Abstract

Berdasarkan hasil wawancara dengan Agus Santoso, S.Pd, diperoleh informasi bahwa soal UAS gasal mata pelajaran IPS  kelas III SD Negeri se-Gugus Pangeran Diponegoro tahun ajaran 2014/2015 diujikan tanpa melalui tahapan analisis butir soal. Oleh karena itu, tujuan penelitian ini yaitu untuk mengetahui kualitas butir soal dan pelaksanaan tes pada soal UAS tersebut. Penelitian ini menggunakan metode expost facto. Sampel dalam penelitian ini yaitu sebanyak 202. Hasil penelitian menunjukkan bahwa, soal yang dianalisis ditinjau dari aspek materi, konstruksi, dan bahasa berkategori sangat tinggi. Jenjang ranah kognitifnya yaitu terdapat 7 (28%) soal berkategori C1, 17 (68%) soal berkategori C2, dan 1 (4%) soal berkategori C3. Aspek validitasnya yaitu terdapat 2 (8%) soal berkategori sangat signifikan, 8 (32%) soal berkategori signifikan, dan 15 (60%) soal berkategori tidak signifikan. Aspek reliabilitasnya memiliki kriteria rendah. Aspek  tingkat  kesukarannya  yaitu terdapat 17 (68%) soal  berkategori mudah, 7 (28%) soal berkategori sedang, dan 1 (4%) soal berkategori sukar. Aspek daya pembedanya yaitu terdapat 7 (28%) soal berkategori baik, 7 (28%) soal berkategori cukup, 10 (40%) soal berkategori jelek, dan 1 (4%) soal berkategori jelek sekali. Aspek efektivitas pengecohnya yaitu terdapat 11 (44%) soal berkategori efektif dan 14 (56%) soal berkategori tidak efektif. Pelaksanaan tes tergolong cukup baik.

 

Based on the interview with Agus Santoso, S.Pd, obtained information that final exam social studies class III Elemantary School in the District Pangeran Diponegoro 2014/2015 academic year without going through the stages of analysis item. Therefore, the purpose of this study is to determine the quality of the items and the implementation of tests on about the final exam. This study uses ex post facto. The sample in the study as many as 202. The results show that, in a matter which analyzed in terms of the aspect of material, construction, and the language is very high category. Levels of cognitive domains there are 7 (28%) categorized as C1, 17(68%) categorized as C2, and 1 (4%) categorized as C3. The aspects of validity there are 2 (8%) categorized as very significant, 8 (32%) categorized as significant, and 15 (60%) categorized as not significant. Aspects reliability is low criteria. The aspects of difficulty index there are 17 (68%) categorized as easy, 7 (28%) categorized as medium, and 1 (4%) categorized as difficult. The aspects of discrimination power there are 7 (28%) the good category, 7 (28%) the enough category, 10 (40%) the ugly category, and 1 (4%) the very ugly category. The aspects of effectiveness swizlling there are 11 (44%) categorized as effective and 14 (56%) categized as ineffective. Implementation of the test is good enaugh.