Empowering Parents in the Digital Age: Does Good-Practices Information Shift Parenting Behaviour? — Follow-Up Replication Study

2026-06-28

2026-09-01

Intention I — willingness to pay for a premium digital-safety app (£ slider)
Intention II — priority given to social media management vs. other parenting goals (1–5)
Donating to Charity — amount donated out of a £10 lottery prize to a digital-safety charity
Signing the Petition — whether the parent signs the real petition to Parliament, with and without the full social media ban.
Parental Beliefs about Social Media Risks— perceived prevalence of 6 social media harms (beliefs update)
Parental Worry about Their Own Child — emotional worry about child's exposure to those 6 harms
Change in practices — Supervision, rules, discussion, and change related to the treatment.
Child Well-Being I— child well-being index (SDQ)
Child Well-Being II— child well-being ONS measures (follow-up Survey 2 only)
Valuation of School and Home Policies I— estimated probability child has severe mental health issues under 4 school/home scenarios
Valuation of School and Home Policies II — estimated probability child graduates from university under the same 4 scenarios
Parent-Children Communication — estimated number of hours per week having a conversation
Parent-Children Activities — estimated number of hours spent doing activities with the child that do not involve screens
Parents Mental Health — Generalized Anxiety Questionnaire Index and Scale
Social Media Use — Parents' reported hours on social media by child

Life Satisfaction — parent life satisfaction (Cantril ladder, 0–10)

Participants are recruited via Prolific, an online panel provider, targeting UK-resident parents of at least one child aged 10–15. All recruitment, consent, randomisation, and survey administration occur within Qualtrics, with respondents accessed through the Prolific panel.

The treatment consists of exposure to a condensed video summarising the key good-practices guidance developed across the four weeks of Intervention 2.

The video is embedded directly in the Qualtrics survey flow. Treated participants see the video before answering the first set of outcome questions. Control participants proceed directly to the outcome questions without video exposure. All other survey elements are identical across conditions.

Not available

Randomisation is performed within Qualtrics using its built-in branching randomiser. Fifty per cent of respondents are assigned to treatment and fifty per cent to control. The randomisation is not stratified in advance; balance on pre-specified covariates will be verified ex post and any imbalance will be accounted for in robustness checks.

Randomisation is at the individual level, with no clustering.

No

Clustered at the individual level

Planned total completes:1600 respondents. For the follow-up survey, we anticipate re-contact rates of approximately 25%, yielding approximately 400 matched responses (200 per arm). This is consistent with attrition observed in prior waves of this project.

Planned total completes: 1600 respondents. With 50% assignment to each arm, we expect approximately 800 treated and 800 control respondents.

For the follow-up survey, we anticipate re-contact rates of approximately 25%, yielding approximately 400 matched responses (200 per arm). This is consistent with attrition observed in prior waves of this project.