Dukungan Keluarga sebagai Prediktor Kecerdasan Emosional Siswa Sekolah Menengah

Nurhaliza, Nurhaliza
Arsyad, Muhammad
Makaria, Eklys Cheseda

Nusantara: Jurnal Pendidikan Indonesia
Vol. 6 No. 2 (2026)

Purpose – Emotional intelligence is an important aspect of adolescent development that influences how students manage emotions, interact socially, and respond to challenges. However, emotional instability is still observed among secondary school students, indicating the need to examine contextual factors, particularly family support. This study aims to analyze the role of family support as a predictor of students’ emotional intelligence. Methods – This study employed a cross-sectional quantitative survey design. The sample consisted of 551 students from SMAN 1 Gambut and SMKN 1 Gambut selected using non-probability quota sampling. Data was collected using Likert-scale questionnaires measuring family support based on Friedman’s theory and emotional intelligence based on the Wong and Law Emotional Intelligence Scale. Data analysis was conducted using descriptive statistics and non-linear quadratic polynomial regression with the assistance of IBM SPSS Statistics version 30. Findings – The results showed that family support among students was generally in the high category. The regression model was statistically significant, indicating that family support serves as a meaningful predictor of students’ emotional intelligence. The coefficient of determination (R²) was 0.116, indicating that family support is associated with 11.6% of the variance in emotional intelligence, which can be interpreted as a small to moderate effect size.Research Implications – These findings suggest that support is one of several factors related to students’ emotional intelligence, although its role appears relatively limited. Practically, schools and parents can collaborate to support students’ emotional development through consistent emotional support, appreciation, and effective communication at home. A limitation of this study is that it was conducted in only two schools using non-probability sampling, which may limit generalizability. Future studies are recommended to involve more diverse samples and examine additional related factors.