The Impact of Beliefs on Treatment Efficacy

How to cite

Toma, Mattie. 2020. "The Impact of Beliefs on Treatment Efficacy." AEA RCT Registry. AEARCTR-0005151. https://doi.org/10.1257/rct.5151-1.1

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

Mattie Toma*

* Primary investigator

Registry ID
AEARCTR-0005151
DOI
https://doi.org/10.1257/rct.5151-1.1
First registered
Last updated
Status
In development
Trial dates
2020-01-24 – 2021-01-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5151

Abstract

There exists ample evidence documenting the role of a "placebo" (often a sugar pill) in improving objective outcomes in medical trials, including a smaller literature identifying beliefs as the channel through which this effect operates (Malani, 2006; Kamenica et al, 2013; Benedetti et al, 2005). In this paper I plan to test whether a congruent effect exists in the social sciences. That is, do recipients of government interventions or research treatments experience better outcomes if they have more positive beliefs about the efficacy of the treatment they are receiving? To identify the role of beliefs in determining treatment efficacy, I am partnering with a non-profit organization, Wholesome Wave, to run an experiment in which I measure the impact of exogenous changes in beliefs. Wholesome Wave provides fruit and vegetable gift cards to low-income individuals in the United States to encourage healthy eating. I plan to introduce an intervention in a new Wholesome Wave program in Corpus Christi, TX in which participants will be asked to set a goal for the amount they want to spend on fruits and vegetables. I will randomly assign individuals to receive either optimistic or no information about the efficacy of goal setting. My primary outcome of interest is the impact of this exogenous variation in beliefs on subsequent goal achievement, measured through card expenditures. In the event that beliefs do indeed affect treatment efficacy, such a finding would allow researchers to better understand the generalizability of observed treatment effects and also provide a tool for policymakers looking for a cost-effective means of improving intervention outcomes.

Intervention

I will ask all participants to set a goal for fruit and vegetable consumption. Half of these participants will be randomly assigned to the treatment group and will receive information indicating that research has shown that goal-setting is effective in helping people stick to healthy habits like eating fruits and vegetables. To further shape beliefs regarding the efficacy of goal setting, treatment participants will also answer a comprehension question about the information they just received. Control participants, meanwhile, will not receive information about the efficacy of goal setting. All participants will also receive follow-up text reminders regarding the goal they set; reminders sent to treatment participants will reiterate that goal-setting has been shown to be effective.

Experimental design

See the Intervention section above.

Design details

Randomization method
Treatments are assigned within the survey platform Qualtrics using their random number generator.
Randomization unit
Treatments are assigned at the individual level.
Clusters
700 individuals
Observations
700 individuals.
Arms
350 individuals in each beliefs treatment arm.
Primary outcomes
I will use purchase-level card expenditure data to determine whether those who receive the optimistic information about the efficacy of goal setting get closer to achieving their goal over the subsequent month.

Cited by

No citation data available for this registration yet.

Keywords