OPTIMASI VALIDITAS DAN RELIABILITAS TES PILIHAN GANDA BUATAN GURU MATA PELAJARAN IPS SD

ABSTRACT This development research aims to improve the quality of measurement results from evaluation tests of learning outcomes made by teachers on social studies subjects. Optimization activities are carried out to improve the validity and reliability of tests. This research uses level 3 R&D methods, namely researching, developing and improving existing products. The model used in the instrument development guide is the richey model with 4 stages: research, design, production and evaluation. The empirical instrument testing was carried out on the fifth grade elementary school students. The testing activities were carried out in two stages, with a total sample of 140 students. The results of this development research are valid and reliable evaluation instruments so that the measurement results can provide correct estimates of student learning outcomes. The results of the evaluation of the developed test instrument inform that 1) multiple choice tests have passed content evaluations by evaluation experts and social studies, 2) item validity analysis shows that all multiple choice test items designed have been statistically valid. Based on the results of the Pearson correlation calculation, the calculated R value is greater than the R table at alpha 0.05 which is> 0.20, meaning that the test has a good correlation between the item scores and the total test scores. 3) the value of the reliability of the test is quite high at 0.84 which means that the measurement results of the developed test are reliable