Age and the labor market for Hispanics in the United States

How to cite

Mosquera, Roberto; Lahey, Joanna. 2022. "Age and the labor market for Hispanics in the United States." AEA RCT Registry. AEARCTR-0009609. https://doi.org/10.1257/rct.9609-1.0

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

Roberto Mosquera* , Joanna Lahey

* Primary investigator

Registry ID
AEARCTR-0009609
DOI
https://doi.org/10.1257/rct.9609-1.0
First registered
Last updated
Status
Completed
Trial dates
2019-09-02 – 2022-06-01
Countries
United States of America ()
JEL codes
J71, J15
Registry record
http://www.socialscienceregistry.org/trials/9609

Abstract

We explore the labor market for Hispanic high school graduates in the United States by age using information from the US Census, American Community Survey, Current Population Survey, and three laboratory experiments. We find, in general, that the differences in outcomes for Hispanic and non-Hispanic high school graduates do not change across the lifecycle. Moving to a laboratory setting, we provided participants with randomized resumes for a clerical position that are on average equivalent except for name and age (as indicated by the date of high school graduation). In all three experiments, hypothetical applicants with Hispanic and non-Hispanic names were generally treated the same across the lifecycle by a student population, a population of human resources managers, and a more general population from mTurk. These results stand in contrast to earlier results that find strong differences by age in how resumes with Black and White names are treated.

Experimental design

First, we re-analyze data from two previous eye-tracking experiments that asked students (Experiment 1) and human resources (HR) managers (Experiment 2) to rate resumes with randomized inputs for a clerical position while tracking their eyes with an eye-tracking device. We look for differences in ratings between candidates with a Hispanic last name and candidates with White last names. We test for these differences using a local polynomial smoother (lpolyci in STATA) and regressions to look for differences in means.

Second, used a large sample of mTurk managers, clerical workers, and the general population as our participant population and again asked participants to rate resumes for a clerical position. In this study, we expanded indicators of Hispanic status to include Hispanic first names in addition to last names, we added indicators on resumes to contradict or reinforce Hispanic stereotypes, and we asked questions about specific aspects of each worker such as the need for supervision and communication skills.

Finally, we complement Experiment 3 with another intervention in mTurk to ask more specific questions about the Hispanic status and socioeconomic status signaled by names.

Design details

Randomization method
We created the resumes using Lahey and Beasley’s (2009) resume randomizer program and used a database of resume inputs drawn from actual resumes. The program randomly chooses from the database high school graduation date (to signal age), first name (to signal gender and race), last name (to signal ethnicity), home address, email address, high school attended previous work experience texts, additional training, and volunteer experience.
Randomization unit
Randomization occurs in two steps. First, we randomly create resumes from a pool of resume characteristics. Then we randomly assign 30 resumes to each participant.
Clusters
Experiment 1: 152 individuals Experiment 2: 67 HR managers Experiment 3: 1,068 individuals from mTurk Experiment 4: 200 individuals from mTurk
Observations
Experiment 1: 6,080 resumes Experiment 2: 2,680 resumes Experiment 3: 32,040 resumes Experiment 4: 8,000 names
Arms
Experiments 1 and 2: 13 percent of the resumes had Hispanic last names and non-Hispanic first names Experiments 3 and 4: 80 percent of the resumes have non-Hispanic first and last names, 10 percent have Hispanic first and last names, 5 percent have a non-Hispanic first name and Hispanic last name, and 5 percent have a Hispanic first name and non-Hispanic last name
Primary outcomes
Resume ratings: participants were asked to rate the ability of the candidate to fulfill the position, using a 7-point Likert scale, with 7 as the most “hireable”

Cited by

No citation data available for this registration yet.

Keywords