Sara Brandão

Development of Listening Comprehension Tests with Narrative and Expository Texts for Portuguese Students

This investigation aimed to develop and collect psychometric data for two tests assessing listening comprehension of Portuguese students in primary school: the Test of Listening Comprehension of Narrative Texts (TLC-n) and the Test of Listening Comprehension of Expository Texts (TLC-e). Two studies were conducted. The purpose of study 1 was to construct four test forms for each of the two tests to assess first, second, third and fourth grade students of the primary school. The TLC-n was administered to 1042 students, and the TLC-e was administered to 848 students. The purpose of study 2 was to test the psychometric properties of new items for the TLC-n form for fourth graders, given that the results in study 1 indicated a severe lack of difficult items. The participants were 260 fourth graders. The data were analysed using the Rasch model. Thirty items were selected for each test form. The results provided support for the model assumptions: Unidimensionality and local independence of the items. The reliability coefficients were higher than .70 for all test forms. The TLC-n and the TLC-e present good psychometric properties and represent an important contribution to the learning disabilities assessment field.

Evaluating listening comprehension in children using a dynamic assessment approach: An exploratory study

The aim of this study was to investigate the psychometric properties of the Dynamic Assessment Test of Listening Comprehension4–6 (DATLC4–6) for Portuguese preschoolers aged four to six years old. The test construction was based on the principles of dynamic assessment, specifically the gradual prompts model. The test includes 40 multiple-choice items, each of which includes three types of prompts. A total of 108 Portuguese preschool children participated in the study. High corrected item-total correlations were obtained. The principal component analysis results suggested that a single factor explains a high percentage of the results’ variance. The value of Cronbachs alpha coefficient indicates a notably high internal consistency. The preliminary analysis of the DATLC4-6 suggests adequate psychometric properties. Limitations and directions for future studies are discussed.