| dc.description.abstract | This study aims to evaluate the aitem subtests of IN 1 (Number Series) and IN 2 (Unfitting Number) on the UGRT. Evaluation of a test must be carried out routinely to check the validity and reliability of the test. This research is psychometric research and uses sampling, namely convenience sampling. Analysis on aitem sets A, B, C and D on IN 1 and IN 2 using Rasch modeling which includes reliability coefficient analysis, validity coefficient analysis (fit), aitem difficulty index analysis and unidimensionality. Based on the results of the analysis, the Cronbach alpha value and person reliability in IN 1 and IN 2 are categorized as unsatisfactory reliability coefficient values for all sets but on all reliability items in IN 1 and IN 2 get satisfactory reliability values. While the validity coefficient obtained information that there are only 4 items that do not meet Rasch criteria. The unidimensionality value in the entire set has met the unidimensionsl criteria of > 20%. The results of this study indicate that the validity of the dominant test is good, but it still has to be re-evaluated, especially the items on IN 1 set B aitem 6, set D aitem 6 and also IN 2 set B aitem 5. | en_US |
