Analysis of the psychometric properties of the Zambia achievement test : reading recognition

Abstract

This study set out to analyze the psychometric properties of the Reading Recognition part of the Zambia Achievement Test. This test was developed by a team of researchers at the Yale University and was adapted to the Zambian situation by the University of Zambia Researchers. The test was then translated into Chinyanja. The main purposes of this study were: (1) to analyse properties of the two versions of the test and: (2) to find out if the two tests were comparable in terms of their ability to discriminate between good and bad readers. A sample of 240 children from grades 4 to 6 were assessed, 121 were assessed in English and 119 were assessed in Chinyanja. The two groups of children were drawn from two schools in Lusaka, Ng'ombe Basic School and Olympia Basic School. The pupils were assigned to either the Chinyanja condition or the English condition depending on which language they spoke at home and were most comfortable with. The pupils were taken through training instructions to make sure that they understood the task. The test, which consisted of letter-matching, soundpicture matching letter-sound matching and word identification, was given to the pupils individually. There were no stoppage rules and all the pupils had to try all the items in the test. The responses were recorded on a bubble sheet. The results from the two groups of children were subjected to several levels of analysis. Item difficulty and discrimination, reliability and the means and standard deviations were all computed. T - tests were applied at each grade level to compare the reading achievement in the two groups of children. Analysis of variance was also applied