Estimation of Test Item Parameters with Polytomous Item Response Using Partial Credit Model (PCM)
Suciati Suciati, Suciati, Sudji Munadi, Sugiman, Sugiman · Advances in Social Science, Education and Humanities Research/Advances in social science, education and humanities research · 2021
Assessment activities would provide useful information if instruments developed and used have good quality.One of the standards for a quality test instrument is having acceptable item parameters.This research aims to analyze the quality of the test instrument in terms of the item parameters' quality with difficulty level on polytomous scoring.This research is explorative research by using a quantitative approach.The data used in this research is the test participant responses data to the items of numeracy.The participants have involved as many as 273 students of class VIII SMP in DI.Yogyakarta Province.The polytomous item response theory model used in this research is the Partial Credit Model (PCM).The estimation of item parameters is carried out using the R Program.The results showed that of the ten test items analyzed, only six items had the right criteria based on the item location parameters (level of difficulty of the stage) and Item Characteristic Curve (ICC).The information function value obtained is in the range -3.00 to 3.00, representing the instrument developed is suitable for test participant with abilities from -3.00 to 3.00 or from low abilities to high abilities.