The difference in the return to Swedish language skills across immigrant groups – ethnicity or region of origin?

How to cite

Rooth, Dan-Olof; Eriksson, Stefan; Carlsson, Magnus. 2021. "The difference in the return to Swedish language skills across immigrant groups – ethnicity or region of origin?." AEA RCT Registry. AEARCTR-0007524. https://doi.org/10.1257/rct.7524-1.0

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

Dan-Olof Rooth* , Stefan Eriksson , Magnus Carlsson

* Primary investigator

Registry ID
AEARCTR-0007524
DOI
https://doi.org/10.1257/rct.7524-1.0
First registered
Last updated
Status
In development
Trial dates
2021-04-15 – 2022-06-15
Countries
Private
JEL codes
J24, J71
Registry record
http://www.socialscienceregistry.org/trials/7524

Abstract

The purpose of the research project is to use field experiments (correspondence tests) to investigate whether language skills in Swedish affect the probability of getting a callback for a job interview for different groups of immigrants. To this end, we set up field experiments in which fictitious job applications with different language levels are sent to employers with advertised jobs in the Swedish labor market. The applicants’ language skills and country of origin in randomly assigned to the job applications, and hence we can measure the causal effects of these factors.

Experimental design

The correspondence study will be conducted in the three largest cities of Sweden: Stockholm, Goteborg and Malmo. The population consists of all employers that post vacancies in 18 low-, medium-, and high-skilled occupations at Sweden's largest vacancy website Platsbanken in the geographic regions during the study period. For additional details, see the document preplan.

Design details

Randomization method
We will send three unique fictitious applications to each posted vacancy in the sample. Randomization will be performed by a computer program. For additional details, see the document preplan.
Randomization unit
The main variable (language level) is randomly assigned to the job applications. For additional details, see the document preplan.
Clusters
No clustering.
Observations
Around 9,000 fictitious job applications will be sent to around 3,000 employers with an advertised vacancy.
Arms
We have four language levels, hence ¼ of the job applications for each language level. We have three applicant groups (country of origin), hence 1/3 of the job applications per group. For additional details, see the document preplan. See also planned total number of observations.
Primary outcomes
The main variable of interest is an indicator taking the value one if the fictitious job applicant receives a callback from the employer.

Cited by

No citation data available for this registration yet.

Keywords