Claim Missing Document
Check
Articles

Found 3 Documents
Search

Panelist Analysis of Higher Education Governance Instruments Based on Good University Governance Principles Using Many Facet Rasch Measurements Iriyadi, Deni; Muhajir; Nikmatullah, Cecep; hartati, Hevriana
JISAE: Journal of Indonesian Student Assessment and Evaluation Vol 9 No 2 (2023): JISAE (Journal of Indonesian Student Assessment and Evaluation)
Publisher : LPPM Universitas Negeri Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21009/jisae.v9i2.39043

Abstract

This research aims to develop instruments related to measuring higher education governance so that it can fulfill the principles of good governance. This research used 3 panelists to measure the content validity of the instrument developed. The criteria for selecting experts/validators are adjusted to the expertise and needs of the instrument development being carried out. The study on testing the content validity of panelists uses Rasch Model analysis with the Facets application which aims to test subjectively rated statement items by looking at several aspects including unidimensional assumptions, testing model fit (item fit), and rater/expert consistency. The results of the content validity analysis carried out show that the unidimensional criteria (the raw variance explained by measure value is more than 20%), the item fit value (the outfit value is in the range of 0.5-1.5), and the rater consistency assessment show that all raters/experts meet the fitness criteria (outfit and infit value categories are in the range 0.5-1.5). The results of the content validity analysis carried out show that the quality of the instrument developed is suitable for use to measure higher education governance.
Comparison Equating Method Based on Sample Size Iriyadi, M.Pd., Dr. Deni; Hartati, Hevriana; Rustam, Ahmad
Almufi Journal of Measurement, Assessment, and Evaluation Education Vol 3 No 2: Desember (2023)
Publisher : Yayasan Almubarak Fil Ilmi

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The purpose of this research is to compare the linear technique to the equipercentile method in terms of the variance of equalization scores using a sample size of one hundred and three different test lengths of twenty-five, twenty-nine, and thirty minutes respectively. In this particular investigation, the value of the variance of the equalization score serves as the dependent variable, while the equalization technique and the duration of the exam serve as the independent variables. A learning outcome exam for physics was used as the instrument of assessment in this particular research project. The participants were high school students. The replies of all of the people who took part in the study and worked on the several test sets that the researcher had put together constituted the study's population. The research sample consisted of the replies provided by the students for each package, for a total of one hundred responses. These responses were selected using a random selection method that included returns, and the process was repeated thirty times. We put the study hypothesis through its paces by employing comparative hypothesis testing in conjunction with t-test analysis. The findings indicated that the variance values of the equalized results obtained via the application of the linear technique were lower than the variance values obtained through the application of the equipercentile approach for test durations of 20, 25, and 30 when there was a sample size of 100.
Equating by Using Circle Equation Approach: Applied Mathematics Formula for Prevent Discrimination Iriyadi, Deni; Rustam, Ahmad; Hartati, Hevriana
JTAM (Jurnal Teori dan Aplikasi Matematika) Vol 8, No 3 (2024): July
Publisher : Universitas Muhammadiyah Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31764/jtam.v8i3.22195

Abstract

This study aims to determine the accuracy of the equating method that uses a circle equation approach in terms of its circular arc (Simplified Circle Arc). This research uses 2015 National Examination data from two questions packages. Using the number of preliminary samples as many as 2135 on the X and 2271 test devices on the Y test. After doing a Rasch analysis using a Mean Square Outfit (MNSQ), the data was acquired and analyzed. Following this, replication was performed up to a maximum of 50 times for each kind of data distribution. For each replication, up to a maximum of 50 respondents were selected from the original data set to be used as data for score equalization. The Root Mean Square Error (RMSE) statistic is then used to analyze the outcomes of the equating score. The results showed that the average RMSE group that has the same distribution will provide a lower RMSE value compared to groups that have different data distributions. The low average RMSE value indicates the accuracy of the equal of the scores performed. Thus, the use of the SCA method is highly recommended to equalize scores, especially in small samples in classes at school to prevent discrimination in grading.