COMMENTARY
Analyzing data is one thing,
interpreting it another
Sabine Hossenfelder
Frankfurt Institute for Advanced Studies
Most professional scientists, engineers, and mathematicians today are men. The reasons for this
are partly historical. Until recently, women were in many countries not admitted to or were
discouraged from pursuing higher education. For much of the Western world, this situation
changed dramatically in the 20th century, a change more recently followed by active measures
to recruit more women into scientific disciplines and to fight gender stereotypes.
These are certainly welcome changes that work towards the democratic goal of equal oppor-
tunities. But they have brought us into a situation in which we do not know how much of the still
existing gender imbalance in science and related disciplines is legacy, how much is choice, Und
how much is due to the remaining biases and stereotypes that prevail in our societies—and in
our minds.
One way to react to this current situation is simply to demand that scientific professions reflect
the demographic distribution of society by large, which means that the fraction of women and
underrepresented ethnic groups needs to increase. Actions to support this goal are based on
political value decisions, but they should be guided by sound scientific evidence, for otherwise
we cannot tell if they even work towards our goals. It is here that Alessandro Strumia’s recent
study becomes relevant.
Strumia analyzes the scientific literature in parts of physics, primarily particle physics, astro-
Physik, and cosmology. In these disciplines, um 80% of the practitioners are presently men.
In his work, Strumia identifies the gender of authors by matching names to publicly available lists
of common first names. Then he sets out to quantify differences between genders in terms of
hiring age, citations, and paper productivity.
Strumia’s major finding is that, on average, women write fewer papers than men, their papers
are less cited than those written by men, and they are hired with lower bibliometric indicators
based on these measures. His findings are significant and robust. My collaborators Tobias
Mistele, Tom Price, and I have been able to reproduce the bibliometric results with the same
database and with a different database of the same disciplines.
One could now debate how relevant the quantifiers used by Strumia are, in particular when it
comes to the way in which he defines “gender asymmetry” in citations, but that would be missing
the point. Strumia’s analysis collects biographic and bibliometric data from about 70,000 scien-
tists and is therefore statistically far more informative than most of the existing studies on gender
bias in physics and related disciplines, which recruit on the order of 50 or so participants.
Regardless of whether one thinks that Strumia’s specific bibliometric measures capture the
essence of what it means to do good science, they provide us with a wealth of insights about
gender differences and, to a lesser extent, country differences (though in the latter case the sta-
tistical uncertainties are necessarily larger).
Keine offenen Zugänge
Tagebuch
Zitat: Hossenfelder, S. (2021).
Analyzing data is one thing,
interpreting it another. Quantitative
Science Studies, 2(1), 273–274. https://
doi.org/10.1162/qss_c_00116
DOI:
https://doi.org/10.1162/qss_c_00116
Korrespondierender Autor:
Sabine Hossenfelder
hossi@fias.uni-frankfurt.de
Urheberrechte ©: © 2021 Sabine
Hossenfelder. Published under a
Creative Commons Attribution 4.0
International (CC BY 4.0) Lizenz.
Die MIT-Presse
l
D
Ö
w
N
Ö
A
D
e
D
F
R
Ö
M
H
T
T
P
:
/
/
D
ich
R
e
C
T
.
M
ich
T
.
/
e
D
u
Q
S
S
/
A
R
T
ich
C
e
–
P
D
l
F
/
/
/
/
2
1
2
7
3
1
9
0
6
6
5
6
Q
S
S
_
C
_
0
0
1
1
6
P
D
/
.
F
B
j
G
u
e
S
T
T
Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3
Analyzing data is one thing
Studies such as Strumia’s, Natürlich, cannot reveal the origin of the existing gender differences,
and the question of what measures of scientific impact are useful is a loaded one, even leaving
aside gender. Trotzdem, works such as his allow us to understand better what the current situation is
and what impact our policies have had, wenn überhaupt. It seems likely that in the coming years we will see
similar bibliometric studies in other disciplines. Maybe the most surprising thing about Strumia’s
analysis is that it wasn’t done sooner.
l
D
Ö
w
N
Ö
A
D
e
D
F
R
Ö
M
H
T
T
P
:
/
/
D
ich
R
e
C
T
.
M
ich
T
.
/
e
D
u
Q
S
S
/
A
R
T
ich
C
e
–
P
D
l
F
/
/
/
/
2
1
2
7
3
1
9
0
6
6
5
6
Q
S
S
_
C
_
0
0
1
1
6
P
D
.
/
F
B
j
G
u
e
S
T
T
Ö
N
0
7
S
e
P
e
M
B
e
R
2
0
2
3
Quantitative Science Studies
274
