Analisis kualitas instrumen penilaian materi keanekaragaman hayati melalui tes klasik dan Rasch model

Sari

EN
Learning evaluation functions as an indicator. Assessment is an activity that involves the interpretation of measurement data in accordance with certain criteria or rules. An effective assessment instrument must meet the requirements including validity, reliability, level of difficulty, discriminatory power, and distractor effectiveness. This study aims to compare the results of the analysis of assessment instruments between classical test theory and the Rasch model on the assessment of Biodiversity material. The study was conducted in class VII at one of the junior high schools in Sukoharjo. This study is a quantitative descriptive study with random sampling techniques. The data were analyzed using Microsoft Excel (and Winsteps. The results of the validity analysis through the classical test theory were categorized as sufficient/moderate quality (5 valid) from the Rasch model analysis obtained good question quality (6 valid). The reliability analysis of the questions through the classical test theory was 0.458 (moderate) and the Rasch model obtained a person with a weak value (0.34) and a question with a good value (0.86). The level of difficulty, according to the classical test theory and the Rasch model, was distributed into three groups. The discriminatory power of the classical theory test was divided into 5 questions with sufficient value, 4 questions with high value, and 1 question with very high value. Then, in the Rasch model there were four groups of questions that could be identified, while for the person there was only one group. For the effectiveness of the distractors in the classical theory test and the Rasch model, most of them had shown good quality while a small number of distractors did not work. The Rasch model approach is considered better because it is more objective, easy to interpret the results, flexible, and statically strong.

ID
Evaluasi pembelajaran berfungsi sebagai indikator Penilaian merupakan kegiatan yang melibatkan interpretasi data pengukuran yang sesuai dengan kriteria atau aturan-aturan tertentu. Instrumen penilaian yang efektif harus memenuhi persyaratan yang meliputi validitas, reliabilitas, tingkat kesukaran, daya pembeda, dan efektivitas distraktor. Penelitian ini bertujuan untuk membandingkan hasil analisis instrumen penilaian antara teori tes klasik dan Rasch model pada assesment materi Keanekaragaman Hayati Penelitian dilakukan pada kelas VII di salah satu SMP di  Sukoharjo. Penelitian ini berupa penelitian deskriptif kuantitatif dengan teknik sampling acak. Data dianalisis menggunakan Microsoft Excel (dan Winsteps. Hasil analisis validitas melalui teori tes klasik dikategorikan berkualitas cukup/sedang (5 valid) dari melalui analisis Rasch model diperoleh kualitas soal yang baik (6 valid). Analisis reliabilitas butir soal melalui teori tes klasik sebesar 0,458 (sedang) dan Rasch model diperoleh person bernilai lemah (0,34) dan butir soal bernilai bagus (0,86). Tingkat kesukaran, menurut teori tes klasik dan Rasch model terdistribusi dalam tiga kelompok. Daya pembeda tes teori klasik terbagi pada 5 soal bernilai cukup, 4 soal bernilai tinggi, dan 1 soal bernilai sangat tinggi.  Lalu, pada Rasch model terdapat empat kelompok butir soal yang dapat diidentifikasi, sedangkan untuk person hanya terdapat satu kelompok. Untuk efektivitas distraktor pada tes teori klasik dan Rasch model sebagian besar sudah menunjukkan kualitas baik sedangkan sebagian kecil distraktor tidak bekerja. Pendekatan Rasch model dianggap lebih baik dikarenakan lebih objektif, mudah dalam menafsirkan hasil, fleksibel, dan kuat secara statis.