VALIDITY AND RELIABILITY OF TEST QUESTIONS TO MEASURE THE INFORMATION LITERACY SKILLS OF PROSPECTIVE TEACHER STUDENTS

Abstract

This study aims to describe the quality of information literacy instruments as seen from the aspects of validity, reliability, and difficulty level of questions. This research is a quantitative descriptive research. Data were obtained through documentation method in the form of 22 items of multiple choice questions with four answer options and tested on 224 students majoring in teacher education in a university. The data were analyzed using the classical test theory approach and Rasch model with JMetric software. Validity was assessed by content validity which relies on the use of a panel of experts to evaluate the elements of the instrument and rate them based on their relevance and representativeness to the content domain. To measure the experts' judgment, several indices have been discussed in this paper such as content validity ratio (CVR) and content validity index (CVI). From the CVR results, it is known that 20 questions are valid and 2 questions are considered invalid. From the results of the CVR value, the CVI index of 0.935 was calculated. The reliability value using the KR-20 formula through the classical test theory approach is 0.740 in the sufficient category and the reliability value in the Rasch model is 0.720 (sufficient) for Person Reliability Item Reliability of 0.97 (high). The results of the analysis of the level of difficulty based on the Rasch model show that there are 2 items with difficulty levels (10%), 7 items with moderate difficulty (35%), and 11 items with easy difficulty (55%).