Increasing Personal Data Contributions: Field Experimental Evidence from an Online Education Platform

Last registered on January 05, 2022
View Trial History

Pre-Trial

Trial Information

General Information

Title
Increasing Personal Data Contributions: Field Experimental Evidence from an Online Education Platform
RCT ID
AEARCTR-0004604
Initial registration date
September 07, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
September 09, 2019, 9:59 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
January 05, 2022, 1:29 PM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country
Germany
Region

Primary Investigator

Name
Viola Ackfeld
Affiliation
University of Cologne
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Sylvi Rzepka
PI Affiliation
University of Potsdam
Contact Investigator

Additional Trial Information

Status
Completed
Start date
2019-09-09
End date
2020-03-31
Keywords
Education
Additional Keywords
Personal data, Public good, Privacy Costs, Field Experiment
JEL code(s)
C93, D83, H41
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
In this project, we study a new type of contribution to a (impure) public good, personal data. We do so in the context of a German online education platform, which provides free online courses for participants but lacks sufficient data about them to further improve their services. We design soft behavioral interventions that highlight higher benefits and lower costs of contribution. In one treatment, we increase the salience of the public benefit of personal data contributions for the whole user community. In the second treatment, we additionally highlight data protection standards, thereby reducing potentially overestimated privacy costs. If our intervention is effective, we plan to use machine learning methods to study whether our intervention indeed increases the quality of the public online education good. For example, we may explore whether a larger base of personal data generates more precise predictions of course-related outcomes such as drop-out and grades.
External Link(s)

Registration Citation

Citation
Ackfeld, Viola and Sylvi Rzepka. 2022. "Increasing Personal Data Contributions: Field Experimental Evidence from an Online Education Platform." AEA RCT Registry. January 05. https://doi.org/10.1257/rct.4604-1.1
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
In treatment 1, we increase the salience of the public benefit of personal data contributions for the whole user community. In treatment 2, we additionally highlight data protection standards, thereby reducing potentially overestimated privacy costs.
Intervention Start Date
2019-09-09
Intervention End Date
2019-12-20

Primary Outcomes

Primary Outcomes (end points)
a) Intention to provide data: Clicked on link to profile in pop-up
b) Extensive margin: User has any entry in profile (yes/no)
c) Intensive margin: Number of entries filled
Primary Outcomes (explanation)
a) the click is recorded if forwarding page to profile is visited (0 = "profile not visited", 1 = "profile visited")
b) coded as dummy variable (0 = "no profile entries", 1 = ">0 profile entries")
c) count how many profile entries are filled. Categories: date of birth, affiliation, career status, highest degree, background in IT, professional life, position, city, gender, country. New categories: motivation, computer use at work

Secondary Outcomes

Secondary Outcomes (end points)
d) Intensive margin by privacy-sensitivity: Number of entries filled separately for privacy sensitive / insensitive categories
e) Directional intensive margin: Number of entries reduced / added separately
f) Intensive margin: Updates of profile categories that were filled previously
Secondary Outcomes (explanation)
d) Entries are rated by treatment-blind student assistants on a scale from 1 = "not at all sensitive" to 7 = "completely sensitive". We conduct a mean split based on the average rating by the student assistants.
e) as c) but additionally separate whether change is an extension or reduction of profile content
f) Dummy = 1 if entry is still filled but with different content than before (we expect only a few chnages to occur here, but information may be relevant, e.g., for updating the professional status)

Experimental Design

Experimental Design
Our treatments are implemented as simple pop-up messages embedded in the online learning platform in the second course week. We collect pre- and post-intervention profile data 5-6 days after course start (before course week 2) and 21-22 days after course start (or as close to theses dates as possible if days are not workdays), respectively. We experimentally vary the text prompting participants to review their user profile. In the first treatment, we emphasize the public benefit of sharing personal data. In the second treatment, we combine this treatment with emphasizing data protection standards.
Experimental Design Details
Randomization Method
by computer (via infrastructure of the platform)
Randomization Unit
Randomization into treatments takes place at the beginning of the second course week of each course based on platform user IDs. Using platform-wide user IDs avoids assigning participants into one treatment group in one course and to another in another course.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
All participants who are active in the second course week and/ or third week are part of our experimental sample. Due to technical reason, we only consider users who access the course by computer. It is not yet clear how many participants each course attracts. Over all courses, we will aim at reaching at least 13000 participants in total.
Sample size (or number of clusters) by treatment arms
We divide this sample evenly into one control and two treatment groups, i.e., at enrollment we aim at having at least 4300 participants in each group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
With at least 4300 participants at enrollment in each group, we have enough power to identify a 10% effect size at the extensive margin and a 5% effect at the intensive margin (number of profile categories filled out).
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
Ethics Review Board (ERB) of the Department of Management, Economics, and Social Sciences at the University of Cologne
IRB Approval Date
2019-09-04
IRB Approval Number
19023VA
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Yes
Data Collection Completion Date
Final Sample Size: Number of Clusters (Unit of Randomization)
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Abstract
Personal data increasingly serve as inputs to public goods. Like other types of contributions to public goods, personal data are likely to be underprovided. We investigate whether classical remedies to underprovision are also applicable to personal data and whether the privacy-sensitive nature of personal data must be additionally accounted for. In a randomized field experiment on a public online education platform, we prompt users to complete their profiles with personal information. Compared to a control message, we find that making public benefits salient increases the number of personal data contributions significantly. This effect is even stronger when additionally emphasizing privacy protection, especially for sensitive information. Our results further suggest that emphasis on both public benefits and privacy protection attracts personal data from a more diverse set of contributors.
Citation
Ackfeld, V., Rohloff, T., & Rzepka, S. (2021). Increasing personal data contributions for the greater public good: A field experiment on an online education platform. Behavioural Public Policy, 1-27. doi:10.1017/bpp.2021.39
URL
https://www.doi.org/10.1017/bpp.2021.39

Reports & Other Materials