| dc.description.abstract | The U-GRT is a test instrument to measure the reasoning abilities belonging to the USU agency. This study used a descriptive quantitative method using a convenience sampling technique. This study aims to analyze the IF 2 (Matrices) and DF 2 (Spatial Relations) subtests on the USU General Reasoning Test (U-GRT). Analysis was performed on 31 items of the IF 2 subtest (Matrices) and 28 items of the DF 2 subtest (Spatial Relations) using the Rasch Model which includes analysis of the reliability coefficient, analysis of the validity coefficient, and analysis of the item difficulty index. This study uses secondary data obtained from previous research by giving classical tests to 207 samples with an age range of 15 to 25 years. Based on the results of the analysis of the reliability coefficient using person reliability in the IF 2 subtest, the scores were 0.287 in set A, 0.207 in set B, 0.336 in set C, and 0.252 in set D, all sets included in the weak category. The validity coefficient using the MNSQ infit and the MNSQ outfit as a whole is in the range 0.5 – 1.5 so that 31 items are declared valid which is good in measurement. The obtained item measure values range from -1.6845 to 2.9637. The reliability coefficient using person reliability in the DF 2 subtest obtained a value of 0.142 in set A, 0.421 in set B, -0.149 in set C, and 0.364 in set D. All sets fall into the weak category. The validity coefficient using the MNSQ infit and the MNSQ outfit as a whole is in the range 0.5 – 1.5 so that 28 items are declared valid which is good in measurement. The obtained item measure values range from -0.9082 to 2.023. The results of the research on the two subtests show that the validity of the test is good and it is necessary to re-test with a larger sample in order to get better reliability results and item difficulty index. | en_US |
