Analisis kualitas soal di perguruan tinggi berbasis aplikasi TAP

Abstract

Tujuan penelitian ini adalah untuk menemukan butir instrumen yang berkualitas berdasarkan tingkat validitas, reliabilitas, tingkat kesukaran, daya beda dan pengecoh. Jenis penelitian ini adalah jenis penelitian ex post facto. Objek penelitian semua jawaban hasil tes calon mahasiswa baru tahun 2014-2016. Data dikumpulkan dengan metode Observasi, wawancara dan dokumentasi. Teknik analisis data secara kualitatif dan kuantitatif. Hasil penelitian menunjukkan bahwa untuk validitas isi ditemukan nilai validitas isi (vi) sebesar 0,42 termasuk kategori sedang. Selanjutnya tampak bahwa nilai koefisien reliabilitas instrumen sebesar 0,514. Jumlah butir soal yang sukar sebanyak 57,5%, kategori sedang sebanyak 42,5% dan tidak terdapat soal kategori mudah. Selain itu, butir soal yang memiliki daya beda sangat baik sebanyak 5%, baik sebanyak 20%, perlu revisi sebanyak 13,75%, tidak baik sebanyak 61,25%. Sedangkan option yang tidak berfungsi dengan baik pada saat dijadikan sebagai pengecoh sebanyak 5 butir soal, akan tetapi terdapat 40 butir soal yang harus direvisi karena option pengecoh malah dianggap sebagai kunci jawaban oleh peserta yang pintar.Kata kunci: validitas, reliabilitas, tingkat kesukaran, daya beda,pengecoh,TAP ANALYSIS OF EXAMINATION INSTRUMENTS QUALITY AT UNIVERSITY BASED ON TAP APPLICATIONAbstractThis study is aimed at finding the items of quality instruments based on the level of validity, reliability, difficulty, differentiation, and distraction. This study was an ex post facto research. The object of research was all the answers of enrollment test results of prospective students in 2014-2016. The data were collected by using observation, interview, and documentation methods, and the data were analyzed using both qualitative and quantitative technique. The results of the research show that the value of the content validity found is 0.42, which belongs to medium category. Furthermore, it appears that the value of the instrument reliability coefficient is 0.514. The number of the question items which belongs to difficult category is 57.5%, medium category is 42.5%, and no item is in easy category. In addition, the percentage of the question items that belong to the category of very good, good, need a little revision, and not good, in terms of the items’ differentiation, are respectively 5%, 20%, 13.75%, and 61.25%. Further, the options that do not work properly when used as distractors are 5 items, but there are 40 items that must be revised because the distractive option is even considered as the key answer by smart participants.Keywords: validity, reliability, difficulty level, differentiation, distraction, TAP

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 70 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.