Behavioral Intervention to Increase Honest Responses Using Pledges

How to cite

Makki, Fadi. 2020. "Behavioral Intervention to Increase Honest Responses Using Pledges." AEA RCT Registry. AEARCTR-0006097. https://doi.org/10.1257/rct.6097-2.0

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

Fadi Makki*

* Primary investigator

Registry ID
AEARCTR-0006097
DOI
https://doi.org/10.1257/rct.6097-2.0
First registered
Last updated
Status
In development
Trial dates
2020-11-20 – 2021-08-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6097

Abstract

This study aims to investigate if honesty pledges are able to improve honest reporting among applicants to a volunteer program. Participants will be randomly allocated into intervention versus control conditions, where the former will receive the honesty pledge at the beginning of their form, and the latter will not. The honesty pledge will be a single statement asking applicants to click a box if they pledge to respond honestly in their application. Researchers will evaluate if there is an observable difference between the answers of the control and treatment groups, which will indicate that the honesty pledge led to a reduction in dishonest behavior.

Intervention

This randomized controlled trial investigates if commitment pledges improve honest reporting among applicants to a volunteer program in Qatar. The pledge will be a single statement asking applicants to click a box if they pledge to respond honestly in their application, and will hover frozen at the top of the application form as they progress. Any differences observed in the responses between the treatment and control groups will be attributed to the honesty pledge.

Experimental design

This experiment will be a large scale randomized controlled trial that will evaluate if the presence of an honesty pledge on an application form impacts responses. Subjects will be randomly assigned to one of two groups. Subjects in the treatment condition will receive the honesty pledge on their application. Subjects in the control group will not receive the pledge. Afterwards, both groups will answer the same questions on an application form.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
A stratified randomization strategy will be employed. Participants in the database will be randomly allocated into control and treatment groups following stratification on variables of gender and the Corruption Perceptions Index rankings of country of residence. Corruption index is clustered in four different categories: high corruption (0-29), moderately high corruption (30-49), medium corruption (50-69) and low corruption (70-89).
Clusters
8 clusters of applicants (female and male; four categories of the Corruption Perception Index rankings)
Observations
The potential sample consists of 230,000 candidates who had previously registered their interest for volunteering via an online form. Only those who are 18+ years old and who indicate being a resident in one of the countries ranked on Corruption Perceptions Index (CPI) are eligible for the study. The recruitment plan is opportunistic, i.e. without a pre-planned sample size, and researchers will involve as many people as possible from the SC’s current database. Researchers expect approximately 50% to submit an application, or 100,000 applicants (after excluding those deemed ineligible).
Arms
The recruitment plan is opportunistic, i.e. without a pre-planned sample size, and researchers will involve as many people as possible from the SC’s current database.
Primary outcomes
We will evaluate if there is an observable difference between the two groups on four main variables: (1) 'Importance of health' questionnaire item: "How much does your health matter to you?" The outcome is an ordinal variable measured on a 0-4 Likert scale where 0: Not at all, 1: A little, 2: Somewhat, 3: Quite a bit and 4: Very much. (2) Language proficiency, Arabic: 'Please indicate your proficiency in the Arabic language.' This question is measured on a 0-4 Likert Scale where 0: None, 1: Basic, 2: Intermediate, 3: Native and 4: Professional. (3) Language proficiency, English: 'Please indicate your proficiency in the English language.' This question is measured on a 0-4 Likert Scale where 0: None, 1: Basic, 2: Intermediate, 3: Native and 4: Professional. (4) Volunteer preferences, Comfort with disability: 'Are you comfortable working with people with disabilities?' - This variable will measure if subjects are comfortable working with people with disabilities. This will be measured on a binary scale (yes/no).
Secondary outcomes
We also examine differences in responses on the 'Physical activity' questionnaire item: 'Do you make sure that you walk or do other moderate physical activity for at least 30 minutes a day?' This is measured on a binary scale (yes/no).

Cited by

No citation data available for this registration yet.

Keywords