Dominasi Computational Thinking atas Motivasi dalam Kemampuan Pemrograman C Siswa SMA

Abstract

Kemampuan pemrograman tidak hanya ditentukan oleh penguasaan sintaks, tetapi juga oleh cara siswa memformulasikan masalah dan mempertahankan usaha ketika menghadapi kesalahan program. Penelitian ini bertujuan menguji kontribusi computational thinking skill dan motivasi belajar terhadap kemampuan pemrograman Bahasa C siswa SMA, sekaligus mengidentifikasi prediktor yang benar-benar dominan. Penelitian menggunakan desain kuantitatif korelasional ex post facto pada 138 siswa kelas XI SMA Negeri 1 Kubung yang dipilih secara proportionate stratified random sampling dari populasi 203 siswa. Computational thinking diukur dengan 13 soal objektif, motivasi dengan 22 pernyataan skala Likert, dan kemampuan pemrograman dengan 18 soal objektif; koefisien reliabilitas masing-masing sebesar 0,884, 0,852, dan 0,897. Hasil menunjukkan computational thinking berkorelasi sangat kuat dengan kemampuan pemrograman (r = 0,910; p < 0,001; R² = 0,829), sedangkan motivasi belajar tidak berkorelasi signifikan (r = -0,107; p = 0,210). Model simultan signifikan, F(2,135) = 331,81; p < 0,001, dengan R² = 0,831, tetapi penambahan motivasi hanya meningkatkan daya jelas model sebesar 0,2% (ΔR² = 0,002) dan koefisiennya tetap tidak signifikan (β = -0,044; p = 0,212). Kebaruan penelitian terletak pada bukti bahwa signifikansi model gabungan hampir sepenuhnya digerakkan oleh computational thinking, bukan motivasi. Temuan ini menegaskan perlunya pembelajaran Bahasa C yang memprioritaskan dekomposisi, pengenalan pola, abstraksi, perancangan algoritma, tracing, dan debugging, tanpa menafsirkan motivasi sebagai penyebab langsung capaian pemrograman.

Programming ability depends not only on syntax mastery but also on how students formulate problems and sustain effort when code fails. This study examined the contributions of computational thinking skills and learning motivation to high school students’ C programming ability and identified the genuinely dominant predictor. A quantitative ex post facto correlational design involved 138 eleventh-grade students from SMA Negeri 1 Kubung, proportionally stratified and randomly selected from a population of 203. Computational thinking was measured using 13 objective items, motivation using 22 Likert statements, and programming ability using 18 objective items; reliability coefficients were 0.884, 0.852, and 0.897, respectively. Computational thinking showed a very strong correlation with programming ability (r = 0.910, p < 0.001, R² = 0.829), whereas learning motivation was not significantly related (r = -0.107, p = 0.210). The simultaneous model was significant, F(2,135) = 331.81, p < 0.001, with R² = 0.831; however, adding motivation increased explained variance by only 0.2% (ΔR² = 0.002), and its coefficient remained nonsignificant (β = -0.044, p = 0.212). The study’s novelty is evidence that the significant joint model was driven almost entirely by computational thinking rather than motivation. C programming instruction should therefore prioritize decomposition, pattern recognition, abstraction, algorithm design, tracing, and debugging while avoiding causal interpretations from correlational data.