| Summary: | The major purpose of assessment in education is to measure students' achievement, test performance is used to make decision. The reliance on information gained from test scores to make decision has been a major area of concern in teaching and research. Therefore, item analysis is applied to evaluate the quality of items used in measuring students learning to ensure the standard will be achieved. In education and psychology, two competing measurement frameworks, namely Classical Test Theory (CIT) and Item Response Theory (IRT) techniques are applied to evaluate and improve measuring instruments. This study aims to analyse the teacher made Chemistry Achievement Test and compare the item parameters generated by using CIT and IRT models. The developed forty (40) items of Chemistry test was administered to a sample of N=530 senior secondary school students. Subsequently, the obtained data had been analysed using CIT and 2 parameter IRT models to determine the difficulty and discrimination indices of the items and the correlation between the indices derived from the two models. The specialised XCALIBRE 4 and SPSS 20v softwares were used to conduct the item analysis. The result of findings indicate that, both frameworks commonly identified 17(42.5%) items as "rosak" having failed to meet the set items quality standards and 23(57.5%) were "good". Similarly, the results indicate a significantly higher correlation between item parameters derived from the CIT and the 2 parameter IRT models, [(r=-0.985, p<0.05) and (r=0.80 I, p<0.05)] for item difficulty and discrimination respectively. It can be concluded that firstly, the Chemistry Achievement test used in Kano State Qualifying Examination needs to be adequately improved; secondly, CIT and IRT frameworks appeared to be effective and reliable in assessing test items as the two frameworks provide similar and comparable results. The study recommends that the teacher made Chemistry tests used in measuring students' achievement should be made to pass through all the processes of standardisation. Meanwhile, CIT and IRT approaches of item analysis should be integrated in the aspects of item development and evaluation due to their superiority in the investigation of reliability and minimising measurement errors.
|
|---|