- ASA 2022 Data-Driven Decision Making
- A cura di Enrico di Bella, Luigi Fabbris, Corrado Lagazio
The joint estimation of accuracy and speed: An application to the INVALSI data
- Luca Bungaro
- Marta Desimoni
- Mariagiulia Matteucci
- Stefania Mignani
- © 2023 Author(s) |
- CC BY 4.0
- DOI: 10.36253/979-12-215-0106-3.39
In Italy, the National Institute for the Evaluation of the Education and Training System (INVALSI) every year administers standardized tests via computer-based testing (CBT) to students attending grades 8, 10, and 13. The CBT mode allows to collect data not only on the students’ response accuracy (RA) based on item responses, but also on their response times (RT). By using these data, it is now possible to estimate the speed ability of examinees, besides the usual ability (e.g. Italian language, mathematics or English ability). In this study, we use the 2018 mathematics data for grade 10 to estimate the ability and speed of students following the fully Bayesian approach of Fox et al. (2021), who implemented in the R package LNIRT the models of van der Linden (2007) and Klein Entik et al. (2009). In a second step, we use the estimated mathematics ability and speed in a bivariate multilevel model, where the first-level units are represented by students and the second-level units are represented by classes. Covariates such as gender, school type, immigrant status, economic, social, and cultural status, prior achievement, grade retention, student anxiety, class compositional variables, and geographical area are included in the model. The main results show that the ability and speed are inversely proportional, e.g. as ability increases, speed decreases. Also, differences in the students performance by gender and school type are significant for both ability and speed.
- Keywords:
- educational assessment,
- large standardized test,
- mathematics achievement,
- IRT models for response times,
- multilevel models,
University of Bologna, Italy
INVALSI, Italy - ORCID: 0000-0002-3407-0002
University of Bologna, Italy - ORCID: 0000-0003-3404-6325
University of Bologna, Italy - ORCID: 0000-0003-4746-1130
- Charlton, C., Rasbash, J., Browne, W.J., Healy, M., Cameron, B. (2020). MLwiN Version 3.05. Centre for Multilevel Modelling, University of Bristol.
- Fox, J. P., Klotzke, K., & Simsek, A. S. (2021). LNIRT: An R Package for Joint Modeling of Response Accuracy and Times. arXiv preprint arXiv:2106.10144.
- Klein Entink, R. H., Fox, J.-P., van der Linden, W. J. (2009). A Multivariate Multilevel Approach to the Modeling of Accuracy and Speed of Test Takers. Psychometrika, 74(1), pp. 21-48.
- Rasbash, J., Steele, F., Browne, W.J., Goldstein, H. (2017). A User's Guide to MLwiN, v3.00. Centre for Multilevel Modelling, University of Bristol.
- van der Linden, W. J. (2007). A Hierarchical Framework for Modeling Speed and Accuracy on Test Items. Psychometrika, 72(3), 287.
Informazioni sul capitolo
Titolo del capitolo
The joint estimation of accuracy and speed: An application to the INVALSI data
Autori
Luca Bungaro, Marta Desimoni, Mariagiulia Matteucci, Stefania Mignani
Lingua
English
DOI
10.36253/979-12-215-0106-3.39
Opera sottoposta a peer review
Anno di pubblicazione
2023
Copyright
© 2023 Author(s)
Licenza d'uso
Licenza dei metadati
Informazioni bibliografiche
Titolo del libro
ASA 2022 Data-Driven Decision Making
Sottotitolo del libro
Book of short papers
Curatori
Enrico di Bella, Luigi Fabbris, Corrado Lagazio
Opera sottoposta a peer review
Anno di pubblicazione
2023
Copyright
© 2023 Author(s)
Licenza d'uso
Licenza dei metadati
Editore
Firenze University Press, Genova University Press
DOI
10.36253/979-12-215-0106-3
eISBN (pdf)
979-12-215-0106-3
eISBN (xml)
979-12-215-0107-0
Collana
Proceedings e report
ISSN della collana
2704-601X
e-ISSN della collana
2704-5846