Digital Financial Education and Consumer Credit Behaviour: Evidence from a Multi-Arm Randomized Controlled Trial

How to cite

Girolamo, Amalia Di; Atkinson, Adele. 2026. "Digital Financial Education and Consumer Credit Behaviour: Evidence from a Multi-Arm Randomized Controlled Trial." AEA RCT Registry. AEARCTR-0018100. https://doi.org/10.1257/rct.18100-2.0

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

Amalia Di Girolamo* , Adele Atkinson

* Primary investigator

Registry ID
AEARCTR-0018100
DOI
https://doi.org/10.1257/rct.18100-2.0
First registered
Last updated
Status
On going
Trial dates
2026-03-08 – 2026-12-31
Countries
Private
JEL codes
G53, D14, G51, C93
Secondary IDs
G51, D91
Registry record
http://www.socialscienceregistry.org/trials/18100

Abstract

Financial education is widely promoted as a tool to improve household financial outcomes, yet rigorous evidence on whether it translates into measurable improvements in credit behaviour remains limited — particularly for financially vulnerable populations and digital delivery channels. This study evaluates the causal impact of targeted financial education on credit outcomes, financial literacy, and financial wellbeing among users of a financial technology platform. The intervention is implemented as a randomized controlled trial in which eligible FinTech app users are invited to participate. Baseline data on financial capability, financial wellbeing, and socio-demographic characteristics are collected through a survey and combined with administrative credit records.

Approximately 9,000 participants are randomly assigned to one of four groups: three treatment groups receiving financial education through different digital delivery channels (in-app video learning, in-app chat, or virtual workshops) and a control group with standard app access only. Stratified block randomization based on age, education, and baseline credit score ensures balance across experimental arms.

The primary outcomes are changes in credit score and access to affordable credit. Secondary outcomes include: financial literacy, financial wellbeing and financial resilience. These outcomes are measured using administrative credit data and follow-up surveys conducted approximately three months after the intervention. The analysis estimates intent-to-treat and treatment-on-the-treated effects, and enables comparison of the relative effectiveness of alternative delivery channels.

Intervention

The intervention consists of targeted financial education delivered through a UK-based consumer credit score application to users. Participants assigned to the treatment arms receive standardized educational content covering credit management, budgeting, and financial wellbeing, developed by a national financial education charity.
The content is delivered through one of three channels: (i) short, on-demand video modules embedded within the application; (ii) a structured in-app chat interface providing guided responses on the same topics; or (iii) a live, one-hour virtual webinar led by a trained facilitator with opportunities for interaction. The three delivery channels are designed to provide comparable content and require similar time commitments.
Participants in the control group retain standard access to the application and do not receive additional educational content.

Experimental design

This study is a four-arm randomized controlled trial conducted with users of a UK-based consumer credit score application. The objective is to evaluate the causal impact of targeted financial education on credit outcomes, financial literacy, and financial wellbeing among users experiencing short-term financial difficulty.

Participants are recruited through an in-app invitation. Those who consent are included in the study and complete a baseline survey capturing financial capability, wellbeing, and socio-demographic characteristics.

Participants are randomly assigned in equal proportions to one of four groups using stratified block randomization based on age, education, and baseline credit score. The groups are: (i) a control group with standard app access; (ii) financial education delivered via in-app videos; (iii) financial education delivered via a structured in-app chat interface; and (iv) financial education delivered through a live virtual workshop.

The educational content is standardized across treatment arms and covers credit management, budgeting, and financial wellbeing. Delivery channels are designed to provide comparable content and time commitment.

Participants complete a follow-up survey approximately three months after the intervention. Outcomes are measured using survey data and administrative credit records. Engagement with the intervention is tracked to define compliance for secondary analyses.

Design details

Randomization method
randomization done in office by a computer.
Randomization unit
Individual app users.
Clusters
0
Observations
9000 app users
Arms
2000 individuals
Primary outcomes
• Individual Credit Score • Access to credit
Secondary outcomes
Secondary Outcomes • Financial Literacy • Financial wellbeing • Financial resilience • Intervention engagement: channel assigned, time spent on video/chat, number of visits, onward navigation behaviour. • Self-reported actions taken to improve credit score. • Demand for credit

Cited by

No citation data available for this registration yet.

Keywords