Analisis Kualitas Butir Soal Tes Hasil Belajar Materi Metamorfosis di Sekolah Dasar Menggunakan ITEMAN

Nadiya Solastika Andhini (Universitas Sebelas Maret)
Sri Yamtinah (Universitas Sebelas Maret)
Joko Daryanto (Universitas Sebelas Maret)

MIMBAR PGSD FLOBAMORATA
Vol. 4 No. 4 (2026): MIMBAR PGSD FLOBAMORATA

Abstrak: Penelitian ini bertujuan untuk menganalisis kualitas butir soal tes hasil belajar pada materi metamorfosis di sekolah dasar. Metode yang digunakan adalah deskriptif kuantitatif dengan melibatkan 105 peserta didik dari dua sekolah di Kecamatan Magetan. Teknik pengumpulan data dilakukan melalui tes dan nontes, sedangkan analisis data menggunakan pendekatan teori tes klasik berbantuan program MicroCat ITEMAN versi 4.3. Aspek yang dianalisis meliputi validitas isi, reliabilitas, tingkat kesukaran, daya pembeda, serta efektivitas pengecoh. Hasil penelitian menunjukkan bahwa seluruh butir soal telah memenuhi validitas isi berdasarkan penilaian ahli. Nilai reliabilitas tergolong tinggi dengan koefisien alpha sebesar 0,854 pada paket A dan 0,884 pada paket B. Tingkat kesukaran soal bervariasi dari mudah, sedang, hingga sukar. Daya pembeda sebagian besar berada pada kategori cukup hingga baik, meskipun masih terdapat beberapa butir yang perlu diperbaiki. Selain itu, ditemukan beberapa pengecoh yang belum berfungsi secara optimal. Abstract: This study aims to analyze the quality of achievement test items on the topic of metamorphosis in elementary schools. A quantitative descriptive design was utilized, involving 105 third-grade students from two schools in Magetan District. Data collection employed test and non-test techniques, and classical test theory analysis was conducted using the MicroCat ITEMAN software version 4.3 to evaluate content validity, reliability, difficulty levels, discriminating power, and distractor effectiveness. The results revealed that all items achieved content validity based on expert judgment. Reliability coefficients were high, scoring 0.854 for Package A and 0.884 for Package B. Item difficulty varied proportionally across easy, medium, and hard categories, and most discriminating power indices ranged from fair to good. However, several items still exhibited low discriminating power and ineffective distractors. This research impacts educators by demonstrating how item analysis programs can guide targeted test improvements. In conclusion, the test instrument is generally sound and feasible, though minor revisions are necessary to maximize its measurement effectiveness.