The objective of the research is to study the political participation of (international) migrants and we are particularly interested in the impact of messaging on voter registration and turnout. The GOTV literature has focused on testing rational choice and social psychological theories and finds that social pressure, pride, and shaming messages have large effects. However, much of this literature is US- and Europe-centric. On the other hand, there's an active literature that has studied the motivations for turnout in the Indian context but these haven’t been experimentally evaluated.
We hope that our study will lie at the intersection of these two literatures. We wish to empirically examine the causal impact of an informational campaign that not only emphasizes the messages noted in the academic literature but to also test for the effect of patriotism (by invoking the image of the national flag) on participation.
External Link(s)
Citation
Dar, Aaditya and Rahul Verma. 2016. "An online voter registration experiment with Indian citizens living abroad." AEA RCT Registry. November 22. https://doi.org/10.1257/rct.1801-1.0.
The main intervention is to randomly prime randomly selected individuals to register to vote along the following themes:
1) patriotism (positive - love) 2) patriotism (negative - guilt)
3) civic duty 4) government helps in times of need
5) generic message (placebo)
6) CONTROL - no message
Intervention Start Date
2016-11-21
Intervention End Date
2017-01-08
Primary Outcomes (end points)
A) Voter registration
B) Voter engagement C) A proxy for voter turnout
Primary Outcomes (explanation)
Registration - % of respondents who click on “complete Form 6A now” button and % of those who actually register. In India, registration records (name, age and sex) are public and this information will be tracked and used as the primary outcome variable.
Engagement - % of respondents who complete online and offline activities. Note: the engagement index will serve as the intermediate variable and shall be derived by creating a cumulative score based on respondents' action/initiatives.
Turnout proxy - % of respondents who submit an “e-ballot” (An e-ballot is a blank postal ballot paper which is emailed to the voter, who then has to fill it and return it by post to the election management body.)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
The intervention will have 5 treatment arms and one control group that receives no informational messages. A brief baseline survey will be conducted and the following information areas will be covered:
● Knowledge
◦ Awareness of the registration process in general ◦ Awareness of registration process as an overseas elector
● Attitude
◦ Likelihood of participation in elections
● Practice
◦ Having a voter ID card
◦ Whether voted in elections before
◦ Reasons for not voting
● Basic demographic information (country of residence, age, sex, occupation and education)
Experimental Design Details
Randomization Method
Randomization done by a computer
Randomization Unit
Individual
Was the treatment clustered?
No
Sample size: planned number of clusters
NA
Sample size: planned number of observations
The targeted sample size is: 1000 x 6 = 6,000
Sample size (or number of clusters) by treatment arms
1,000 individuals in each of the 5 treatment arms and 1,000 individuals in control.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)