Strategies for Reducing Non-Institutional Fraud and Building Trust in a Digital Market Platform: A Behavioral Lab Experiment in Nigeria

How to cite

Putman, Daniel; King, Michael; Byrne, Shane; Jang, Chaning. 2022. "Strategies for Reducing Non-Institutional Fraud and Building Trust in a Digital Market Platform: A Behavioral Lab Experiment in Nigeria." AEA RCT Registry. AEARCTR-0009470. https://doi.org/10.1257/rct.9470-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Daniel Putman* , Michael King , Shane Byrne , Chaning Jang

* Primary investigator

Registry ID
AEARCTR-0009470
DOI
https://doi.org/10.1257/rct.9470-1.0
First registered
Last updated
Status
In development
Trial dates
2022-07-25 – 2022-10-25
Countries
Private
JEL codes
D18, G41, G53, O12
Registry record
http://www.socialscienceregistry.org/trials/9470

Abstract

The high prevalence of digital financial fraud makes it difficult for businesses to distinguish between real communications from digital service providers and fraudulent communication. This could lead to a lack of trust and usage in digital financial services. We test two strategies for preventing non-institutional fraud: an anti-fraud campaign and a technical intervention – a unique communications code – which verifies the provenance of messages sent from a digital platform. Using a behavioral laboratory experiment along with survey outcomes, we test whether anti-fraud campaigns increase the ability of MSMEs to distinguish between fraudulent and legitimate communications. Additionally, we test how confidence, trust, and usage of digital financial services respond to anti-fraud campaigns. Finally, we test that the deployment of the unique communications code is suitable for businesses using a follow-up experimental exercise.

Experimental design

Before the experiment, participants will take a short survey. Our lab partner will randomize participants into either a control group or one of three treatment groups. Each of the three treatment groups receives a variation of an educational intervention aiming at helping participants distinguish between genuine and fraudulent communication. After the educational intervention is received, participants will complete a task that requires them to determine whether an example communication is fraudulent or not. After this, a second very short survey is administered. Finally, participants will be re-randomized into two groups to receive a unique communication code. The first group will choose their own code and the other will be assigned one. After the end of the experiment, the control group will then receive the educational intervention. Finally, we will follow up with a task that assesses the feasibility of the unique communications code as a security device.

Design details

Randomization method
Participants will be randomly selected to attend sessions, which will feature a single treatment or control. The sessions will be randomly ordered within blocks. All randomization will take place via computer.
Randomization unit
Individual
Clusters
750 Micro, Small, and Medium Enterprises (MSMEs)
Observations
750 MSMEs
Arms
750 participants will be evenly allocated across 4 cells (An average of 187.5 participants per cell - including a control group and three treatment groups).
Primary outcomes
Accurate identification of fraudulent scenarios, Confidence in identification ability, Reported likeliness of using digital financial services in the future, trust in digital financial services

Cited by

No citation data available for this registration yet.

Keywords