Using the Rasch’s Partial Credit Model to Analyze the Quality of an Essay Math Test

Abstract

The instrument in the form of an essay math test is one of the tools to measure the progress of the learning process or learning outcomes. The essay math test needs to be applied in education because students are trained to be independent, creative, innovative, and improve literacy skills. A good measurement model facilitates as much information as possible, for example, qualitative information, the accuracy of the interpretation of the instrument according to its function, provides a linear measure, can overcome missing data, and find data that is incorrect (misfits) or uncommon (outliers). Rasch's Partial Credit Model can identify the quality of the essay math test more accurately. The reliability value in Rasch modeling is indicated by the value of individual separation (person separation) and item separation. Output summary statistics to get information on the person and item reliability and Cronbach alpha. Analysis with the Rasch model produces a statistical analysis of suitability (fit statistics) which provides information that the data obtained ideally illustrates that people who have a high ability provide patterns of answers to items according to their level of difficulty. To analyze the quality of an essay math test using Rasch's Partial Credit Model, ministep software can be used.