NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Bridging the Digital Divide: A Randomized Controlled Trial Providing Internet Access to Low-Income Students
Initial registration date
May 21, 2018
May 22, 2018 2:56 PM EDT
This section is unavailable to the public. Use the button below
to request access to this information.
Teacher's College Columbia University
Other Primary Investigator(s)
Teacher's College Columbia University
Additional Trial Information
A “digital divide” exists in access to technology and internet connectivity. Policymakers argue that this divide impedes lower-income and minority students’ ability to complete schoolwork and succeed academically. Government agencies and private entities have directed large investments toward bridging the digital divide in the hopes of reducing achievement disparities across socio-economic groups. However, there have been no large-scale experiments testing whether access to the internet improves student achievement. This study identifies the causal effect of bridging the digital divide on student achievement by randomizing the provision of 10,000 smart phones, tablets and hot spots, with 4-year data plans, to high school students who lack home internet access. This project is a part of a larger initiative, the 1Million Project (1MP), which provides devices and data plans to one million high school students across the country. This study will inform policies that aim to provide digital devices and internet access to low-income students.
Bergman, Peter, Susha Roy and Elizabeth Setren. 2018. "Bridging the Digital Divide: A Randomized Controlled Trial Providing Internet Access to Low-Income Students." AEA RCT Registry. May 22.
We will randomize the provision of 10,000 devices--smartphones, tablets, and hotspots--with up to 4-year data plans to high school students who do not have home internet access to assess the effects of each type of device on academic engagement, student achievement and college enrollment outcomes. All devices can be used as hotspots to provide home internet access. Students who do not have access to a high-speed, internet-connected device for at least one hour a day will be eligible to participate in the study.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
Internet connectivity and use, academic engagement, secondary student achievement, post-secondary student achievement
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
School recruitment: We will work with large public districts that are already participating in the 1Million Project to identify a list of schools that would like to participate in the study.
Eligibility Survey: We will administer an eligibility survey to two grade levels in participating schools in order to identify which students do not have reliable home internet access and are thus eligible to participate in the 1Million Project.
Randomization: We will randomly assign either the 9th, 10th, or 11th grade of each participating school to be in the treatment group. All eligible students (those without reliable home internet access) in the treatment grade will be given the option to receive a free device and data plan through the 1Million Project subject to their consent. Students in the control group grade will receive the default services available at the school, which includes any connectivity access provided by the school or in their neighborhood.
Follow-up Survey: One year after the initial device distribution, we will administer a follow-up survey to all study participants: those in both the treatment and control groups.
Experimental Design Details
Randomization done in office by a computer.
Was the treatment clustered?
Sample size: planned number of clusters
200 grade levels (100 schools with 2 grade levels in each school)
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
100 grade levels (~10,000) students in control group
100 grade levels (~10,000) students in treatment group (receiving free devices and data plans)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
INSTITUTIONAL REVIEW BOARDS (IRBs)
Teachers College IRB
IRB Approval Date
IRB Approval Number