Explaining Socioeconomic Differences in Support for Smartphone Bans in Schools

Subjects will be randomly exposed to one of 5 conditions:

1) Control
2) Information treatment emphasizing classroom externalities
3) Information treatment emphasizing policy alternatives
4) Information emphasizing childrens‘ self-control
5) Information emphasizing greater affectedness of low-SES students

2026-05-21

2026-06-30

(Question: To what extent do you agree with the following statement?)
• Support for Smartphone bans in schools: “There should be a comprehensive ban on private smartphone use in all schools.”

Secondary dependent variables (different school levels):
(Question: “To what extent do you agree with the following statements about the private use of smartphones at schools?”)
• Support for smartphone bans in primary schools: “There should be a comprehensive ban on private smartphone use in all primary schools.”
• Support for smartphone bans in secondary school up to 10th grade: “There should be a comprehensive ban on private smartphone use in all secondary schools up to 10th grade.”
• Support for smartphone bans in secondary schools from 11th grade upwards: “There should be a comprehensive ban on private smartphone use in all secondary schools from 11th grade upwards.”

Secondary dependent variables (posterior beliefs)
Question: “To what extent do you agree with the following statements about the private use of smartphones at schools?”
• Belief about disruptive effects of smartphones: “Smartphones can disrupt lessons.”
• Belief about Externalities (impact on others): “Even if only a few students use smartphones, this can disrupt other students’ learning.”
• Belief about Availability of alternative policies: “There are better ways than a comprehensive ban on smartphones at schools to reduce disruptions to lessons caused by smartphones.”
• Belief about Students’ self-control ability: “Many children and adolescents are competent in using smartphones and therefore do not benefit from a comprehensive ban on smartphones at schools.”
• Belief about Effectiveness of bans: “Smartphone bans in schools are effective in preventing private smartphone use during lessons.”
All outcomes are measured on a 5-point Likert scale (1 = Strongly disagree; 2 = Somewhat disagree; 3 = Neither; 4 = Somewhat agree; 5 = Strongly agree; 9 = Don’t know).
All outcomes will be binarized: 1 if response ∈ {4, 5}; 0 if response ∈ {1, 2, 3, 9}
“Don’t know” responses are coded as 0 to reflect absence of expressed support.
Secondary dependent variables (posterior beliefs about inequality)
(Question: “To what extent do you agree with the following statements about the private use of smartphones at schools?”)
A) Belief about effects of smartphones on low-SES students: “The academic performance of socially disadvantaged children and adolescents worsens due to smartphone use at school.”
B) “The academic performance of socially advantaged children and adolescents worsens due to smartphone use at school.”
C) “Socially disadvantaged children and adolescents benefit from a smartphone ban at schools.”
D) “Socially advantaged children and adolescents benefit from a smartphone ban at schools.”
All outcomes are measured on a 5-point Likert scale (1 = Strongly disagree; 2 = Somewhat disagree; 3 = Neither; 4 = Somewhat agree; 5 = Strongly agree; 9 = Don’t know).
To measure the social gradient in perceived problem affectedness and perceived policy benefits, the following differences will serve as dependent variables
A)-B) and
C)-D)
“Don’t know” responses are coded as missing.

This study investigates the determinants of public support for paternalistic policies targeting children, focusing on restrictions of private smartphone use in schools. We document differences in preferences for smartphone bans across several socio-economic characteristics and investigate the role of beliefs, values and own smartphone use and sophistication in explaining these differences. In particular, we leverage a survey experiment to examine the importance of beliefs about children’s self-control, externalities imposed on peers, the role of private forces to regulate behaviour, and perceived socioeconomic inequalities in exposure and impact and document differences in updating across the studied SES groups.

Not available

randomization done in survey software

individual

No

approx. 3600

approx. 3600

approx. 720 in each of the 5 treatments