Psychometric Analysis of Dyscalculia Test (Published)
The aim of the research was to carry out a Psychometric Analysis of the Dyscalculia Test. Triangulation research design was utilized in this study. The study was driven by four research questions. The study’s population included all 4,758,800 pupils in Nigeria’s upper primary and junior secondary schools in Nigeria. A total of 2340 students were randomly selected using a multistage sampling procedure. The data for the study was the Dyscalculia Test. Expert judgment and empirical evidence of factor analysis were used to establish the instruments’ face, content, and construct validity. Split half Technique was used to ensure the instrument reliability. The split-half reliability study for the Dyscalculia Test indicates that the first half of the test has a reliability estimate of.894 and the second half of the test has a reliability estimate of.780 and a Spearman Brown Prophecy estimate of .824 was used to evaluate the whole test’s reliability. Research questions were answered using p value and discriminatory indexes, content validity Ratio, factor Analysis, and Split half reliability estimate. Result on analysis revealed that for Dyscalculia Test, the P values for the test obtained under CTT varied from 0.2 to 0.8, whereas the R values obtained ranged from -0.002 to 0.740. It was also found that Reliability, content and construct validity were properly established using factor analysis, and a high content validity Ratio and Split Half Reliability estimate. It was recommended based on findings that assessment instruments employed in the educational system, whether at primary or secondary, institutions, should be subjected to item by analysis since they give adequate information on how effectively particular items operate.
Selection of test items is a vital step in test development procedure. Test items could be selected either through the item card or Item Characteristic Curve (ICC). In recent times ICC, has become a vital tool in the selection of tests items as it shows graphically the psychometric properties of the test items. At a glance the curve gives such information as discriminating, difficulty and guessing value (depending on the model). This study works on modeling of 2- parameter model of Item Response Theory (IRT) to generate the item characteristics curve using a software in addition to the windows Microsoft excel application. It also involves a do-it-yourself guide using the windows excel application.The experimental result shows that the ICC curve could be a vital tool in determining the suitability of item for section in a test.