Acceptability of food policies

How to cite

Espinosa, Romain; Nassar, Anis. 2020. "Acceptability of food policies." AEA RCT Registry. AEARCTR-0006429. https://doi.org/10.1257/rct.6429-1.0

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

Romain Espinosa* , Anis Nassar

* Primary investigator

Registry ID
AEARCTR-0006429
DOI
https://doi.org/10.1257/rct.6429-1.0
First registered
Last updated
Status
In development
Trial dates
2020-09-16 – 2020-09-26
Countries
Private
JEL codes
Q18, Q58, C90, D72, D91
Registry record
http://www.socialscienceregistry.org/trials/6429

Abstract

This experiment aims to test the validity of the results found in an exploratory experiment.

In February - March 2020, we ran a first online survey on Prolific in which we sought to explore the determinants of the acceptability of food policies (information campaigns, labels, taxation, and withdrawal from the market). We focused on three types of products (sugar, palm oil and cage-eggs) contained in snacks like cereal bars. Participants were asked to report the acceptability level of each policy for a given topic (sugar, palm oil or cage-eggs) and whether they would hypothetically vote for their implementation.

We explored 9 factors for the acceptability of food policies:
- Legitimacy of the state to intervene
- Awareness of the externalities at stake
- Scientific norm on the topic
- Social norm on the topic
- Effectiveness of the proposed policy
- Targeting of the consumers of the proposed policy
- Coerciveness of the proposed policy
- Majoritarian support for the proposed policy
- Inequalities resulting from the proposed policy

We found support for all variables but scientific norm. We also found evidence for an interaction effect between awareness and coerciveness: the more the policy is seen as coercive, the more awareness has an impact on the acceptability of the food policies.

Intervention

We will replicate our study on Prolific (see the material for the surveys).
660 participants will be recruted to fulfil our treatment surveys (220 participants for each survey).

Experimental design

Participants are allocated to one of the three questionnaires (see materials for the questionnaire).
220 participants are selected for each survey on Prolific.
Selection rule: born in the UK, UK national, English as first language, did not participate to the previous experiment.

The effectiveness item is asked twice (1 to 7 Likert scale). Participants who give significantly different answers to the same series of questions will be excluded from the analysis. (The average absolute deviation between the two answers should not be greater than 2.)

Design details

Randomization method
Between subject design – Treatment varying survey content (sugar, palm, egg)
Randomization unit
Individual level.
Clusters
3 treatment surveys.
Observations
220 participants per treatment, 660 in total.
Arms
220 participants for each of the three treatments
Primary outcomes
Acceptability score (Likert scale between 1 and 7), Voting decision (Dummy variable)

Cited by

No citation data available for this registration yet.

Keywords