Labor Markets Matching in Indonesia: Estimating Employers’ Preferences with Incentivized Resume Rating

How to cite

Ferrara, Eliana La; Fiorin, Stefano; Shofia, Naila. 2022. "Labor Markets Matching in Indonesia: Estimating Employers’ Preferences with Incentivized Resume Rating." AEA RCT Registry. AEARCTR-0008559. https://doi.org/10.1257/rct.8559-1.1

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

Eliana La Ferrara* , Stefano Fiorin , Naila Shofia

* Primary investigator

Registry ID
AEARCTR-0008559
DOI
https://doi.org/10.1257/rct.8559-1.1
First registered
Last updated
Status
On going
Trial dates
2022-02-21 – 2022-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/8559

Abstract

We collaborate with a large job portal website to understand which characteristics of job seekers are valued by prospective employers in different sectors on the Indonesian labor market. We will use a modified version of the Incentivized Resume Rating (IRR) design proposed by Kessler et al. (2019). This tool allows to elicit employer preferences for job applicants’ characteristics without deception. Employers will be offered the possibility to participate in an experiment in which they will be asked to rate a number of resumes of hypothetical candidates. We will generate random resumes of hypothetical candidates by randomizing socio-demographic features of the candidates, typical characteristics used by employers to assess candidates’ quality, and the picture of the applicant. Employers will be told that the resumes are only hypothetical (thus avoiding deception), but will be incentivized to disclose their true preferences because their rating will later be used to select real job applicants who best matches the employer’s stated preferences through machine-learning algorithms.

Intervention

We will use a modified version of the Incentivized Resume Rating (IRR) design proposed by Kessler et al. (2019). This tool allows to elicit employer preferences for job applicants’ characteristics without deception. Employers active on the website of our partner institution (a large Indonesian job portal) will be recruited via email, WhatsApp, and online. They will be offered the possibility to participate in an experiment in which they will be asked to rate a number of resumes of hypothetical candidates. We will generate random resumes of hypothetical candidates by randomizing socio-demographic features of the candidates, typical characteristics used by employers to assess candidates’ quality, and the picture of the applicant. Employers will be told that the resumes are only hypothetical (thus avoiding deception), but will be incentivized to disclose their true preferences because their rating will later be used to select (through machine-learning algorithms) real job applicants from the job portal who best match the employer’s stated preferences.

Experimental design

We will generate random resumes of hypothetical candidates by randomizing socio-demographic features of the candidates (gender and marriage status), typical characteristics used by employers to assess candidates’ quality (level and field of education, GPA, and work experience), and the picture of the applicant.

Design details

Randomization method
Randomization will be done in STATA.
Randomization unit
Employer/Vacancy/Hypothetical Candidate
Clusters
We plan to send invitations for our experiment to all employers active on the job portal with open vacancies for entry-level and early-carreer positions. The effective sample size will depend on the take-up rate. Our target is to achieve 500 to 1000 completed surveys.
Observations
Each participant is asked to rate 30 hypothetical resumes, so our number of observations will be 30 times the number of participants.
Arms
-
Primary outcomes
We will ask employers: i) about their interest in the candidate; ii) whether they believe that the candidate would accept the job; iii) whether they would like to invite the candidate for an interview.
Secondary outcomes
We will also ask employers: i) what would be an appropriate salary for the candidate; and ii) how well they think the candidate would fit in their company. More specifically, we ask each participating employer, the following questions about each of the hypothetical candidates: 1. “If you decide to hire [candidate’s name], how much salary you would offer to [candidate’s name]?” if the employer answered yes to the previous question, and “If another company decides to hire [candidate’s name] as a [job_title], how much salary do you think they would offer to [candidate’s name]?” in instead the employer answered no. 2. How well would this candidate fit in the work environment of your company (e.g., with other workers and / or with customers)? Both version of question 1 are measured in Indonesian Rupiahs. Question 2 is measured on a 0 to 10 likert scale.

Cited by

No citation data available for this registration yet.

Keywords