Predatory publication has created a headache for the academic community worldwide. I have been asked by a newspaper in Vietnam to comment on the issue, and I have done a bit of research and found out that many journals indexed in Scopus and Pubmed may be ‘predatory’. The presence of these bogus journals masqueraded as legitimate academic journals has seriously compromised the criteria of professorship promotion in developing countries.
In Vietnam, the State Council for Professorship is charged with the responsibility of assessing candidates for professorship promotion nationally. Peer-reviewed publication is one of the key criteria for promotion. However, it has recently become clear that many candidates have published their work in predatory outlets, and this practice has compromised the evaluation process.
The matter is boiled down to differentiating predatory journals from legitimate ones. Last week, in the wake of Professor Nguyen Ngoc Chau’s call for a total review of professorial candidates’ publication records , the Council has convened a meeting to sort out the problem. After a rather quick deliberation, the Council has decided that papers published in journals indexed by Web of Science (WoS), Scopus, Pubmed and Emerging Sources Citation Index (ESCI) are considered legitimate publications .
Well, I am not convinced by the decision. My research suggests that an unconditional acceptance of journals indexed by those citation databases can lead to wrong decision.
Let me elaborate a little bit further.
How many journals in WoS, Scopus, Pubmed?
There are differences in the criteria and procedure for journal selection between Scopus and WoS. To be indexed in Scopus, a journal should meet some minimum criteria such as readable scientific content, peer-review process, regularity of publication, and policy concerning publication ethics.
For WoS, the indexing requirements include 28 criteria relating to scholarly content quality and relevance, article titles and abstracts in English, presence of peer-review policy, timeliness publication, website functionality, editorial board with affiliation details, and information pertaining to authors . Journals that meet the criteria are then indexed in the so-called “Emerging Sources Citation Index” (ESCI). Journals that meet further impact criteria are ‘admitted’ to Science Citation Index Expanded (SCIE), Social Sciences Citation Index (SSCI) or Arts & Humanities Citation Index (AHCI) 
The differences in criteria explain the large difference in the number of journals indexed in the two databases. According to data published by Texas A&M University libraries, in 2019 there were 28,560 journals indexed by WoS; of which, 20,219 (71%) journals were still active . The Scopus database included 37,535 journals, and among which, 23,793 (63%) were still active. As expected, there was a considerable overlap between the two citation databases, the overlap varied according to disciplines. The proportion of coverage overlap ranged between 46% for biomedicine to 62% for natural sciences and engineering .
Predatory journals in Scopus
In any discipline, journals are vastly different in quality and impact. The quality and impact of a journal are often measured by citation metrics. The SCImago group has used the distribution of 3-year citations to group Scopus-indexed journals into 4 quartile based groups: Q1, Q2, Q3, and Q4. According to this grouping, Q1 includes journals in the top 25% of journals in the discipline-specific list; Q2 journals are in the 25 to 50% quartile; Q3 journals are in the 50 to 75% quartile; and Q4 journals are in the bottom 25% quartile.
The problem is that predatory journals have seeped into Scopus database. In an analysis of 944 confirmed predatory journals, the authors found that 56 (or about 6%) of those were indexed in Scopus citation data base . Scopus conducts regular reviews, and after each review a number of journals is excluded from the citation database. An analysis of 299 excluded journals found that 13% of them were in Q1 or Q2 group, 41% were in Q3 group, and 46% were in Q4 group .
Predatory journals in Pubmed
Pubmed is another curated database with a focus on biomedicine. The criteria for indexing in Pubmed include, among others, periodical publication (eg 3 or more times a year), citable content, accessibility for at least 12 months, availability of URL, and compliance with Medline policy on Indexing Electronic Journals.
There are about 7,000 journals in Pubmed, but this number is increasing every year. The problem is that collection has included some predatory journals. Professor Franca Deriu and colleagues identified more than 200 predatory journals across disciplines of neurosciences and rehabilitation in Pubmed . Within the discipline of neurology, more than 10% of the discipline’s journals were predatory .
What about ESCI? Well, the situation is not that better. Actually, the proportion of predatory journals in ESCI was even higher than that in Scopus! Indeed, an analysis by Nguyen Minh Duc and colleagues found that 28 out of 944 predatory journals were indexed in ESCI! It is estimated that the proportion of predatory journals in ESCI was 0.61%, almost two-fold higher than Scopus’ .
Toward a solution
The presence of predatory journals in well known bibliometric databases such as Scopus, Pubmed and ESCI is a cause for concern. Predatory publication can be considered a corruption of publication ethics that can lead to societal harms through multiple mechanisms. These outlets have the potential to propagate false or misleading information that can confuse the public. Unscrupulous operators can use these outlets to promote unorthodox therapies that harm patients and the community at large.
In academia, predatory publication represents a threat to academic integrity and university reputation. There have been instances where postgraduates and PhDs were produced by, and academics were promoted based on, predatory publications. The recent affair in Vietnam  is a case in point.
I argue that the unconditional reliance on Scopus, Pubmed and ESCI as sources of legitimate journals for professorial promotion can lead to blunder.
I recommend that the Vietnam’s State Council for Professorship should consider only official journals associated with learned societies and/or journals published by reputable academic publishers (eg Elsevier, Springer-Nature, Wiley, Sage, Taylor & Francis, Routledge, Oxford, Cambridge, Harvard, MIT, Academic Press, etc).
I strongly recommend that Vietnam universities and research centers develop codes of publication ethics so that their researchers be mindful of the journals they publish and papers they read. The codes should categorically discourage researchers from publishing their work in predatory journals and/or serving as members of their editorial boards. Moreover, Vietnam’s funding agencies should undertake a throughout audit to identify any predatory publications they have funded and have measures to disincentivize any further publication in those outlets.
PS: A longer version of this article (in Vietnamese) has been published in VNexpress.
 Texas A&M University Libraries. Web of Science versus Scopus: Journal Coverage Overlap Analysis.
 Nguyen Minh Duc, et al. Predatory Open Access Journals are Indexed in Reputable Databases: a Revisiting Issue or an Unsolved Problem. Med Arch 2020; 74(4): 318–322.
 Cortegianiaet al. Inflated citations and metrics of journals discontinued from Scopus for publication concerns: the GhoS(t)copus Project. bioRxiv preprint doi: https://doi.org/10.1101/2020.03.26.007435.
 Manca et al. PubMed should raise the bar for journal inclusion. Lancet 2017;390:734–5.
 Fiorini et al. Towards PubMed 2.0. eLife 2017; 6: e28801.
 Somoza-Fernández et al. Presence of alleged predatory journals in bibliographic databases: analysis of Beall’s list. Melcior de Palau, 140. 08014 Barcelona.