Advanced search
Start date
Betweenand

ITAI-PI: computerized adaptive test of academic English for web

Grant number: 14/10404-4
Support Opportunities:Regular Research Grants
Start date: November 01, 2014
End date: October 31, 2016
Field of knowledge:Physical Sciences and Mathematics - Probability and Statistics - Applied Probability and Statistics
Principal Investigator:Mariana Cúri
Grantee:Mariana Cúri
Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

This project aims the transformation of an English Proficiency Test (EPT), from Computer Science and Computational Mathematics graduate program of ICMC-USP, in a computerized adaptive test (CAT) for web, based on item response theory (IRT) models. The EPT pattern since 2002, the data structure collected along its semestral applications are briefly described and will base the CAT proposed to be developed in this work. The software iTAI_PI (web based computerized adaptive test for english proficiency) will be created using Java and PostgreSQL, for data base. The statistical methodology consists in three IRT models (Samejima graded response model, item parameter random effects model for testlets and DINA model), the Kullback-Leibler criteria for item selection, the expected a posteriori bayesian estimation for latent traits and shadow test approach for rigid constraints control (size of the test or content restriction, for instance). Psychometric and computational validation will be studied. As well as the interpretation of the proficiency scale and determination of a classification criteria of the individual in approved or not approved. The project is multidisciplinary, formed by researchers, undergraduate and graduate students of Statistics, Literature and Computational areas. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
More itemsLess items
Articles published in other media outlets ( ):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)