JLIS.

it 10, 2 (May 2019)

ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Who is interacting with researchers on Twitter?

A survey in the field of Information Science
Belén Álvarez-Bornstein(a), Michela Montesi(b)

a) IFS, Centre for Humanities and Social Sciences (CCHS); Spanish National Research Council (CSIC), https://orcid.org/0000-0003-4394-2543
b) Departamento de Biblioteconomía y Documentación, Universidad Complutense de Madrid (UCM), https://orcid.org/0000-0002-5509-2075
__________
Contact: Belén Álvarez-Bornstein, belen.alvarez@cchs.csic.es; Michela Montesi, mmontesi@pdi.ucm.es
Received: 19 November 2018; Accepted: 23 February 2019; First Published: 15 May 2019
__________

ABSTRACT
The social web appears a promising environment to study the societal impact of research, and, although platforms such as
Twitter appear to be popular to share scientific content, little is known about the outreach of scientific communication on
social media. By surveying a sample of Twitter users who are also followers of communication and Library and Information
Science (LIS) researchers, this article aims to understand who is interacting with researchers on Twitter and if the activity
of following allows to get closer to science and research. A survey was sent to all followers of 9 active researchers, and 53
responses were collected and analyzed. According to the results, followers are mostly professionals (49%) although the
higher education sector (36%) accounts for an important part of the audience. Twitter allows respondents to keep updated,
feel as a part of the scientific community and gain more visibility, whereas interactions appear as an important though
secondary facet of these users’ activity on Twitter. Finally, participants consider that they do not have influence on the
scientific process, maybe because the flow of information on Twitter occurs mostly as retweeting in an unidirectional fashion.
More research is needed in order to understand better interactions and engagement on social media before they can be
quantified and measured as indicators of societal impact of research.

ACKNOWLEDGEMENTS
We would like to thank all the researchers that, tweeting the link to our survey, contributed to this research, as well as those
who took part in the survey.

KEYWORDS
Altmetrics; Societal Impact of Research; Communication; Library and Information Science; Twitter; Social Media
Interactions.

CITATION
Álvarez-Bornstein, B., Montesi, M. “Who is interacting with researchers on Twitter? A survey in the field of Information
Science.” JLIS.it 10, 2 (May 2019): 87-106. DOI: 10.4403/jlis.it-12530.

__________
© 2019, The Author(s). This is an open access article, free of all copyright, that anyone can freely read, download, copy, distribute, print, search, or link to the
full texts or use them for any other lawful purpose. This article is made available under a Creative Commons Attribution 4.0 International License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. JLIS.it is a journal of the SAGAS Department,
University of Florence, published by EUM, Edizioni Università di Macerata (Italy).
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

1. Introduction
Altmetrics or social media metrics measure the impact of research when scientific communication
takes place on the social web. Moving from the closed scientific system to the web, altmetrics present
several advantages over traditional citation metrics (Priem, Taraborelli, Groth et al. 2010). They can
be harvested sooner than citation-based indicators and they can capture the impact of research outside
the academic community, allowing in this way to measure impact of research products different from
the research article. Since 2010, much research has been carried out. Many studies, if not most of
them, have relied on correlation analysis in order to find meaningful relationships between mentions
of scientific articles on social media and their respective citations in citation indexes. According to
Bornmann (2014), however, the results obtained by correlation studies are “hardly meaningful and
can be interpreted fairly loosely” (p. 901). When positive correlations are found, meaning that
altmetrics have something to do with scholarly activity, it is argued that social web metrics should not
be considered alternative metrics because they align with traditional metrics. On the other hand, when
correlations between social web mentions and citations are weak, and consequently their ties to
scientific communication appear to be looser, it is argued that altmetrics measure a different dimension
of impact, specifically the societal impact of research (Bornmann 2015). Unfortunately, the concept
itself of societal impact of research is still vague and difficult to measure (Bornmann 2013), whilst
metrics from social media appear to be more reliable when they originate in relatively closed
environments such as Mendeley. Indeed, the most robust source of social metrics seems to be
Mendeley (Thelwall and Kousha 2015), a social reference manager, whose users are mostly young
academics, i.e. PhD students, postgraduates and postdocs across different disciplines (Mohammadi,
Thelwall, Haustein et al. 2015). Correlations between readers on Mendeley and citations seem to have
even improved over time, probably as a consequence of the growing readership of this platform
(Mohammadi, Thelwall, Haustein et al. 2015). Being strongly related to citations, readership counts
on Mendeley allow to obtain similar impact data to citations, though ahead of time. The basically
academic nature of Mendeley and similar reference managers such as Figshare limits its application
to measure impact on a broader non-academic audience. Bornman (2015a) found that metrics coming
from Facebook and Twitter were significantly higher than metrics obtained from Mendeley and
Figshare for articles published in the Public Library of Science and tagged as “good for teaching” −
and thus more likely to appeal to a broader audience.
In addition to the limitations discussed above, correlation studies alone cannot prove that there exists
a connection between altmetrics and research impact (Sud and Thelwall 2014), and that makes it
difficult to build meaningful theories. The lack of theoretical foundations for altmetrics by itself is a
frequent claim (Priem 2014; Holmberg 2015, 9.2). Haustein (2016) mentions the lack of a conceptual
framework to understand the actions behind the metrics, among the other challenges of altmetrics
research, whilst Haustein, Bowman and Costas (2015) try to make sense of different actions occurring
on the social web, specifically access, appraise, and apply, but on the basis of pre-existing adapted
theories. Different methodological approaches are needed in order to build significant theories and
understand altmetrics.

88
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

1.1 Alternative metrics and societal impact of research

A related question which is still to be answered concerns what dimension of societal impact scientific
communication on the web should reflect. Bornmann (2012; 2013) reviews different approaches to
measure societal impact of research, concluding that, at the moment and considering the extreme
complexity of the issue, qualitative case studies assessment seems to be the best practical approach.
Greenhalgh, Raftery, Hanney, et al. (2016), after reviewing several approaches to assess impact within
the health sciences, conclude that research impact per se can have many definitions and different
approaches might be necessary depending on the purpose. On the other hand, alternative metrics can
play a role in measuring societal impact of research because they make it possible to measure, to a
certain extent, public engagement with research (Bornmann 2014). According to Ozanne, Davis,
Murray et al. (2017), altmetrics allow to introduce measurement in what is still difficult to express in
numbers, i.e. the engagement of the research community with society. The social web is an informal
context in which engagement and interactions are made possible (Holmberg, Bowman, Haustein, et
al. 2014). The advantages of altmetrics for measuring societal impact include the coverage of different
audiences from academia, the diversity of measurable research products, the speed with which they
are generated, and their openness (Bornmann 2014). On the other hand, limitations span from the
dependence on commercial platforms to several issues regarding the quality of data, such as little
knowledge of audiences and users, multiple versions of papers, or lack of normalization (Bornmann
2014).
Addressing the problem of the societal impact of research, some suggest we should look at the
processes by which research contributes to social change more than at results. Spaapen and van
Drooge (2011) consider that productive interactions, as a type of engagement, between scientists and
different stakeholders may foster knowledge transfer and application. What the authors call “[...]
exchanges between researchers and stakeholders in which knowledge is produced and valued that is
both scientifically robust and socially relevant” (p. 212) allow to move the focus of evaluation from
actual results, which may take time to produce and might be difficult to attribute, to the process
leading to impact. Looking at the process does justice to basic research, that rarely produces visible
and tangible results in applied projects (de Jong, Barker, Cox et al. 2013). Dilling and Lemos (2011)
speak also of iterativity, as “the purposeful and strategic interaction between climate knowledge
producers and users so as to increase knowledge usability” (p. 681). In their opinion, the demand of
usable scientific knowledge to address environmental problems can only be met through iterative
interactions between scientists and users of knowledge. Pedrini, Langella, Battaglia et al. (2018), from
the health sciences, suggest implementing a multi-stakeholder perspective in health research processes
in order to maximize its societal implications. In the same line of thought, Ozanne et al. (2017) retake
the concept of productive interactions defining the “relational engagement approach” to assessing
societal impact of research. This perspective emphasizes engagement with researchers of other
disciplines and social stakeholders beyond actual results and products as a measure of impact on
society, which points to an important limitation of altmetric research so far. Most studies carried out
on altmetrics have been looking for mentions of research articles on the social web, paying attention
to the most traditional research output. Conversely, the process itself of engaging or interacting on the
social web has received little attention. It is unlikely that the layperson can understand the scientific
literature as it is produced for other researchers and the context itself in which it has been produced

89
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

(Zuccala 2009). In the area of environmental research, Archie, Dilling, Milford et al. (2014), after
conducting a survey about the role of information for public lands managers and municipal officials,
concluded that peer-reviewed information was considered the best available knowledge, though it
was used regularly by less than half of the population studied. In many cases, there might not exist
relevant scientific literature for particular problems, or it might be too difficult to access and
understand. The limitations of traditional scientific literature to address the needs of a no specialized
audience put into question the fact that mentions of research articles on the social web should be
measured as an indicator of societal impact of research. Additionally, it remains to be determined how
often authors mention their own articles online as a means to gain visibility and boost impact. A recent
study of an Information Science community on Twitter found that almost half of the times articles
were tweeted by the authors themselves or by closed colleagues (Álvarez-Bornstein and Montesi
2016). It takes very little effort to tweet an article, whilst it requires commitment and time to really
engage and interact with no academic audiences.

1.2 Engagement on social media

If repeated productive interactions with different stakeholders might be a way to promote societal
impact of research, then it is important to understand such interactions within specific contexts. In
the case of Mendeley, engagement with the community and participation in groups have been found
to be less frequent activities than research-related activities (Jeng, He, and Jiang 2015). In the case of
Twitter, several studies try to characterize it as for its interactive affordances. Boyd, Golder and Lotan
(2010) study retweets as a way of interaction. Quan-Haase, Martin and McCay-Peet (2015), after
interviewing 25 Digital Humanities scholars, conclude that being on Twitter per se is a conversation,
that “takes place through a single tweet or a series of interactive tweets, and in relation to other
discourses online and offline, both inside and outside their disciplines” (p. 9). On the other hand,
Holmberg, Bowman, Haustein, et al. (2014) look at mentions, as a way of describing interactions, of
Twitter users in the tweets published by a sample of 37 astrophysicists. They found that the
astrophysicists studied interacted with a variety of user types including other colleagues, science
communicators, such as scientific journalists, other researchers and educators.
In order to understand interactions on social media around scientific contents, it is important to know
who takes part. We know little about users of Twitter for scientific and research purposes, though the
academic community has been surveyed in a few studies in order to know their attitude towards social
media or their presence on social platforms. Haustein, Peters, Bar-Ilan et al. (2014) surveyed all
attendants at the 2012 STI conference of Montréal, finding that the most popular platform for these
scientists was LinkedIn, whilst half of respondents reported having a Twitter account and only ⅕ had
a ResearchGate or and Academia.edu account. Mas-Bleda, Thelwall, Kousha et al. (2014) measured
the uptake of the social web as a communication tool among highly cited researchers of European
institutions, finding that only 28% had an account on any of the social sites analyzed: Google Scholar,
Mendeley, Academia, LinkedIn, and SlideShare. Mohammadi, Thelwall and Kousha (2016) surveyed
Mendeley users in order to know if they read or intended to read the articles stored as references. It
was found that 55% of Mendeley users actually had read or wanted to read at least half of the articles
saved, which would mean that altmetrics extracted from Mendeley can be counted as reads. Many
studies of users interacting on social media have dealt with Twitter. However, according to Budge,

90
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Lemon, and McPherson (2016), who describe their personal experience on Twitter as academics and
artists, there is a paucity of research on Twitter use by academics. Veletsianos (2012) characterizes
seven main activities of scholars using Twitter on the basis of a sample of tweets from 45 scholars in
several disciplines. Among these activities, he mentions sharing information both for their
professional practice and classroom, requesting and offering suggestions, engaging in social
commentary, or shaping their digital identity. Haustein and Costas (2015) define automatically the
profiles of approximately 80000 Twitter accounts tweeting academic articles. They find that users
tweeting academic articles describe themselves as organizations or interest groups, professionally,
personally or both. Na (2015) describes users’ motivations when they tweet research articles in
psychology, relying on content analysis of the short messages exchanged on this microblogging
platform. Studying a group of 37 astrophysicists on Twitter, Haustein, Bowman, Holmberg et al.
(2014) found moderately negative correlations between tweets per day and the number of publications
in a five year timespan (2008-12), meaning that those who tweet more intensely tend to publish less
scientific literature.
Despite all this literature on Twitter users, little do we know about followers of researchers on Twitter.
With no doubt it is a difficult topic to explore, because, due to the asymmetric nature of Twitter in
which anyone can follow anyone, many users do not know all their followers (Gruzd, Wellman and
Takhteyev 2011), and it is difficult to reach to them. This knowledge gap about users of media
platforms makes it difficult to estimate how biased they are as sources of altmetrics data (Bornmann
2014). The purpose of this article is to survey a sample of Twitter users who are also followers of
Library and Information Science (LIS) researchers, in order to know if the fact of following
researchers get them closer to science and research. Twitter appears to be the most popular space in
which to share scientific content (Holmberg et al. 2014). Pooling results of research into altmetrics in
a meta-analysis, Bornmann (2015, 27) concluded that: “Twitter and blog citations seem to measure
something different from traditional citations, because the pooled coefficients are very low.”
Haustein, Costas and Larivière, (2015) look for traces on the social web of 1.3 millions documents
extracted from Web of Science, finding that the most popular platform they were mentioned on was
Twitter (22% of documents had at least a tweet). Both blogs and Twitter are spaces in which
interactions between the academia and different social actors are possible.

2. Method
Researchers studying altmetrics often claim that more qualitative research designs, such as interviews
and surveys, are needed to understand scientific communication on the web (Holmberg 2015, 14.3).
In this research article, we present results of a web survey with Twitter users, though we understand
that surveys are quantitative research methods that allow to describe a population based on a
representative sample (Pickard 2013, 111). A conceptual difference that this type of contribution can
make from previous altmetrics research is that we switch from objects or events (such as mentions,
tweets, citations, articles…) to persons, in this specific case, Twitter users. Focusing on research
articles in order to study societal impact of research might lead to dead ends provided they seem to
be shared among small communities that are little connected with the general public (Alperin, Gomez
and Haustein 2018). In information behavior research, a similar transition from systems to persons
occurred in the seventies when research started to deal with needs and uses instead of artifacts and

91
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

venues of information seeking (Case 2016, 7−10). In this sense, we interpret the qualitative approach
that many altmetrics researchers are asking for as a claim to look at behaviors instead of objects, or at
processes instead of results, as commented previously in the introduction. This is something that
implicitly appears also in (Haustein, Bowman and Costas 2015) when they try to explain altmetrics
through pre-existing theories by focusing on acts leading to online events.

2.1 The sample

A short survey was designed to collect information about users’ identities and professions, reasons
and expectations for following a particular group of twenty active researchers in Library and
Information Science on Twitter. This specific group was obtained from two previous studies, (i.e.
Torres-Salinas and Milanés-Guisado 2014 and Álvarez-Bornstein and Montesi 2016). Torres-Salinas
and Milanés-Guisado (2014) studied the presence on social media of very active researchers who
published at least 3 articles in the scientific journal El Profesional de la Información between 2009 and
2013. El Profesional de la Información is a scientific journal which is highly regarded in Spain within
the Communication and LIS community. On the other hand, Álvarez-Bornstein and Montesi (2016)
selected from this sample only those who had a Twitter profile in order to analyze their
communication activities (26, though 6 were not active on Twitter). Thus, the sample studied here
includes all followers of the 20 active researchers with a profile on Twitter that the previous literature
just commented on has studied.
The 20 researchers were contacted via email, on July 5th, 2016, and, after the purpose of the research
was explained, they were asked to publish a link to our survey on Twitter. Only nine of them tweeted
the link. When the survey was sent, the 9 researchers who tweeted the link gathered, together, a
population of 24,239 followers. It was not determined how many of these followers followed more
than one of the 9 researchers who tweeted the link to the survey. Though following more than one of
the 9 researchers is a very likely scenario.
Finally, 53 Twitter followers responded to the survey.

2.2 The survey

The survey consisted of 15 questions organized in two parts. In the first part, a set of open-ended and
close-ended questions examined the use of Twitter and the reasons for following researchers.
Additionally, participants were questioned about the advantages of following this group of researchers
and about the way in which this activity of following contributed to their career. The second part
included questions about respondents’ personal and professional profile, such as sex, age, main
occupation and area of specialization. These demographic data were collected in order to answer the
question of who is following researchers on Twitter. We understood that it is important to know if
they are professionals or if they work in the academia, and if they belong to LIS or other disciplines.

2.3 Limitations and difficulties of conducting a survey with followers on Twitter

The non-random convenience sample of 53 respondents is clearly too small to be representative of
the entire population. On the other hand, it is probably impossible to obtain a representative sample
of Twitter users who follow a specific group of people due to its uncertain size and its changing nature

92
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

as Carpenter and Krutka remind (2014). Marwick (2014) identifies several challenges of conducting
survey through Twitter, considering that representativeness is one of the most problematic. Indeed,
the operation of defining an audience on Twitter may be a problem by itself, and different solutions
can be proposed. Bombaci, Farr, Gallo et al. (2016), for instance, in order to define an audience for
tweets published along conference presentations at a Conservation Science Congress, decide to take
retweeters as members of such an audience, because, at least, retweeters have read the message.
However, this is but one of several possible approximations to Twitter audiences. In addition to
Twitter changing nature, another complicating factor is immediacy (Marwick 2014), that in this study
can explain, partly at least, the low rate of participation. On Twitter users are receiving constantly a
lot of information and, with older tweets getting lost quickly in the timeline, potential respondents to
our survey may have missed easily the link to the survey. Conversely, counting on researchers’ support
prevented that our message might be classified as spam.

2.4 Data analysis

Data were analyzed using Excel and SPSS 25, an IBM package for statistical analysis. Correlation tests
were carried out when they were considered relevant. The Chi square test allows to measure
correlations between nominal and ordinal data − such as those obtained in this survey − by comparing
expected vs. observed frequencies in contingency tables. Unfortunately, the sample studied did not
satisfied all requirements for the application of the Chi square test; specifically, in some cases the
expected frequencies for some classes were not large enough (under 5) (Vaughan 2001, 88−91). As
an alternative, variables were edited so that 2x2 contingency tables could be created, and the Fisher’s
exact test was carried out instead (Weinberg and Abramowitz 2008, 499). In order to find
relationships between participants’ occupation, on the one hand, and reasons for and consequences
of following researchers on Twitter, on the other, occupations firstly were reduced to three categories:
a) professionals, which included both those working in public administrations and in the private
sector, b) the higher education sector, which covered researchers, teachers, and students, and finally
c) a mixed group, with people working as professionals and in the education sector at the same time.
Secondly, this last category that accounted for 11% of all participants was excluded from calculations
in order to obtain a 2x2 contingency table. Similar mergings were conducted on the 4-level scales of
the questions addressing the reasons for following and consequences of following researchers on
Twitter. As a result, participants’ answers were reduced to two groups, not or slightly applicable, and
moderately and strongly applicable. In this way, Fisher’s exact test could be carried out.

3. Results

3.1 Demographic data and general use of Twitter

The distribution of participants according to their sex is even, as 51% of respondents are male and
49 % female. Conversely, when grouped by age range, participants tend to concentrate in two out of
five categories, specifically between twenty-six and forty years of age (41,5%) and between forty-one
and fifty years of age (28,3) (Table 1).

93
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

AGE % Σ%
Under 25 5.7 5.7
Age 26-40 41.5 47.2
Age 41-50 28.3 75.5
Age 51-65 24.5 100
Over 65 0 100
Table 1. Participants distribution by age range

Participants were asked to grade their use of Twitter on a scale from 1 (“mostly personal use”) to 10
(“mostly professional use”). Twitter resulted as used mostly for professional (mean=7.4, median=8,
mode=10, SD=2.4) than personal reasons. Participants were also asked to grade their participation on
Twitter on a scale from 1 (“passive, I do not take part”) to 10 (“very active”), and they reported a
moderately active participation (mean=6.6, median=7, mode=7, SD=2.1).

3.2 Participants’ main occupation and area of specialization

Participants could select more than one option in order to describe their main occupation, and the
greatest percentage (60.4%) introduced themselves as professionals, working either in the public
administration or in the private sector and as freelance (Table 2). Only 24.5% of participants
introduced themselves as researchers, 18.9% as professors, and 20.7 as either postgraduate or
undergraduate students.
Occupation %
Researcher 24.5
Professional – public administration 41.5
Professional – private sector 13.2
Professional – freelance 5.7
Professor 18.9
Undergraduate student 7.5
Postgraduate student 13.2
Other 3.8
Table 2. Participants distribution by occupation

A subsequent classification based on participants’ responses allowed to produce 4 categories: 1)

“Professionals”, which includes either those working in the public sector, the private sector or as
freelance; 2) “Higher education/research”, which contains those who selected the option
“researchers” or “professors”, and all students; 3) “Mixed”, which includes those that worked as
professionals and at the same time as researchers or university professors or were also college students;
4) and finally, “Others” , i.e. all answers that could not be identified (Figure 1). Once more

94
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

professionals (49%) emerged above researchers (36%) as the most prominent occupation category in
this sample.

Figure 1. Participants distribution by main occupation

Finally, as for participants’ main area of specialization, 81% of survey respondents indicated LIS,
while the rest were experts in closely related fields such as journalism, economics, computer science
or digital marketing.

3.3 Reasons for following and interacting with researchers on Twitter

Asked about the reasons for following researchers on Twitter, participants declared that they use
Twitter basically to keep up to date (66%) and to obtain ideas for their work (50.9%), as shown in
Figure 2 below. Also Twitter affords participants the possibility of showcasing their own work
(35.8%), though it is less frequently used to interact and debate with other members, either
professionals (17%) or scientists and researchers (9.4%). Interactions with professionals appear more
common reasons for following researchers on Twitter than interactions with scientists and
researchers. “Interacting and debating with professionals in my specialization” was never or slightly
applicable for 45% of participants, and moderately and strongly applicable for 54.7% of participants.
Conversely, “interacting and debating with scientists and researchers” was never or slightly applicable
for 66% of participants, and moderately and strongly applicable for 33.9% of participants. “Getting
closer to the world of science and research”, however, is slightly or moderately applicable for 67.9%

95
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

of participants. Comments appended in a subsequent open-ended question underpinned Twitter’s

role as a source of ideas and information.

Figure 2. Reasons for following researchers on Twitter

The most common way of interaction is retweeting followed by mentioning, as 86.8% and 64%
respectively of participants declared they did so often or almost always (Figure 3).

Figure 3. Types of interaction with other researchers on Twitter

96
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Other communication channels are never or rarely used by 62.3% of participants, whilst sending
private messages is something quite unusual (83% of participants used them never or rarely). The
other channels of communication were specified in a subsequent open-ended question. Participants
mentioned Facebook, email, WhatsApp, and other professional or academic social networks, such as
LinkedIn, ResearchGate and Mendeley.
Figure 4 shows participants’ reasons for interacting with researchers and other professionals on
Twitter. Usually they do so to solve doubts and ask questions (54.7%) and provide more information
(41.5%). On the other hand, becoming better known within the community appears to be a less
common reason for interacting, as only 22.6% of respondents agreed on this. In a subsequent open-
ended question, some of them considered that Twitter gave them the opportunity to find employment
and make new contacts for future works or collaborations.

Figure 4. Reasons for interacting with researchers on Twitter

3.4 Results of following researchers or professionals on Twitter

Most of the participants agreed to different extents that, as a result of following other users on Twitter,
it is easier to keep up to date (85%) as well as implement new scientific and technical advances in
their professional areas (79%). Respondents also reported that they felt more motivated with their
field of specialization (81.1%), increased their professional circle (77.4%) and perceived themselves

97
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

as members of a professional or scientific community (75.5%). Strikingly, most respondents (67.9%)

felt that following researchers on Twitter did not allow them to influence research trends (Figure 5).

Figure 5. Consequences of following researchers on Twitter

3.5 Correlation tests

The Fisher’s exact test was run in order to test the null hypothesis that there was no differences
between professionals and members of the higher education sector as for the reasons and
consequences of following researchers on Twitter. However, in no case could the null hypothesis be
rejected, which means that participants’ occupation appears to have no influence on the reasons for
following researchers on Twitter nor on the consequences of doing so. Similarly, Fisher’s exact test
was run to look for a possible relationship between the channel preferred for interacting on Twitter
(retweet, mention, private message, or other channels) and occupation. As in the previous cases, the
null hypothesis could not be rejected.

4. Discussion and conclusions

4.1 Limitations
In the introduction of this research paper we wondered what kind of audience followed researchers
on Twitter, and whether the interactions between researchers and their followers could be looked at
as “productive interactions”, i.e. as encounters in which science somehow makes its way to society.
The answers provided here proceed from a survey with 53 Twitter users who were also followers of

98
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

active Spanish researchers in the area of LIS. This sample is indeed little representative, though we
already explained that certain characteristics of Twitter by themselves make it difficult to reach
followers as potential respondents, whilst poor response rates appear in other similar Twitter surveys
such as Mohammadi, Thelwall, Kwasny et al. (2018). Another limitation is that we studied a single
interdisciplinary discipline, LIS, and research-society interactions depend highly on the field studied
(De Jong, Barker, Cox et al. 2014). According to Chikoore, Probets, Fry, et al. (2016), academics
working in pure sciences look at the general public as a more relevant audience than academics
working in applied disciplines. Notwithstanding these limitations, considering that little research has
been published on this topic (Ke, Ahn, and Sugimoto 2017; Mohammadi et al. 2018) and considering
as well the demand for more survey research as explained in the introduction, this study adds some
knowledge about the identities, reasons and expectations of those following researchers on Twitter,
and allows to describe, to a certain extent, the interactions taking place between researchers and their
followers on Twitter.

4.2 Who is interacting with researchers on Twitter?

According to the results obtained, it can be concluded that the researchers studied here are followed
mostly by professionals (49%), though the higher education sector (including researchers, professors,
and students) accounts for an important part of the audience (36%). This means at least that
professionals are interested in knowing what researchers are saying on social web media, confirming
the results of Mohammadi et al. (2018) survey in which practitioners resulted as a secondary though
important group of people tweeting academic articles only preceded by the scientific community
itself. Budge et al. (2016), reflecting on their participation on Twitter as academics, use the concept
of “messy identities” to describe the fact that, on Twitter, they can perform indifferently as artists and
academics. Conversely, in this study, participants presented themselves mostly as professionals,
secondarily as members of the higher education sector, and only to a minor extent as both of this
(11%), which means that they did not perceive their identity as “messy”. Additionally, they look at
Twitter more as a tool for professional than personal communication. The prevalence of professionals
might be a characteristic of LIS more than a feature of the sample studied here. A 2002 survey (Powell,
Baker and Mika 2002) found that 89% American and Canadian LIS practitioners read scientific
literature on a regular basis, and almost 40% had engaged in actual research occasionally or
frequently. LIS practitioners might be more inclined than practitioners in other areas to have an
interest for research, and this might have affected results.

4.3 What kind of interactions?

The kind of activity that participants perform on Twitter consists mostly of keeping updated (strongly
or moderately applicable in 92% cases), obtaining ideas to implement in one’s own work (strongly or
moderately applicable in 83% cases), finding out the most interesting and novel research themes in
their area (strongly or moderately applicable in 81% cases), and working for one’s own visibility
(strongly or moderately applicable in 73% cases). These reasons for following researchers on Twitter
differ only to a certain extent from the reasons for publishing on academic or professional journals
that Clapton (2010) described studying 100 UK-based LIS practitioners. Clapton (2010) found that

99
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

practitioners publish mainly to share their ideas with others (84%), secondly for professional
development (78%), and thirdly to raise their profile (47%), something that, transferred to the social
web, can be seen as shaping one’s own visibility. Being on Twitter appears as something similar,
probably alternative, to publishing academic literature, with the difference that on Twitter this activity
is public. Haustein et al. (2014) found that publishing on Twitter and publishing scientific literature
were to some extent alternatives for a sample of very active physicists on Twitter, who tended to be
less productive in terms of scientific publications.
Interactions appear as an important though secondary facet of these users’ activity on Twitter. They
take place more often with other professionals (strongly or moderately applicable in 54% cases) than
with researchers or scientists (strongly or moderately applicable in 33% cases). The types of
interactions performed on Twitter are highly dependent on the affordances of the platform.
Interaction occurs mostly by retweeting (used often or almost always by 86% of participants) or
mentioning (used often or almost always by 64% of participants), i.e. two Tweeter-specific
communication devices. Other channels of communication are used often or almost always in 37.7%
of cases. There are no differences in the form of interaction chosen between professionals and the
higher education sector. Nor are there differences between professionals and the higher education
sector as for the reasons for following researchers on Twitter and for the consequences of following
them. The interactions described by the sample studied here are clearly Twitter-shaped interactions.
Draucker (2015) characterizes communication on Twitter as an interaction between “broadcasters”
and “followers”, contending that the characteristics of the medium affect the nature of
communication, and that participation frameworks need to be studied in order to understand the
specificities of communication on new media. Following Draucker (2015), the possibility of
addressing them converts broadcasters into participants in the communication. However, the
audience studied here, which is mainly made up of professionals, reported that interactions occur
more with other professionals than scientists and researchers, who appear as a less approachable
target.
As a consequence of this activity as Twitter followers, more than three quarters of participants agreed
to different extents that they could more easily keep updated and implement technical and scientific
advances in their daily practice, they felt more motivated, and they felt as part of a community. On
other dimensions more related to the social reach of science and research, responses were less clear-
cut, though meaningful. For instance, almost 70% and 54.7% of participants felt that they understand
better, respectively, the scientific terminology and the scientific method. Conversely only 32.1%
considered that they can influence research trends − and 67.9% straightly disagreed on this. It is
difficult to say if this perceived little influence on research trends is really a consequence of being a
follower in a mostly “unidirectional” flow of information, in which participants interact mostly by
retweeting, i.e. forwarding information to their own followers. Whatever the reason, communication
flows on Twitter sometimes appear to be at least “mismatched”. Bombaci et al. (2016), for example,
studied tweets published during a conference of Conservation Biology. The authors, mainly
academics, intended to reach, in order of importance, other academics, government agencies and
policy makers, though actually it appeared that most tweets reached twice as much as intended to
other academics, less than one-fourth as much as intended government agencies, whilst policy makers
were not reached at all. In Bombaci et al.’s study broadcasters appeared to have little control over

100
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

their followers, whereas in our study followers interacted back with researchers and scientists rarely.
Müllerleile (2014) warns that social media alone may fail as a strategy for engagement with the public,
though they work well for disseminating contents in combination with other tools. The results of
Bornmann (2018) point to the same direction concluding that altmetrics might reveal the degree of
unknown attention paid to research more than the link between research and action in society.
It is difficult to say if the followers’ activity described here qualify as a type of “productive interaction”
because the concept itself of productive interaction is still undetermined and mostly analyzed in case-
study approaches (Spaapen and Van Drooge 2011; De Jong et al. 2014). The results of this study allow
to conclude that the knowledge acquired by following researchers on Twitter definitively is relevant,
considering participants’ claims about the consequences of following researchers on Twitter.
However, it cannot be said if it is robust. Concerns about the veracity and the proper interpretation
of the scientific knowledge on social media exist, especially in disciplines such as conservation science
(Darling, Shiffman, Cote et al. 2013; Bombaci et al. 2016). On the other hand, the interaction
described here appears scarcely interdisciplinary, as only 19% of participants belong to disciplines
others than LIS, confirming previous results. De Jong et al. (2014) contend that, when societal impact
of research is under evaluation, researchers in other research fields should be considered as important
stakeholders. In this sense, the monodisciplinary interactions described here do not appear
meaningful in a societal perspective, as they don’t in similar studies of research communities on
Twitter. The monodisciplinary nature of communities around scientific profiles on Twitter appears
also in Ke, Ahn and Sugimoto (2017), who identify more than 45000 scholarly profiles on Twitter,
and conclude that scholars tend to follow profiles of their same discipline across all disciplines.
Participating on Twitter makes participants more acquainted with science and research in terms of
implementation of research findings, and comprehension of research terminology and research
methods. Indeed, knowledge acquired on Twitter has an application in the participants’ professional
sphere, a condition that Spaapen y van Drooge (2011) emphasized for interactions to be productive.
However, the results of this study do not allow to describe exhaustively the type of relational
engagement taking place among the researchers broadcasting and the audience following them. Some
of the results point to a mainly unidirectional flow of information in which followers feel that they
have little influence on the activities performed by broadcasters (the researchers). Conversely, when
the assessment of societal impact of research is concerned, it is understood that knowledge
dissemination occurs in complex multidirectional networks (Spaapen and Van Drooge 2011; Ozanne
et al. 2017) or at least in bidirectional interactions (Greenhalgh et al. 2016). The question we should
address in future research is whether being on Twitter has some implications on research activities as
well, in the sense of reflecting societal concerns and problems as discussed and proposed by followers.
Recent research points to the fact that the research articles discussed on Twitter tend to be more
broadly oriented than research articles in general, at least within the area of climate change
(Haunschild, Leydesdorff, Bornmann et al. 2018). If it might reveal certain awareness of the audience,
it does not guarantee that a real communication with the public occurs.
As a conclusion, it can be said that being a follower on Twitter accomplishes for the professionals
surveyed here a similar role to publishing, in the sense of exchanging ideas and gaining visibility
through participation and presence in a monodisciplinary community. The scientific communication
system allows scientists to participate in knowledge production abiding by specific rules, such as citing

101
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

previous relevant work. In this study, participation rules are defined to an important extent by the
affordances of the platform itself. From this point of view, interactions on Twitter do not allow easy
engagement because information flows mostly in a unidirectional way and researchers’ followers act
mainly as retweeters of the information they receive. There are clear advantages for the participants
who follow our sample of researchers, which derive from being part of a Twitter community.
However, their ability to truly interact is limited and as a consequence it seems that they have little
influence on research and researchers. Future research should address this question from the point of
view of researchers, and better document in LIS and other areas how interactions on social media
occur and with whom. Only when researchers’ interactions on the social web are really proved to
reveal some kind of societal impact, an important issue to address in future altmetrics research will be
how to quantify and measure interactions and engagement.

References

Alperin, Juan Pablo, Charles J. Gomez, and Stefanie Haustein. 2018. “Identifying diffusion patterns
of research articles on Twitter: A case study of online engagement with open access articles”. Public
Understanding of Science, 0963662518761733.
Álvarez-Bornstein, Belén and Michela Montesi. 2016. “La comunicación entre investigadores en
Twitter. Una etnografía virtual en el ámbito de las ciencias de la documentación”. Revista Española
de Documentación Cientifìca 39, 4:e156. https://doi.org/10.3989/redc.2016.4.1352.
Archie, Kelli M., Lisa Dilling, Jana B. Milford, and Fred C. Pampel. 2014. “Unpacking the
‘information barrier’: Comparing perspectives on information as a barrier to climate change
adaptation in the interior mountain West”. Journal of Environmental Management 133:397−410.
https://doi.org/10.1016/j.jenvman.2013.12.015.
Bombaci, Sara P., Cooper M. Farr, H. Travis Gallo, Anna M. Mangan, Lani T. Stinson, Monica
Kaushik, and Liba Pejchar. 2016. “Using Twitter to communicate conservation science from a
professional conference”. Conservation Biology 30, 1:216−225. https://doi.org/10.1111/cobi.12570.
Bornmann, Lutz. 2012. “Measuring the societal impact of research”. EMBO Reports 13, 8:673−676.
Doi: 10.1038/embor.2012.99.
Bornmann, Lutz. 2013. “What is societal impact of research and how can it be assessed? A literature
survey”. Journal of the American Society of Information Science and Technology 64, 2:217−233.
https://doi.org/10.1002/asi.22803.
Bornmann, Lutz. 2014. “Do altmetrics point to the broader impact of research? An overview of
benefits and disadvantages of altmetrics”. Journal of Informetrics 8, 4:895−903.
https://doi.org/10.1016/j.joi.2014.09.005.
Bornmann, Lutz. 2015. “Alternative metrics in scientometrics: A meta-analysis of research into three
altmetrics”. Scientometrics 103, 3:1123−1144. https://doi.org/10.1007/s11192-015-1565-y.

102
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Bornmann, Lutz. 2015a. “Usefulness of altmetrics for measuring the broader impact of research: A
case study using data from PLOS and F1000Prime”. Aslib Journal of Information Management 67,
3:305−319. https://doi.org/10.1108/AJIM-09-2014-0115.
Bornmann, Lutz, Robin Haunschild, and Jonathan Adams. 2018. “Convergent validity of altmetrics
and case studies for assessing societal impact: an analysis based on UK Research Excellence
Framework (REF) data.” In Proceedings of Science & Technology Indicators (STI), September 12-14,
Leiden, 41−48.
Boyd, Danah, Scott Golder, and Gilad Lotan. 2010. “Tweet, tweet, retweet: Conversational aspects of
retweeting on twitter.” 43rd Hawaii International Conference on System Sciences.
Budge, Kylie, Narelle Lemon and Megan McPherson. 2016. “Academics who tweet: ‘messy’ identities
in academia”. Journal of Applied Research in Higher Education 8, 2:210−221.
https://doi.org/10.1108/JARHE-11-2014-0114.
Carpenter, Jeffrey P., and Daniel G. Krutka. 2014. “How and why educators use Twitter: A survey of
the field”. Journal of Research on Technology in Education 46, 4:414−434.
Case, Donald O., and Lisa M. Given (eds.). 2016. Looking for information. A survey of research on
information seeking, needs and behavior. Bingley: Emerald Group Publishing Limited.
Chikoore, Lesley, Steve Probets, Jenny Fry, and Claire Creaser. 2016. “How are UK academics
engaging the public with their research? A cross-disciplinary perspective”. Higher Education
Quarterly 70, 2:145−169. https://doi.org/10.1111/hequ.12088.
Clapton, Janet. 2010. “Library and information science practitioners writing for publication:
motivations, barriers and supports”. Library and Information Research 34 106:7−21.
Darling, Emily S., David Shiffman, Isabelle M. Cote, and Joshua A. Drew. 2013. “The role of Twitter
in the life cycle of a scientific presentation”. Ideas in Ecology and Evolution 6:32–43.
De Jong, Stefan, Katharine Barker, Deborah Cox, Thordis Sveinsdottir, and Peter Van den Besselaar.
2014. “Understanding societal impact through productive interactions: ICT research as a case”.
Research Evaluation 23, 2:89−102. https://doi.org/10.1093/reseval/rvu001.
Dilling, Lisa, and María Carmen Lemos. 2011. “Creating usable science: Opportunities and constraints
for climate knowledge use and their implications for science policy”. Global environmental change 21,
2:680−689. https://doi.org/10.1016/j.gloenvcha.2010.11.006.
Draucker, Fawn. 2015. “Participation structures in Twitter interaction: Arguing for the broadcaster
role”, in Participation in Public and Social Media Interactions, edited by Marta Dynel and Jan
Chovanec. Amsterdam/Philadelphia: John Benjamin Publishing Company, 49−66.
https://doi.org/10.1075/pbns.256.03dra.
Greenhalgh, Trisha, James Raftery, Steve Hanney, and Matthew Glover. 2016. “Research impact: a
narrative review”. BMC Medicine 14, 1. Doi: 10.1186/s12916-016-0620-8.

103
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Gruzd, Anatoliy, Barry Wellman and Yuri Takhteyev. 2011. “Imagining Twitter as an imagined
community”. American Behavioral Scientist 55, 10:1294−1318.
https://doi.org/10.1177/0002764211409378.
Haustein, Stefanie. 2016. “Grand challenges in altmetrics: heterogeneity, data quality and
dependencies”. Scientometrics 108, 1:413−423. https://doi.org/10.1007/s11192-016-1910-9.
Haustein, Stefanie, and Rodrigo Costas. 2015. “Identifying Twitter audiences: who is tweeting about
scientific papers?” ASIS&T SIG/MET Metrics 2015 Workshop, available at:
https://www.asist.org/SIG/SIGMET/wp-content/uploads/2015/10/sigmet2015_paper_11.pdf.
Haustein, Stefanie, Timothy D. Bowman, and Rodrigo Costas. 2015. “Interpreting “altmetrics”:
Viewing acts on social media through the lens of citation and social theories”. In Theories of
Informetrics and Scholarly Communication, edited by Cassidy Sugimoto. Berlin: Walter de Gruyter,
372−406.
Haustein, Stefanie, Timothy D. Bowman, Kim Holmberg, Isabella Peters, and Vincent Larivière.
2014. “Astrophysicists on Twitter: An in-depth analysis of tweeting and scientific publication
behavior”. Aslib Journal of Information Management 66, 3:279−296. https://doi.org/10.1108/AJIM-
09-2013-0081.
Haustein, Stefanie, Rodrigo Costas, and Vincent Larivière. 2015. “Characterizing social media metrics
of scholarly papers: the effect of document properties and collaboration patterns”. PloS one 10, 3:
e0120495. https://doi.org/10.1371/journal.pone.0127830.
Haustein, Stefanie, Isabella Peters, Judit Bar-Ilan, Jason Priem, Hadas Shema, and Jens Terliesner.
2014. “Coverage and adoption of altmetrics sources in the bibliometric community”. Scientometrics
101, 2:1145−1163. https://doi.org/10.1007/s11192-013-1221-3.
Haunschild, Robin, Loet Leydesdorff, Lutz Bornmann, Iina Hellsten and Werner Marx. 2018. “Does
the public discuss other topics on climate change than researchers? A comparison of networks based
on author keywords and hashtags”. arXiv preprint arXiv:1810.07456. Available at:
https://arxiv.org/abs/1810.07456 [Last access 03.01.2019].
Holmberg, Kim. 2015. Altmetrics for Information Professionals: Past, Present and Future. Amsterdam
etc.: Chandos Publishing.
Holmberg, Kim., Timothy D.Bowman, Stefanie Haustein, and Isabella Peters. 2014.
“Astrophysicists’ Conversational Connections on Twitter”. PloS one 9, 8:e106086.
https://doi.org/10.1371/journal.pone.0106086.
Ke, Qing, Yong-Yeol Ahn, Cassidy R. Sugimoto. 2017. “A systematic identification and analysis of
scientists on Twitter”. PloS one 12, 4:e0175368.
Jeng, Wei., Daqing He, and Jiepu Jiang. 2015. “User participation in an academic social networking
service: A survey of open group users on Mendeley”. Journal of the Association for Information Science
and Technology 66, 5:890−904. https://doi.org/10.1002/asi.23225.

104
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Marwick, Alice E. 2014. “Ethnographic and Qualitative Research on Twitter.” In Twitter and Society,
edited by Katrin Weller, Axel Bruns, Cornelius Puschmann, Jean Burgess, and Merja Mahrt. New
York: Peter Lang, 109−122.
Mas-Bleda, Amalia, Mike Thelwall, Kayvan Kousha, and Isidro F. Aguillo. 2014. “Do highly cited
researchers successfully use the social web?” Scientometrics 101, 1:337−356.
https://doi.org/10.1007/s11192-014-1345-0.
Mohammadi, Ehsan, Mike Thelwall, and Kayvan Kousha. 2018. “Academic information on Twitter:
A user survey”. PloS one, 13, 5, e0197265.
Mohammadi, Ehsan, Mike Thelwall, and Kayvan Kousha. 2016. “Can Mendeley bookmarks reflect
readership? A survey of user motivations”. Journal of the Association for Information Science and
Technology 67, 5:1198−1209. https://doi.org/10.1002/asi.23477.
Mohammadi, Ehsan, Mike Thelwall, Stefanie Haustein, and Vincent Larivière. 2015. “Who reads
research articles? An altmetrics analysis of Mendeley user categories”. Journal of the Association for
Information Science and Technology 66, 9:1832−1846. https://doi.org/10.1002/asi.23286.
Müllerleile, Andreas. 2014. “European Studies and Public Engagement: A Conceptual Toolbox”.
Journal of Contemporary European Research 10, 4:505−517.
Na, Jin-Cheon. 2015. “User Motivations for Tweeting Research Articles: A Content Analysis
Approach”. In Allen, R.B., Hunter, J., Zeng, M.L. (Eds.), Digital Libraries: Providing Quality
Information. Proceedings of the 17th International Conference on Asia-Pacific Digital Libraries, ICALD,
2015, Seoul, Korea, December 9-12, 197−208.
Ozanne, Julie L., Brennan Davis, Jeff B. Murray, Sonya Grier, Ahmed Benmecheddal, Hilary
Downey, et al. 2017. “Assessing the Societal Impact of Research: The Relational Engagement
Approach”. Journal of Public Policy & Marketing 36, 1:1−14. http://dx.doi.org/10.1509/jppm.14.121.
Pedrini, Matteo, Valentina Langella, Mario A. Battaglia and Paola Zaratin. 2018. “Assessing the health
research’s social impact: a systematic review”. Scientometrics 114, 3:1227−1250.
Pickard, Alison. J. 2013. Research methods in information. London: Facet publishing.
Powell, Ronald. R., Lynda M. Baker, and Joseph J. Mika. 2002. “Library and information science
practitioners and research”. Library & Information Science Research 24, 1:49−72.
https://doi.org/10.1016/S0740-8188(01)00104-9.
Priem, Jason. 2014. “Altmetrics”, in Beyond Bibliometrics: Harnessing Multidimensional Indicators of
Scholarly Impact, edited by Blaise Cronin and Cassidy Sugimoto. MIT Press, 263−287.
Priem, Jason, Dario Taraborelli, Paul Groth, and Cameron Neylon. 2010. “Altmetrics: A manifesto”,
available at: http://altmetrics.org/manifesto/ (last access 1.31.2017).
Quan-Haase, Anabel, Lori McCay-Peet. 2015. “Networks of digital humanities scholars: The
informational and social uses and gratifications of Twitter”. Big Data & Society 2, 1.
http://dx.doi.org/10.1177/2053951715589417.

105
JLIS.it 10, 2 (May 2019)
ISSN: 2038-1026 online
Open access article licensed under CC-BY
DOI: 10.4403/jlis.it-12530

Spaapen, Jack, and Leonie Van Drooge. 2011. “Introducing ‘productive interactions’ in social
impact assessment”. Research Evaluation 20, 3:211−218.
https://doi.org/10.3152/095820211X12941371876742.
Sud, Pardeep, and Mike Thelwall. 2014. “Evaluating altmetrics”. Scientometrics 98, 2:1131−1143.
Doi: 10.1007/s11192-013-1117-2.
Thelwall, Mike, and Kayvan Kousha. 2015. “Web indicators for research evaluation. Part 2: Social
media metrics”. El profesional de la Información 24, 5:607−620.
https://doi.org/10.3145/epi.2015.sep.09.
Torres-Salinas, Daniel, and Yusnelkis Milanés-Guisado. 2014. “Presencia en redes sociales y
altmétricas de los principales autores de la revista El Profesional de la Información”. El Profesional
de la Información 23, 4:367−372. https://doi.org/10.3145/epi.2014.jul.04.
Vaughan, Liwen. 2001. Statistical methods for the information professional: A practical, painless
approach to understanding, using, and interpreting statistics. Medford, New jersey: Information Today,
Inc.
Veletsianos, G. 2012. “Higher education scholars’ participation and practices on Twitter”. Journal of
Computer Assisted Learning 28, 4:336−349. https://doi.org/10.1111/j.1365-2729.2011.00449.x.
Weinberg, Sharon L., and Sarah k. Abramowitz. 2008. Statistics using SPSS: An integrative approach.
Cambridge: Cambridge University Press.
Zuccala, Alesia. 2009. “The lay person and Open Access”. Annual Review of Information Science and
Technology 43, 1:1−62. https://doi.org/10.1002/aris.2009.1440430115.

106