Societal preferences for minimum wage reform: a choice experiment

How to cite

Dickens, Eleanor; Meango, Romuald. 2023. "Societal preferences for minimum wage reform: a choice experiment." AEA RCT Registry. AEARCTR-0012013. https://doi.org/10.1257/rct.12013-1.1

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

Eleanor Dickens* , Romuald Meango

* Primary investigator

Registry ID
AEARCTR-0012013
DOI
https://doi.org/10.1257/rct.12013-1.1
First registered
Last updated
Status
In development
Trial dates
2023-12-08 – 2023-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/12013

Abstract

This study aims to explore different ways in which support for reform to the minimum wage changes based on a variety of trade-offs in the labor market and broader economy. By employing a choice experiment, we will investigate societal preferences for minimum wage reform and estimate the willingness to pay. Furthermore, this study will examine whether the willingness-to-pay changes based on a variety of trade-offs in the labor market and broader economy.

Intervention

The intervention is a randomised survey. Respondents are first asked questions about their views on politics, redistribution, and the labour market. The first group is asked questions on their views of how the minimum wage policy is set ("policymaking block"), and then asked to participate in the choice experiment (and follow-up block). The second group is asked to participate in the choice experiment (and follow-up block), and then asked to participate in the policymaking block. Finally, all respondents are asked demographic questions.

Experimental design

Each person will be randomised into two groups: policymaking block before or after the choice experiment. The policymaking block asks opinions on how the minimum wage policy is and should be set.

The choice experiment is structured as follows: There are 6 attributes, each of which have either four or five levels. The respondents are shown two scenarios at a time (in one choice set) that vary randomly in levels of severity of attributes, and asked to indicate a choice probability between choosing each option. They will be shown 9 choice sets in three blocks. The order of choice sets is randomised within block, as are the order of each option within a choice set. Some choice sets are required while others are presented as a randomised subset. Two additional examples of the same scenario are shown, which serve as comprehension checks.

Finally, respondents are asked to answer question that asks their preference for a minimum wage policy change based on trade-offs for every attribute they have been shown, which vary randomly in levels of severity and redistributional attribute. Respondents see a randomised subset from the pool of possible sliders.

All choices, both in the choice experiment and in all other blocks, have a randomised order (or randomly flipped order) of responses to minimise order bias (where possible).

There are two attention checks near the beginning of the survey. If respondents fail both they are rejected from the study, in-line with survey platform guidelines.

Design details

Randomization method
Randomisation done within-survey by Qualtrics.
Randomization unit
Individual
Clusters
The treatment is not clustered.
Observations
820 respondents
Arms
410 per treatment group
Primary outcomes
Reported valuation of hypothetical increases in the minimum wage and associated trade-offs in the economy. WTP will be calculated for each attribute and compared between attributes if possible. Results will be analysed by group--whether WTP varies based on whether the policymaking block (questions on views of minimum wage policy) are asked before or after the choice experiment. Self-assessed political views, attitude to redistribution, demographics, views and knowledge of economy and labor market, and views on beneficiaries of minimum wage will also be analysed, both on their own and alongside WTP estimates.. Misperception will also be analysed. Respondents will also be classified into social preference archetypes and analysed by type. Results will also be analysed by level of attention and response time to the survey tasks. For example and in-line with current research, based on attention and response time of respondents, the sample may be trimmed for outliers to response time and analysed.

Cited by

No citation data available for this registration yet.

Keywords