RESEARCH ARTICLE - Specialized Research AI at MIT

RESEARCH ARTICLE

The effect of academic mobility on research
performance: The case of Italy

Giovanni Abramo1

, Ciriaco Andrea D’Angelo1,2

, and Flavia Di Costa1

1Laboratory for Studies in Research Evaluation, Institute for System Analysis and Computer Science (IASI-CNR),
National Research Council of Italy (Italy)
2Department of Engineering and Management, University of Rome “Tor Vergata” (Italy)

Keywords: bibliometrics, FSS, mobile academics, research performance, universities

ABSTRACT
This work investigates the effects of researchers’ mobility on their research performance. The
reference context is that of national intrasector mobility, in a country, Italy, characterized by a
research system lacking the typical elements of an academic labor market. In particular, the
analysis was conducted on 568 academics working at national universities and affected by
mobility in the period 2009–2014. The effect of mobility on the variation of performance at the
turn of the transfer was analyzed considering the interplay of demographic/sociological
characteristics of the researchers, as well as contextual factors related to both the organization
of origin and destination. Results show that it is the less productive academics who represent
the larger share of those who move, and more than half of the mobile academics worsen their
performance after the transfer.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

INTRODUCTION

1.
Mobility is a hallmark of a scientist’s professional life. The phenomenon, especially in recent
decades, has acquired important implications and has been the subject of numerous studies by
scholars and interventions by policymakers.

The basic idea is that “researcher mobility” is a mechanism of knowledge diffusion able to
generate positive spillovers on organizations, sectors, and territories (Song, Almeida, & Wu,
2003). In the private sector, inventor mobility underlies the performance dynamics of innova-
tion processes of firms and regions (Breschi, Lenzi et al., 2010). Interfirm and interregion
mobility have positive effects on inventors’ performance, which in turn increases the innova-
tion rates of companies (van der Wouden & Rigby, 2021). By determining the diffusion of ideas
and new knowledge, the mobility of researchers can therefore be considered as a factor for the
development of the whole research system (Fernández-Zubieta, Geuna, & Lawson, 2016) and,
as a consequence, of countries’ economies.

The policy implications are such that the European Commission has funded a number of
studies over the years to collect detailed information and data on the mobility patterns and
career paths of EU researchers, the latest being the MORE4 survey (European Commission
Directorate-General for Research and Innovation, 2021). Also, within the EU, enhancing
balanced mobility of researchers is a priority policy objective under the European Research

a n o p e n a c c e s s

j o u r n a l

Citation: Abramo, G., D’Angelo, C. A.,
& Di Costa, F. (2022). The effect of
academic mobility on research
performance: The case of Italy.
Quantitative Science Studies, 3(2),
345–362. https://doi.org/10.1162/qss_a
_00192

DOI:
https://doi.org/10.1162/qss_a_00192

Peer Review:
https://publons.com/publon/10.1162
/qss_a_00192

Received: 19 November 2021
Accepted: 4 April 2022

Corresponding Author:
Ciriaco Andrea D’Angelo
dangelo@dii.uniroma2.it

Handling Editor:
Ludo Waltman

Copyright: © 2022 Giovanni Abramo,
Ciriaco Andrea D’Angelo, and Flavia Di
Costa. Published under a Creative
Commons Attribution 4.0 International
(CC BY 4.0) license.

The MIT Press

Academic mobility on research performance

Area framework. The recent 2020 Communication on the ERA emphasizes the need to
improve access to excellence and broaden talent capacity (e.g., by supporting mobility)1.

Many scholars have investigated the relationship between mobility and academic careers, as
the former impacts on the breadth and standing of a researcher’s professional network; the quality
of work experiences and scientific challenges faced; and the intensity of production of new scien-
tific advances and their impact (Cañibano, D’Este et al., 2020; Horta, Jung, & Santos, 2020).

There are different types of mobilities, such as national vs. international, intrasector vs.
cross sector, and single vs. multiple (El-Ouahi, Robinson-García, & Costas, 2021; Horta,
Meoli, & Santos, 2022). Our study analyzes the effects of researchers’ national intrasector
mobility on research performance. Many scholars claim that these effects are influenced by
the demographic/social characteristics of the researchers (Aksnes, Rørstad et al., 2013), as well
as by contextual factors related to both the place of origin and the place of destination (Bäker,
2015; Deville, Wang et al., 2014). The effects of mobility on research performance are found
to depend also on such other aspects as the field of research, or the frequency and duration of
mobility (Franzoni, Scellato, & Stephan, 2014).

Some studies on the relationship between mobility and research performance distinguish
between domestic and international mobility. As regards international mobility, a literature
review by Guthrie, Lichten et al. (2017), which mainly analyzes studies of researchers moving
to and from the United Kingdom, highlights the presence of a relation between international
mobility and improvements in academic performance, although it is difficult to establish a true
direct or reverse causal relationship. A more recent literature review by Netz, Hampel, and
Aman (2020) reports that 19 out of 34 studies highlight positive effects of international mobility
on scientific performance. Inbred scientists tend to pursue cozy research programs, while mobile
scholars are risk takers and pioneer new scientific trajectories (Horta et al., 2021). At the same
time there are several studies that find heterogeneous, no or negative effects (Gyorffy, Nagy et al.,
2018; Halevi, Moed, & Bar-Ilan, 2016). The authors point out that the various studies use differ-
ent indicators and nonhomogeneous measurement approaches, so it is clear that the framework
for analyzing the relationship between mobility and performance is far from being consolidated.

Referring, instead, to the context of national mobility, a distinction must be made between
those countries, typically Anglo-Saxon, where the researcher operates in a free labor market,
and others, typically in continental Europe, where the researcher is a civil servant who moves
within a strongly regulated and centrally governed context, in which individual organizations
are in fact precluded from customizing the job offer. Italy falls into this second category. Its
higher education system is a long-standing, classic example of a public and highly centralized
governance structure, with low levels of autonomy at the university level and a very strong role
played by the central state. The Ministry of Universities and Research (MUR) recognizes 96
universities as having the authority to issue legally recognized degrees. 94.9% of faculty are
employed in public universities, which are largely financed by the government through mostly
noncompetitive allocation of funds. There are no “teaching-only” universities, as all aca-
demics are required to carry out both research and teaching. At the end of 2020, there were
56,500 faculty members in Italy (full, associate, and assistant professors).

There is no tenure-track system in place. The recruitment and career advancement of
academics are regulated by specific law and occur through public competition. Academics
can transfer from a university to another, holding the same academic rank, upon clearance

1 https://ec.europa.eu/info/research-and-innovation/strategy/strategy-2020-2024/our-digital-future/era_en;

accessed April 4, 2022.

Quantitative Science Studies

346

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

by both institutions. Salaries are regulated at the central level and are calculated according to
role (administrative, technical, or professorial), rank within the role (e.g., assistant, associate, or
full professor), and seniority. An academic’s salary does not depend on merit.

These conditions altogether create an environment and a culture that are completely noncom-
petitive, yet favoritism in recruitment and career progress flourishes. The overall result is a system of
universities that are almost completely undifferentiated for quality and prestige, with the exception
of the tiny Schools for Advanced Studies and a very small number of private special-focus univer-
sities. Top scientists are dispersed more or less uniformly among all Italian universities, along with the
low performers, so that no single university reaches the critical mass of excellence necessary to
develop as an elite university and compete at the international level (Abramo, Cicero, & D’Angelo,
2012b). Academics’ mobility then is not the effect of competition among universities, trying to
attract talents; rather, it occurs for career progress and/or family reasons.

For its peculiar characteristics, Italy is an interesting case study for the analysis of the rela-
tionship between mobility and research performance. Furthermore, observing the Italian case
will allow us to apply a sophisticated research performance indicator at the individual level, as
we can count on an extremely accurate authors’ name disambiguation algorithm, and on a
fine-grained field classification scheme of researchers by which we avoid distortions in com-
parative performance analysis.

The paper is organized as follows: in the next section, we review the relevant literature.
Section 3 presents the methodology and describes the data and methods. In Section 4, we
show the results of the analysis. Section 5 concludes the work, discussing some policy impli-
cations and identifying further research developments.

2. THEORETICAL BACKGROUND

The literature on the subject under investigation is vast. We will try to describe and comment
on the main contributions, organizing them around each of the research hypotheses presented
in Section 1.

2.1. Mobility Positively Impacts an Academic’s Research Performance

The topic of the link between scientists’ performance and science mobility (interorganizational)
can be approached from two different perspectives. On the one hand, one can consider per-
formance as a driver explaining mobility, as supported by the theory of competition in higher
education (Jongbloed, 2004), universities search for a competitive advantage over competitors
by attracting top-quality research staff. On the other hand, one can investigate if it is mobility
that influences research performance, as supported by the theory of knowledge externalities
(Ciccone & Peri, 2006), whereby institutions with higher research quality positively impact the
performance of new recruits. The topic is still very open because the results are conflicting and
have not yet come to a consolidated view.

Much of this variability may be attributable to the manner in which these studies were con-
ducted and the chosen field of observation. Some of them focus on a single country (Aksnes
et al., 2013; Allison & Long, 1990; Cañibano, Otamendi & Andújar, 2008; Cruz-Castro &
Sanz-Menéndez, 2010; De Filippo, Casado, & Gómez, 2009; Ejermo, Fassio, & Källström,
2020; Fernández-Zubieta et al., 2016; Jonkers & Tijssen, 2008; Tartari, Di Lorenzo, & Campbell,
2020). Others, however, have analyzed a single discipline (Albarrán, Carrasco, & Ruiz-Castillo,
2017; Bolli & Schläpfer, 2015; Dubois, Rochet, & Schlenker, 2014; Jonkers & Tijssen, 2008;
Ryazanova & McNamara, 2019; Tartari et al., 2020).

Quantitative Science Studies

347

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Analyses conducted on the effects of mobility taking place on a global scale show mixed
results. Franzoni et al. (2014) showed that foreign academics and those who have spent long
periods abroad make higher impact publications; Jonkers and Tijssen (2008) found a positive
correlation between international mobility and publication intensity of Chinese researchers, as
labor mobility contributes to the scientific and technical human capital of scientists to
the extent that it increases the number of collaborations and strengthens existing relationships.
Halevi, Moed, and Bar-Ilan (2016), observing the top 100 authors between 2010 and 2015 in
terms of publications in each of seven disciplines2, showed that domestic interorganizational
mobility induces an increase in both output and impact, while intercountry mobility does not
seem to produce the same effects.

An in-depth analysis of the duality of the link between scientists’ productivity and science
mobility was carried out in particular by Fernández-Zubieta et al. (2016) on a sample of 171
UK research active academics, in “tenured type” positions, affiliated in 2005 to 53 different UK
universities in four scientific fields (chemistry, physics, computer science, and mechanical, aero-
nautical, and manufacturing). The authors come to nondefinitive results, as they find an overall
positive effect of mobility, although not significant, and a short-term negative effect (i.e., a drop in
performance in the years following the job change, probably due to adjustment costs). De Filippo
et al. (2009) analyzed the relationship between the mobility of researchers and their scientific
performance in a medium-sized Spanish university (Carlos III University of Madrid-UC3M), which
had 1,800 researchers distributed in the engineering, social sciences, and humanities disciplines.
The results of their analyses show that there is a strong association between mobility, performance,
and visibility: researchers “with mobility” show better performance indicators (higher average
impact factor, a greater number of citations per document, lower percentages of noncited docu-
ments, and a higher rate of international collaboration) than their “without mobility” colleagues.

The most interesting studies for the purposes of this paper are those that address the issue by
focusing on an entire national academic system. Included in this typology are the works of
Aksnes et al. (2013) and Ejermo et al. (2020) on the Norwegian and Swedish academic systems
respectively. In the first case, the analysis, conducted on a sample of about 11,000 scientists,
revealed that researchers affected by mobility episodes tend to have slightly higher publication
and citation rates than other researchers. However, the results are not unambiguous, so there is
no strong evidence that mobility has beneficial effects on research performance. In the second
case Ejermo et al. (2020) studied the effects of interuniversity mobility on the productivity of
over 35,000 Swedish academic researchers for the period 2002–2012. The results indicate that
the effect of mobility on performance is not found when considering career progress. In
essence, career progression positively impacts a researcher’s performance, whether this
involves a physical transfer or not. It is interesting to test whether in Italy as well, mobility
has slight or no positive impact on research productivity.

2.2. The Impact of Mobility on a Researcher’s Performance Is Greater if the Transfer Is to
Universities/Groups with Better Research Performance

As Yan, Zhu, and He (2020) found in their recent study of US academia, academics tend to move to
institutions with higher research intensity. Earlier, Allison and Long (1990) had already showed that
in correspondence with a change of institution, academic scientists who moved to more presti-
gious departments registered an increase in both the number of publications and the number of

2 Neuroscience, Mechanical Engineering, Arts and Humanities, Oncology, Environmental Geology, Business,

and Infectious Diseases.

Quantitative Science Studies

348

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

citations, as opposed to their colleagues who underwent a downgrade. Fernández-Zubieta et al.
(2016) also showed that researchers affected by upward mobility register significant increases in
productivity, but not on citations, while downward mobility negatively affects the researcher’s
overall research performance. These results were confirmed by Tartari et al. (2020), who found
that mobility has a positive effect on individual performance and this effect is strengthened if
the movement takes place to better endowed institutions, although their analysis is limited to a
sample of 348 academic scientists working in the biology departments of UK universities. Accord-
ing to Bolli and Schläpfer (2015), however, this hypothesis is rejected in the analysis they con-
ducted with respect to a single field (economics). Deville et al. (2014) have instead observed that
there is stratification in science whereby movement is predominantly horizontal rather than
vertical: People from elite institutions are more likely to move to other elite institutions;
similarly, people from lower rank institutions are more likely to move to peer institutions. In
addition, when movement between groups occurs, it is observed that moving from elite to
lower ranking institutions results in a modest decrease in scientific impact on average; The
opposite event, however, does not result in gains in impact. Differently from the Anglo-Saxon
higher education system, the Italian one is scarcely competitive and hardly differentiated.
Therefore, to what extent the performance of universities/groups of destination impact on that
of mobile academics is difficult to predict.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

2.3. The Productivity of a Mobile Researcher Increases if He or She Joins a Larger Group

In addition to the prestige of the institution of origin and/or destination, some scholars have also
investigated the influence of other factors related to the context of origin and destination of the
mobile researcher, including the size of the department in which he or she works. The link
between department/university size and research performance has been extensively dissected
in the literature (Abramo, Cicero, & D’Angelo, 2012a; Aksnes, Piro, & Rørstad, 2018; Bauer, Schui
et al., 2013), both in general terms and related to researcher mobility. Several studies have ana-
lyzed precisely this link between scientific performance and destination department following a
mobility event, as it is believed that larger institutions and/or departments may offer more oppor-
tunities in terms of facilities and scientific collaborations. From the analysis of a large sample of
German-speaking economists and management researchers, Bäker (2015) found that the larger the
starting department, the more negative the impact on performance will be in the short run. This
change determines a risk of having their social capital reduced for those who have more col-
leagues in their old department of origin. Deville et al. (2014), in their analysis of the relationships
between career, mobility, geography, and scientific impact, observed that the institution size has
little influence on publication intensity but is positively correlated with publication impact. This
suggests that large institutions offer more opportunities for internal collaborations: Complex
research needs complex collaborations, and the resulting publications, with more coauthors in
the byline, may attract a higher number of citations. Dubois et al. (2014) found that for mathe-
maticians, department size does matter: Larger departments attract better researchers, even after
controlling for author-fixed effects. Moreover, specialized departments seem to stimulate the
quality rather than the quantity of the scientific production of their researchers.

2.4. The Performance of a Mobile Researcher Increases if Associated with an Academic

Rank Advancement

Scientific performance and mobility represent two of the determinants that influence the
achievement of a permanent position and career progression towards the top of the academic
hierarchy (Pezzoni, Sterzi, & Lissoni, 2012; Sanz-Menéndez, Cruz-Castro, & Alva, 2013).
Particularly in continental European countries, including Italy, academics are civil servants,

Quantitative Science Studies

349

Academic mobility on research performance

so that mobility within national borders very often corresponds to career advancement. As men-
tioned earlier, Ejermo et al. (2020), in their study on the effects of national mobility on the
performance of Swedish university researchers, showed that a move coupled with a promotion
leads to a greater impact on productivity. However, these researchers do not have a statisti-
cally significant different publication rate compared to other moving researchers, so the
observed effects are mainly attributable to the move. Furthermore, the results suggest that
those who only achieved career advancement without being “mobile” are also able to
increase their publications over time, although a slightly smaller effect on citations is observed.
Finally, such an effect differs significantly across disciplines: It is rather evident in the
technical-scientific disciplines, but absent in the social science and humanities. The different
effects on publications and citations call for further investigation.

2.5. The Impact of Mobility on a Researcher’s Productivity Is Confounded by Certain Personal Traits of
the Individual

The impact of mobility on a researcher’s productivity is confounded by certain personal char-
acteristics (age, gender, previous academic rank, etc.) that need to be taken into account when
modeling the phenomena and analyzing the results. Aksnes et al. (2013) note that in the Nor-
wegian academic system, mobile researchers tend to have slightly higher publication and cita-
tion rates than their peers. However, regression analyses where such personal characteristics
as gender, age, and academic position of the researchers were accounted for removed most of
the independent effect that mobility may have on publication and citations. Our investigation
should add more evidence on the role of personal factors on productivity variations.

In particular, concerning national intrasector mobility, the hypotheses that we find more

interesting to test are as follows:

(cid:129) Mobility positively impacts an academic’s research productivity.
(cid:129) The impact of mobility on a researcher’s productivity is greater if the transfer is to

universities/groups with better research performance.

(cid:129) The productivity of a mobile researcher increases if he or she joins a larger group.
(cid:129) The productivity of a mobile researcher decreases with the geographic-administrative

distance of the transfer.

(cid:129) The productivity of a mobile researcher increases if it is accompanied by academic rank

advancement.

(cid:129) The impact of mobility on a researcher’s productivity is confounded by certain personal

characteristics, including age and gender.

If, on the one hand, we can affirm that these hypotheses are relevant to adequately support
national policies and to evaluate their effects ex post, on the other hand, we can also affirm
that only two studies related to similar contexts and, in particular, to Scandinavian countries,
can be found in the literature (Aksnes et al., 2013; Ejermo et al., 2020). The major difference
between ours and all previous studies, however, is the indicator and method of comparative
measurement of research performance. As we will explain in more detail below, our indicator
does not separate quantity and impact of scientific production, but includes both.

3. DATA AND METHODS

Our field of observation is made of all Italian academics on national mobility between 2009
and 2014. This time period will allow us to measure the bibliometric research performance

Quantitative Science Studies

350

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

of individuals in the 5-year periods before and after the transfer. As an example, the perfor-
mance of an academic who moved in 2014 is measured both in 2015–2019 and in 2009–
2013, to contrast the two periods.

Data on the faculty at each university were extracted from the database of Italian university
personnel, kept by the MUR3. This source contains, at December 31 of each year, information
on each academic, including affiliation, academic rank, and field of research. All academics
are classified in one, and only one, field, called a Scientific Disciplinary Sector (SDS), for
370 SDSs which are grouped into disciplines (called University Disciplinary Areas (UDAs),
for 14 UDAs.

For the measurement of scientific productivity at individual level, we will consider the

fractional scientific strength (FSS)4. defined as

FSS ¼

i¼1

ci
(cid:1)c

(1)

where:

t = number of years of work of the academic in the period under observation
N = number of publications by the academic in the period under observation
ci = citations received by publication i
(cid:1)c = average of distribution of citations received for all cited publications in the same year

and subject category of publication i

fi = fractional contribution of the academic to publication i.

The fractional contribution equals the inverse of the number of authors in those fields where
the practice is to place the authors in simple alphabetical order, but assumes different weights
in other cases. For the life sciences, widespread practice in Italy is for the authors to indicate
the various contributions to published research according to the order of names in the listing of
the authors. Hence, according to Abramo, D’Angelo, and Rosati (2013), for life science SDSs
we give different weights to each coauthor according to their position in the list of authors and
to the character of the coauthorship (intramural or extramural).

The bibliometric data set was extracted from the Italian Observatory of Public Research
(ORP), a database developed and maintained by the authors and derived under license from
the WoS. Beginning from the raw data of the WoS and applying a complex algorithm to
reconcile author’s affiliation and disambiguation of the true identity of the authors, each
publication (article, review, letter and conference proceeding) is attributed to the academic
that produced it (D’Angelo, Giuffrida, & Abramo, 2011). Thanks to this algorithm, we can
measure the FSS at the individual level on a national scale. For reasons of significance, we
will limit the analysis to academics:

(cid:129) working in hard sciences SDSs (201 in all, concerning 11 UDAs), where publications
in international journals serve as a reliable proxy for overall research output5; and

(cid:129) on staff for at least 1 year in the 5 years before and after the transfer.

3 https://cercauniversita.cineca.it/php5/docenti/cerca.php; accessed April 4, 2022.
4 A thorough explanation of the theory and assumptions underlying FSS can be found in Abramo and

D’Angelo (2014).

5 Evaluative bibliometrics should not be applied to the arts and humanities, due to the scarce coverage of

these areas in bibliographic repertories (Archambault, Vignola-Gagné et al., 2006; Hicks, 1999).

Quantitative Science Studies

351

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Applying this restriction, the final data set is composed of 568 academics involved in a
transfer, 152 of which are characterized by career progression (from assistant to associate pro-
fessor, or from associate to full professor).

With the aim of filtering out the effect of possible time patterns on the data, the individual
productivity value will be normalized to the data expected at SDS level (i.e., to the average
value of the FSS measured for all national SDS academics). From now on, the term FSS is to
be interpreted as the normalized value of Eq. 1.

In turn, the productivity of the group of origin (composed of the academics from the SDS
and university in which the academic was placed prior to transfer) and that of the destination
group (composed of the academics from the SDS and university to which the academic trans-
ferred) will be measured through the average FSS of the individuals that compose the group.

4. RESULTS

The distribution of productivity measured over the 5-year period before transfer for the 568
academics in the data set is extremely skewed (Figure 1), given the presence of several outliers
and, in particular, a maximum value of almost 23. The average value is 0.990, therefore
slightly lower than the expected (unit) value, but the median is significantly lower and equal
to 0.513, because 71% of academics show an FSS lower than the unit, among which 60 (over
10% of the total) are even unproductive. Thus, good research productivity would not seem to
be a significant predictor of the likelihood of intrasector mobility of Italian academics. That
could be the case, as Nascia, Pianta, and Zacharewicz (2021) argue, for young researchers
at the beginning of their careers who migrate abroad in search of new and better opportunities.

Indeed, one would expect that the destination university would rather welcome high
performers. At the same time, universities cannot force low performers to move to other
institutions.

If we then analyze the difference in productivity at the turn of the transfer (Figure 2), we see
that 54.8% of academics worsen their performance, while the remaining 45.2% improve it. So

Figure 1. Distribution of the productivity of the mobile academics in the data set in the 5 years
preceding their transfer.

Quantitative Science Studies

352

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Figure 2. Distribution of the difference in productivity of the mobile academics in the data set,
between the 5-year period following and that preceding their transfer.

not only are the least productive people moving out, but most of those people are also wors-
ening their performance. Of the initial 60 unproductive academics, 20 remain so. The distri-
bution of performance variation shows an almost normal trend (confirmed by the Shapiro-Wilk
W test), with a negative mean and median (equal to −0.258 and −0.041, respectively) and
particularly long tails due to the presence of several outliers. Among these, two academics
stand out, one who goes from an FSS of 22.9 in the 5-year period preceding the transfer to
5.8 in the one that follows it and, conversely, a second who presents an FSS of 5.0 before
the transfer and 26.9 after it.

The change in productivity at the turn of the transfer and the productivity in the period
before the transfer are inversely correlated, as shown by the scatterplot in Figure 3. The

Figure 3. Scatterplot of the difference in productivity of academics in the data set, as a function of
productivity in the five years preceding their transfer.

Quantitative Science Studies

353

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Figure 4. Scatterplot of the difference in productivity of academics in the data set at the turn of
the transfer, as a function of the difference in productivity between the destination (D) and origin
(O) groups.

Pearson’s correlation coefficient ρ is equal to −0.700 ( p-value 0.0000). It must clearly be dif-
ficult for the most productive academics to maintain their research performance, posttransfer,
at levels similar to those recorded in the period before the transfer. Of course, there are excep-
tions: In particular, of the 164 academics with a pretransfer FSS of more than one, 24 (i.e., less
than 15%) show an increase in productivity in the subsequent period.

In a nutshell, empirical evidence from national academic transfers seems to show no pos-

itive links between mobility and a scientist’s research productivity after the transfer.

Coming to the second hypothesis of the work, Figure 4 represents the dispersion of the data
on the change in productivity of each academic at the turn of the transfer, and the difference in
productivity between the group in which he or she is being inserted and the one from which
he or she comes. It should be remembered that by “group” we mean colleagues from the same
SDS. The diagram shows data for 384 observations of the total 568, as we exclude those
transfers involving source/destination locations with less than one FTE staff unit in the moving
academic’s SDS.

There is a weak correlation between the two variables (Pearson ρ = 0.163, p-value 0.0014),
indicating that transfers accompanied by an increase in productivity concern destinations that
perform better than those of origin and vice versa. Therefore, there is a positive link between
the variation in productivity of a researcher and the difference in standing between the context
that he or she leaves and the one that he or she moves to.

Regarding the link between performance and size (in terms of FTE) of the group in which
mobile academics are inserted, the scatterplot in Figure 5 shows a total absence of correlation
between the sizes of the groups of origin and destination and their variation in productivity in
the periods at the turn of the transfer. The result is in line with the absence of returns to scale in
Italian university research (Abramo et al., 2012a).

The question now arises of whether the possible effect of the change in performance
following a transfer can be related to the fact that the transfer takes place following the award
of a competition involving a change of academic rank (from assistant to associate professor, or

Quantitative Science Studies

354

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Scatterplot of the difference in productivity of the academics in the data set at the turn of
Figure 5.
the transfer, as a function of the difference in size between destination (D) and origin (O) groups.

from associate to full professor). There are 152 such cases in the data set, and in Table 1 we
report the outcome of a t-test to check for differences in the productivity gap for such aca-
demics and the remaining 416 affected by a transfer without a change of academic rank.
Although there is a smaller average reduction in productivity for the former (−0.138), the
difference with respect to that recorded for the latter (−0.306) is not statistically significant.
The same result is reached with the nonparametric Wilcoxon rank-sum Mann-Whitney test
(z = 1.239; Prob > |z| = 0.215).

The same result is reached by considering the FSS as a function of

(cid:129) a period dummy, which takes the value of 0 before the transfer and 1 after;
(cid:129) a “treatment” dummy, which takes the value of 1 for transfers characterized by an aca-

demic rank change and 0 for others; and

(cid:129) an interaction effect between the two dummies above.

The coefficient of this interaction effect takes the value of 0.168, but the relative p-value is
0.423. This means that academics in the treatment group (transferred because of an academic
rank change) tend to record positive changes in FSS compared to their peers transferred with-
out upgrade of academic rank, but this outcome is statistically nonsignificant.

Table 1.
academics moving for a career progression (vertical move) or not (horizontal move)

Two-sample t-test with equal variances for FSS variation at the turn of the transfer between

Type of move
Horizontal

Vertical

Combined

Difference

Obs
416

152

568

Mean
−0.306

−0.138

−0.261

−0.168

Std Err.
0.071

0.178

0.071

0.159

Std Dev.
1.447

[95% Conf. Interval]
−0.167
−0.446

2.198

1.680

−0.490

−0.400

−0.481

0.214

−0.123

0.145

t = −1.0567; Pr(T < t) = 0.1455. Quantitative Science Studies 355 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u q s s / a r t i c e - p d l f / / / / 3 2 3 4 5 2 0 3 1 8 8 7 q s s _ a _ 0 0 1 9 2 p d / . f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Academic mobility on research performance We then tried to investigate the correlation, if any, between the variation in academic per- formance and the geographic-administrative distance of the transfer that affected him or her. In fact, the transfer may have taken place between universities (cid:129) in the same city (46 cases, equal to 8.1% of the total); (cid:129) from different cities but from the same region (18.5% of total cases); (cid:129) from different regions but from the same macro geographical area of the country (i.e., north-east, north-west, central, south and islands) (24.8%); or (cid:129) from different macro areas (48.6%). The geographical dimension of the transfer could have an impact on the so-called adjust- ment costs for the moving academic, who will face increasing “difficulties” as the geographical, administrative, and cultural distance between the place of origin and the place of destination increases. However, analysis of the data reveals the total absence of correlation between trans- fer distance and performance variation. Finally, we investigated the influence of some personal characteristics, such as age, gender and academic rank at the time of transfer. To do so, we conducted an OLS regression in which the response variable (Y ) is represented by the FSS variation recorded for the mobile academic at the turn of the transfer. The independent variables considered are (cid:129) Gender (X1), specified by a dummy variable (1 for female); (cid:129) Age (X2–4), specified with four classes through three dummies (baseline “40 or less”); (cid:129) Academic rank (X5–6), specified by two dummies (baseline “Assistant professor”); (cid:129) Productivity before transfer (X7), FSS measured for the mobile academic in the 5-year period preceding the transfer; (cid:129) Productivity difference of the destination/origin group (X8) (i.e., difference in the average FSS of the research group (SDS) that the mobile academic joins with respect to the one the academic left); and (cid:129) Career progress (X9), specified by a dummy variable (1 if the transfer is associated with career progression). The results are reported in Table 26 and indicate that gender is not statistically significant. The coefficient of this variable is negative and could indicate that, for a woman, adaptation to the new work environment is more complex than for a man; however, the p-value is very high and, therefore, this conclusion is not supported by statistically significant evidence. Age, on the other hand, shows a significant coefficient, for all classes considered. In particular, all other things being equal, compared to a young academic (under 40 years of age), one between 41 and 46 years old experiences a significant reduction in productivity after the transfer, and the reduction is further exacerbated for even older academics. The data therefore seem to show greater “flexibility” for young people, who are less affected by change than their older colleagues. Regarding academic rank at the time of transfer, the coefficients of the two variables con- sidered are both positive and significant. This indicates that compared to an assistant, the transfer of an associate (and even more in the case of a full professor) is accompanied by 6 The footnote indicates that the analysis is referred to 345 observations, because compared to the initial 568 transfers, we excluded those concerning places of origin/destination with less than one unit of FTE staff in the SSD of the moving professor and 39 transfers relating to professors whose age is not known. Quantitative Science Studies 356 l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u q s s / a r t i c e - p d l f / / / / 3 2 3 4 5 2 0 3 1 8 8 7 q s s _ a _ 0 0 1 9 2 p d . / f b y g u e s t t o n 0 7 S e p e m b e r 2 0 2 3 Academic mobility on research performance Table 2. OLS regression on the possible drivers of FSS variation at the turn of the transfer of academics in the data set _cons Gender Age: 41–46 Age: 47–53 Age: 54 or more Rank: Associate Rank: Full FSS before the move FSS difference of the O/D groups Career progress X1 X2 X3 X4 X5 X6 X7 X8 X9 Coeff. 0.587 −0.165 −0.572 −0.641 −0.961 0.370 0.694 −0.657 0.137 0.201 Std Err. 0.193 0.189 0.234 0.262 0.290 0.223 0.277 0.043 0.069 0.211 t 3.04 −0.87 −2.44 −2.44 −3.31 1.66 2.5 −15.46 1.99 0.95 P > |t|
0.003

0.384

0.015

0.001

0.098

0.013

0.000

0.048

0.341

Number of obs = 345; F(9, 335) = 28.96; Prob > F = 0.000; R-squared = 0.438; Adj R-squared = 0.423; Root MSE = 1.451.

[95% Conf. Interval]
0.207

0.967

−0.537

−1.033

−1.157

−1.532

−0.069

0.149

−0.741

0.001

−0.214

0.207

−0.111

−0.124

−0.390

0.810

1.239

−0.574

0.272

0.616

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

an increase in performance. Perhaps this could be explained by the fact that the higher the
academic rank, the higher the value that the transferred academic brings to the new university
in terms of social, professional, and instrumental capital, which the assistant professor has per-
haps to (re)build after his or her transfer to the new university.

Finally, the data for the last three variables shown in Table 2 confirm what has already
emerged in the univariate analyses presented earlier. The scientific standing of the academic
matured before his or her transfer is not easily maintained and developed downstream of the
transfer: The coefficient of this variable is negative (−0.657) and significant. Similarly, the sig-
nificance of the contextual conditions in which the moving academic operates is confirmed.
The positive and significant value of the variable X8 (0.137) indicates that the transfer can
have a positive link to the academic’s productivity if it takes place to a more productive groups
than the source group, and vice versa. Finally, it is confirmed that any career progress accom-
panying the transfer does not have a significant link to the change in productivity of the
mobile academic.

5. DISCUSSION AND CONCLUSIONS

The mobility of a researcher represents a mechanism of knowledge diffusion, able to generate
relevant spillovers in organizations, sectors, and countries. As it leads to the dissemination of
ideas and new knowledge, mobility is considered an important factor for the development of
research systems and, consequently, of the economies of countries. All this explains the inter-
est of policymakers, who are increasingly engaged in defining and implementing interventions
to facilitate and support mobility, as well as the wealth of studies in the literature aimed at
investigating motivations, drivers, and effects. In this paper, we have focused on this last aspect
and, in particular, on the relationship between mobility and research performance, consider-
ing the confounding effect of the researchers’ personal traits, as well as contextual factors
related to the origin/destination of their transfer. We limited the analysis to transfers between
national universities and therefore to a particular type of mobility (i.e., one within a domestic
domain, the Italian one, strongly regulated and centrally governed). A context in which the

Quantitative Science Studies

357

Academic mobility on research performance

researcher is in fact a civil servant, whose mobility is not determined by incentives typical of
a real labor market, individual organizations being in fact precluded from the possibility of
customizing the job offer (salaries, for example, are only a function of academic rank and
seniority).

Data on Italian mobile academics in the 5-year period 2009–2014 reveal that good
research productivity is not a significant predictor of the likelihood of transfer; rather, the oppo-
site is true. The research performance of mobile academics is quite heterogeneous, but it is the
less productive ones that represent the larger share of those who move, and cases of transfers of
unproductive academics are very numerous. This evidence contrasts with the results of those
who have analyzed the flows of young researchers going out of the country in search of better
opportunities to exploit their potential (Nascia et al., 2021), especially to countries with world-
class universities. This apparent contradiction is partly explained by rigged competitions for
recruitment and career advancement, involving extensive favoritism: Talented young scientists
who cannot enter the higher education system or whose career progress is halted by “favored”
candidates are likely to emigrate to countries that better appreciate and reward merit (Abramo,
D’Angelo, & Rosati, 2014, 2015). National mobility entrusted to researchers without a high
scientific profile, combined with the presence of such consistent outflows and, at the same
time, very modest inflows (less than 1% of academics in Italy are foreigners, with an average
research productivity only slightly higher than that of Italians)7, does not bode well for the
development of the Italian research system.

To complicate the picture, there is the fact that more than half of the academics affected by
mobility worsen their performance after the transfer. This evidence is in line with that part of
the literature that detects significant productivity decreases in the years after job changes, due
to the so-called “adjustment costs” (Fernández-Zubieta et al., 2016), which do not seem to
differ with the “geographical” relevance of the move. The absence of positive effects of mobil-
ity on research performance for the Italian case confirms what emerged from Aksnes et al.
(2013) for the Norwegian academic systems, featured by low levels of mobility and differences
in publication and citation rates of mobile researchers (as compared to other researchers) van-
ishing when demographic characteristics are accounted for. It must be said, however, that
comparability between studies is always made complex by the different methodologies
employed, in terms of both indicators and empirical strategy. In particular, the Norwegian
study involves pure cross-sectional data (mobile vs. nonmobile academics’ performance),
while ours is a longitudinal study (mobile academics’ performance before vs. after transfer).
With respect to the second hypothesis of the paper, the analyses revealed a weak correlation
between the change in performance at the turn of the transfer and the difference in perfor-
mance of the two locations of origin and destination. Thus, it is not possible to establish a
causal link, according to which upward mobility is associated with a significant increase in
performance and vice versa; a link that is indicated in the literature in several papers, but
almost all of which refer to Anglo-Saxon academic systems (Allison & Long, 1990; Fernández-
Zubieta et al., 2016; Tartari et al., 2020; Yan et al., 2020).

Certainly, there is no correlation between the variation in performance of a mobile
researcher and the size of the groups that he or she leaves/joins with his or her transfer, con-
firming the absence of returns to scale in Italian research (Abramo et al., 2012a). It should be
noted that the result may be flawed by the inability to stratify the analyses by subject area,
given the low number of observations. The overall analysis overlooks the possible existence
of area effects and, in particular, the differences existing between scientific fields of an

7 For details, see Abramo, D’Angelo, and Di Costa (2019).

Quantitative Science Studies

358

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

experimental nature and others of a theoretical nature, in terms of technologies, necessary
infrastructural endowments, critical financial mass, etc. (He, Zhen, & Wu, 2019).

The presence of concurrent career advancement in an academic’s transfer does not appear
to impact his or her research performance, at least not in a statistically significant way. Related
to this, the observation of the Italian case seems to return a similar result to that obtained by
Ejermo et al. (2020) in their study on the effects of national mobility on the performance of
Swedish researchers. More precisely, the authors report a positive effect of mobility on some
bibliometric indicators, but not when mobility implies career advancement.

Regarding the confounding effect of the personal characteristics of the mobile researcher,
the multivariate analysis conducted reveals that gender is not a variable that significantly
impacts the variation in productivity at the turn of the transfer, while age is, all others being
equal. Compared to an academic less than 40 years old, an older academic of the same aca-
demic rank shows a significant reduction in productivity after the transfer, a reduction that
increases for the older age groups, indicating a greater “plasticity” in adapting to the new con-
text on the part of young people compared to older colleagues. The effect of academic rank,
all others equal, is also significant and positive. It is plausible that higher academic ranks cor-
respond to infrastructural endowments and social capital that the mobile academic “transfers”
to the destination location with less difficulties. An assistant professor needs more time/costs
to adapt to the new working environment than a full professor, whose transfer is probably
preceded by informal interactions with colleagues in the new university, which speed up
the start-up.

The results of the work represent a relevant stimulus of certain interest for policymaking.
The fact that mobile academics are not the most productive ones reveals the absence of ade-
quate competitive mechanisms in the Italian higher education system, which should elicit
competition among universities to attract talented academics. Furthermore, combining our
results with those by Nascia et al. (2021) showing the brain drain reveals the failure of the
national level competition exams (concorsi) for career progress. The ultimate question is
whether national research mobility should be favored or not, given that it does not seem to
be beneficial for research performance. Of course, there are benefits other than research
performance (e.g. the territorial transfer of knowledge). Future research might investigate
the effects of mobile academics on the research performance of colleagues in the hosting
university.

By nature, this study is inevitably domestic in scope, as the cultural, sociological, and
normative context of the country heavily affects results. Consequently, a certain caution is rec-
ommended in generalizing the results, or even comparing them with those of other national
contexts. Nevertheless, compared to the state of the art, which can be composed through the
merging of contributions typically referring in any case to individual national contexts or
individual fields of interest, the contribution presented here is characterized by the robustness
of the approach used to measure performance, compared instead to the heterogeneity of the
approaches proposed in the literature, especially with reference to the partiality and nonho-
mogeneity of the indicators used, which could, at least in part, explain the nonconvergence of
the evidence that has so far emerged on the subject.

We are aware of a few limits of the study. First of all, the intrinsic limits of evaluative bib-
liometrics approaches apply to this work: Publications are not representative of all knowledge
produced (tacit knowledge is not captured); bibliographic repositories do not cover all
publications; and citations are not always certification of real use and representative of all
use. Furthermore, the scope of the work is limited to national mobility, as information on

Quantitative Science Studies

359

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

international mobility was not available to us. Finally, results might be affected by confounding
variables that we do not consider in our study (e.g. the frequency of mobility, or the variation
in the disciplinary focus that mobility might entail).

AUTHOR CONTRIBUTIONS

Giovanni Abramo: Conceptualization; Methodology; Project administration; Supervision;
Writing—Review & editing. Ciriaco Andrea D’Angelo: Data curation; Formal analysis; Inves-
tigation; Methodology; Validation; Visualization; Writing—original draft. Flavia Di Costa: Data
curation; Formal analysis; Investigation; Visualization; Writing—original draft.

COMPETING INTERESTS

The authors have no competing interests.

FUNDING INFORMATION

The research project received no funding by third parties.

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

DATA AVAILABILITY

Being subject to Clarivate-WoS license restrictions, the raw data cannot be made publicly
available but are available from the authors upon request by the reader for personal interest
only.

REFERENCES

Abramo, G., & D’Angelo, C. A. (2014). How do you define and
measure research productivity? Scientometrics, 101(2),
1129–1144. https://doi.org/10.1007/s11192-014-1269-8

Abramo, G., Cicero, T., & D’Angelo, C. A. (2012a). Revisiting
size effects in higher education research productivity. Higher
Education, 63(6), 701–717. https://doi.org/10.1007/s10734-011
-9471-6

Abramo, G., Cicero, T., & D’Angelo, C. A., (2012b). The dispersion
of research performance within and between universities as a
potential indicator of the competitive intensity in higher educa-
tion systems. Journal of Informetrics, 6(2), 155–168. https://doi
.org/10.1016/j.joi.2011.11.007

Abramo, G., D’Angelo, C. A., & Di Costa, F. (2019). A nation’s
foreign and domestic professors: Which have better research per-
formance? (The Italian case). Higher Education, 77(5), 917–930.
https://doi.org/10.1007/s10734-018-0310-x

Abramo, G., D’Angelo, C. A., & Rosati, F. (2013). The importance
of accounting for the number of co-authors and their order when
assessing research performance at the individual level in the life
sciences. Journal of Informetrics, 7(1), 198–208. https://doi.org
/10.1016/j.joi.2012.11.003

Abramo, G., D’Angelo, C. A., & Rosati, F. (2014). Career advance-
ment and scientific performance in universities. Scientometrics,
98(2), 891–907. https://doi.org/10.1007/s11192-013-1075-8
Abramo, G., D’Angelo, C. A., & Rosati, F. (2015). The determinants of
academic career advancement: Evidence from Italy. Science and
Public Policy, 42(6), 761–774. https://doi.org/10.1093/scipol/scu086
Aksnes, D. W., Rørstad, K., Piro, F. N., & Sivertsen, G. (2013). Are
mobile researchers more productive and cited than non-mobile

researchers? A large-scale study of Norwegian scientists.
Research Evaluation, 22(4), 215–223. https://doi.org/10.1093
/reseval/rvt012

Aksnes, D. W., Piro, F. N., & Rørstad, K. (2018). Does size matter?
An investigation of how department size and other organiza-
tional variables influence on publication productivity and cita-
tion impact. NIFU Working Paper, 2018, 14. https://www.nifu
.no/en/publications/1649778/

Albarrán, P., Carrasco, R., & Ruiz-Castillo, J. (2017). Are migrants
more productive than stayers? Some evidence from a set of
highly productive academic economists. Economic Inquiry,
55(3), 1308–1323. https://doi.org/10.1111/ecin.12430

Allison, P. D., & Long, J. S. (1990). Departmental effects on scien-
tific productivity. American Sociological Review, 55(4), 469–478.
https://doi.org/10.2307/2095801

Archambault, É., Vignola-Gagné, É., Côté, G., Larivière, V., &
Gingras, Y. (2006). Benchmarking scientific output in the social
sciences and humanities: The limits of existing databases. Scien-
tometrics, 68(3), 329–342. https://doi.org/10.1007/s11192-006
-0115-z

Bäker, A. (2015). Non-tenured post-doctoral researchers’ job
mobility and research output: An analysis of the role of research
discipline, department size, and coauthors. Research Policy,
44(3), 634–650. https://doi.org/10.1016/j.respol.2014.12.012
Bauer, H. P., Schui, G., von Eye, A., & Krampen, G. (2013). How
does scientific success relate to individual and organizational
characteristics? A scientometric study of psychology researchers
in the German-speaking countries. Scientometrics, 94(2),
523–539. https://doi.org/10.1007/s11192-012-0760-3

Quantitative Science Studies

360

Academic mobility on research performance

Bolli, T., & Schläpfer, J. (2015). Job mobility, peer effects, and
research productivity in economics. Scientometrics, 104(3),
629–650. https://doi.org/10.1007/s11192-015-1625-3

Breschi, S., Lenzi, C., Lissoni, F., & Vezzulli, A. (2010). The geog-
raphy of knowledge spillovers: The role of inventors’ mobility
across firms and in space. In R. Boschma & R. Martin (Eds.), The
handbook of evolutionary economic geography (pp. 353–369).
C h i c h e s t e r : E d w a r d E l g a r. h t t p s : / / d o i . o r g / 1 0 . 4 3 3 7
/9781849806497.00025

Cañibano, C., D’Este, P., Otamendi, F. J., & Woolley, R. (2020).
Scientific careers and the mobility of European researchers:
An analysis of international mobility by career stage. Higher
Education, 80(6), 1175–1193. https://doi.org/10.1007/s10734
-020-00536-z

Cañibano, C., Otamendi, J., & Andújar, I. (2008). Measuring and
assessing researcher mobility from CV analysis: The case of the
Ramón y Cajal programme in Spain. Research Evaluation, 17(1),
17–31. https://doi.org/10.3152/095820208X292797

Ciccone, A., & Peri, G. (2006). Identifying human-capital externali-
ties: Theory with applications. Review of Economic Studies, 73(2),
381–412. https://doi.org/10.1111/j.1467-937X.2006.00380.x
Cruz-Castro, L., & Sanz-Menéndez, L. (2010). Mobility versus job
stability: Assessing tenure and productivity outcomes. Research
Policy, 39(1), 27–38. https://doi.org/10.1016/j.respol.2009.11
.008

D’Angelo, C. A., Giuffrida, C., & Abramo, G. (2011). A heuristic
approach to author name disambiguation in bibliometrics
databases for large-scale research assessments. Journal of the
American Society for Information Science and Technology,
62(2), 257–269. https://doi.org/10.1002/asi.21460

De Filippo, D., Casado, E. S., & Gómez, I. (2009). Quantitative
and qualitative approaches to the study of mobility and scien-
tific performance: A case study of a Spanish university. Research
E v a l u a t i o n , 1 8 ( 3 ) , 1 9 1 – 2 0 0 . h t t p s : / / d o i . o rg / 1 0 . 3 1 5 2
/095820209X451032

Deville, P., Wang, D., Sinatra, R., Song, C., Blondel, V. D., &
Barabási, A.-L. (2014). Career on the move: Geography, stratifica-
tion and scientific impact. Scientific Reports, 4(1), 1–7. https://doi
.org/10.1038/srep04770, PubMed: 24759743

Dubois, P., Rochet, J. C., & Schlenker, J. M. (2014). Productivity and
mobility in academic research: Evidence from mathematicians.
Scientometrics, 98(3), 1669–1701. https://doi.org/10.1007
/s11192-013-1112-7

El-Ouahi, J., Robinson-García, N., & Costas, R. (2021). Analyzing
scientific mobility and collaboration in the Middle East and
North Africa. Quantitative Science Studies, 2(3), 1023–1047.
https://doi.org/10.1162/qss_a_00149

Ejermo, O., Fassio, C., & Källström, J. (2020). Does mobility across
universities raise scientific productivity? Oxford Bulletin of Eco-
nomics and Statistics, 82(3), 603–624. https://doi.org/10.1111
/obes.12346

European Commission Directorate-General for Research and Inno-
vation. (2021). MORE4: Support data collection and analysis
concerning mobility patterns and career paths of researchers:
Annexes to the final report. Publications Office. https://data
.europa.eu/doi/10.2777/117072

Fernández-Zubieta, A., Geuna, A., & Lawson, C. (2016). Productiv-
ity pay-offs from academic mobility: Should I stay or should I go?
Industrial and Corporate Change, 25(1), 91–114. https://doi.org
/10.1093/icc/dtv034

Franzoni, C., Scellato, G., & Stephan, P. (2014). The mover’s
advantage: The superior performance of migrant scientists. Eco-
nomics Letters, 122(1), 89–93. https://doi.org/10.1016/j.econlet
.2013.10.040

Gyorffy, B., Nagy, A., Herman, P., & Torok, A. (2018). Factors
influencing the scientific performance of momentum grant
holders: An evaluation of the first 117 research groups. Sciento-
metrics, 117(1), 409–426. https://doi.org/10.1007/s11192-018
-2852-1, PubMed: 30220748

Guthrie, S., Lichten, C., Harte, E., Parks, S., & Wooding, S. (2017).
International mobility of researchers. Santa Monica, CA and
Cambridge, UK: RAND Europe. https://doi.org/10.7249/RR1991
Halevi, G., Moed, H. F., & Bar-Ilan, J. (2016). Researchers’ mobility,
productivity and impact: Case of top producing authors in seven
disciplines. Publishing Research Quarterly, 32(1), 22–37. https://
doi.org/10.1007/s12109-015-9437-0

He, Z., Zhen, N., & Wu, C. (2019). Measuring and exploring the
geographic mobility of American professors from graduating
institutions: Differences across disciplines, academic ranks, and
genders. Journal of Informetrics, 13(3), 771–784. https://doi.org
/10.1016/j.joi.2019.05.001

Hicks, D. (1999). The difficulty of achieving full coverage of inter-
national social science literature and the bibliometric conse-
quences. Scientometrics, 44(2), 193–215. https://doi.org/10
.1007/BF02457380

Horta, H., Birolini, S., Cattaneo, M., Shen, W., & Paleari, S. (2021).
Research network propagation: The impact of phd students’ tem-
porary international mobility. Quantitative Science Studies, 2(1),
129–154. https://doi.org/10.1162/qss_a_00096

Horta, H., Jung, J., & Santos, J. M. (2020). Mobility and research
performance of academics in city-based higher education sys-
tems. Higher Education Policy, 33, 437–458. https://doi.org/10
.1057/s41307-019-00173-x

Horta, H., Meoli, M., & Santos, J. M. (2022). Academic inbreeding
and choice of strategic research approaches. Higher Education
Quarterly, 76, 76–101. https://doi.org/10.1111/hequ.12328

Jongbloed, B. (2004). Regulation and competition in higher educa-
tion. In P. Teixeira, B. Jongbloed, D. Dill, & A. Amaral (Eds.),
Markets in higher education. Higher education dynamics, vol. 6.
Dordrecht: Springer. https://doi.org/10.1007/1-4020-2835-0_5
Jonkers, K., & Tijssen, R. (2008). Chinese researchers returning
home: Impacts of international mobility on research collabora-
tion and scientific productivity. Scientometrics, 77(2), 309–333.
https://doi.org/10.1007/s11192-007-1971-x

Nascia, L., Pianta, M., & Zacharewicz, T. (2021). Staying or leav-
ing? Patterns and determinants of Italian researchers’ migration.
Science and Public Policy, 48(2), 200–211. https://doi.org/10
.1093/scipol/scab007

Netz, N., Hampel, S., & Aman, V. (2020). What effects does inter-
national mobility have on scientists’ careers? A systematic
review. Research Evaluation, 29(3), 327–351. https://doi.org/10
.1093/reseval/rvaa007

Pezzoni, M., Sterzi, V., & Lissoni, F. (2012). Career progress in cen-
tralized academic systems: Social capital and institutions in
France and Italy. Research Policy, 41(4), 704–719. https://doi
.org/10.1016/j.respol.2011.12.009

Ryazanova, O., & McNamara, P. (2019). Choices and consequences:
Impact of mobility on research-career capital and promotion in
business schools. Academy of Management Learning & Education,
18(2), 186–212. https://doi.org/10.5465/amle.2017.0389

Sanz-Menéndez, L., Cruz-Castro, L., & Alva, K. (2013). Time to
tenure in Spanish universities: An event history analysis. PLOS
ONE, 8(10), e77028. https://doi.org/10.1371/journal.pone
.0077028, PubMed: 24116199

Song, J., Almeida, P., & Wu, G. (2003). Learning-by-hiring: When is
mobility more likely to facilitate interfirm knowledge transfer?
Management Science, 49(4), 351–365. https://doi.org/10.1287
/mnsc.49.4.351.14429

Quantitative Science Studies

361

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d

f
/

3
2
3
4
5
2
0
3
1
8
8
7
q
s
s
_
a
_
0
0
1
9
2
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Academic mobility on research performance

Tartari, V., Di Lorenzo, F., & Campbell, B. A. (2020). “Another roof,
another proof”: The impact of mobility on individual productivity
in science. Journal of Technology Transfer, 45(1), 276–303.
https://doi.org/10.1007/s10961-018-9681-5

van der Wouden, F., & Rigby, D. L. (2021). Inventor mobility and
productivity: A long-run perspective. Industry and Innovation,

28(6), 677–703. https://doi.org/10.1080/13662716.2020
.1789451

Yan, E., Zhu, Y., & He, J. (2020). Analyzing academic mobility of
US professors based on ORCID data and the Carnegie Classifi-
cation. Quantitative Science Studies, 1(4), 1451–1467. https://
doi.org/10.1162/qss_a_00088

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
q
s
s
/
a
r
t
i
c
e
–
p
d