David Arthur
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
David Arthur has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Hua-Hua Chang
Department of Educational Studies, Purdue University, West Lafayette, 47907, Indiana, USA
2 shared publications - Nielsen Pereira
Purdue University, West Lafayette, IN, USA
2 shared publications - Yuxiao Zhang
Department of Educational Studies, Purdue University, West Lafayette, 47907, Indiana, USA
2 shared publications - Hernán Castillo-Hermosilla
Purdue University, West Lafayette, IN, USA
1 shared publication - Yukiko Maeda
Purdue University, West Lafayette, USA
1 shared publication - Zafer Ozen
Purdue University, West Lafayette, IN, USA
1 shared publication
Latest Publications
- 1 January 2026
- Published by Psychometric Society
A Joint Factor–Topic Model for Mixed-Format Response Data
Mixed-format response data that combine rating-scale items and open-ended text are increasingly common, yet these modalities are often analyzed separately. We propose a joint factor–topic model that integrates confirmatory factor analysis for rating-scale responses with an LDA-style topic model for text. The two components are coupled through a multinomial logistic regression in which latent factor scores predict document-level topic proportions, enabling coherent joint inference across modalities. The model is implemented in Stan and estimated via Hamiltonian Monte Carlo. We demonstrate the approach using survey data from 577 middle school students, where rating-scale responses measure academic and social giftedness and open-ended responses describe career interests. The model recovered two interpretable career-interest topics (technology/design-oriented vs. people-centered/helping professions) and revealed clear associations between the giftedness factors and topic prevalence. This framework extends topic models by allowing latent covariates inferred under a measurement model and broadens psychometric latent-variable modeling by incorporating text evidence.
- 1 January 2026
- Published by Psychometric Society
A Joint Factor–Topic Model for Mixed-Format Response Data
Mixed-format response data that combine rating-scale items and open-ended text are increasingly common, yet these modalities are often analyzed separately. We propose a joint factor–topic model that integrates confirmatory factor analysis for rating-scale responses with an LDA-style topic model for text. The two components are coupled through a multinomial logistic regression in which latent factor scores predict document-level topic proportions, enabling coherent joint inference across modalities. The model is implemented in Stan and estimated via Hamiltonian Monte Carlo. We demonstrate the approach using survey data from 577 middle school students, where rating-scale responses measure academic and social giftedness and open-ended responses describe career interests. The model recovered two interpretable career-interest topics (technology/design-oriented vs. people-centered/helping professions) and revealed clear associations between the giftedness factors and topic prevalence. This framework extends topic models by allowing latent covariates inferred under a measurement model and broadens psychometric latent-variable modeling by incorporating text evidence.
- 21 November 2025
- Published by SAGE Publications in Gifted Child Quarterly
- Vol. 70 (2) , 257-269
Using Topic Modeling in Gifted Education Research: Drawing Insights From Open-Ended Survey Responses
Despite the rich insights they offer, open-ended survey responses remain underutilized in educational research, often due to a lack of awareness of emergent methods that can efficiently analyze such data. This methodological brief addresses this gap by introducing the topic modeling technique, particularly the latent Dirichlet allocation, as a tool for analyzing open-ended responses and facilitating the investigation of relationships between textual responses and measurable outcomes. This brief outlines the rationale behind topic modeling and details the implementation process through the analysis of open-ended responses on students’ career interests as an example. In addition, we discuss the advantages and limitations of this technique and provide practical considerations for its application in educational research.
- 21 November 2025
- Published by SAGE Publications in Gifted Child Quarterly
- Vol. 70 (2) , 257-269
Using Topic Modeling in Gifted Education Research: Drawing Insights From Open-Ended Survey Responses
Despite the rich insights they offer, open-ended survey responses remain underutilized in educational research, often due to a lack of awareness of emergent methods that can efficiently analyze such data. This methodological brief addresses this gap by introducing the topic modeling technique, particularly the latent Dirichlet allocation, as a tool for analyzing open-ended responses and facilitating the investigation of relationships between textual responses and measurable outcomes. This brief outlines the rationale behind topic modeling and details the implementation process through the analysis of open-ended responses on students’ career interests as an example. In addition, we discuss the advantages and limitations of this technique and provide practical considerations for its application in educational research.