Gender discrimination against next-generation researchers: Evidence from a field experiment

How to cite

Berger, Sebastian; Gerber, Rafael; Wyss, Annika M.; Doerrenberg, Philipp; Feldhaus, Christoph. 2020. "Gender discrimination against next-generation researchers: Evidence from a field experiment." AEA RCT Registry. AEARCTR-0006309. https://doi.org/10.1257/rct.6309-1.0

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

Sebastian Berger* , Rafael Gerber , Annika M. Wyss , Philipp Doerrenberg , Christoph Feldhaus

* Primary investigator

Registry ID
AEARCTR-0006309
DOI
https://doi.org/10.1257/rct.6309-1.0
First registered
Last updated
Status
Completed
Trial dates
2018-03-06 – 2018-04-09
Countries
Austria (), Switzerland (), Germany ()
JEL codes
Discrimination, Gender, Academic Economics, Field Experiment
Secondary IDs
DOI 10.17605/OSF.IO/GUZKE
Registry record
http://www.socialscienceregistry.org/trials/6309

Abstract

We conduct a natural eld experiment to study gender discrimination against
prospective next-generation researchers (pre-PhD level students) in the economics
profession. We send emails to members of the professional association of German-
speaking economists in which we invite the members to participate in a survey.
The survey invitations are part of an actual research project of two non- ctitious
Master's-degree students at a Swiss university, one of whom is female and one is
male. We randomly vary between email recipients if the email is sent and signed
by the male or female student. We nd, on average across all email recipients, that
the overall response rate is higher if the email sender is female. However, we nd
strong evidence for same-sex discrimination. Female economists strongly discrim-
inate against the female email sender, while male economists discriminate against
the male email sender. Same-sex discrimination among women is prevalent among
tenured and non-tenured recipients. Same-sex discrimination among men is only
observed for tenured recipients. The latter result implies that the average overall
e ect (favorable treatment of female sender) is driven by tenured male recipients.
Our results have implications for policies supporting female economists.

Intervention

The project is laid out in detail in the Open Science Framework
https://osf.io/guzke/

Experimental design

We conducted a natural field experiment with mem-
bers of the association of German-speaking economists in countries Austria, Germany
and Switzerland (Verein fuer Socialpolitik ; VfS henceforth). During the time of our
experiment, the VfS had approximately 3,900 members at all career stages (incl. PhD
students). We contacted all members who, upon signing up for membership, consented to
be contacted for research purposes and therefore appeared on a list with email contacts.
This list, which contained the email contacts of about 59% of all members, was used to
implement our field experiment and to approach the members of the association. Overall,
2,356 members of VfS were part of our field experiment.

The experimental variation was embedded in the emails
that we sent to the VfS members and in which they were invited to take the survey.
The objective of our field experiment is to investigate if response rates of professional
economists depend on the gender of the support-requester. A key part of our field ex-
periment is that one of the two students was female and the other student was male.
To address the objective of our study, we randomly varied if the emails were sent and
signed by the female student or the male student: one randomly selected half of VfS
members on the email list received the email from the female student and the other ran-
domly selected half received the email from the male student. The emails were identical
except for the name of the sender. The treatment variation appeared at two points in
the treatment letter: once in the first sentence where the respective student introduces
her/him-self, and once at the end of the letter where the email is signed in the name of
the respective student. In addition, the name appeared in the email address of the sender
(email addresses had the name of the respective student before the @-symbol, while the
part after @-symbol was identical). These three appearances imply that the name, and
therefore gender, of the email sender was salient to email recipients.

Design details

Randomization method
randomization done in office by computer to decide which email adress to enroll in which treatment.
Randomization unit
individual
Clusters
all members of the Verein für Socialpolitik having consented to be contacted via email.
Observations
3500 emails sent.
Arms
emails were sent in equal numbers to both treatments.
Primary outcomes
Prevalence of discrimination following invitation to supply real effort.

Cited by

No citation data available for this registration yet.

Keywords