Validity and Reliability Analysis in Supporting Mastery Learning STEM Model

Abstract

Research with invalid and unreliable instruments will greatly affect the results of the study. The lack of complete exposure of the results of data analysis and the validity and reliability analysis of the daily tests have not been found, the researchers conducted a study that aims to describe the results of the validity and reliability of colloidal material daily test items in supporting the completeness of learning the STEM model. This type of quantitative descriptive research. The researcher wants to describe the validity of the chemistry questions on the colloidal material daily test instrument in terms of validity, reliability, level of difficulty, and discriminating power. The content validity test results were obtained that the instrument was valid. The instrument questions were 18 multiple-choice items and 6 description items, all valid and reliable results were obtained. Multiple-choice questions only have questions 6 and 14 which have good distinguishing power, while questions 1, 2, 3, 10, 12, 13, 15, 16, 17, and 18 have good distinguishing power. Explanation questions from numbers 1 to 6 have good discriminating power. Based on the criteria for the level of difficulty, the results of multiple-choice questions 6, 7, 8, 10, 13, and 17 were questions with a moderate level of difficulty, while questions numbered 1, 2, 3, 4, 5, 12, 14, 15, 16, and 18. Instruments that can be used at least have valid and reliable criteria.