Penerapan Ordinal Logistic Regression untuk Skrining Tingkat Stres Mahasiswa Berbasis Skala DASS-21

Khoirun Nisa

Abstract


Penelitian ini bertujuan untuk memodelkan tingkat keparahan kesehatan mental yang meliputi depresi, kecemasan, dan stres berdasarkan instrumen DASS-21 menggunakan pendekatan Ordinal Logistic Regression. Dataset menggunakan data kuesioner DASS-21 dengan 367 responden dari satu kelompok mahasiswa. Kemudian dilakukan proses pembersihan dan penyaringan data. Skor DASS-21 dikonversi ke dalam lima tingkat keparahan ordinal, yaitu normal, ringan, sedang, berat, dan sangat berat. Untuk menghindari circularity, setiap subskala diprediksi menggunakan dua subskala lainnya sebagai variabel independen. Hasil pemodelan menunjukkan bahwa seluruh prediktor memiliki pengaruh positif dan signifikan secara statistik terhadap peningkatan tingkat keparahan pada masing-masing subskala. Evaluasi kinerja model dilakukan menggunakan metrik yang mempertimbangkan sifat ordinal data, yaitu akurasi, mean absolute error, dan weighted Cohen kappa.  Model mencapai kinerja terbaik pada prediksi tingkat kecemasan dengan akurasi sebesar 75,5% dan MAE sebesar 0,33. evaluasi model tidak hanya difokuskan pada ketepatan klasifikasi semata tetapi juga tingkat kesesuaian ordinal yang kuat. Hasil menunjukan nilai quadratic weighted Cohen kappa masing-masing sebesar 0.850 untuk depresi, 0.883 untuk kecemasan, dan 0.905 untuk stres. Hal ini menegaskan bahwa Ordinal Logistic Regression merupakan pendekatan yang interpretatif dan sesuai untuk memodelkan tingkat keparahan kesehatan mental berbasis DASS-21.


References


Z. Hyseni Duraku, H. Davis, A. Arënliu, F. Uka, and V. Behluli, “Overcoming mental health challenges in higher education: A narrative review,” Frontiers in Psychology, pp. 1–14, 2024, doi: 10.3389/fpsyg.2024.1466060.

A. El-Zayat, S. Sultan, S. Alharthi, D. Jamal, A. Abdullah, and N. Albusati, “The relationship between perceived stress and emotional eating among university students in Saudi Arabia,” Discover Mental Health, vol. 5, no. 107, 2025, doi: 10.1007/s44192-025-00215-4.

Z. Zhang, W. Liu, C. Zhang, and L. Sun, “The relationship between family health, stress, and self-efficacy on depression among university students: A large-scale national cross-sectional study,” Frontiers in Public Health, pp. 1–9, 2025, doi: 10.3389/fpubh.2025.1625269.

J. Luo et al., “Further validating the Depression Anxiety Stress Scales-21 (DASS-21) in Chinese adolescents: Factor structure and measurement invariance,” BMC Public Health, vol. 25, no. 207, pp. 2–12, 2025, doi: 10.1186/s12889-025-23287-x.

P. Adu, T. Popoola, N. Iqbal, O. N. Medvedev, and C. R. Simpson, “Validating the Depression Anxiety Stress Scales (DASS-21) across Germany, Ghana, India, and New Zealand using Rasch methodology,” Journal of Affective Disorders, vol. 383, pp. 363–373, 2025, doi: 10.1016/j.jad.2025.04.099.

R. I. Zahra and A. D. Kalifia, “Prediction of student stress levels based on Random Forest and the DASS-21 questionnaire,” Binary Digital, vol. 8, no. 2, pp. 2114–2124, 2025, doi: 10.32877/bt.v8i2.3208.

S. A. Sulak and N. Koklu, “Analysis of Depression Anxiety Stress Scale (DASS-42) with methods of data mining,” European Journal of Education, vol. 59, no. 12778, pp. 1–18, 2024, doi: 10.1111/ejed.12778.

K. S. Srinath, K. Kiran, S. Pranavi, M. Amrutha, P. D. Shenoy, and K. R. Venugopal, “Prediction of depression, anxiety, and stress levels using DASS-42,” in 2022 IEEE 7th International Conference on Convergence Technology (I2CT), pp. 1–6, 2022, doi: 10.1109/I2CT54291.2022.9824087.

M. Richter et al., “Generalizability of clinical prediction models in mental health,” Nature, 2025, doi: 10.1038/s41380-025-02950-0.

P. F. Lovibond and S. H. Lovibond, “The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories,” Behaviour Research and Therapy, vol. 33, no. 3, pp. 335–343, 1995.

K. Xu et al., “Application of ordinal logistic regression analysis to identify the determinants of illness severity of COVID-19 in China,” Epidemiology and Infection, 2020, doi: 10.1017/S0950268820001533.

C. J. Selman, K. J. Lee, S. Y. C. Tong, M. Jones, and R. K. Mahar, “Evaluating the performance of Bayesian cumulative logistic models in randomised controlled trials: A simulation study,” arXiv preprint, pp. 1–25, 2025.

A. C. Mathew, E. Siby, A. Tom, and S. Kumar, “Applications of proportional odds ordinal logistic regression models and continuation ratio models in examining the association of physical inactivity with erectile dysfunction among type 2 diabetic patients,” Physical Activity and Nutrition, vol. 25, no. 2, pp. 30–34, 2021.

M. Ali, E. W. Andangsari, W. Syahputra, G. Paramita, and A. Kemala, “Data set of depression (DASS-21) and emotional reactivity (ERS) – Indonesian,” Mendeley Data, 2024. doi: 10.17632/sk27t97ymb.1.

T. Shadbahr et al., “The impact of imputation quality on machine learning classifiers for datasets with missing values,” Communications Medicine, pp. 1–15, 2023, doi: 10.1038/s43856-023-00356-z.

F. Pargent, F. Pfisterer, J. Thomas, and B. Bischl, “Regularized target encoding outperforms traditional methods in supervised machine learning with high-cardinality features,” Computational Statistics, vol. 37, no. 5, pp. 2671–2692, 2022, doi: 10.1007/s00180-022-01207-6.


Refbacks

  • There are currently no refbacks.