Understanding Grievance Redressal Behaviour Through a Randomised, Vignette-Based Survey Experiment

Last registered on October 17, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

Understanding Grievance Redressal Behaviour Through a Randomised, Vignette-Based Survey Experiment

RCT ID

AEARCTR-0012311

Initial registration date

October 17, 2023

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

October 17, 2023, 2:43 PM EDT

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

Locations

Country

India

Region

Uttar Pradesh

Primary Investigator

Name

Aayush Agarwal

Affiliation

Centre for Social and Behaviour Change at Ashoka University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Dr. Pavan Mamidi

PI Affiliation

Director, Centre for Social and Behaviour Change, Ashoka University

Contact Investigator

PI Name

Dr. Sharon Barnhardt

PI Affiliation

Research Director, Centre for Social and Behaviour Change, Ashoka University

Contact Investigator

PI Name

Dr. Luis Artavia-Mora

PI Affiliation

Faculty Affiliate, Centre for Social and Behaviour Change, Ashoka University

Contact Investigator

PI Name

Prof. Enrique Fatas

PI Affiliation

Director, Institute for Behavioral Economics

Contact Investigator

Additional Trial Information

Status

On going

Start date

2023-09-28

End date

2023-10-25

Keywords

Behavior, Finance & Microfinance, Gender

Additional Keywords

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Through a randomised, vignette-based survey experiment, we intend to establish a causal link between a belief in karma and gendered differences in the likelihood of seeking redressal for an individual’s digital financial services (DFS)-related issues. As a secondary goal, we also investigate a causal link between the prevalence of discriminatory gender norms (normative and empirical expectations of raising complaints and redressal supply-side gender discrimination) and the likelihood of seeking redressal through formal mechanisms. Vignette group 1 is a 2x2 design vignette that varies the gender of the protagonist and their belief in karma. All survey respondents are randomly assigned to one of the four vignettes. After being exposed to their respective vignette, respondents are asked to answer a common set of outcome measures regarding their perceptions of the protagonist’s likelihood of seeking redressal through formal and informal complaint mechanisms, preferred mode of informal redressal, and supply-side discrimination. We hypothesise that after observing the treatment vignettes in vignette group 1 wherein the community believes in karma, respondents would rate the respondent the protagonist as less likely to seek redressal for their problem, as compared to those vignettes which do not mention that the community believes in karma.

External Link(s)

Registration Citation

Citation

Agarwal, Aayush et al. 2023. "Understanding Grievance Redressal Behaviour Through a Randomised, Vignette-Based Survey Experiment." AEA RCT Registry. October 17. https://doi.org/10.1257/rct.12311-1.0

Sponsors & Partners

Interventions

Intervention(s)

Intervention Start Date

2023-09-28

Intervention End Date

2023-10-25

Primary Outcomes

Primary Outcomes (end points)

Our primary outcome measures include the likelihood of raising complaints through various formal and informal means, resolving issues by oneself, inaction, support from community for seeking redressal, likelihood of treating customer care agent disrespectfully.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Our secondary outcome variables include trust in groups (customer service agents, family members, CSP agents) and gendered empirical expectations of DFS prevalence in Uttar Pradesh.

Secondary Outcomes (explanation)

Experimental Design

The primary experimental vignette follows a 2x2 design which varies belief in karma (present and absent) and gender of protagonist (male or female). All survey respondents are randomly assigned to one of the four vignettes. After being exposed to their respective vignette, respondents are asked to answer a common set of outcome measures. Later in the survey, after the primary vignette, two secondary vignettes with similar 2x2 designs follow, which test the causal impact of discriminatory gender norms (specifically gender bias in how consumers are treated by customer service agents) on women’s redressal behaviour.

Experimental Design Details

Randomization Method

Individual-level randomised through an in-built digital randomiser in SurveyCTO, the survey data collection software.

Randomization Unit

Individual

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

Sample size: planned number of observations

752

Sample size (or number of clusters) by treatment arms

188 per arm in the primary vignette

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

We expect that participants in the control group will report an average 60% likelihood of seeking redressal, while there is a negative minimum detectable effect size of 10% of the karma treatment in the question B1(a), “How likely do you think the protagonist is to raise a formal redressal complaint?”. The required sample size is of 752 subjects at 80% power and 5% significance. The karma versus no karma conditions require 376 samples each, with 50% split by gender (188 each per condition). No study has examined this question before. Previous research in honesty and cheating has found effect sizes ranging from zero to 22 percentage points in samples gathering between 900 and 2,000+ participants. Wiese (2023) finds no effect of implicit or explicit karma (writing tasks) in a coin toss task (n=1,045), but as large as 19% and 22% in a real die roll (n=2,149, p-value<1%). STATA command used for power calculation: power twoproportions 0.7 (0.45(0.05)0.65), test(chi2) power(0.8) continuity

Supporting Documents and Materials

IRB