Content uploaded by Sayed-Amir Marashi
Author content
All content in this area was uploaded by Sayed-Amir Marashi on Aug 05, 2017
Content may be subject to copyright.
RT. A Journal on Research Policy & Evaluation 5 (2017)
Submitted on 16 January 2017, accepted on 26 July 2017, published on 4 August 2017
Doi: 10.13130/2282-5398/7997
How accessibility influences citation counts: The case of
citations to the full text articles available from ResearchGate
Mohammad Sababi*, Sayed Amir Marashi, Moohebat Pourmajidian, Sana S.
Pourtabatabaei, Faezeh Darki*, Mohammad Reza Sadrzadeh,
Mohsen Dehghani, Amirali Zandieh*, Mohammad Kazem Zim,
Meisam Yousefi, Monireh Jamalkhah, S. Kasra Tabatabaei, Fatemeh Safaeifard,
Andisheh Talaei, Motahareh Sobat, Faezeh Moakedi, Pouyan Nejadi
Abstract: It is generally believed that the number of citations to an article can positively be correlated
to its free online availability. In the present study, we investigated the possible impact of academic
social networks on the number of citations. We chose the social web service “ResearchGate” as a case.
This website acts both as a social network to connect researchers, and at the same time, as an open
access repository to publish post-print version of the accepted manuscripts and final versions of open
access articles. We collected the data of 1823 articles published by the authors from four different
universities. By analyzing these data, we showed that although different levels of full text availability
are observed for the four universities, there is always a significant positive correlation between full
text availability and the citation count. Moreover, we showed that both post-print version and
publisher’s version (i.e., final published version) of the archived manuscripts receive more citations
than non-OA articles, and the difference in the citation counts of post-print manuscripts and
publisher’s version articles is nonsignificant.
Keywords: academic social web services; gold open access; green open access; full text; visibility;
vitation analysis
*School of Biology, College of Science, University of Tehran, Tehran, Iran
Department of Biotechnology, College of Science, University of Tehran, Tehran, Iran
Institute of Biochemistry and Biophysics, University of Tehran, Tehran, Iran
School of Mathematics, Statistics and Computer Science, College of Science, University of Tehran, Tehran, Iran
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
2
1. Introduction
Citation analysis is commonly used as an approach to evaluate quality of research articles and
performance of researchers and journals (Eysenbach, 2006; Garfield, 2006a). Most of the
scientometric measures of publishing performance, including impact factor and h-index, are based on
citation statistics (Garfield, 2006b; Braun et al, 2006). Consequently, it is important to understand how
factors other than scientific merit may influence the number of article citations (Moed, 2009; Falagas
et al, 2008).
Many studies have previously shown the positive impact of free online availability of articles on
their citations (Antelman, 2004; Niyazov et al, 2016). For this reason, it is not surprising that the total
number of open access (OA) scholarly journals and the number of research articles published in these
journals have increased significantly during the last decade (Kamat, 2015; Laakso et al, 2011). In
contrast, some studies showed that there may not be a significant correlation between citation count
and free accessibility of articles (Davis, 2008; Craig et al, 2007) and other factors could affect citation
counts too. For example, Gaule and Maystre suggested that the higher number of citation of OA
articles could be due to self-selecting of higher quality articles to be open-access by authors (Gaule et
al, 2011).
Today, in the era of social networking websites, a striking increase in communication and
collaboration among scientists has occurred (Wren, 2005; Moed et al, 2016). This new way of scientific
communication is now so popular that some scientometric measures, e.g., Altmetrics, are introduced
based on the citations and bookmarkings within social web services (Niyazov et al, 2016; Ale Ebrahim
et al, 2014; Van Noorden, 2015). There are also a number of academic social network websites
designed for people in the field of science and technology, including Academia.edu, Mendeley and
ResearchGate. Such web services are primarily developed to help researchers in communicating and
sharing scientific texts with other scholars in an efficient way (Ovadia, 2014), although they are also
suggested to be used for evaluation purposes (Hoffmann et al, 2016; Min-ChunYu et al, 2016).
ResearchGate, as one of the most popular social networks among academics, currently has over 10
million registered members and has indexed more than 100 million articles. Moreover, ResearchGate
also acts as an open access repository and encourages the researchers to archive a post-print version
of their manuscripts (following the “green OA” model). Despite the fact that this website asks the
members to abide by the laws of copyright, in practice many authors publish the publisher’s version
of their works without permission from publisher (Kamat, 2015; Thelwall et al, 2015; Jamali, 2015).
In the present work, we collected the data of 1823 articles published by the authors from four
different universities, namely King’s College London, University of Cape Town, Peking University,
and University of Tehran. We study whether publishing OA articles or posting manuscripts on
ResearchGate depends on the researcher’s academic affiliation. Furthermore, we investigate how
being OA or being freely available can influence the number of citations to an article.Among
researchers, it sometimes happens that practitioners of a heterodox methodology complain about the
(either real or alleged) tyranny of a mainstream school of thought. Whereas some of these quarrels
depend on private rivalries, it is almost indisputable that, within many disciplines, there actually are
two or more rival schools of thought whose followers periodically engage in disputes and try to
ostracize each other. I argue that, as far as these rivalries are genuine (i.e., as far as they are rooted in
some genuinely epistemological disagreement), such disputes highlight social phenomena that have
important epistemic consequence.
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
3
2. Materials and Methods
2.1 Analysis of the papers published by four universities
Based on Alexa ranking, United Kingdom, China and Iran are among the top countries with the
highest number of ResearchGate visits.
i
South Africa is a developing country in which English is the
official scientific language. Therefore, we included this country in our study for comparing its
characteristics to the United Kingdom on the one hand, and to Iran and China on the other hand. In
our study, we chose one representative university from each country (Table 1). One should keep in
mind that based on one single university, it is not possible to draw any general conclusion among the
profiles of these countries.
University name
Abbreviation
Country
Number of analyzed
documents
King’s College London
KCL
United Kingdom
505
University of Cape Town
UCT
South Africa
228
Peking University
PU
China
480
University of Tehran
UT
Iran
610
Table 1. List of selected universities in the present study
In June 2015, the profile of the universities was searched at the ResearchGate website
(https://www.researchgate.net), and researchers from different research areas were randomly chosen.
For the University of Cape Town, we considered all of the authors. Altogether, the publication data
of all chosen researchers were obtained, which comprises 1823 articles published from 2012 to 2014.
In some cases, we observed an inconsistency between the publication year reported in the journal
website and the publication year reported in ResearchGate. In such cases, the date reported in the
journal website was considered as correct.
For each article, the following data were collected:
Impact factor of the journal in which the paper is published;
The number of authors;
Availability as an OA article via the journal published website;
Availability of the full text article (as the post-print or publisher’s version) via ResearchGate;
Total citation count based on Google Scholar website (in June 2015)
In the present study, we adopted the following definitions. An article was considered as a gold OA
article if its full text is downloadable from the journal website. If an article is not available as gold OA,
but its full text is available via ResearchGate, then it is considered as a green OA article (Craig et al,
2007; Harnad et al, 2004). Such a green OA article is either available in ResearchGate as a post-print
(that is, the final author’s draft after refereeing ) or as a publisher’s version (the final copy of the peer
reviewed edited full text in the journal layout) (Craig et al, 2007). Finally, if the full text article is
available neither from the journal website not from ResearchGate, then it is considered as a non-OA
article. It should be mentioned that the full text of gold OA articles are almost always (automatically)
available in ResearchGate. Additionally, our definition of green OA might be an underestimation of
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
4
the real definition, as full texts might be available from other websites and repositories (e.g., ArXiv),
but not ResearchGate.
2.2. Statistical Analysis
In the present work, the data were generally not normally distributed and most of the variables were
integer-valued. Therefore, for comparing two distributions, we used Wilcoxon rank-sum test. To
determine whether there is a significant difference between two observed frequencies in two
categories, we used chi-squared test.
3. Results
Figure 1 shows the number of articles whose full texts are available in ResearchGate (i.e., the circles)
and other articles (i.e., the non-OA articles). Note that the articles which are available as OA in the
journal website were also available from the ResearchGate. Additionally note that in the present work,
the term “green OA” refers to the articles whose full texts are freely available via ResearchGate (but
not the journal website).
Figure 1. Venn diagram representing the number of OA and non-OA articles published by
the four universities. In each panel, the circle represents the articles which are available as full
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
5
text in ResearchGate, either as gold OA articles (i.e., the internal, strong-yellow circle) or green
OA articles (i.e., the big, green circle). The green OA articles are either archived in the post-
print format or publisher’s version format. Those articles whose full texts are not available in
ResearchGate are counted as non-OA.
3.1. Increased citation count is linked to online availability of the article
In the first part of our analysis, we investigated if the number of citations to gold OA articles is
significantly greater than the number of citations to non-OA ones. A significant difference was found
between these two variables (p=2.5×10-9 in Wilcoxon test; OA mean rank=752 ±591.8; Non OA
mean rank=624.09 ±483.3; SI Table 1) which confirms that gold OA articles attract more citations
compared to non-OA articles.
Full text version of gold OA articles are almost always available in ResearchGate. However, we also
tested whether availability of green OA full text articles (in both post-print and publisher’s version
formats) influences the number of citations. The difference between the citation count to green OA
articles and the citation count of non-OA articles was significant (p=1.1×10-8 in Wilcoxon test; Green
OA mean rank=710.95 ±577.02 Non OA mean rank=601.74 ±440.94; SI Table 2). Such a significant
difference was also present if this analysis was independently repeated for each of the four universities
(data not shown).
We observed that the authors of non-OA articles may publish their works as post-prints or
publisher’s versions. Many journals allow online archiving of the manuscript in post-print form to
make it a green OA article. In contrast, archiving publisher’s version of papers is often not allowed.
We compared the number of citations to post-print and publisher’s version articles. No significant
difference was observed between the medians of the two distributions (p=0.12 in Wilcoxon test; Post
print mean rank=170.7 ±260.4 publisher’s version mean rank=240.8 ±62.8; SI Table 3) , which means
that it is not possible to conclude that publisher’s version of articles attract more citations compared
to the post-print versions. It should be mentioned that because of the low count of the post-print
articles and consequently low power of this test, these results are only preliminary results. Therefor a
more comprehensive study is required to obtain conclusive results.
3.2. Factors which influence full text availability of non-OA articles
In the next step we studied the factors which can influence full text availability of non-gold OA articles.
One important parameter is impact factor of the journal. Briefly, there was conjecture that one
motivation to archive and distribute the full text of an article on ResearchGate might be to
communicate low-quality articles with other researchers with the hope to attract more attention. On
the other hand, researchers may simply want to proudly share their high-quality findings with others.
To test which of the above scenarios might be stronger, the difference between impact factor (IF) of
Green OA and Non OA articles was explored. No significant difference was observed between the
two variables (p=0.079 in Wilcoxon test; Green OA IF mean rank:665.73±577.02; Non OA IF mean
rank=628.14 ±440.94; SI Table 4), suggesting that impact factor, in general, is not a major determinant
of full text archiving and availability. A similar analysis was performed for each of the universities, and
interestingly, only in case of UT and PU a significant difference between impact factor of Green OA
and NonOA articles observed(For PU: p=0.03 in Wilcoxon test; Green OA IF mean
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
6
rank:201.52±169.6; Non OA IF mean rank=165.95±102.77) and ( For UT: p=0.0001 in Wilcoxon
test; Green OA IF mean rank= 295.40±228.26; Non OA IF mean rank=255.98±209.42; SI Table 5).
It has been previously suggested that there is a positive correlation between number of authors of
an article and its citation count (Borsuk, 2009). In this study we showed that green OA articles may
attract more citations than non-OA articles. Here, we tested the difference between the number of
authors of green OA and non-OA articles to see if such significant difference in the number of authors
of green OA vs. non-OA articles exists. A significant difference was observed in this case (p=1.4×10-
5 in Wilcoxon test; Number of authors for Green OA mean rank=700.11±577.02; Number of authors
for Non OA mean rank=608.06±440.94; SI Table 6).
3.3. Pairwise comparison of the four universities
Nowadays, different ranking of universities are published every year. Some of these university ranking
systems include citation-dependent measures, e.g. h-index, in their analysis. Here, as a side result, we
show that citations to the papers published from a certain university might be linked with the OA-
related policies of that university/country, and the scientific merit of the papers is not the only
determinant of the citation count of each university. Note that due to the small sample size (i.e., only
four universities) and the nature of our analysis (which are to detect correlations and not causes), one
cannot extend the findings of the present analysis to draw explicit conclusions on OA rate for each
institution or their OA advantage.
The gold OA publication rate seems to be considerably different among the four universities (44%
in KCL, 55% in UCT, 27% in PU and 10% in UT). In this step, we compared each pair of universities
to see if OA publication rate is significantly different over these universities. The results are
summarized in Table 2. Interestingly, all of the pairwise differences were found to be significant.
University pair
Chi-square
p
-value
Test Statistic
KCL vs. UCT
6×10-3
7.01
KCL vs. PU
3×10-8
29.80
KCL vs. UT
1×10-37
162.48
UCT vs. PU
7×10-13
50.42
UCT vs. UT
6×10-43
186.29
PU vs. UT
6×10-13
50.61
Table 2. Comparing universities for their OA publication rate. In each case, a 2×2 contingency
table was constructed to summarize the number of OA vs. non-OA articles which are published by
University X vs. University Y. Then, the significance was computed as the chi-square p-value.
The archiving rate was also found to be comparable over the universities except for the case of UT
(17% in KCL, 18% in UCT, 20% in PU and 41% in UT). In the next step, we repeated a similar
analysis to compare the archiving rate of the non-gold OA articles. The results of this analysis are
shown in Table 3.
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
7
University pair
Chi-square
p
-value
Test Statistic
KCL vs. UCT
0.13†
2.198
KCL vs. PU
0.27 †
1.173
KCL vs. UT
6×10-5
15.889
UCT vs. PU
1×10-2
5.774
UCT vs. UT
0.32†
0.992
PU vs. UT
2×10-8
30.273
† Nonsignificant at the level of 0.05.
Table 3. Comparing universities for archiving rate of their non-gold OA articles. In each case,
a 2×2 contingency table was constructed to summarize the number of archived vs. not-
archived articles which are published by University X vs. University Y. Then, the significance
was computed as the chi-square p-value.
4. Discussion
4.1. Free full text availability is correlated with increased citation rate
It is suggested that free online availability of an article increases its citation potential (Eysenbach,
2006). In the present study, we showed that a similar pattern is observed for the dataset of articles
selected from ResearchGate. Moreover, some other studies had shown that archiving a manuscript
can increase its citation potential (Kurtz et al, 2005; Davis et al, 2007). Here, we show that full text
availability via ResearchGate has a similar positive effect. This finding can be explained by the fact
that the full text articles are not only available to ResearchGate members, but also accessible to others
via Google or Google Scholar (Jamali et al, 2015). It should be mentioned that in the present article
we only considered those OA articles whose full text version is available from the journal website or
from ResearchGate. In fact, the significant difference that was observed between the citation counts
of OA and non-OA articles might be even stronger, as many articles which are counted as non-OA
here might be in fact available via other platforms (like ArXiv). Therefore, such articles might attract
higher number of citations due to their actual OA availability.
As mentioned above, archiving manuscripts can increase the citation potential of the article. We
showed that such an increase is comparable in case of post-print and publisher’s version of the article.
Therefore, it is reasonable for the traditional non-gold OA publishers to encourage researchers to
archive their works in post-print format, as such availability can increase the citation rates, and
eventually, the journal impact factor.
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
8
4.2. The relationship between archiving non-gold OA articles and impact factor
In the present work, we showed that at least in case of Peking University and University of Tehran,
there is a significant correlation between impact factor and tendency for archiving their (otherwise
non-OA) articles in ResearchGate (section3.2). Interestingly, these two universities have the lowest
rate of gold OA publication among the four studied universities. This finding can be explained by the
fact that researchers in PU and UT, like other researchers, want to share their high-quality findings
with other investigators. In fact, a 2011 survey on author attitudes toward OA (Kenneway, 2011)
reported that 81.3% and 72.2% of participants from Iran and China, respectively, believed that OA
publishing is important. Interestingly, in developed countries like Japan and the United States, the
values are not as high (69% and 70.8%, respectively). Consequently, researchers in Iran and China
may tend to publish in gold OA journals, but due to non-scientific reasons (e.g., lack of financial
support for gold OA publication) they are not able to do so. In different surveys, it has been shown
that in developing countries, where institutional support is limited, serious concerns may exist for
those who cannot afford to pay normal OA publication fees (Schroter et al, 2005) and this may turn
such a fee into an important factor in selecting the journal (Solomon et al, 2012). Consequently, they
might use ResearchGate to freely share their work with others as gold OA articles.
In the present study, we showed that there is a significant positive correlation between full text
availability and the citation count. Moreover, we showed that both post-print version and publisher’s
version (i.e., final published version) of the archived manuscripts receive more citations than non-OA
articles. A possible drawback of our methodology is that the citation impact of full text and non-full
text articles are not compared among papers of the same discipline. This issue may potentially
influence the results, as both the shares of different access models and citation rates might be dissimilar
among different subject areas. This drawback, however, may be alleviated to some extent by the fact
that in our dataset, the ratio of papers within each subject area is fairly comparable (see Table 4).
Therefore, we believe that our analysis can still provide insight about the citation trends of articles.
Subject area
Percentage
Medical Sciences
19.7%
Agriculture and Biological Sciences
17.6%
Physical and Chemical Science
18.1%
Engineering
13.6%
Art, Social sciences and Humanities
13.1%
Other
17.9%
Table 4. The ratio of papers in our dataset in each major subject area. Note that if all ratios
were the same, one would expect each subject area to include 16.7% of all articles.
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
9
Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations
The following abbreviations are used in this manuscript:
OA: open access
non-OA: non-open access
IF: impact factor
CC: correlation coefficient
KCL: King’s College London
UCT: University of Cape Town
PU: Peking University
UT: University of Tehran
References
Ale Ebrahim, N., Salehi, Hadi, Embi, Mohamed A., Danaee, Mahmoud, Mohammadjafari, Marjan,
Zavvari, Azam, Shakiba, Masoud, Shahbazi-Moghadam, Masoomeh M. (2014), "Equality of Google
Scholar with Web of Science citations: Case of Malaysian engineering highly cited papers" Modern
Applied Science, 8, (5), 63.
Antelman, Kristin (2004), "Do open-access articles have a greater research impact?" College & research
libraries, 65, (5), 372-382. DOI: 10.5860/crl.65.5.372
Böck, Barbara (2015), "ResearchGate und andere Archive" Chemie Ingenieur Technik, 87, 1279. DOI:
10.1002/cite.201590074
Borsuk, R., Budden, A., Leimu, R.; Aarssen, L., Lortie, C. (2009), "The influence of author gender,
national language and number of authors on citation rate in ecology" Open Ecology Journal, 2, 25-28.
DOI: 10.2174/1874213000902010025
Braun, Tibor, Glänzel, Wolfgang, Schubert, Andràs (2006), "A Hirsch-type index for journals"
Scientometrics, 69, 169-173. DOI: 10.1007/s11192-006-0147-4
Craig, Iain D., Plume, Andrew M., McVeigh, Marie E., Amin, Mayur (2007), "Do open access
articles have greater citation impact?: a critical review of the literature" Journal of Informetrics, 1,
(3), 239-248. DOI: 10.1016/j.joi.2007.04.001
Crawford, Mark (2011), "Biologists using social-networking sites to boost collaboration" BioScience,
61, (9), 736-736. DOI: 10.1525/bio.2011.61.9.18
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
10
Davis, Philip M., Lewenstein, Bruce V., Simon, Daniel, Booth, James G., Connolly, Mathew J.
(2008), "Open access publishing, article downloads, and citations: randomised controlled trial" BMJ,
337, a568. DOI: 10.1136/bmj.a568
Davis, Philip, Fromerth, Michael (2007), "Does the arXiv lead to higher citations and reduced
publisher downloads for mathematics articles?" Scientometrics, 71, 203-215. DOI: 10.1007/s11192-
007-1661-8
Eysenbach, Gunther (2006), “Citation advantage of open access articles”, PLoS biology, 4, (5), 692.
DOI: 10.1371/journal.pbio.0040157
Falagas, Matthew E., Alexiou, Vangelis G. (2008), "The top-ten in journal impact factor
manipulation" Archivum Immunologiae et Therapiae Experimentalis, 56, 223-226. DOI: 10.1007/s00005-
008-0024-3
Garfield, Eugene (2006a), "Citation indexes for science. A new dimension in documentatio through
association of ideas" International journal of epidemiology, 35, (5), 1123-1127. DOI: 10.1093/ije/dyl189
Garfield, Eugene (2006b), "The history and meaning of the journal impact factor" Journal of the
American Medical Association, 295, 90-93. DOI: 10.1001/jama.295.1.90
Gaule, Patrick, Maystre, Nicolas (2011), "Getting cited: does open access help?" Research Policy, 40,
(10), 1332-1338. DOI: 10.1016/j.respol.2011.05.025
Hajjem, Chawki, Harnad, Stevan, Gingras, Yves (2006), "Ten-year cross-disciplinary comparison of
the growth of open access and how it increases research citation impact" arXiv preprint cs/0606079
Harnad, Stevan, Brody, Tim, Vallieres, Francois, Carr, Les, Hitchcock, Steve, Gingras, Yves,
Oppenheim, Charles, Stamerjohanns, Heinrich, Hilf, Eberhard R. (2004), "The access/impact
problem and the green and gold roads to open access" Serials Review, 30, 310-314. DOI:
10.1016/j.serrev.2004.09.013
Hoffmann, Christian P., Lutz, Christoph, Meckel, Miriam (2016), "A relational altmetric? Network
centrality on ResearchGate as an indicator of scientific impact" Journal of the Association for Information
Science and Technology, 67, 765-775. DOI: 10.1002/asi.23423
Jamali, Hamid R.; Nabavi, Majid (2015), "Open access and sources of full-text articles in Google
Scholar in different subject fields" Scientometrics, 105, 1635-1651. DOI: 10.1007/s11192-015-1642-2
Kamat, Prashant V. (2015), "Open Access Debate: On the Flip Side" The Journal of Physical Chemistry
Letters, 6, (7), 1238-1239. DOI: 10.1021/acs.jpclett.5b00362
Kenneway, Melinda (2011), "Author attitudes towards open access publishing" Survey Report by TBI
Communications and InTech
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
11
Kurtz, Michael J., Eichhorn, Gunther, Accomazzi, Alberto, Grant, Carolyn, Demleitner, Markus,
Henneken, Edvin, Murray, Stephen S. (2005), "The effect of use and access on citations" Information
Processing & Management, 41, (6), 1395-1402. DOI: 10.1016/j.ipm.2005.03.010
Laakso, M.; Welling, P.; Bukvova, H.; Nyman, L.; Björk, B.-C.; Hedlund, T. (2011), "The
development of open access journal publishing from 1993 to 2009" PloS one, 6, (6), e20961. DOI:
10.1371/journal.pone.0020961
Lawrence, Steve (2001), "Free online availability substantially increases a paper's impact" Nature, 411,
(6837), 521-521. DOI: 10.1038/35079151
McCabe, Mark J., Snyder, Cristopher M. (2015), "Does online availability increase citations? Theory
and evidence from a panel of economics and business journals" Review of Economics and Statistics, 97,
(1), 144-165. DOI: 10.1162/REST_a_00437
Min-ChunYu, Yen-Chun JimWu, WadeeAlhalabi, Hao-YunKao, Wen-HsiungWu (2016),
"ResearchGate: An effective altmetric indicator for active researchers?", Computers in Human Behavior,
55, 1001-1006. DOI: 10.1016/j.chb.2015.11.007
Moed, Henk F. (2007), "The effect of “open access” on citation impact: An analysis of ArXiv's
condensed matter section" Journal of the American Society for Information Science and Technology, 58, (13),
2047-2054. DOI: 10.1002/asi.20663
Moed, Henk F. (2009), "New developments in the use of citation analysis in research evaluation"
Archivum Immunologiae et Therapiae Experimentalis, 57, 13-18. DOI: 10.1007/s00005-009-0001-5
Moed, Henk F., Bar-Ilan, Judith, Halevi, Gali (2016), "A new methodology for comparing Google
Scholar and Scopus" arXiv preprint, arXiv:1512.05741v2.
Niyazov, Yuri, Vogel, Carl, Price, Richard, Lund, Ben, Judd, David, Akil, Adnan, Mortonson,
Michael, Schwartzman, Josh, Shron, Max (2016), "Open access meets discoverability: Citations to
articles posted to Academia.edu" PLOS One, 11, e0148257. DOI: 10.1371/journal.pone.0148257
Ovadia, Steven (2014), "ResearchGate and Academia.edu: Academic Social Networks" Behavioral and
Social Sciences Librarian, 33, 165-169. DOI: 10.1080/01639269.2014.934093
Pickard, Thomas K. (2012), "Impact of open access and social media on scientific research" Journal of
Participatory Medicine, 4, e15.
Piwowar, Heather A., Day, Roger S.; Fridsma, Douglas B. (2007), "Sharing detailed research data is
ssociated with increased citation rate" PLoS One, 2, (3), e308. DOI: 10.1371/journal.pone.0000308
Schroter, Sarah, Tite, Leanne, Smith, Richard (2005), "Perceptions of open access publishing:
interviews with journal authors" BMJ, 330, 756. DOI: 10.1136/bmj.38359.695220.82
How accessibility influences citation counts:
the case of citations to the full text articles available from ResearchGate
12
Solomon, David J., Björk, Bo-Christer (2012), "Publication fees in open access publishing: Sources
of funding and factors influencing choice of journal" , 63, 98-107. DOI: 10.1002/asi.21660
Thelwall, Mike; Kousha, Kayvan (2015), "ResearchGate: Disseminating, communicating, and
measuring Scholarship?" Journal of the Association for Information Science and Technology, 66, (5), 876-889.
DOI: 10.1002/asi.23236
Van Noorden, Richard (2014),"Online collaboration: Scientists and the social network" Nature, 512,
126-129.
Wren, Jonathan D. (2005), "Open access and openly accessible: a study of scientific publications
shared via the internet" BMJ, 330, (7500), 1128. DOI: 10.1136/bmj.38422.611736.E0
