017). The proposed method can be used for the development of an adaptive recommendation system for online e-learning platforms, intelligent tutoring systems (VanLehn, 2011), etc.…”

Evaluating Intervention Effects in a Diagnostic Classification Model Framework

“…In contrast to the continuous score-based modeling approaches mentioned above, a diagnostic classification model (DCM; Rupp, Templin, & Henson, 2010; see also Bradshaw, 2016) can be used when continuous latent traits (i.e., abilities) are replaced by categorical latent traits or attributes. Recent developments have combined latent transition analysis (LTA; Collins & Wugalter, 1992) with a DCM as the measurement model to assess change in attribute mastery over time (Kaya & Leite, 2017;Li, Cohen, Bottge, & Templin, 2016;. Li et al (2016) used LTA in conjunction with the deterministic-inputs, noisy-and-gate (DINA; e.g., Haertel, 1989;Junker & Sijtsma, 2001) model to form what they refer to as the LTA-DINA model.…”

“…Li et al (2016) used LTA in conjunction with the deterministic-inputs, noisy-and-gate (DINA; e.g., Haertel, 1989;Junker & Sijtsma, 2001) model to form what they refer to as the LTA-DINA model. Kaya and Leite (2017) combined LTA and the deterministic inputs noisy or gate model (DINO; Templin & Henson, 2006) to form a longitudinal DINO model. Madison and Bradshaw (2016) developed a more general version of these models by using a subsuming DCM, the log-linear cognitive diagnosis model (LCDM; Henson, Templin, & Willse, 2009), to form what they refer to as the transition diagnostic classification model (TDCM).…”