The Welfare Effects of Woodwork Effects

How to cite

Anders, Jenna; Rafkin, Charlie. 2020. "The Welfare Effects of Woodwork Effects." AEA RCT Registry. AEARCTR-0005566. https://doi.org/10.1257/rct.5566-1.0

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

Jenna Anders* , Charlie Rafkin

* Primary investigator

Registry ID
AEARCTR-0005566
DOI
https://doi.org/10.1257/rct.5566-1.0
First registered
Last updated
Status
In development
Trial dates
2020-03-18 – 2020-05-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5566

Abstract

This experiment tests whether beliefs about means-testing of a social program affects individuals' attitudes towards the program.

Intervention

The experiment will be embedded in a survey about the stigma around SNAP. Respondents will be given information intended to change their beliefs about the share of Americans who are eligible for SNAP.

Experimental design

The experiment has a 2x2 randomization design. All respondents will be asked to report their beliefs about the share of Americans who had low enough income to qualify for SNAP in 2016. All respondents will be provided with a “hint,” which provides this statistic for “one U.S. state.” The first randomization (50/50) affects whether the hint corresponds to a state with a relatively high or relatively low eligibility share. The second randomization (50/50) involves a beliefs correction intervention, where we tell treatment respondents the true share of Americans who had low enough income to qualify for SNAP.

We will incentivize accurate belief reporting with the promise of a donation made to the charity of the respondent’s choice if their response is the closest to the truth (or, if they tie for the most accurate response, if they are chosen at random among those whose response is closest to the truth).

Design details

Randomization method
Computer/survey platform
Randomization unit
Individual
Clusters
No clustering
Observations
2,500 respondents
Arms
625 in each of four treatment arms (2x2 randomization with equal probability)
Primary outcomes
Beliefs: We will ask survey respondents what they believe is the share of Americans overall who were eligible for SNAP in 2016. Stigma index: We will ask respondents for the extent of their agreement or disagreement to eight statements which are intended to measure the stigma they perceive around food stamps.
Secondary outcomes
Attention check: We will ask survey respondents what the acronym “SNAP” stands for. This is a multiple choice survey question with four choices. We will use this to measure and adjust for inattention. We will also consider respondents inattentive if they report beliefs of overall eligibility outside the interval [1%,99%]. We will also evaluate effects on individual items in the stigma instrument.

Cited by

No citation data available for this registration yet.

Keywords