Peer-reviewed article
Effects of requiring active parental consent on participation, sample composition, and mental health outcomes in school-based adolescent surveys: A cluster randomized trial.
Brunborg GS, Biele G, Askeland RB, Krogstad IE, Tveit OB et al. ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001This study tested whether requiring active parental consent affects how many adolescents participate in school-based surveys about mental health in Norway. Schools were randomly assigned to require either parental consent or allow adolescents to consent themselves. Active parental consent reduced participation by about a third, but did not significantly change which types of students participated or their mental health responses, except that students with immigrant backgrounds were less likely to participate.
Key points
- Active parental consent reduced participation by 32.2 percentage points compared to adolescent self-consent
- Despite lower overall participation, sample composition remained similar across groups for sex, age, and socioeconomic status
- Students with two parents born abroad were 7.1 percentage points less likely to participate under active parental consent conditions
- Mental health outcome estimates showed no statistically significant differences between consent conditions
- The findings suggest parental consent requirements do not meaningfully bias mental health estimates despite reducing participation
Limitations
The study was conducted only in Norwegian lower secondary schools in 2025, so results may not generalize to other countries, age groups, or school systems with different cultural or institutional contexts. The analysis cannot explain why immigrant families had lower participation rates or how this might affect conclusions about populations most affected by mental health challenges.
Abstract
Background Requiring active parental consent in school-based adolescent surveys may reduce participation rates and introduce response bias, particularly in studies assessing sensitive outcomes such as mental health. However, experimental evidence on its impact remains limited. Methods A cluster randomized controlled trial within 20 Norwegian lower secondary schools: 259 classes were randomized to either (1) active parental consent, or (2) adolescent self-consent. Data were collected with online questionnaires administered during school hours in the autumn of 2025. Differences in class-level participation rates were examined, as well as sample composition (sex, age, immigrant background, subjective socioeconomic status) and a broad range of mental health-related measures. The data were analyzed using fixed-effects regression models. Results Requiring active parental consent substantially reduced participation (average marginal effect: -32.2 percentage points, 95% CI -35.5, -28.8). Despite this large difference, sample composition was highly similar across conditions with respect to sex, age, and socioeconomic status. However, a lower proportion of students with two parents born abroad participated in the active parental consent condition (average marginal effect: -7.1 percentage points, 95% CI -9.4, -4.7). For mental health outcomes, no statistically significant differences were observed. Conclusions Requiring active parental consent can markedly reduce participation in school-based adolescent surveys, but it does not appear to meaningfully bias sample composition or estimates of mental health outcomes, although students with immigrant backgrounds may be underrepresented. These findings can inform ethical and methodological debates regarding adolescent autonomy, survey validity, and consent procedures in public health research.
Findings the Hub has extracted
Each claim below is labelled by its type of design and links back to this report.
- Intervention
- Requiring active parental consent for participation in school-based mental health surveys
- Population
- Adolescents in 259 classes across 20 Norwegian lower secondary schools
Key finding
Active parental consent reduced participation by 32.2 percentage points (95% CI -35.5, -28.8)