Peer review scams – in which reviews were submitted under the names of real researchers but with fake, non-institutional emails administered by those involved – have heightened curiosity about the email addresses used by researchers in their publications. Ronald Rousseau reports on research examining the prevalence of commercial email addresses in scholarly articles, their distribution per country, and whether there is a difference in visibility between articles with institutional and commercial emails, as measured by citations. Use of commercial emails is on the rise, particularly among researchers from Brazil, Russia, India, and China, but those articles with institutional emails were shown to receive more than double the number of citations.
Last year 107 papers were retracted from the Journal of Tumor Biology because they were accepted based on fraudulent reviews. How was this possible? It turned out these reviews were submitted under the names of real researchers but with fake, non-institutional emails administered by the researchers involved in this scam. It seems the preferred technique was to replace Famous.Researcher@pku.edu.cn (hypothetical example of a Peking University email address) with Famous.Researcher@gmail.com (hypothetical example of an email address used by the researchers involved in the scam). Reading about these events, my co-authors and I became interested to learn more about researchers’ email addresses.
In our investigation, non-institutional emails are defined as emails that can be used by almost anyone, such as those with the suffix @gmail.com, @hotmail.com or @yahoo.com. Institutional emails refer to emails that require a confirmed identity, for example, a university staff member or student, such as @harvard.edu, a member of an academic or scientific institution, such as @cabi.org, a government agency such as @ars.usda.gov, or a company such as @microsoft.com. So, requiring a confirmed identity is the main difference between the two types of email addresses we distinguished.
Once this distinction was established we wanted to find out what the most popular commercial emails were; their distribution per country; the percentage of each type in research articles; and whether there was a difference in visibility between each type, as measured by citations. In compiling our dataset, we only considered the email address of the corresponding authors of articles included in the Web of Science (WoS). We found that, between 2008 and 2012, 13% of corresponding authors used a non-institutional email address, with the annual percentage during that period increasing from 10% (2008) to 16% (2012). This increase is likely attributable to the increase of Chinese publications indexed in the WoS (which have about 34% non-institutional email addresses).
Among these, @gmail addresses were most common, followed by different versions of yahoo mails, hotmail, 163.com, and 126.com (the latter two being Chinese commercial email providers). It’s worth noting that @gmail is (officially) not available in China so very few Chinese people have a gmail address (it is only available via a VPN connection and even then may be blocked by the government). Of the 20 most popular institutional email addresses, 19 belonged to universities and one was governmental (ars.usda.gov).
Among the countries we investigated, Sweden and Switzerland had the highest ratio of institutional versus non-institutional email addresses (a ratio higher than 30:1; or about 3% non-institutional ones). Publications from the BRICK countries (Brazil, Russia, India, China, and South Korea) all feature a relatively large (i.e. larger than for Western countries) percentage of non-institutional email addresses, with all at more than 5%. In fact, India actually has more publications with non-institutional email addresses than institutional ones.
When it comes to citations, articles with an institutional email address received more than double the number of citations than those with a non-institutional email address. For China and India these ratios are 1.7 and 1.3 respectively.
This brings us to the discussion of these results. First, one may say that, based on the citation results and the peer review scam, an email address – institutional or non-institutional – could be considered as a proxy for institutional prestige or publication quality, albeit a rather poor one.
Further, we may wonder why colleagues use a commercial email address instead of an institutional one, especially in China and India. Having consulted with colleagues in China and India, almost all said their institutional emails often did not work, with messages either bouncing back or, worse, not even generating a “failed to deliver” message. My colleague Liying Yang from the National Science Library, Chinese Academy of Sciences, suggested the following reasons why authors might opt for a non-institutional email:
- Commercial emails provide more functionality compared to institutional ones
- Commercial emails provide more storage space
- Commercial emails support uploading or accepting larger attachments
- Students and researchers prefer commercial emails as they do not have to change when moving to a new institution
- Institutional email servers apply stricter safety rules when filtering email from outside, filtering out legitimate messages.
Yet it is my personal feeling that colleagues from well-known universities or institutes tend to use their institutional email. Combining this observation with the fact that the use of email tends to reflect research quality, I would recommend “metric-wise” colleagues to use their institutional email in their publications.
This blog post is partly based on the author’s co-written article, “Do papers with an institutional email address receive more citations than those with a non-institutional one?”, published in Scientometrics (DOI: 10.1007/s11192-018-2691-0).
Note: This article gives the views of the author, and not the position of the LSE Impact Blog, nor of the London School of Economics. Please review our comments policy if you have any concerns on posting a comment below.
About the author
Ronald Rousseau is the former president of the International Society for Scientometrics and Informetrics (ISSI). Although formally retired he is still an active researcher at the universities of Leuven and Antwerp in Belgium. Ronald’s ORCID ID is 0000-0002-3252-2538; his institutional email addresses are: email@example.com and firstname.lastname@example.org; but he also has a commercial address: email@example.com which, however, he does not use in formal publications.