What Do Jobseekers Want? Comparing Methods to Estimate Reservation Wages and the Value of Job Attributes
How to cite
Osman, Adam; Feld, Brian. 2022. "What Do Jobseekers Want? Comparing Methods to Estimate Reservation Wages and the Value of Job Attributes." AEA RCT Registry. AEARCTR-0009957. https://doi.org/10.1257/rct.9957-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Understanding jobseeker preferences---including their reservation wages and how much they value different non-wage amenities---is difficult because they are not directly observable. We test four different methods for estimating these preference parameters using an experiment in a job-matching center. We find large and important differences between the methods. We also estimate jobseekers' valuations of several job attributes, and explore how those valuations differ by gender. Using a follow up survey for validation and comparing the consistency of estimates with prior literature we find that Discrete Choice Experiments perform best. We show how these methods can improve our understanding of labor market frictions and help policymakers and employers develop targeted policies and compensation bundles to address inequities in the labor market.
Intervention
The intervention was a randomized survey. We collected data about job seekers' preferences in collaboration with the National Employment Pact (NEP), an NGO based in Cairo that provides job matching services. We randomized respondents into three different groups. The first group was shown ``open ended'' questions, the second group was given ``payment card'' questions, and the third group got a set of ``double-bound dichotomous choice'' questions. Afterwards, all 3 groups were then asked to go through the discrete choice experiment.
Experimental design
We randomized respondents into three different groups. The first group was shown ``open ended'' questions, the second group was given ``payment card'' questions, and the third group got a set of ``double-bound dichotomous choice'' questions. Afterwards, all 3 groups were then asked to go through the discrete choice experiment. This was randomized at the individual level on the spot using SurveyCTO.
Design details
Cited by
No citation data available for this registration yet.