Early career researchers and their publishing and authorship practices

Abstract

This study presents findings from the first year of the Harbingers research
project, a 3-year longitudinal study of early career researchers (ECRs),
which sought to ascertain current and changing habits in scholarly communication.
The study recruited 116 science and social science ECRs from
seven countries who were subject to in-depth interviews, and this paper
reports on findings regarding publishing and authorship practices and attitudes.
A major objective was to determine whether ECRs are taking the
myriad opportunities proffered by new digital innovations, developing
within the context of open science, open access, and social media, to publish
their research. The main finding is that these opportunities are generally
not taken because ECRs are constrained by convention and the
precarious employment environment they inhabit and know what is best
for them, which is to publish (in high impact factor journals) or perish.

Full text

RESEARCH ARTICLE
(wileyonlinelibrary.com) doi: 10.1002/leap.1102 Received: 11 December 2016 | Accepted: 7 March 2017
Early career researchers and their publishing and authorship
practices
David Nicholas ,
1
*Blanca Rodríguez-Bravo ,
2
Anthony Watkinson ,
1
Cherifa Boukacem-Zeghmouri ,
3
Eti Herman ,
1
Jie Xu ,
4
Abdullah Abrizah ,
5
and
Marzena Świgo
n
6
1
CIBER Research Ltd, Newbury, Berkshire
RG147RU, UK
2
Biblioteconomía y Documentación, Universidad de
León, 24071 León, Castilla y León, Spain
3
Department of Computer Science, Université de Lyon,
Université Lyon 1, 69100 Villeurbanne, France
4
School of Information Management, Wuhan
University, Wuhan, Hubei 430072, China
5
Department of Library & Information Science, Faculty
of Computer Science & Information Technology,
University of Malaya, 50603 Kuala Lumpur, Malaysia
6
WydziałHumanistyczny, Uniwersytet Warminsko-
Mazurski, 10-719 Olsztyn, Poland
ORCID:
D. Nicholas: 0000-0001-8046-2835
B. Rodríguez-Bravo: 0000-0002-9476-7602
A. Watkinson: 0000-0002-2317-6557
C. Boukacem-Zeghmouri: 0000-0002-0201-6159
E. Herman: 0000-0001-8526-9081
J. Xu: 0000-0002-9820-8066
A. Abrizah: 0000-0002-8224-5268
M. Świgo
n: 0000-0003-3600-8349
*Corresponding author: David Nicholas
E-mail: Dave.Nicholas@ciber-research.eu
Abstract
This study presents findings from the first year of the Harbingers research
project, a 3-year longitudinal study of early career researchers (ECRs),
which sought to ascertain current and changing habits in scholarly com-
munication. The study recruited 116 science and social science ECRs from
seven countries who were subject to in-depth interviews, and this paper
reports on findings regarding publishing and authorship practices and atti-
tudes. A major objective was to determine whether ECRs are taking the
myriad opportunities proffered by new digital innovations, developing
within the context of open science, open access, and social media, to pub-
lish their research. The main finding is that these opportunities are gener-
ally not taken because ECRs are constrained by convention and the
precarious employment environment they inhabit and know what is best
for them, which is to publish (in high impact factor journals) or perish.
AIMS AND OBJECTIVES
This paper investigates the publishing and authorship practices of
early career researchers (ECRs), the newest and biggest wave of
scholarly researchers. In doing so, it sets out to fill a long-felt gap
in the state of knowledge concerning the practices characterizing
ECRs’tackling this crucial aspect of their initiation into the pro-
fession. To date, knowledge on the topic is derived mostly from
the few major quantitative studies that have investigated young
researchers as part of the research population as a whole to see
how different or similar they are. CIBER studies on social media
use (Nicholas & Rowlands, 2011; Rowlands, Nicholas, Russell,
Canty, & Watkinson, 2011) and trustworthiness (Jamali et al.,
2014; Nicholas et al., 2014; Nicholas, Watkinson et al., 2015;
Nicholas, Jamali et al., 2015; Tenopir et al., 2015) and Tenopir
et al.’s (2016) study into the motivations of academic authors in
selecting a journal in which to publish are part of this camp. Mül-
ler’s (2012) interviews-based qualitative exploration of the role
Learned Publishing 2017 www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
1

that authorship plays within the everyday contexts of young
researchers’collaborative work is the closest to the study pre-
sented here, albeit more limited in scope (one country, 15 post-
docs and just life sciences).
Setting out to draw a more comprehensive and detailed pic-
ture of ECRs’research dissemination practices, we sought to dis-
cover how productive they are in publishing terms, what the
authorship policies/practices they are subject to are and which
they might have issues with, what influence they have on where
research is published, whether they have a publishing strategy,
and –if so –what it is. We also wanted to establish whether
ECRs are taking the myriad opportunities proffered by new digital
innovations, developing within the context of open science, open
access (OA), and social media, in order to publish their research.
Or are they shackled by conventions within their discipline or by
the pressures within their research group? In this regard, we are
interested in: (1) practices, preferences, and policies regarding OA
publishing; (2) whether ECRs are using the new social technolo-
gies to disseminate their research more informally; and
(3) whether they make their data available/share it.
The data upon which the paper is built come from the first
year of a 3-year longitudinal study of ECRs, which sought to
establish whether they are going to be the harbingers of change
when it comes to scholarly communications (http://ciber-
research.eu/harbingers.html). This paper constitutes one of the
foundation stones for the project.
SCOPE AND DEFINITIONS
There are various definitions of ECRs (Poli, 2016; for more detail
see http://ciber-research.eu/download/20160901-Harbingers-
ECRs_literature_review.pdf), and they vary from country to coun-
try. After an examination of the literature and consultations with
international partners, this definition was adopted:
Researchers who are generally not older than 35 and who
either have received their doctorate and are currently in a
research position or have been in research positions, but
are currently doing a doctorate. In neither case are they
researchers in established or tenured positions. In the case
of academics, they are non-faculty research employees of
the university. In addition, included were a small number
of ECRs who had come from or were in positions that
were essentially servicing clinicians or researchers.
The focus is on ECRs in the sciences and social sciences,
which is where the funders’(Publishing Research Consortium)
main priorities lie and also where the majority of ECRs are based.
A wide geographical reach was sought as we wished to support
research on issues facing the STM industry globally. Balancing
the need for representativeness (with regard to size, importance,
level of development, and language) with Publishing Research
Consortium national interests and the availability of interviewers
on the ground, China, France, Malaysia, Poland, Spain, UK, and
USA were selected.
METHODOLOGY
Full details can be found in an article published in an earlier issue
of this journal (Nicholas et al., 2017), with an abbreviated version
provided here. Long, structured interviews were conducted face-
to-face and/or remotely (Skype or telephone). A detailed inter-
view schedule was compiled and sent to interviewees ahead of
the interview. The interview schedule contained around 60 ques-
tions, and the interview took between 60 and 90 minutes to
complete. For the purposes of this paper, we are concerned with
11 of the questions:
Q1. What contributions have you made to the papers, which
you have co-authored?
Q2. Does your research team/department/university have an
authorship policy?
Q3. Would you do things differently if you had a say in this?
Q4. What influence (if any) have you had on the choice of
journal in which to publish your research?
Q5. Does your research team/department/university/funder
have a policy in regard to OA publishing?
Q6. What do you think are the advantages and disadvan-
tages of OA publishing from the point of view of the author?
Q7. Do you think OA publishing advances science and
research, or are you worried that it will dilute the quality of publi-
cations, or do you agree/disagree with both propositions? If
so, how?
Q8. Do you have a preference for journals with innovative
features, such as video articles (e.g. Jove), when placing your
research?
Q9. Is there pressure on you to publish in particular top-
ranked journals and, if so, how do you think this affects scholarly
communications, in general, and your career?
Key points
•Early career researchers (ECRs) are constrained by conven-
tion and the precarious environment they inhabit and
know that to publish in top journals is the right thing to do
for their careers.
•ECRs voice few complaints about authorship policies and
are often influential in choosing where to publish.
•When selecting journals in which to publish, being indexed
by Scopus and Web of Science counts most highly.
•Being innovative or open does not make journals more
attractive for ECRs.
•ECRs do see the potential of social media and community
platforms as valuable publishing outlets.
•ECRs think open access publishing is a good idea, but they
do not practice what they believe.
2D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

Q10. Do you have a conscious publication strategy relating
to your research and is that to do with obtaining a tenured/
established position and, if so, please describe?
Q11. Would you prefer to make public your research findings
in less formal ways, such as blogs, which could make them more
visible?
Data have also been sourced from the CVs of ECRs, which
they provided ahead of the interviews, and from elsewhere in the
interview schedule. The full list of questions can be found on the
CIBER website (http://ciber-research.eu/downloads/20160916-
harbingers-research_instruments.pdf).
Interviews were conducted by national interviewers, largely
in their own languages. The proceedings of the interviews were
taken down in note form. A transcript of the interview was
returned to the interviewee for validation. The record was then
translated into English for all non-English-speaking countries and
then manually coded using a heuristic approach and a standar-
dized thematic framework (see http://ciber-research.eu/
downloads/20160916-harbingers-research_instruments.pdf).
A total of 116 ECRs from the sciences and social sciences
were recruited from seven countries (Table 1). To reach this
number, interviewers for the case countries were given a
recruitment quota of 20–29 for the UK and USA (the larger
number a reflection on the importance of these communities to
STM publishers) and 10–15 for the other countries. Within
countries, the general guidance was to build the sample along
the following lines: (1) two-thirds science and one-third social
sciences, (2) a representative balance of men and women, and
(3) a range of ages within the 20s and 30s age groups. ECRs
came from 81 institutions.
Details of recruitment methods can be found in the afore-
mentioned article (Nicholas et al., 2017).
RESULTS
The findings presented here are based on three broad sets of
data, which are woven together to create the narrative:
•Information from the 116 ECRs, which is variously summar-
ized, paraphrased, and directly quoted. This is what ECRs say
they behave like and how they view things, and it is not neces-
sarily wholly informed. Its value is that it shows us what they
think.
•Contextual comments from the national interviewers –all sen-
ior researchers themselves, who provide important background
data in order to understand what was being said and why.
•Findings from the published literature, which provide support
and authority for the information above.
Publishing productivity
Journal publishing is the main medium by which ECRs dissemi-
nate research. This is because it is by paper publishing that they
are judged, given tenure, and promoted. This is also true for their
more established colleagues (Borrego & Anglada, 2016; Nicholas,
Herman, & Jamali, 2015; Nicholas, Watkinson et al., 2015; Van
Dalen & Henkens, 2012; Wolff, Rod, & Schonfeld, 2016a,
2016b), but ECRs are, arguably, even more driven because of
their precarious job position (Müller, 2012, 2014a, 2014b).
TABLE 1 Country-wise ECR profiles broadly compared.
Country No. Subject Gender Age PhD Institutions
China 13 Science: 70% Female: 46% Twenties: 46% 8% Doctoral students 6
Soc.Sci.:30% Male: 54% Thirties: 54%
France 14 Science: 82% Female: 35% Twenties: 65% 100% Postdocs 4
Soc.Sci.: 18% Male: 65% Thirties: 35%
Malaysia 12 Science: 58% Female: 50% Thirties: 100% 100% Postdocs 5
Soc.Sci.: 42% Male: 50%
Poland 10 Science: 80% Female: 40% Twenties: 40% 50% Doctoral students 1
Soc.Sci.: 20% Male: 60% Thirties: 60%
Spain 18 Science: 78% Female: 44% Twenties: 40% 28% Doctoral students 16
Soc.Sci.: 22% Male: 56% Thirties: 60%
UK 21 Science: 62% Female: 38% Twenties: 24% 33% Doctoral students 21
Soc.Sci.: 38% Male: 62% Thirties: 76%
USA 28 Science: 79% Female: 41% Twenties: 27% 34% Doctoral students 28
Soc.Sci.: 21% Male: 59% Thirties: 73%
ECR, Early career researcher.
3ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org

Between them, the 116 ECRs published an impressive 1,178 jour-
nal articles in peer reviewed journals over their short research
lifetime, with the average ECR having published over 10 papers.
After papers, conferences are a popular venue utilized by
ECRs for delivering research findings. Conferences are no less
important for more senior researchers (Watkinson et al., 2016;
Wolff et al., 2016a, 2016b), but for ECRs, the rewards of dissemi-
nating the results of their work in this way are greater: confer-
ences are generally easier to get into and therefore offer a useful
proving ground and a leg up the publishing ladder. There is a hier-
archy of conference contributions of course, presentations
(speakers, keynotes, etc), publishing in proceedings, and posters.
Altogether, ECRs produced 711 published proceedings, most of
which –crucially for them –are indexed in the Web of Science
(WoS) and Scopus. Conferences count more in certain disciplines,
such as computer science and physics, as well as in certain coun-
tries. Thus, in Poland, partly because of the language barrier faced
when publishing in high impact factor (IF) journals and partly
because they obtain reputational points for them, conferences
and presentations/posters are much favoured. Malaysian ECRs
also publish widely in conference proceedings, encouraged by the
fact that participation in conferences attracts institutional finan-
cial support for ECRs from research-intensive universities. How-
ever, in Malaysia, unlike papers, conference papers do not count
towards reputational scores.
AUTHORSHIP
As Müller (2012) points out, today’s ECRs deal with the demand
to be both a highly competitive individual and a member of
research teams. With publications, as important for ECRs’career
development as they are, questions of authorship inevitably play
a pivotal role within these negotiations of collaboration and com-
petition. Nevertheless, ECRs’authorship experiences align with
what we know of academic authorship in general, which is hardly
surprising given that publications establish reputation, signal pro-
ductivity, and serve as yardsticks in deliberations of appoint-
ments, promotions, and funding for senior and novice researchers
alike. Thus, for example, the order in which author names appear
in the by-line of an article, assumed as it is to reflect the indivi-
dual’s level of contribution to both research and manuscript
development, may be hotly contested among members of a
research team, be they established researchers or ECRs (Brand,
Allen, Altman, Hlava, & Scott, 2015; Dance, 2012; Elliott et al.,
2017; Kornhaber, McLean, & Baber, 2015; Watkinson et al.,
2016). Indeed, the majority of our ECRs in all countries say they
are aware of –although not invariably happy with –their univer-
sity’s or department’s formal/informal authorship policies as to
who can be named as authors and the order in which they are
named. However, there are differences between countries in
regard to how formal or informal policies and practices are.
In China, where the ranking of authors is viewed as being
very important –counting as it does towards reputation and pro-
motion –there are very strict rules. Thus, the first author is the
person who undertakes most of the research and writes most of
the paper. To complicate matters somewhat, there can be joint
first authors where each joint first author splits the normal first
author’s reputational credit. The corresponding author’s contribu-
tion is considered almost the same as the first author, so most
institutes give them similar reputational credit. A collaborating
author, the third authorship rank, is the researcher who assists
the research by collecting and analysing data, and translating and
polishing the paper. Minor contributions might merit the reward
of authorship. Despite these formal arrangements, ECRs report
encountering problems with coauthorship, which arise from
group work. The sharing of credit is a particularly thorny problem
in domestic collaborations because Chinese ECRs so desperately
want to be the first or corresponding author, so their work can
be recognized by their institutes and funders. Complaints also
arise about mentors who do not shoulder research responsibilities
but insist on being listed as an author. Most interviewees have
published papers on their research with their mentors who are
listed as first authors. As we shall learn, the authorship problems
young Chinese researchers encounter are in fact globally
experienced.
In Malaysia, the rules of authorship are fairly cut and dry: To
be listed as an author, we must meet these criteria: you must be
involved in project design and research protocols, collect data and
analyze results, and participate in drafting and writing the paper.
They are also transparent: rules are now polished and published,
the problem with who should sit in the author list is not a big issue
in my field. Order is decided like this: To earn authorship, all in the
team must be intellectually engaged and that the order is decided by
the corresponding author. Authorship policy can vary, however, by
discipline. For instance:
•In my lab, the author list is strictly ranked. The top researcher is
normally at the end of the list, and sometimes the person who
owns the lab, owns the research grant is also given credit as
author. The person who actually did the work goes first, nor-
mally the research student or the postdoc. The research super-
visors came after the student authors (ECR, sciences).
•I am always the first author if I write the paper entirely, but I
will always put in my supervisor’s name, although I work on
my own, I don’t rely on him anymore, but I always, invite him
to give input (ECR, social sciences).
In France, things are less prescriptive, with French ECRs
saying that there is no authorship policy as such, but there are
informal rules that are accepted by the community and differ
according to broad disciplines, as in Malaysia. For mathemati-
cians, research is often solitary, and papers can have just one
author, but where this is not the case, alphabetical order is the
rule as all contributions are held to be equal. For most of the
others (biology, physics, computer science and chemistry, and
interdisciplinary research), the main contributor is the first
author. The head of the research group who obtains the funds
is the last author. Between the first and the last author, in the
4D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

‘middle’as it is called in French, come the other contributors
who undertook experiments or specific tasks, and they are
mainly displayed in alphabetical order. All ECRs feel comfortable
with these rules, and most would not change anything. A few
did suggest changes, and they mostly concern packing people
into the author statement: Provide the precise contribution of
every author in order to know who did what and avoid ‘parasites’.
Stop adding too many people.
At Polish universities as well, there are only unwritten and
variable practices. Every group, department, and so on has their
own house rules. Sometimes, mentors or heads of the groups are
placed in first position, but in other cases, they are simply contri-
buting authors. It all depends on the individual or group/depart-
ments. There are few complaints among ECRs, other than not
wanting professors to be added as a matter of course to their
publications, even if they did not contribute to them.
It is a similar situation in Spain where there are no fixed
authorship policies. Group heads decide on their own criteria.
However, there are typical practices. As in France, they differ
by discipline. In STM fields, all the members of the group appear
as authors. First position goes to those who did the experimen-
tal work and last position and the status of corresponding
author goes to mentors. In the social sciences, where ECRs are
part of groups, usually small ones, mentors or heads of the
groups are usually placed in first position. Again, the main con-
cern of ECRs is that too many people are seen to be taking a
free ride on the back of ‘their’papers. Thus, eight ECRs –mak-
ing the Spanish the most critical group –said they would do
things differently if they were fully in charge and that they
would not include those who did not work directly on the
paper, and the order would reflect the importance of the
contribution made.
In the UK, very few (two) ECRs admit to having, or being
aware of, an authorship policy that governed or guided their
publication efforts. A total of 15 say there is no policy, and
another 4 are not sure but not aware of one. This, of course,
does not mean they have loose working arrangements, and
these matters are discussed. Typically, they were listed first on
more than half their articles. The deal was that if you wrote it
and did most of the work, you went first. A few were sole
authors. There is little sense that they are very junior partners.
Asked whether they would do things differently if it is up to
them, six say they would, and their suggestions include collabo-
rate rather than compete; authors must have made a contribu-
tion (senior people get listing when it is not merited); order should
be according to the amount of input; and the main contributor
should come first.
A far higher proportion was aware of policies in the USA,
with 12 out of 28 saying they have one. These policies originate
from a whole range of places, including the laboratory, depart-
ment, university, and professional associations. For example, one
ECR reports that the research group used the National Science
Federation’s (NSF) responsible conduct of research guidelines,
and another group said that Institute of Medicine guidelines are
the default. However, several ECRs report difficulties in a
multidisciplinary group where a disciplinary set of standards,
which they are used to, was different from the set of standards
accepted by the rest of the group. These examples are taken
from replies made by ECRs who did not know of a general
policy.
In spite of the wording of the question, the comments asso-
ciated with practices in the UK and the USA suggest that just
because there was no author policy at a university or depart-
mental level, there was a policy or at least a norm governing
the order of authors and related conventions at group level
even if it was written down nowhere.
As we have seen, being first author is seen as an important
reputational asset for ECRs, a sign of how important their contri-
bution is, and an indicator as to whether the journal choice is
theirs or not. So, how often are ECRs first authors? In fact, sur-
prisingly perhaps, for ECRs to be first author is, on the whole, not
that difficult. Thus, it is normal for them to be first author on any
scholarly outputs, including papers, based on their dissertation.
Most principal investigators (PIs) allow postdocs to be first
authors when they are the ones who have undertaken most of
the research for a paper. It can also depend on country, and
China is an interesting and complex case. Thus, in China, most
universities acknowledge that postdocs can be first authors if
they put their supervisors’name before theirs! So, an ECR might
be the first author who did the main research, but their name is
placed second, a case of having two first authors. In most cases,
ECRs do this to please their mentors or –because their mentors
are more influential and well-known –in the belief that it will
help them to get published.
ECRs are, typically, first authors in around one-third to one-
half of all the papers to which they contribute, but it can vary,
and in France, the proportion is more like 51–75%.
JOURNAL CHOICE
ECRs’replies to the question What influence (if any) have you had
on the choice of journal? indicated that, often, they do have a sig-
nificant influence, with, for instance, 11 out of 14 French, 17 out
of 18 Spaniards, 10 out of 13 Chinese and 8 out of 10 Polish
ECRs saying they did. Influence seems to be a little less in the UK
and USA however, where a quarter say they had no influence at
all. The extent of influence depends on several things:
•Colleagues: Other members of the group to which they belong
and their supervisors/mentors. It is a well-known feature of
ECR life that their choices must suit the other parties involved
in their undertakings (Brechelmacher, Park, Ates, & Campbell,
2015; Cusick, 2015). Thus, for the great majority of those
ECRs in the USA and the majority in the UK who research in a
group, journal choice is a group decision. Where ECRs are not
in a group, as is, for example, the case for many social scien-
tists, they do make decisions on their own but not always as
there are always supervisors and mentors to consult. In the
5ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org

case of multidisciplinary research, the choice is more that of
the person in whose discipline they wish to publish the
research. In France, the choice of a journal is always the result
of a ‘group’discussion. The journal chosen is the result of a
consensus between the ECRs and their colleagues, supervi-
sors, and the head of the department. During this discussion,
the experience and the knowledge of every member contri-
butes to ensuring that the right and the best journal is
selected. Some French (and Polish) ECRs did not want to make
the choices because they fear they are too young and too
inexperienced to take the right decision. It is a precarious thing
to choose a journal: not too high, not too low.As I lack experi-
ence, I prefer to let senior researchers to choose. It is also true of
one UK ECR (and several Spaniards) as well, who said, Always
take advice from people who know best, even when first author.
•Underscoring previous evidence on the strategic importance
accorded to publishing in high IF journals (Borrego & Anglada,
2016; Pepermans & Rousseau, 2014; Rowlands & Nicholas,
2005), which holds true for novice researchers just as much as
for their senior counterparts (Müller, 2012; Nicholas, Jamali
et al., 2015), the universal directive to do so is found to be the
overriding consideration for our ECRs. Add to this the assess-
ment criteria, mandates, and lists that abound, and we find
that the choice is constricted. In several countries, they must
refer to lists of acceptable journals. A case in point is Poland,
where ECRs are formally directed towards a government list
(moreover, the top of the list) and appear to have limited room
for manoeuvres. This is also the case in China, France, Malay-
sia, and Spain, but their lists tend to be the proprietary ones,
such as the journals indexed by the WoS and Scopus.
It is difficult to imagine ambitious ECRs in the UK and the
USA even considering submitting to a journal not indexed in
WoS. In these countries, ECRs appear to have more freedom,
but this is still only relative. The question was not directly
asked in the interview, and the role of indexing was not raised
by any of the interviewees, but an examination of publications
listed in the CVs shows that every one of them is WoS-listed.
In these countries, ECRs appear to have more freedom in the
sense that there was no mention of prescriptive lists, but that
is still only relative. In the UK, the Research Excellence Frame-
work (REF) exerts a special pressure on all researchers, but in
practice, as far as choice of journals is concerned, ECRs
believe this pressure only reinforces the existing pressure to
go for top journals. This is in spite of the fact that the REF
(UK REF) documentation makes it clear that the article itself is
assessed, not the journal in which it is published; such asser-
tions are just not believed.
•ECR choice: The ability of an ECR to select the journal depends
on status, a finding that echoes Müller’s (2012) observation
that publications have different career value to different
authors according to their position in the list of authors, with
first authorship being by far the most valuable career asset for
young scientists. Thus, in the case of Chinese ECRs, if they are
the first author, they have a lot of influence; if they are the
corresponding author, not much influence, but they can make
suggestions. If they are simply a coauthor, they have no influ-
ence at all. It is not the same in France. So, while French ECRs
are first author in the cases where they do the most work, they
still do not have the main choice as to where to publish as that
decision is very much that of their supervisor or head of the
research team. French ECRs appear not to mind as they see
their mentors/supervisors helping them publish in top journals
and mainly because they are first authors. Indeed, they regard
the practice as a kind of ‘help’or a chain of solidarity between
those who are in a secured position and those who are not. In
the UK/USA, it is sometimes the practice for the principal
investigator of the research project to have the choice,
although the ECRs disapprove of this practice.
Criteria used when determining where
to publish
When they are in a position to choose or influence where their
research is published, ECRs’overriding preference is for top jour-
nals, that is, journals perceived to be of high quality and stellar
reputation, which in academe translates to high IF (Nature Pub-
lishing Group and Palgrave Macmillan, 2015). Regarding this, as
Müller (2012) and Nicholas, Jamali et al., 2015 find, ECRs follow
in the footsteps of their more experienced colleagues, whose list
of priorities for choosing a publication venue has repeatedly been
found to feature high journal quality and reputation (Nature Pub-
lishing Group and Palgrave Macmillan, 2015; Tenopir et al., 2016;
Wolff et al., 2016a, 2016b). Interestingly, however, both Watkin-
son et al. (2016) and Tenopir et al. (2016) find senior faculty rate
IF as slightly less important than their younger colleagues, a phe-
nomenon that can be explained by the inexperienced researchers’
greater reliance on quantifiable metrics in selecting appropriate
publication outlets. This is very much in line with Nicholas, Jamali
et al., 2015 conclusion in their study that younger researchers
rely more on trust markers and proxies, such as citations.
In addition to looking for markers of journal quality and repu-
tation, ECRs employ a wide and overlapping range of criteria
(Table 2), which –although differing by country –are mostly tra-
ditional, with ‘indexed and rated in WoS or Scopus’being the
main criteria for all. Still, not all things traditional hold sway;
although researchers have been found to prefer journals with
highly regarded editors over journals with unknown editors
(Rousseau & Rousseau, 2012), for our ECRs, the editor and edito-
rial board count very little. This is probably explained by the fact
that ECRs are quite ignorant of the publisher background
(Nicholas et al., 2017). Non-traditional criteria, such as being
innovative (e.g. Journal of Visualized Experiments) or open (with
respect to access and peer review), do not count much either.
Chinese ECRs, as a matter of course, check whether the jour-
nal is indexed in Science Citation Index Expanded (SCIE), Social
Science Citation Index (SSCI), Arts & Humanities Citation Index
(A & HCI), and Engineering Village, or Chinese indexes, such as
Chinese Social Sciences Citation Index and Chinese Science
6D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

Citation Index. This is because ECRs need to reach the particular
assessment requirements of their institutes, which can be very
demanding, and the standing and rank or impact factor of the
journal is very important. Second, they will determine whether
the journal is related to their research field, and within this, they
generally prefer journals that are specific to their ‘small field’
because that way they can have maximum impact. However, a
few do opt for the journal with the highest journal impact factor
(IF) and a big rejection rate (this is a desirable factor) because
they want to ‘challenge’themselves.
French ECRs are similar but a little more pragmatic. Whether
it is their choice or that of their supervisors, they select the jour-
nal based on its relevance to the topic and its IF. In some cases,
the fact that a journal is ranked in the first quartile (WoS) is an
important criterion for ECRs, their supervisors, and the head of
department. In France, publishing in ‘indexed journals’, an impor-
tant reputational requisite, is implicitly understood as indexed in
WoS, which has a unique position in France, especially for eva-
luators. A journal indexed in Scopus (and not indexed in WoS) is
less highly regarded or not considered at all (as in the case of
physics).
If the choice of a journal is the result of a discussion, the
consensus between ECRs, the supervisors, and the head of
department is based on the appropriate IF level (not necessarily
the highest). However, in the case of articles having the ECR as
first author, the ‘high’IF argument takes prominence as it
improves their chances of getting a job. Four ECRs also mention
that when choosing a top ranked journal, they factor in how likely
they are to be accepted. Above all, they do not want to waste
time as it is a commodity they do not have in abundance. Rejec-
tions waste valuable time (but not, interestingly, for Chinese
ECRs, who say rejections produce valuable advice, which
improves the product).
French ECRs also consider journals acknowledged by the
community as the place to publish. It is not so much a matter of
prestige as having a kind of ‘local’authority. They are journals
that are important for their contributions to the history of the
discipline at a local scale, even though they are less visible at an
international level.
Malaysian ECRs are unanimous in stating that they choose
journals on the basis of their relevance to their research field, and
this can probably be considered true for all ECRs everywhere,
TABLE 2 Criteria used for choosing a journal for publishing research.
Criteria for choosing China France Malaysia Poland Spain UK USA
Indexed in WoS, Scopus 5 5 5 4 5 5 5
High JIF 2. For some
disciplines, Chinese
journals preferred
53 25 5 5
Most relevant to the field 3 5 5 3 4 5 5
General prestige 3 4 4 0 3 5 5
Covers specialist audience 4 4 4 0 3 4 4. For second
attempts
Fast processing 2 4 2 0 3 3 3
Trusted and used in past 0 0 2 0 3 3 3
Interdisciplinary journal 0 2 3 2 1 2 2
Journal approved by university/
Ministry
414510 0
Open access 1 1 2 0 3 3 3
No page charges 2. Depends on
discipline
24 03 0 0
Easier to get into 1 3 0 3. Local
Polish
2 2. For second
attempts
2. For second
attempts
Innovative features 1 1 2 0 2 3 3
High level of peer reviewing 3 2 4 0 4 4 4
Practices open peer review 0 1 0 0 2 1. Not
applicable to
all
1. Not
applicable to
all
Has both copy and online variants 0 1 2 0 2 0 0
Familiar editor or editorial board 0 1 2 0 0 1 1
Note. Averages, 1 = not important to 5 = extremely important. WoS, Web of Science; JIF, journal impact factor.
7ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org

even if not made explicit. Three-quarters of ECRs say this. Inter-
estingly, after that comes choosing multidisciplinary journals
related to their discipline (nearly half say that). Then, in order of
popularity, come factors such as IF journals (WoS), journals listed
by Scopus, journals with no page charges or submission fees,
journals with an early view online function (a sure sign of rapid
publication), journals approved by the university and the Ministry
of Higher Education, and journals that review quickly.
In Poland, matters are highly prescribed. Firstly, and most
importantly, ECRs take account of the list provided by the
Ministry of Higher Education, which is updated annually. See
http://www.nauka.gov.pl/ujednolicony-wykaz-czasopism-
naukowych/wykaz-czasopism-naukowych-zawierajacy-historie-
czasopisma-z-publikowanych-wykazow-za-lata-2013-2016.
html. The list comprises three parts: part A (the best journals,
with JIFs and indexed by WoS, mostly international), part B
(mainly Polish journals that do not have an IF but have a Pol-
ish score/points), and part C (Polish and international journals
without JIFs but with Polish scores). List A is the best,
thenC,andlastofallisB.Thelistisveryimportantnotonly
for individual researchers but for the universities/departments
as well. Every year, the sum of points for every university is
counted, and grants are provided according to sum of points.
After the list, the relevance to the topic and appropriate audi-
ence are considerations.
As with their Chinese, Malaysian, and Polish counterparts,
for whom metrics are all important, Spanish ECRs will first con-
duct an authentication check of the rankings and determine
whether the journal is indexed in WoS or Scopus, but mainly
WoS, as in France. They will look for journals in the first or sec-
ond Journal Citation Report (JCR) Quartile. They choose top-
ranked journals, and if their paper is not accepted by the first
choice, they will try a slightly lower one and so on until the article
finally finds its acceptance level. They accept the fact that the
publishing process can be long. As in China, Spanish ECRs favour
specialist journals because such journals bring their research to
their target audience. The quality of reviews and the duration of
the publication process are two other factors they take into
account. A few say that if they had the funding, they would
choose OA journals, but few do.
When making (or influencing) a selection, UK and US ECRs
consider the journal’s IF as the most important. However, it is not
quite as simple as that because there always has to be a plan B if
the research cannot get published in top journals, for whatever
reason, and a plan for papers, which are already acknowledged as
being less important (and there are always plenty of these). In
fact, there is always tension between a wish to get into the very
top journal and the need to be more pragmatic. Even in prestig-
ious research groups, ECRs are only expected, for instance, to
publish one paper out of every three or four in a top journal.
Much of the research, therefore, inevitably goes to less highly
ranked journals. Many ECRs emphasize that it is the research
itself –and what it tells us –that is important and not any one
publication. Other criteria (and possible plan Bs) mentioned are
good chances of acceptance, familiar territory, a ‘quick journal’
(more applicable to social scientists) and ‘efficient journal’, which
presumably (these are terms used by ECRs), from evidence
gleaned in other parts of the interview, is not only quick but also
responsive to authors. There are a small number of ECRs –less
than three in both the USA and UK –who think largely just in
terms of the number of publications and a similar small number
who (still) aim for second- or third-rank journals that are just right
for their audience. These ECRs are, to some extent, out of the rat
race and can do what they like. They seem to be happy to either
be in a ‘service’capacity or enjoy working in a less demanding
environment in a less important university, where teaching is a
big component. In France, these people are typically those who
are working out their time before moving to industry.
It is interesting to find that the ranking of journals in WoS is
more important a criterion in the UK/USA than Scopus (as it is in
Poland and China), although there is not much discussion about
this –it is taken as read. For USA and UK ECRs in the medical
sciences, being indexed by PubMed is quite important, and in any
case, research funded by the US National Institutes of Health
‘must’be submitted to PubMed.
OPEN ACCESS
Having established in a previous study (Nicholas, Jamali et al.,
2015) that (1) there is strong interest in the topic of OA among
young researchers, (2) the potential value of OA to ECRs is that
(at least in theory) it offers more/alternative routes to publication
and hence helps them to establish their reputations more quickly,
and (3) that they are generally sympathetic to the ‘open’princi-
ples involved, the topic merited three questions. They were ques-
tions on OA policies/mandates, perceived advantages/
disadvantages, and whether OA publishing advances science and
research or, on the other hand, whether it would dilute the qual-
ity of papers. There was also a hypothesis to test and that was
whether OA journals are conceived as easier to get into.
It might, then, have been expected that ECRs would wel-
come OA with open arms, especially as the results of both the
Ithaka S + R US Faculty Survey (Wolff et al., 2016a) and the
Ithaka S + R/Jisc/RLUK Surveys of UK Academics (Wolff et al.,
2016b) indicate that OA has become important to researchers,
with 64% of US and 57% of UK respondents strongly agreeing
that they would be happy to see the traditional subscription-
based publication model replaced entirely by an OA system.
However, it seems that whilst the principle of OA is growingly
endorsed, that is not the case where it comes to practice; indeed,
a recent survey (Nature Publishing Group and Palgrave Macmil-
lan, 2015) has found that the option to publish via OA was
among the least important factors driving authors’choice of
where to submit. By the same token, according to Borrego and
Anglada (2016), OA is a factor of marginal interest for authors
when deciding where to publish. Similarly, in the aforementioned
study of the motivations that influence US academic authors in
selecting a journal in which to publish (Tenopir et al., 2016), OA
was rated the lowest in importance, even if graduate student
8D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

researchers and postdoctoral researchers rated it as slightly more
important than faculty.
Our results also indicate that the principles of OA are
accepted by ECRs, but in practice, they are not focused on OA
publishing, and they are especially lukewarm about depositing
papers in institutional repositories. This is because they are pre-
occupied with chasing the high IFs that typically are not associ-
ated with OA journals and also because they obtain no
reputational credit for making their publications OA. There is also
some confusion in some countries, but not so much in the
UK/USA, about what OA is. Thus, for instance, 8 out of 13 Chi-
nese ECRs, initially anyway, told us that they had never heard of
OA, although, when it was explained, it turns out that they had,
but they perceived OA journals to be predatory journals. There
are, however, discipline differences, with scientists being more
familiar with the mechanics and details of OA.
Nevertheless, the hypothesis that ECRs publish in OA journals
because they are easier to get into, posed to ECRs in all countries,
was roundly rejected. There appears to be a lot less distrust of
OA than there once was, which is wholly in line with the findings
of both of the aforementioned Nature Publishing Group and Pal-
grave Macmillan (2015) survey and the CIBER study of trustwor-
thiness in the digital age (Watkinson et al., 2016), according to
both of which a decreasing number of authors are concerned
about perceptions of the quality of OA publications.
Policies/mandates
According to Ithaka Surveys (Wolff et al., 2016a, 2016b), aca-
demics on both sides of the Atlantic strongly support government
policy mandating that publicly funded scholarly research be made
freely available, with 72% of US and 67% of UK respondents say-
ing so. However, OA seems to be more popular in theory than in
practice and not only among ECRs, as CIBER findings indicated.
Asked whether research policy directives and mandates influence
where they publish their research, only 20% said yes, heavily;
36.3% said yes, somewhat; about a quarter said not at all; and
the rest said they were not aware of any policies (Jamali
et al., 2014).
Thus, very much in line with the state of affairs described by
their senior counterparts, most of our ECRs believe they are
either not subject to OA mandates or they are not aware of
them. The US Office of Science & Technology Policy directive
that all federal agencies must issue policies for public access to
their funded research was not mentioned by the interviewees.
Moreover, a high proportion of the US interviewees (23 out of
28) say that their university has no policy on OA publishing. One
of the few ECRs, who is aware of policies, provided one of the
very few mentions of institutional repositories/green OA: The
university has started a Harvard-type system to obtain copies for
the IR. They look for an OA journal when placing their research.
By the same token, in France, 10 out of 14 ECRs say there are
no OA policies in their institutions, with the rest saying they did
not know. This is not surprising as only a very few research insti-
tutions have OA publishing policies in France. There are no
mandates at all in favour of Gold. Similarly, in Malaysia, only 4 of
the 12 ECRs say they knew of policies. Spain is an interesting
case with a similar proportion (one-third) saying they knew about
mandates but only half obeying them. In Poland also OA is not
widely practiced, and there are hardly any mandates. Only one
ECR said that funds are available to support OA publishing.
Only in the UK is OA mandated by many institutions and
most funders as well as being the subject of a comprehensive
government policy. Half of the ECRs are aware that their institu-
tion has a policy, and a quarter made the point that their funders
had one. Policies include publishing in hybrid journals, and most
spoke about placing papers in an institutional repository. Some
mentioned that older professors are very wary of OA policies and
OA itself. Moreover, even if UK ECRs are not always sure
whether their institutions have a policy or not, they are much
more aware of their funders’policies.
Advantages/disadvantages of OA publishing
The fact that ECRs are not familiar with mandates, or are not
subject to any, does not mean they have no opinions about the
strengths and weaknesses of OA as the following comments
demonstrate:
Advantages
•Opens the closed world of publication (a reference to fire-
walls) to more researchers.
•New ideas can be dispersed more rapidly, widely, and, in turn,
this triggers further research.
•Provides more immediate and increased visibility.
•Gives more personal control over research work as it can be
disseminated more freely.
•Easier to re-use data.
•Provides a larger audience for a paper.
•Obtain more citations and, hence, an improvement in
reputation.
•It is ethical to do so because of the public money involved.
Disadvantages
•Tend to be less-established journals that are OA.
•Predatory journals that inhabit the OA space can give a wrong
impression of the status of OA journals.
•Quality is low or wholly missing because anyone can publish
anything as long as they can pay.
•It is not a level playing field; only groups with funding can
publish in OA journals and so obtain more citations.
•Open Archive Repositories do not have embedded peer
review systems.
•Easier to steal information.
•Fears (no more than that) of light touch peer review.
•It is not a sustainable model, with author publishing fees being
too expensive with the risk that Gold OA means many authors
will not be able to publish anymore.
9ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org

So, how does all this manifest itself in terms of actual OA
publishing at the national level? It does of course depend, to an
extent, on the existence of mandates. Thus, in Spain, publishing
in OA journals is not common, and most ECRs do not publish in
them. This is because they do not have enough funds to do so
and because they do not trust OA completely, although, they
generally trust hybrid journals more. Chinese researchers think
similarly, although they did not mention hybrid journals. In Malay-
sia, ECRs’choices are constrained by the fact that APCs are only
paid for publishing in high IF WoS OA journals. French ECRs do
not always have the money to pay APCs. Only those who lead or
who are part of funded research projects have APCs paid for
them. Research team budgets do not allow for more than that,
and ECRs are particularly disadvantaged because they do not lead
projects and depend on others to pay the APC for them. UK
ECRs are much more positive about OA and depositing papers in
IRs but do not necessarily publish more in OA journals. In the
OA-‘savvy’UK, just two ECRs said it impacts their decision where
to publish, mentioning that they publish in ‘hybrid’journals as a
consequence.
Does OA advance or dilute the quality of
science?
The large majority, typically two-thirds of ECRs, feel that OA
publishing advances science, but around half of these felt it also
has negative consequences. This is the case in the UK, where
18 out of 21 researchers agree with the first part of the state-
ment –that OA helps advance science –and 12 agree with the
second, negative statement about dilution. French ECRs are fence
sitters, tending to believe that OA advances science but see a risk
to science coming from the predatory publishers. Nearly a third
of Spanish ECRs also thought the same.
The reasons given for believing OA will advance science tend
to be that it is democratic, allowing researchers to read each
other’s papers; furnishes easy access to research; and provides
them all with the same chance to gain knowledge. Some thought
OA was not the issue because quality is more important than the
business model, and the real key is the quality of peer review.
DATA/SOFTWARE
As Tenopir et al. (2011) state, with science becoming increasingly
more data intensive and collaborative, data sharing assumes
greater importance. Findings demonstrate that there is a willing-
ness to share data, but in actual fact, only a minority make their
data electronically available to others: less than 6% of those who
responded to the question on data sharing made all of their data
available; 46% reported they made at least some of their data
available somehow; and as many reported that they did not make
their data available to others at all.
When questioned whether they would like to be recognized
more for their data and software and make it more visible
(weavoidedtheterm‘open data’in case of unfamiliarity), our ECRs
also thought highly of the practice but did little about it. They essen-
tially say that they produce data but do not make it available,
although many see the benefits of doing so. The case of China is
illustrative, where a majority (11) produced data, and one created
software. But no one made their data and software available. Five
say they will make the data open only if a journal required them to
do so. In the case of France, all the ECRs produce data and computer
science software. However, only three ECRs (Bio-medics) made their
data available when publishing their articles because it was required
by the publisher and the peer reviewing process. They did so on
PubMed, where you load supplementary data as part of an article.
Many just did not want to provide visibility for, or share, the
data. In the case of the French physicists, it was considered a
useless exercise to make data open because, as one ECR said,
Nobody would understand the data or would be able to exploit it as
they do not have the context of the research, the methodology and
so on. In Spain, only a minority (five) of ECRs produce software,
and they think they would have to get credit for it if they are to
make it available. A majority (10), however, see benefits in making
data more available, but they are mostly unaware of the useful-
ness of publishing data separately.
In the USA and UK, the use of the word ‘visible’in the ques-
tion did not prevent the ECRs understanding the question as
about ‘open data’. In the USA, 21 said yes in principle, with only
3 taking a negative view and 4 uncertain. The UK response was
similar, with 16 saying yes, 4 being negative, and one response not
relevant. There were a number of qualifications in both countries
mostly concerned with ‘ethical’issues (such as human subjects)
and confidentiality issues relating to industry. Others, although
they answered yes, worried that they would put themselves at a
disadvantage if others did not adopt the same open policies.
Making research findings more widely available
The question posed was Would you prefer to make public your
research findings in less formal ways, such as blogs, which could
make them more visible? This was really an invitation for ECRs to
talk about social media as a publishing medium without being
seen to push the social media agenda. The general answer, some-
times defensively put, is that they do not use social media chan-
nels for disseminating their research because they lack the time
(busy as they are writing papers for high IF journals) and are not
given any recognition for these activities (e.g. not admissible on
CVs). Other reasons also given are: (1) they lack the know-how;
(2) some journals forbid the authors to do so and thus put in peril
their all-important papers; (3) there are dangers of misinterpreta-
tion, a comment levelled against the media; and (4) dissemination
in formal channels is thought to be all important because it is the
only guarantee of research quality.
ECRs only confirm the prevailing, if unwritten, norms of aca-
deme in feeling the way they do about utilizing social media for
research dissemination purposes; as Wolff et al. (2016a, 2016b)
find, academics believe that when their work is assessed for
appointment, promotion, or research funding, more recognition
should be awarded for traditional research publications as
10 D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

compared to alternative research products. That, of course, is not
to say that ECRs do not want to disseminate more widely and
innovatively, and at least half are tempted or interested in
doing so.
Thus, in France, although no ECR publishes research via
social media channels, a sizeable number (8) say they are tempted
to do so. This intention is related to the original purpose of doing
their PhD, which is to disseminate their research to a larger audi-
ence. For the French and all the other countries, there is a caveat
and that is dissemination should take place after publication in a
peer reviewed journal.
Where social media is encouraged (as it is in the UK by the
REF in order to promote outreach and impact), there are higher
levels of activity, with over one-third giving an unequivocal yes
that they did use social media. Significantly, the drive to use
social media, especially to reach practitioners and policy makers,
appears to be coming from university management, marketing
departments, and –less so –their senior colleagues.
PUBLISHING STRATEGIES
The question asked was Do you have a conscious publication strat-
egy relating to your research and is that to do with obtaining a
tenured/established position and, if so, please describe? It was asked
because there was some doubt whether ECRs had a strategy,
something that is thought to be good for their career progression.
In fact, most (three-quarters) ECRs do have a publishing strategy
of sorts or profess they do. There are, however, variations with,
for instance, nearly all ECRs in China having one but only half in
Malaysia.
Not surprisingly, given the importance accorded to the vol-
ume of papers published in high-ranking journals and the previ-
ous evidence, ECRs overwhelmingly adopt just one strategy and
that is to publish as many articles as they can in high IF journals
because that enhances their reputation and increases the proba-
bility of tenure. A French ECR spoke for many by saying: ‘My
research activity and career progression is only seen through the
prism of publication’.
A small proportion (20% in France) spoke out against such a
strategy, preferring a more qualitative approach, which focuses on
producing fewer, better papers that have a greater impact on the
community. For them, the publish or perish regime was overloading
and harming science. This approach is illustrated by a French com-
puter scientist: Fundamentally, I do not adhere to the rules, it is
damageable to the scholarly publication system that we publish so
many articles.We could publish less and better articles.We are all in
competition, one against the other and this leads us to publishing even
embryo results.In this competitive race in the past I’ve published arti-
cles that were not mature. Other alternative, minority publishing
strategies included: obtaining maximum attention; always choosing
the most appropriate journal, not ignoring journals that were pres-
tigious but did not have high JIFs; producing practical and influen-
tial work; obtaining copyright protection for their work; and
balancing high IFs with high readership and speed of publication.
Those who did not have a strategy gave reasons such as
being unfussy, opportunistic, that they ‘operate on automatic
pilot’, that it is too much of a lottery to have one article pub-
lished, and the fact that they so disliked what an ECR from Spain
called the ‘paper engineering’(salami slicing).
CONCLUSION
ECRs are productive, if not prolific, having published, on average,
10 journal articles and six conference proceedings in their
research careers. Most say their institution and/or their depart-
ment have a formal or informal authorship policy, which largely
concerns who can be named as an author and in what order. Poli-
cies are formal in China and Malaysia and informal in Poland and
Spain, and they can vary by discipline. ECRs voice few complaints
about the policies and are first author in around one-third to one-
half of all the papers to which they contribute. They also have a
significant influence in choosing where a paper is published, with
three-quarters saying this. Choices are, however, constrained by
the fact that most ECRs are part of a research group and their
status in the groups.
When ECRs are in a position to choose or influence where
papers are published, they employ a complex and overlapping
range of criteria, although traditional criteria hold sway, with
being indexed/ranked in WoS or Scopus the most important.
Non-traditional criteria, such as being innovative or open (with
respect to access and peer review), do not count much. As to
making research findings more widely available through social
media and online community platforms, ECRs largely elect not to
do this because they lack the time (busy as they are writing
papers for high IF journals) and are not given any recognition for
these activities. Some would like to, however, if things were dif-
ferent or they were in charge.
ECRs think OA publishing is a good idea, but they do not
practice what they believe, unless mandated to do so and, even
then, not all abide by them –depositing papers in IRs is a case in
point. This, again, is all because they are too preoccupied chasing
the high IF journals, which typically are not OA, and because they
obtain no reputational credit for making their publications
OA. Similarly, ECRs produce data but do not make it available,
although many see the benefits of doing so.
Most ECRs profess to having a publishing strategy, but this
strategy boils down to publishing as many articles as they can in
high IF journals because that is what enhances their reputation
and increases the probability of them obtaining tenure. Despite
the siren voices of Science/Web 2.0 for ECRs, it is a case of ‘pub-
lish or perish’.
Finally, ECRs’room for manoeuvre is limited by the fact that
many work in research groups where the need to consider the
wishes of others whose favours they need to win, are subject to
institutional policies and restrained by the fact that they have to
keep an eye on doing what is best to get them a job (publishing
papers in high IF journals). They are shackled by convention. But
what evidence do we have, then, for thinking that if they were
11ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org

free and independent, they would do things differently and what
things would they do? In fact, many ECRs do not have major pro-
blems with the publication-based system they live in and are
measured by. This tends to be very much the case for UK and US
ECRs. That is not to say they are unaware of the system’s weak-
nesses because many ECRs emphasize that it is the research itself
–and what it tells us –that is important and not any one publica-
tion. Nevertheless, they follow the publishing route. Most likely,
this indicates that they would much rather have tenure commit-
tees and grant-giving bodies actually judge them on the research
they have undertaken and not on the basis of what journal they
published in. French and Spanish ECRs are the most critical and
outspoken, especially regarding their inability to exploit the legion
opportunities there are today for sharing, collaboration, and out-
reach. French ECRs are particularly critical about the lack of
thinking time because they are always struggling to publish, and
Spanish ECRs are concerned about the reputational damage of
utilizing social media.
Limitations
This study is based on a sample of 116 ECRs and might not be
representative of the ECR population as a whole. The ECR pro-
files with regard to subject, age, gender, age, and whether they
are a postdoc/PhD (see Table 1) are very relevant to the interpre-
tation of the results, and analyses from this perspective will be
provided as the longitudinal data builds over the next 24 months.
The study is also based on personal interpretation of the ques-
tions and ECRs willingness to report honestly, knowledgeably,
and objectively, all of which may introduce bias.
ACKNOWLEDGEMENTS
The research reported here has been funded by the Publishing
Research Consortium, the National Social Science Foundation of
China, No. 15CTQ026, and Ministry of Higher Education Malay-
sia (UM.C/HIR/MOHE/FCSIT/11).
REFERENCES
Borrego,

A., & Anglada, L. (2016). Faculty information behaviour in
the electronic environment: Attitudes towards searching, publish-
ing and libraries. New Library World,117(3/4), 173–185.
doi:10.1108/NLW-11-2015-0089
Brand, A., Allen, L., Altman, M., Hlava, M., & Scott, J. (2015). Beyond
authorship: Attribution, contribution, collaboration, and credit.
Learned Publishing,28(2), 151–155. doi:10.1087/20150211
Brechelmacher, A., Park, E., Ates, G., & Campbell, D. F. (2015). The
rocky road to tenure –Career paths in academia. In T. Fumasoli,
G. Goastellec, & B. M. Kehm (Eds.), Academic work and careers in
Europe: Trends, challenges, perspectives (pp. 13–40). Cham, Switzer-
land: Springer International Publishing.
Cusick, A. (2015). Research training as occupational socialization:
Doing research and becoming researchers. Asian Social Science,
11(2), 252. doi:10.5539/ass.v11n2p252
Dance, A. (2012). Authorship: Who’sonfirst? Nature,489(7417),
591–593. doi:10.1038/nj7417-591a
Elliott, K. C., Settles, I. H., Montgomery, G. M., Brassel, S. T.,
Cheruvelil, K. S., & Soranno, P. A. (2017). Honorary authorship
practices in environmental science teams: Structural and cultural
causes and solutions. Accountability in Research,24(2), 80–98.
doi:10.1080/08989621.2016.1251320
Jamali, H. R., Nicholas, D., Watkinson, A., Herman, E., Tenopir, C.,
Levine, K., …Nichols, F. (2014). How scholars implement trust
in their reading, citing and publishing activities: Geographical
differences. Library & Information Science Research,36(3),
192–202. doi:10.1016/j.lisr.2014.08.002
Kornhaber, R. A., McLean, L. M., & Baber, R. J. (2015). Ongoing
ethical issues concerning authorship in biomedical journals: An
integrative review. International Journal of Nanomedicine,10,
4837. doi:10.2147/IJN.S87585
Müller, R. (2012). Collaborating in life science research groups: The
question of authorship. Higher Education Policy,25(3), 289–311.
doi:10.1057/hep.2012.11
Müller, R. (2014a). Racing for what? Anticipation and acceleration in
the work and career practices of academic life science postdocs.
Forum Qualitative Sozialforschung/Forum Qualitative Social
Research,15(3).
Müller, R. (2014b). Postdoctoral life scientists and supervision work in
the contemporary university: A case study of changes in the cul-
tural norms of science. Minerva,52(3), 329–349. doi:10.1007/
s11024-014-9257-y
Nature Publishing Group and Palgrave MacMillan. (2015). Author
insights –Initial findings. Retrieved from https://figshare.com/
articles/Author_Insights_2015_survey/1425362
Nicholas, D., Boukacem-Zeghmouri, C., Rodríguez Bravo, B., Xu, J.,
Watkinson, A., Abrizah, A., …Swigon, M. (2017). Where and how
early career researchers find scholarly information. Learned Pub-
lishing,30(1), 19–29. doi:10.1002/leap.1087
Nicholas, D., Herman, E., & Jamali, H. R. (2015). Emerging reputation
mechanisms for scholars: A literature-based theoretical framework
of scholarly activities and a state-of-the-art appraisal of the social
networking services used by scholars, to build, maintain and show-
case their reputation. Seville, Spain: European Commission, Joint
Research Centre, Institute for Prospective Technological Studies.
Retrieved from http://publications.jrc.ec.europa.eu/repository/
bitstream/JRC94955/jrc94955.pdf.
Nicholas, D., Jamali, H. R., Watkinson, A., Herman, E., Tenopir, C.,
Volentine, R., …Levine, K. (2015). Do younger researchers assess
trustworthiness differently when deciding what to read and cite
and where to publish? International Journal of Knowledge Content
Development and Technology,5(2), 67–92. doi:10.5865/
IJKCT.2015.5.2.045
Nicholas, D., & Rowlands, I. (2011). Social media use in the research
workflow. Information Services and Use,31(1–2), 61–83.
doi:10.3233/ISU-2011-0623
Nicholas, D., Watkinson, A., Jamali, H. R., Herman, E., Tenopir, C.,
Volentine, R., …Levine, K. (2015). Peer review: Still king in the digital
age. Learned Publishing,28(1), 15–21. doi:10.1087/20150104
Nicholas, D., Watkinson, A., Volentine, A., Allard, S., Levine, K.,
Tenopir, C., & Herman, E. (2014). Trust and authority in schol-
arly communications in the light of the digital transition: Set-
ting the scene for a major study. Learned Publishing,27(2),
121–134. doi:10.1087/20140206
Pepermans, G., & Rousseau, S. (2014). The decision to submit to a
journal: Another example of a valence-consistent shift? Journal of
12 D. Nicholas et al.
www.learned-publishing.org © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
Learned Publishing 2017

the Association for Information Science and Technology,67(6),
1372–1383. doi:10.1002/asi.23491
Poli, S. (2016, 10 May). The annual conference of the German
Association for HE research [Web log post]. Retrieved from
https://srheblog.com/2016/05/10/the-annual-conference-of-
the-german-association-for-he-research/
Rousseau, S., & Rousseau, R. (2012). Interactions between journal
attributes and authors’willingness to wait for editorial decisions.
Journal of the American Society for Information Science and Technol-
ogy,63(6), 1213–1225. doi:10.1002/asi.22637
Rowlands, I., & Nicholas, D. (2005). Scholarly communication in the
digital environment: The 2005 survey of journal author behaviour
and attitudes. Aslib Proceedings,57(6), 481–497. doi:10.1108/
00012530510634226
Rowlands, I., Nicholas, D., Russell, B., Canty, N., & Watkinson, A.
(2011). Social media use in the research workflow. Learned Pub-
lishing,24(3), 183–195. doi:10.1087/20110306
Tenopir, C., Allard, S., Douglass, K., Aydinoglu, A. U., Wu, L., Read, E.,
…Frame, M. (2011). Data sharing by scientists: Practices and per-
ceptions. PLoS One,6(6), e21101. doi:10.1371/journal.
pone.0021101
Tenopir, C., Dalton, E., Fish, A., Christian, L., Jones, M., & Smith, M.
(2016). What motivates authors of scholarly articles? The
importance of journal attributes and potential audience on publica-
tion choice. Publications,4(3), 22. doi:10.3390/publications4030022
Tenopir, C., Levine, K., Allard, S., Christian, L., Volentine, R.,
Boehm, R., …Watkinson, A. (2015). Trustworthiness and authority
of scholarly information in a digital age: Results of an international
questionnaire. Journal of the Association for Information Science
and Technology,67(10), 2344–2361. doi:10.1002/asi.23598
Van Dalen, H. P., & Henkens, K. (2012). Intended and unintended
consequences of a publish-or-perish culture: A worldwide survey.
Journal of the American Society for Information Science and Technol-
ogy,63(7), 1282–1293. doi:10.1002/asi.22636
Watkinson, A., Nicholas, D., Thornley, C., Herman, E., Jamali, H. R.,
Volentine, R., …Tenopir, C. (2016). Changes in the digital schol-
arly environment and issues of trust: An exploratory, qualitative
analysis. Information Processing & Management,52(3), 446–458.
doi:10.1016/j.ipm.2015.10.002
Wolff, C., Rod, A. B., & Schonfeld, R. C. (2016a). Ithaka S+R US faculty
survey 2015. New York, NY: Ithaka S+R. Retrieved from http://sr.
ithaka.org/?p=277685.
Wolff, C., Rod, A. B., & Schonfeld, R. C. (2016b, 15 June). UK survey
of academics 2015: Ithaka S+R|Jisc|RLUK [Web log post].
Retrieved from http://www.sr.ithaka.org/publications/uk-survey-
of-academics-2015/
13ECRs publishing and authorship practices
Learned Publishing 2017 © 2017 The Author(s).
Learned Publishing © 2017 ALPSP.
www.learned-publishing.org