Rasch Model Analysis: Validity and Reliability of Context-Based Geometry Performance Assessment Instruments Jakarta Cultural Heritage

The objective of this study is to examine the validity and reliability of performance evaluation rubrics when applied to Jakarta's cultural heritage structures, specifically focusing on the geometrical material. The Rasch Model, namely the Winstep version 3.73, will be utilized for analysis. The Rasch model offers benefits due to its ability to provide more precise estimations and identify flaws within the model. The rubric-based performance evaluation instrument is utilized to evaluate students' problem-solving abilities, with each problem being assigned a distinct rubric. A total of 381 students were sampled from two distinct schools located in North Jakarta and East Jakarta. The study's results indicate that the Cronbach's alpha value attained is 0.96, which is classified as excellent. The individual's dependability score is 0.91, which falls inside the excellent range. Similarly, the item's reliability score is 0.94, also classified as excellent. The item fit test conducted on the instrument indicates that there is no need to reject any items. The components in this performance evaluation instrument have strong validity and reliability, rendering it appropriate for evaluating student performance