Using AI to Expand the Job Search of Displaced Workers in the Aftermath of the Covid-19 Crisis

Last registered on January 29, 2021


Trial Information

General Information

Using AI to Expand the Job Search of Displaced Workers in the Aftermath of the Covid-19 Crisis
Initial registration date
January 28, 2021

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 29, 2021, 8:44 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.



Primary Investigator

Boston College

Other Primary Investigator(s)

PI Affiliation
Harvard Business School
PI Affiliation
Ross School of Business, University of Michigan
PI Affiliation
Ross School of Business, University of Michigan

Additional Trial Information

In development
Start date
End date
Secondary IDs
To date, more than 38 million workers have filed for unemployment during the COVID-19 crisis. Early evidence suggests that the worst economic impacts will be concentrated among entry-level service workers, who are disproportionately young, female, and from marginalized groups. These populations are often slow to be reintegrated into the economy after economic contractions due at least in part to discrimination and imperfect information, with full recovery from the last decade’s financial crisis, for example, taking more than 7 years.
This inequality in employment outcomes is caused in part by job search frictions that disproportionately affect the ability of low-income minority workers to find good matches. In the proposed study, we investigate whether AI-assisted algorithmic matching of skills and psychometric profiles to overlooked vacancies can help displaced workers from entry-level service occupations in their search for employment. To do so, we propose a large-scale randomized controlled trial (RCT) in the United States building on our prior work that focuses on workers who were displaced from their prior jobs as a result of the COVID-19 pandemic.
The RCT will: 1) Use trained machine learning algorithms from prior experiments to predict job performance across a broad array of recovering occupations from psychometric profiles of job seekers; 2) Share a combination of information on best matched occupations and real vacancies,
highlighting deviations from prior search experience.
We will recruit job seekers from online job portals to complete the psychometric survey and answer questions about demographics, recent work, and job search history (including if they have been recently laid off due to Covid-19). After randomized assignment to one of several experimental groups, we will track employment status, job search success rates, time to reemployment, and compensation.
External Link(s)

Registration Citation

Adhvaryu, Achyuta et al. 2021. "Using AI to Expand the Job Search of Displaced Workers in the Aftermath of the Covid-19 Crisis." AEA RCT Registry. January 29.
Experimental Details


Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)

Hazard of reemployment, employment status, hours spent searching for job openings, number of searches, number of applications, number of call backs, number of interviews, number of offers, offer satisfaction, yearly earnings.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants will be randomized to receive a list of local vacancies according to the treatment arm in which they are allocated. Participants in the control group will see a random list of URLs linking to locally available job vacancies. In the first treatment arm we reduce the respondents’ search cost. In the second treatment arm, participants receive the same local list of randomly sorted vacancies as the control group, but we provide information to participants about the entry-level occupations that we predict may be a good fit for them. In the last treatment arm, we reduce both search cost and provide occupational fit information.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
1,640 adults
Sample size: planned number of observations
1,640 adults
Sample size (or number of clusters) by treatment arms
410 adults per arm
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
University of Michigan IRB #3 - Behavioral Science - Health Sciences and Behavioral Sciences
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials