Evaluation of the Computer-based Battery for Oral Language (BILOv3) through the Rasch model for a Brazilian Sample

Abstract

There are few tests in Brazil to evaluate the comprehension of oral language, especially the ones that are computer-based. Thus, this study aimed at investigating psychometric characteristics for a computer-based battery for Oral Language, Bilov3 using the Rasch Model. 569 children collectively answered BILOv3, which is constituted of 6 tests. A general analysis of the battery indicated that the test Completing Sentences was the one with the highest index of difficulty, whereas the test Completing Stories was the lowest. However, in the latter, there are difficult items regarding Morpho-syntactic Comprehension, Logical-Verbal Organization and Story Interpretation. As far as precision was concerned, the Kuder-Richardson coefficient varied from 0.80 to 0.97, which shows that the test is consistent. These results are evidences of validity and precision for BILOv3.