# Longitudinal changes in Stress Responses, Stressors, and Social Supports among adolescent students: A latent growth curve modeling approach

## TL;DR

This study tracks how stress responses, stressors, and social support change over three years in high school students, finding that early stress and lack of social support predict higher stress later.

## Contribution

The study introduces a longitudinal analysis using latent growth curve modeling to assess stress dynamics in adolescents via a web-based system.

## Key findings

- Higher initial stressors and their increase over time significantly predict higher stress responses in third year of high school.
- Initial social support, especially from friends, significantly predicts lower stress responses in third year.
- Friend support specifically reduces depression/anxiety, irritability/anger, and helplessness in later years.

## Abstract

To examine longitudinal changes in stress responses (SR), stressors (ST), and social supports (SS) among high school students using latent growth curve modeling.

This study investigated patterns of stress responses (SR), stressors (ST), and social supports (SS) among high school students over a three‐year period, using data from a web‐based stress check system.

After ethical review approval, informed consent was obtained from both students and their parents. A longitudinal analysis was conducted using data collected from high school students (n = 605) over three consecutive years, from their first year (10th grade) to their third year (12th grade). High school students were assessed using the Public Health Research Foundation‐Type Stress Inventory via the web each year for three years. This inventory consists of items assessing SR, ST, and SS. Data were analyzed using latent growth curve modeling (LGCM).

In the LGCM, both the intercept (first‐year level) and slope of ST significantly predicted higher SR in the third year (β = 0.591 and 0.916, respectively; p < 0.001). By contrast, the intercept of SS (first‐year level) significantly predicted lower SR in the third year (β = –0.279, p < 0.001). Within the SS subscales, the intercept of support from friends significantly predicted reductions in all three SR subscales in the third year: depression/anxiety (β = –0.248, p < 0.05), irritability/anger (β = –0.254, p < 0.01), and helplessness (β = –0.223, p < 0.05).

Our findings suggest the utility of a web‐based stress check administered in the first year of high school to assess SR, ST, and SS.