Associations between extracurricular arts activities, school-based arts engagement, and subsequent externalising behaviours in the Early Childhood Longitudinal Study

Abstract

Adolescent externalising behaviours are associated with numerous long-term negative outcomes, although most research is intervention-based as opposed to risk reduction. Arts engagement has been associated with numerous beneficial factors linked to externalising behaviours, yet direct evidence linking them in longitudinal studies is lacking. Data from the Early Childhood Longitudinal Study were used, with baseline at 5th grade and outcomes measured at 8th grade. Ordinary least squares (OLS) regression was used to examine individual-level associations between extracurricular and school-based arts engagement with externalising behaviours. OLS regression was also used to examine associations between school-level arts classes and facilities with an administrator-reported index of externalising behaviours in the school. All models were adjusted for sociodemographic factors. Individual-level analyses were clustered by school. At the individual level, engaging in a greater number of extracurricular arts activities was associated with fewer externalising behaviours, although there was no association for school-based arts engagement. There were no school-level associations between arts classes or adequate arts facilities and externalising behaviours. Our results suggest extracurricular arts activities may be beneficial in reducing the risk for externalising behaviours, but the relationship is seen at an individual-level of engagement rather than based on school-level provision or facilities.

Figure 1

Individual-level associations of extracurricular and school-based arts engagement in 5th grade with externalising behaviours in 8th grade. Circles represent coefficients, which show the difference in externalising behaviours (ranging from 0 to 20) for each one-unit increase in arts engagement and are from ordinary least squares regression models in 50 imputed datasets. Horizontal lines show 95% confidence intervals. Number of extracurricular arts activities engaged in over the past 12 months ranged from 0 to 4, number of school arts classes offered 1–2 times weekly or more ranged from 0 to 6, and adequacy of school arts facilities ranged from 0 to 3. Models were adjusted for gender, first language, ethnicity, parental education, location, family structure, household income, use of food stamps, and eligibility for free/reduced school meals.

Figure 2

School-level associations of arts classes and adequacy of arts facilities in 5th grade with externalising behaviours in 8th grade. Circles represent coefficients, which show the difference in school externalising behaviour index (ranging from 0 to 21) for each one-unit increase in arts engagement and are from ordinary least squares regression models in 50 imputed datasets. Horizontal lines show 95% confidence intervals. Number of school arts classes offered 1–2 times weekly or more ranged from 0 to 6, and adequacy of school arts facilities ranged from 0 to 3. Models were adjusted for school-level covariates (school type, percentage of students from ethnic minority groups, school overcrowding, location, area safety).