Item Analysis using a Derived Science Achievement Test Data

Item analysis is useful in both the development, evaluation of assessments tools and in computing standardized measures of student performance. In item analysis, item statistics (difficulty, discrimination) for each item or question provide a means of assessing the quality of the test items. This study demonstrates the use of the classical item analysis to evaluate the quality of multiple choices Chemistry test items used in Kano state qualifying examination in July 2014. Forty items Chemistry test administered to students at the end of their senior secondary school two (SSII) was subjected to the classical analysis using sample of 530 students, the Statistical Packages for the Social Sciences, version 20 (SPSS 20V) was used to determine the discrimination and difficulty indices of the items and the classification of the items according to their item characteristics. Findings revealed that, out of the 40 items in the test, 12 (30 %) items failed to meet the set criteria of item quality and are therefore needs to be revise or improve for further administration.28 items based on the established standards has been considered as -good- items. Similarly the results indicate a significant positive correlation between item difficulty and item discrimination indices. It is recommended that, the teacher made Chemistry achievement tests use to examine sciences secondary school students- achievement should be made to pass through all the processes of standardization and validation by conducting psychometric analysis to improve their quality.