ITEM ANALYSIS FOR TEST TO EXAMINE THE EFFECT OF E-MODULE ON THE ACADEMICPERFORMANCE OF 7TH CLASS SCIENCE STUDENTS IN ISLAMABAD

Abstract

The main aim of this research was to investigate quantitative characteristics ( difficulty level, discrimination index and distractor efficiency) of research tool regarding the first four units of 7th class general science to be used for pretest on the research on “effect of e-module on the academic performance of 7th class general science students in Islamabad”. The researchers got the data from 210 randomly selected students both from urban and rural area schools of Islamabad Capital Territory. Researchers used 33% students from high achievers and 33% from the lower achievers by taking 70 students from HE and 70 students from LEs. In difficulty level (Diff I) and discrimination index (DI) of total 140 students were taken both from group of high marks from top corner and group of lower marks for the bottom corner. Distractor analysis (DE) was done on the data of 210 students. Descriptive statistics was used to analyses the data. On the basis of findings from the analyses of difficulty level and discrimination index that two test items (16 & 17) were eliminated from the tool due to very low values of Diff. I ( <29) and DI values (><20). Four items were difficult (revised and improved) and 17 items were moderately difficult retained and sustained. There was only one item which was easy. Discrimination index (DI) of 22 items was discriminatory and retained after minor improvements. Distractor analysis (DE) determines the functionality of distractors. If some distractors are not selected by the students/respondents they are considered for revisions. In this research only one distractor was found to have DE value less than 5%, and three distractors with DE value less than 7%. It was recommended to eliminate and change distractor with value of DE less than 5% and revise and improve distractors with DE value less than 7%. It was also found that except two test items all items were valid and reliable. It was further recommended that item analyses is very useful technique to find the reliability and validity of a tool .>