Claim Missing Document
Check
Articles

Found 31 Documents
Search

Estimating The Probability of Depression in Adolescent Based on Internal Family Factors using Binary Logistic Regression and Naïve Bayes Muhammad Hasan Sidiq Kurniawan; Achmad Fauzan
Jurnal Matematika Sains dan Teknologi Vol. 25 No. 2 (2024)
Publisher : LPPM Universitas Terbuka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33830/jmst.v25i2.10312.2024

Abstract

Depression is a common mental health issue affecting approximately 300 million people globally, with severe cases potentially leading to death. Adolescents with depression are reported to have a 30-fold increased risk of suicide compared to other age groups, making early identification and intervention in this age group essential. Internal family factors play a crucial role in influencing adolescent mental health, including variables such as family type, parenting style, residence status, birth order, and parental occupation. This study aims to identify which internal family factors significantly impact the likelihood of depression and to estimate the probability of depressive episodes in adolescents. By understanding these factors, preventive measures can be better tailored to reduce adolescent depression in the future. The data utilized in this study is primary data obtained through a sampling process employing the simple random sampling method applied to first-year university students. A binary logistic regression model was employed to analyze the significance of each family-related factor. Findings indicate that parenting style and parental occupation are among the most significant factors associated with adolescent depression. The novelty of this study lies in the exploration of internal family factors that are rarely examined comprehensively in previous research, such as the effects of family type, birth order, and parental occupation. Additionally, the study adopts a dual-method approach, combining logistic regression and Naive Bayes, to provide a robust and comparative analysis of predictive accuracy. Probability estimates were conducted using both binary logistic regression and Naive Bayes methods. Results from these analyses suggest that a democratic parenting style tends to foster more stable mental health in adolescents, while adolescents with parents employed in the private sector or similar occupations face a higher likelihood of depression. Both methods demonstrated high predictive accuracy, with 96.31% for logistic regression and 96.93% for Naive Bayes.