I was recently asked to review for an MDPI open access journal. Is this a reputable publishing company? Their website suggests that they are 15 years old, and only do open-access publishing. The journal has a 3-year impact factor of around 2, which is not unreasonable for my field.
The peer review process generally differs between predatory and non predatory journals.
- Given that they have solicited a review from you is a positive sign.
- If you don't know the authors and the work before hand, then that is another positive sign.
- If the article you are reviewing is good, then that would be a third positive sign.
- Finally, if you make comments in your review and the authors address them in the published version, that would be a final positive sign.
If on the other hand you know the authors and work, it is crappy, and the comments are ignored, that would be a bad sign.
Based on the amount of spam that I receive from them (mostly calls for papers in special issues), the fact that their automated emails do not feature a “unsubscribe me” link, and the fact that I did not manage to be removed from their lists after several complaints, I would say that they are not a respectable publisher.
I don't know how long they have been around, but most of their journals in my field (chemistry) were very recently created: see there for a full list, which you can sort by journal creation date. Their older journal, Molecules, has a less-than-stellar impact factor of 2.4. Its editorial board, apart from the occasional celebrity or two, is unremarkable.
As a conclusion: I think they surf on the popularity of open access journals, but I don't consider them serious players.
tl; dr: probably reputable, but they are controversial, and a significant number of academics don't like MDPI.
I used to work in publishing so this answer cuts much closer to the publisher's world than the academic one. It's also going to be very long, so I'm splitting it into sections.
Size: MDPI has grown very large. It's approximately the fifth largest publisher in the world by paper volume, publishing >100k papers a year. It's still growing very quickly and might have overtaken Taylor & Francis at this point. This number is large enough that it's improbable they are all bad papers based only on number. E.g. for comparison OMICS, which is much more commonly regarded as disreputable, is not nearly the size of MDPI. How did MDPI grow so fast? The main reason appears to be ...
Review time: MDPI claims a median time from submission to publication of only 39 days. From the perspective of other publishers this time is absurdly fast, in fact many publishers won't even calculate this number because the time taken for peer review is not something that's under their control. MDPI got this fast by accelerating their production time to nearly zero (more on this later), and by attempting to control the time taken for peer review anyway. From my conversations with an MDPI employee & by looking at the blogosphere of academics, they appear to do this by:
- Requesting reviews very quickly. When reviewing for MDPI you could be asked for a review in 7(!) days. This doesn't mean they are compromising on review quality however: after all, most students will always submit their assignments just before the deadline, and most reviewers will do the same with their reviews. In other words, the request effectively says "are you free to give a quick review? If not, we'll ask someone else." MDPI are able to ask this question because their reviewers are invited by the editorial staff, not editorial board members. Full-time editorial staff can do things like react instantly when a review is submitted, or invite new reviewers the moment one declines. MDPI also have a staggering number of editorial staff per journal - I'm told in the vicinity of ~30 for a journal that publishes ~3000 articles per year - to provide 24-hour coverage. Adding fuel to the fire is that the editorial staff will usually hold advanced degrees in the field of the journal. See also the review process bullet point below.
- Authors are also requested to revise their articles quickly. Their editorial staff tailors the amount of revision time to the difficulty of revisions requested. They can do this because they usually hold advanced degrees in the field of the journal. The time given to revise can be as short as 2 (!) days, although it can also be significantly longer if new experiments are requested.
- Finally, they accelerated their production time to nearly zero - taking only a few days from acceptance to uploading XML files online. This timetable is crazy; for comparison at other publishers I worked at, the target is 25 working days. The only way to get it to go faster is by having more manpower, which MDPI appears to have done - each journal also has hundreds of production staff that literally start working on the paper the instant it arrives on their desk. The MDPI employee who told me the above claimed MDPI has >10,000 employees. This doesn't match their history page, implying that most of these >10,000 employees are freelance and/or part-time, which makes sense given that the workload in journal publishing ebbs and flows.
Their astonishing speed does mean they can claim a legitimate competitive advantage over other publishers. It also means that if you submit/review/edit for MDPI, you can expect very fast response times. It's fairly common on Academia.SE to get questions about how long peer review seem to take at some journals (example). The way MDPI is set up means this will never happen to them.
Summary of this section: I don't see any evidence to think MDPI compromised their peer review process to achieve this speed. They're trying to work around the process in ways that defy industrial standards, but can conceivably work.
Review process: The MDPI employee I talked to said the typical review process goes like this:
- Paper is submitted.
- Paper is shown to an editorial board member specializing in the field of the paper (remember the journal staff have advanced degrees), who makes the decision whether to send it for peer review. If the decision is no, desk reject.
- The journal staff invites reviewers (remember the journal staff have advanced degrees).
- Editorial board member makes the decision based on reviews received.
The argument for doing it this way is that MDPI wants to reduce the workload of their editorial board members as much as possible. To do #2 and #3, they hire at advanced level. Other publishers I worked at tended to leave #2 to the editor-in-chief and #3 to the editorial board member assigned to the paper.
This kind of arrangement definitely cuts the time taken for peer review, but the impact on review quality is hard to quantify. I want to think it isn't worse (because I've done it before and the reviews received didn't seem worse), but some academics will be critical.
This also explains why MDPI's editorial boards often have hundreds of people without an editor-in-chief, and can also explain why people seem to have divergent experiences with MDPI's peer review process (because if my experiences elsewhere are anything to go by, journal staff are more variable than editorial board members).
Summary of this section: I don't see anything organically wrong with how MDPI do their peer review. At least, their intentions are benign. However, some people will object to the editorial board not handling everything (example).
Production process: Same as standard, except they engage advanced degree holders in the relevant field to do copyediting. This is expensive, but doable; many professional copyediting companies offer this. From my conversation with the MDPI employee, I know they are outsourcing their copyediting a lot (apparently there was even one editorial board member who wanted to be involved with copyediting). I don't know if they are engaging these copyediting companies, however.
Article processing charges: MDPI charges about 1000-2000 Swiss Francs per article. Relatively speaking this is somewhat lower than average among big publishers.
They are rather open about how they spend the APC. Based on my knowledge of publication fees, their numbers are believable: there are fixed costs that cannot change, but journals that charge less also get less of a marketing budget, and they also have less leeway to provide discounts/waivers. The numbers indicate that the less expensive journals are making a net loss per paper, and the more expensive journals are subsidizing the less expensive ones - which is also my experience. Production costs are a bit higher than they have to be, but given that they're hiring an army of advanced degree holders, it's not surprising.
For the people who associate "predatory/disreputable" with "high profit margin", the numbers indicate their profit margin is 1-6% per article, which is definitely on the low side.
Special issues: the MDPI employee I talked to appeared to have a different conception of "special issue" than the common one, which is an extra "special issue" of a subscription-based journal with a set number of issues per year. MDPI's special issues simply appear to be collections of papers on that topic with an invited guest editor. The special issues are often very generic (e.g. this special issue with theme "Focus on Dark Matter" could very well be its own journal, and indeed is).
From my perspective the special issues are simply a way to get more academics to collaborate with MDPI.
Publishing metrics: MDPI does well on many of the standard metrics publishers look at. All numbers in this section are from this source.
- Overall rejection rate of 60%, which is not high but also not low.
- Their citability is improving. "In 2016, only 27 of its 169 titles were indexed on SCIE (Science Citation Index of Web of Science) and were on track to get an Impact Factor. By 2019, its leading journals were generally as citable as the average articles in the fields where they compete (more on that below). In summer 2020, 71 of MDPI’s 250 titles had an Impact Factor." (For those unfamiliar with this, to get an impact factor the journal needs to be indexed by the Science Citation Index, which is very hard; clearing this hurdle puts any new journal on very solid footing to be self-sustaining.)
- "All ten titles improved their citability for content published in the same year (Immediacy Index) from 2015-19, and eight of them improved their citability from 2018 to 2019. Six of the titles had a better Immediacy Index in 2019 in their leading research category than articles of other journals, and two of them had a better Immediacy Index than articles of the selective ERA 2018 journals (25,017 journals in the Excellence in Research for Australia 2018 journal list)." (Immediacy Index here is how often the articles in the journal are cited in the same year they are published)
- They don't have a higher retraction rate than the rest of the industry. "[MDPI] reported 19 retractions in 2019, equivalent to 0.5 retractions per 1,000 papers (assuming that retractions refer to year t-2). As a point of contrast, I could locate 352 papers on Elsevier's ScienceDirect that included the phrase 'this article has been retracted' in 2019, implying 0.5 retractions per 1,000 papers (again, assuming that retractions refer to year t-2)."
- They are not reliant on either a single country or a single subject area (see Fig. 6 & 7 of source).
Summary: from other publishers' point of view, as long as one neglects the name, MDPI is a healthy publisher, maybe even one to be envied.
Controversies: MDPI is controversial primarily because Jeffrey Beall blacklisted them back in 2013. Although they were removed in 2015 after appeal, Beall still called them a "borderline case". I wrote more about Beall's list in this answer; the brief summary is that it's not a given that any unbiased observer will agree with Beall's assessment.
One can still look at the controversies themselves on MDPI's Wikipedia page. In general it seems like MDPI have increased their publication standards and have gotten less controversial since 2013. Post 2015, the only really controversial incident one could associate with predatory publishing is the 2018 resignation of several editorial board members. Examining this in more detail the board members argue they were being pressured to accept mediocre papers. However, from the article it's clear that the argument is a well-known one in publishing, and has to do with whether one should demand novelty in articles. I wrote as much in the answer above:
Many OA journals do indeed review lightly. For example I once attended a talk by a Springer spokesperson who talked about a journal which reviews for correctness, not novelty (can't find the journal now, but PLOS ONE has the same policy). Viewed one way this is laudatory - it makes peer review less random by eliminating one completely subjective facet! Viewed another way, this is terrible - it makes it seem as though the journal will publish old results known for hundreds of years as long as the author is willing to pay. Which is closer to the truth? You'll have to come to your own conclusions.
If you believe that this style of reviewing is terrible, as the Nutrients editors apparently did, then you'll conclude MDPI is predatory. If you take the other view that this style of reviewing is laudatory, you'd ignore the entire incident as much ado over nothing. It's not clear-cut; some academics have even published an article defending another journal that outright didn't conduct peer review.
Aside from the above there is also this:
Simen Andreas Ådnøy Ellingsen questioned the quality of MDPI's peer review based on his experiences as a reviewer for the publisher; he wrote that he was only given one week to review a paper, that he recommended rejection, that the paper was then simply published without further comment, and that he never was in contact with any editor.
But this also doesn't look like very problematic, because editors can choose to accept a paper that reviewers recommend reject, the one week review time is part of MDPI's standard operating procedure, and given that the journal staff are inviting peer reviewers, it's not surprising that Simen Andreas Ådnøy Ellingsen was never in contact with a member of the editorial board.
There doesn't seem like much else in the controversies section post-2015, so I conclude by argumentum ex silencio that there is no smoking gun that they're disreputable. At least, they've raised their standards since Beall blacklisted them in 2013.
Email spam: This is perhaps the most annoying thing about MDPI. Submit or review a paper and they send you all sorts of emails about X or Y, some of which there is no realistic chance you'll be interested. Odds are this contributed to MDPI's growth actually, simply because it's how spam works.
<See "update" section below for more about this>
If you believe that only disreputable publishers spam, then MDPI are clearly disreputable. On the other hand if you believe that "disreputable" = "bad peer review", then their spamming doesn't seem related. At least they seem to respect requests to unsub (but see lighthouse keeper's comments below).
Finally: Many people approve of MDPI - their sheer size is an indication. They have hundreds of thousands of authors, and tens of thousands of editors. Equally, many people disapprove. I have no numbers for this, but here are some examples. Therefore however one looks at it, MDPI is controversial. You might want to make up your own mind on them.
Update: a former colleague of mine recently joined MDPI. From conversation with her, there's no question that MDPI are seriously reviewing all papers that are submitted. In fact they might have gone overboard, because they've implemented strict guidelines for editors when inviting reviewers. Some, such as "you must have at least three reviewers per paper", are quite normal. Others are more controversial: "your reviewer must have h-index at least 5", or "your reviewer must be at least an associate professor", or even "your reviewers must be from different countries - and some countries are blacklisted". (MDPI are clearly contributing to the discrepancy for many countries in number of papers submitted and number of papers reviewed.)
My former colleague tells me that it is a hassle to register new reviewers, because the editorial management system they use requires lots of things such as the reviewer's research interests, their institutional webpage, an institutional email address (generic ones like gmail are not permitted) and so on. My colleague further says that this is unrewarding because there're apparently a lot of academics who 1) don't believe in author-pays open access, or 2) believe MDPI is predatory. Furthermore, management keeps a close eye on what their editors are doing: they are expected to write daily (!) reports about their progress. My colleague has more than 10 years of experience in journal publishing, but she says she's still very stressed by the work.
I think this explains why MDPI seems to send spam reviewer invitations - when the editors are stressed and busy, when they spend a couple of hours researching new reviewers who all decline, and they are struggling to meet their daily targets, then the easiest thing to do is to invite previous reviewers regardless of whether the research interests match. However, I'm also told that MDPI have protocols about this: you cannot invite someone who has been invited in the previous month, for example, and the data of who has been invited is stored in a common database. In other words, it's very possible someone else in the company has already invited the person you want.
In conclusion, I think MDPI are aware of the problem and taking steps to fix them. Their methods result in less flexibility for their editors (my colleague cited this as a major drawback of working at MDPI), but I don't see any other realistic alternative. If they want to address the problem, then they have to do something like this. At least with these protocols in place, the next time there is a peer review failure (and those are virtually inevitable) they'll be able to defend their work (recent example when one of their journals accepted a paper that claimed COVID vaccines lacked clear benefit).
In my very personal opinion MDPI is an reputable publisher, but their journals are not first-class. I did a review for them once, and the peer review process was smooth. Also, I knew the field of the reviewed paper very well, so I guess they selected me as a reviewer carefully. But I really can speak only for one of their journals.
On the other hand, I won't publish in their journals since they charge quite a bit for the open access, and there are other options in my field which are free for authors and subscribers. Also I don't like to be spammed with their "newsletters".
Added: I have recently seen this post reporting a problematic peer-review case with an MDPI journal. This lets me doubt, if my initial judgement was right. I suspect, it might depend on the journal.
About a year ago, I was asked to become editor-in-chief for one of their journals. I asked myself the same question. After doing some background research, I came to the conclusion that MDPI was, albeit rather "young", a reputable publisher.
I have worked with them for a little while now and have just started as editor-in-chief on another journal under MDPI with a specific focus on my area of research. My interaction with the editorial manager is very efficient. Surely, we will not compete with Nature and Science for high profile manuscripts, but I am absolutely convinced that both journals will do well with high quality papers and manuscripts.
I recently reviewed a review article about a specific disease for an MDPI journal. I did not know the authors, who were from an institution in China. I was absolutely an appropriate reviewer, as most of my publications in the last 8 years or so are on the same disease.
I won't say that the article was great, but it did carefully cover the topics well and pulled in some of the current challenges. The journal and authors were certainly responsive to review comments and efficient in processing them.
In summary, everything about the process indicated that MDPI is acting like a reputable journal--making a real effort to do good work and provide a venue. What I hope new journals can do is build a good set of publications and solid review process that is perhaps less subject to some of the biases in big-name established journals. Bias, for example that lets a big name in the field get by with an easy review, but is unreasonable or simply rejects new players in a field.
MDPI is an academic publisher with a relatively unique model (among major publishers) that fits a specific need in the marketplace. (Full disclosure - I have mid-authored several papers that ended up in MDPI journals, although I won't submit there)
That need is volume. There are several countries which disproportionately weight volume of publications for promotion and funding. The large majority of MDPI papers are published from these countries. MDPI publishes quickly, with a very small minimum publishable unit, and reasonable apcs.
As you can imagine, the quality is overall poor. However, they do peer review, and at least attempt to avoid publishing complete garbage (there are highly reputable journals which aren't any better at keeping out utter nonsense PNAS I see you over there). The editorial staff isn't strong scientifically but do a good job of moving manuscripts through (they are very author friendly). Many of the journals are indexed, but I don't think I've ever cited anything from a MDPI journal - not because I won't or can't, just because there's nothing worth citing in my field.
Should you publish there? Well, if you're looking for funding from the NIH your biosketch is limited to a handful of papers, which means publishing a bunch of MDPI papers is pretty useless for you. If you included an MDPI paper in your biosketch it probably would hurt your investigator score. Mostly I see people dump dead projects into MDPI journals so students can get some publication. If you're in one of the funding systems where a large volume of papers is good for your career, go nuts. If you have actually good science you think other people will want to read, you can and should send it to better journals where it will be better peer reviewed.
Also the comments on the other answer about getting spammed with rfcs and review or editorial requests once you're in their system is spot on. I ended up just making it a spam filter problem cause I kept getting added to journals (most of which I could contribute nothing to because they're like oil exploration or something that's entirely out of my field).
In my humble opinion, and quite extensive experience in academia, MDPI is not a reputable publisher, in general, and on average, in the sense that:
- having papers in an MDPI journal does not strengthen one's CV (and tenure, and promotion cases, on average), and also
- publications in MDPI do not correlate high enough to papers that constitute important or even reputable scientific progress.
My opinion is based on the following data/information:
I am not aware of any result of even slight importance or relevance to contemporary research in my field of study (STEM, basic research, mathematics) that appeared in an MDPI journal.
I am not aware of any reputable academic in my community serving as an editor there.
I have very bad experience with spam received from MDPI.
Having more than 10% of your journals in ISI is more than an achievement. Most open-access journals don't even care because it is impossible to them. I think it is wrong to take open-access as fake or low quality, traditional journals are also becoming open-access and there are very serious open-access journals like PLOS. MDPI journals fall short compared to PLOS but they are respectable and are building their reputation. Charges are proportional to the journal reputation and most times can be waived if the author makes its case of lack of funding and it is completely independent of the review process.
The recent overview of the impact factors for various journals in MDPI: impact factors , show that there is significant credible scientific scholarship being disseminated. (eg. 'Catalysts' 3.4, 'Energies' 2.6, 'Entropy' 2.3). As of 2017, 'IEEE Access' has an impact factor of 3.5, impact factor on front page and its reputation is accepted and acknowledged.
I'm adding another answer because I recently decided to join MDPI, giving me an insider view. As of time of writing, I've been here three months. I won't edit my previous answer which was written from the outside, but I will refer to it.
tl; dr: not predatory in the traditional definition of the word (i.e., does not conduct peer review and accepts everything). Potentially predatory if you use a different definition.
tl; dr #2: Most of the processes are fine, but the execution can be incompetent.
I'll start by adding comments on statements in my previous answer, which was surprisingly accurate considering it was written from the outside, although some of its statements are wrong.
Review process: This is actually described on MDPI's website, which agrees with what I've seen from the inside. The traditional definition of "predatory" does not apply; we work very hard to adhere to the stated review process. This honestly seems obvious enough that I won't spend more time on it.
There are parts of the "standard operating procedure" that can be mildly discomfiting (example), but nothing I have seen that would raise a red flag. This shouldn't be surprising, because there's a "science committee" made up of advanced degree holders doing repair work each time something goes wrong (and a lot of things have gone wrong in MDPI's past). These people then leave instructions to avoid the error happening again. The result is that there are noticeable scars at MDPI caused by previous errors. More on this later.
Production process: The production process moves very quickly. It's common for me to send an "accept" decision, and then be contacted by the production staff within a few hours. The names of the people who handle each step are listed. I Googled for a person doing copyediting and it seems they are a PhD student, so the statement in my other answer that MDPI outsources copyediting to advanced degree holders is partially correct (since PhD students are not yet advanced degree holders). However, I also met a MDPI English editor who says there are about 100 English editors worldwide, most of which are fresh graduates in non-technical fields, so the other answer is also partially wrong. Some papers are outsourced, some are not.
The English editors work at a rate of 40k words per day, which is shockingly fast, and implies they can't be doing much editing. However, it's not really feasible to do non-trivial edits without specialist knowledge anyway (see example in another answer of mine) so I can't be too critical.
Advanced degree holders: My previous answer said that MDPI has 1) lots of advanced degree holders managing their journals, and 2) 24-hour coverage. Both these statements are incorrect. For the journals I handle, the great bulk of the editorial staff are in the East Asian timezone. We have night shifts, but not 24-hour coverage. Most of the other staff also use Mr. or Mrs./Ms. as their salutations, implying they do not have PhDs, and interacting with them it's clear that many (most?) of them don't have significant expertise in the field of the journal - one person told me they studied postgraduate English, while others are not very confident answering questions like "is this paper a review or a research article?" even though it's apparent from reading the introduction. However, several of my colleagues are passionate, conscientious, and do surprisingly well even without subject knowledge. Recently they've started asking me if so-and-so paper is worth sending out for review, and every time they do so, 1) I can see why they are concerned the paper might be a crank paper and 2) they research the authors' background, affiliation, list of publications. It's impressive.
Also, although it's not the case that every journal staff has an advanced degree, it is true that there are many advanced degree holders in the company handling papers. Most of these are PhD holders, with some MSc holders as well. I just don't know what proportion. It seems like in my local office the proportion is more than 50%, but most of us are handling different journals.
As for why everyone is in the East Asian timezone: I'm not in HR, but I was told that Americans are expensive, don't work very hard, and repeatedly say X cannot be done when it can. Journals assigned to them would consistently struggle. Apparently MDPI used to have an office in the US, but closed it as a result (there is an office in Canada, however). South Americans are allegedly even worse. East Asians on the other hand are willing to work on weekends, providing 7-day coverage for the journal (although during the weekends coverage is still weaker since fewer people are working). I dislike the stereotyping, but I can appreciate the difficulty with getting good employees. More on this later.
(East Asian does not mean Chinese - MDPI has offices in Japan, Thailand, and Singapore.)
Now to the new stuff: one way MDPI could be predatory is that MDPI compromised its peer review process somewhere. However, I don't see evidence for that.
Speed: MDPI is very fast. This is evident from how they measure the MPT (median publication time). With the exception of IEEE Access, MDPI is comfortably faster than all other major publishers (Fig 3 of link). MPT figures heavily in the leadership's thinking, with some justification, since it's apparently a major reason (Fig 4 of same link) why some researchers choose to publish with MDPI.
MDPI is this fast because reviews are requested quickly, because the many full-time editors means the journal office monitors every submission very closely, and because of certain time-saving techniques. I'll address the requested review times first. As mentioned in my other answer, the standard is to ask for reviews in 10 days. Here my other answer is right in that MDPI editors generally think this is OK and reviewers should decline to review if they cannot make the deadline. It's a view I've come around to, because:
- Research indicates that setting shorter review deadlines does not decrease review quality. This meshes with the reviews I've seen personally at MDPI.
- Research also indicates the time taken to actually conduct a review is a few hours to a day. It might take 10+ days to find the time to review, but the review itself does not take 10+ days.
- If we get enough reviews, it's common to ask other agreed reviewers if they want to cancel their reviews. Most of the time, these other reviewers will not have started their reviews (even if it's due tomorrow) and they are OK with cancelling. This meshes with what I've seen at other publishers. Most students will submit their assignments just before the deadline, and most reviewers will also submit their reviews just before the deadline.
- Some reviewers will feel (with some justification) they are being abused, and MDPI expects them to drop everything to review at once for them. If so, they should decline. There is no obligation or expectation to review. If one is not available to do a quick review, decline.
What about equally fast deadlines for authors to submit revisions? This is actually a mirage, since if the authors ask for more time we routinely grant it. It's just a matter of asking for it. (To some extent this also applies for reviewers, although if we get enough reviews we will cancel the late review and move on with the paper.)
As for time-saving processes, there are unsettling things that MDPI will do in the quest to reduce MPT. Examples:
- We will make decisions without waiting the go-ahead from the editorial board, with the intention to backtrack if the editorial board takes an unexpected position. For example, after two revise reviews are submitted, it's possible we will send a revise decision while simultaneously asking the editorial board member for a decision. I've yet to see what happens if the editorial board member says "reject". I imagine we retract the decision and send apologies to the authors.
- In the same vein, we will invite reviewers before the editorial board member does the pre-check, with the assumption that editorial board member will approve the paper for review.
- We also send reviews to the authors/EBMs while there are still reviews promised and ask them for a revision/decision. I had one author reply with the very natural question: "what am I supposed to do if the third review contradicts one of the first two reviews?". From what I've seen, the hope is to either cancel the missing reviews, or to send the review to the authors when it arrives (+ change the revision due date if the authors ask for it).
The time saving from these techniques can be substantial. I just saved four days on one of my papers this way.
Although these things can be unsettling, none of these things actually compromise the quality of peer review. We are very good at predicting Editorial Board decisions. As it turns out, Editorial Board decisions are very predictable if you read the reviews, which we definitely do. Hell, when the reviewer says cite X paper, we check that paper to see if X paper is written by the reviewer (we send requests for more detailed reviews if the answer is "yes").
Accuracy: I use this term to describe reaching the appropriate reviewers for the paper.
Here MDPI leaves a lot to be desired. One example I encountered while trawling the SuSy database was a physicist being invited to review a Cancers paper. That probably happened because the physicist had Bremsstrahlung listed as a keyword in their profile (Bremsstrahlung is a physical process where a charged particle emits electromagnetic radiation such as X-rays). Although this explains why the invitation was sent, it doesn't excuse it. The person who sent the invitation ought to have realized the reviewer was inappropriate, but they sent the invitation anyway. It's troubling.
The scale of the problem seems pretty severe. While searching through the profiles in SuSy it's not uncommon to encounter reviewers who received 5+ out-of-expertise papers in a row, not to mention distressing messages from academics about the never-ending spam they're getting. I've also heard colleagues say their 25% "out of expertise" rate is OK, when it is many times worse than the rate [other editors get].
One thing I will point out is that it's not really fair to criticize the entire journal. An analogy is with banks. Here the bank (publisher) might have several branches (journals), each of which might have several tellers (journal staff). Getting a bad teller does not imply everyone other teller at the branch is bad, and a bad branch does not imply every other branch is bad. For big journals, there will be multiple editors assigned and any of them can send the bad invitation (we do not discuss with each other before sending the invitation). The invitation email does name the person who's sending it. If one is going to criticize, that is the person to direct it at, not the journal.
Another thing I will point out is that no matter how low you get the "out of expertise" rate, as long as it is not zero, you will occasionally reach someone who is inappropriate. That someone can easily get a worse impression of the problem than is actually there. The problem might be bad at MDPI, but it's not the case that MDPI spams random people hoping one of them will accept.* In fact my guess is that most people have some idea what caused the bad reviewer invitation (e.g. being the fourth author on a paper, and the first author would indeed have been an appropriate reviewer).
This suggests that the "solution" is "hire better people". More on this later.
Update: recently there was an initiative to use an AI (as far as I can tell, based on SPECTER) to recommend reviewers. The AI appears to be fairly decent, although sometimes the recommended reviewers are only tangentially working on the paper's topic. Still, that's better than hitting completely unrelated reviewers. I don't know how widely-adopted the tool will be (I'm certainly using it liberally especially for papers I don't really understand), but if it is widely adopted, it should greatly reduce/eliminate this problem.
*Most of the time. If one is sitting on 30 reviewers where most have declined saying they don't have time, then one might start getting desperate and spamming tangentially-related reviewers.
Spam: this is closely related to the above, since some people feel that review invitations that miss the mark qualify as spam. Here I'll deal with other kinds of invitations. Without question there's some level of spam going around, as well as some amount of anti-spam.
MDPI has strict policies on how often one can invite someone to review an article. SuSy will automatically stop reviewer invitations if the reviewer was invited within the past ~10? days (I don't know the exact time, but it's more than a week & less than two weeks). It is possible to override the block, but it takes advanced permissions (i.e., only a senior editor can do it). The official guidelines also say that if a reviewer declines because of lack of time, one should wait longer than two weeks before inviting them again.
The reviewer's activity is logged in SuSy (when they were invited, for which paper, did they actually review, how good was the review, etc.) and all that information is available to the next editor. Distressingly, some editors seem to use this information as a sign to send more invites. Like, if a reviewer reliably responds, then they would absolutely send that person more invitations as long as they have some reason to believe the reviewer's field is related. The upshot is that if you build a track record of writing good reviews for MDPI, people will notice and send you more reviews. Doing this is actually good for the editor on a personal level, since responsive reviewers drive down the MPT, which is a key performance indicator for editors. This is why I've heard conversations alluding to the "competition" between editors for good reviewers. I've even heard of editors who use an alarm clock to warn them (right down to the minute) when the cooldown is up, so they can invite the reviewer before anyone else can. If you reliably get one review request every two weeks from MDPI, you are such a reviewer.
I've yet to encounter editor invitations firsthand (as opposed to reviewer invitations), but I'm told there are two basic styles: either you choose a narrow topic and then find an editor, or you choose a broad topic and ask an editor to narrow it. The latter style can definitely feel like spam. However, I'm also told they are similar to reviewer invitations - there is a "cooldown" after the previous invitation when you cannot send new invitations to that person.
For Call for Papers, MDPI editors definitely try not to spam. I've attended training sessions where the trainer speaks about how to identify appropriate people. The methods (such as searching Web of Science with keywords, checking what papers are citing an on-topic paper, etc.) are fine, but as with reviewer invitations, are heavily dependent on the skill of the person applying the methods. Since many MDPI editors struggle with identifying appropriate reviewers, many MDPI editors also struggle with identifying appropriate people to invite.
Finally, I can say with some confidence that MDPI will stop spamming you if you request it. You need to say clearly you do not want to receive any more such emails, and then MDPI should block you. An example is this image. The person who wrote it is blocked not just by the journal in question, but publisher-wide. Only editors with advanced permissions can override the block. The caveat is that it still takes the person reading the request to implement the block. If they forget, are in a bad mood, don't know how to do it and doesn't ask, etc., then they might not do it.
Quality: I use this word to mean "is the reviewer actually qualified to review the paper?". Here MDPI has some hard guidelines and some soft ones. Hard guidelines include "the reviewer must have a PhD". Soft guidelines I can't reveal, but suffice to say they're controversial.
These guidelines are the only real thing I dislike about MDPI. I disagree with the "reviewer must have PhD" guideline because I think senior PhD students ought to be able to review papers (I'm not the only one who thinks so). As for the soft guidelines, I actively hate several of them. It's especially silly because MDPI knows full-well the guidelines are controversial (that's why I'm barred from naming them), and yet they are there. I don't get it. If we're going to have a policy on who can review, we should have one that can be defended in public.
I'm told the reason why these guidelines exist is that there is a correlation between the data we look at and how comprehensive the submitted reviews are. Which doesn't change how it's gross discrimination. Again, I can't name what the soft guidelines are, but they make me want to tear my hair out.
Another possible interpretation is that MDPI is traumatized by the amount of controversies that have happened in the past, and are relying on these guidelines to avoid them in the future. Some signs of this can definitely be seen in the explanation given whenever the guidelines are changed. I have to admit they are helpful whenever something goes wrong. An example is in 2021, when the MDPI journal Vaccines published a paper claiming a lack of clear benefit in Covid vaccines. The media got involved, and there was an investigation. When this happens, it is really helpful to be able to say that one reviewer for the article is a director of a research institute in Germany, as opposed to a PhD student at an Iranian university nobody has heard of. It might not be fair, but subconscious bias isn't fair.
A third possible interpretation (which I think has some degree of truth) is that MDPI management is aware their employees have invited unqualified reviewers in the past, and are taking action to stop that from happening again. Because many MDPI editors aren't very good, the guidelines need to be clear and unambiguous. As unfair as these guidelines are, they are clear and unambiguous, and if adhered to, they should exclude the worst mismatches.
Finally, one thing I like about the quality aspect is that we will check the review's contents. If a review is poor (for example, if it says nothing except for some English fixes or requests for references), then we will reject it and look for another review. This is quite costly, because inviting new reviewers increases the MPT, but we will do it.
Controversies: MDPI hasn't stopped attracting controversy. This is the most recent controversy, which was widely discussed internally. Pretty much everyone I know independently decided to disregard it, albeit for different reasons. The one I subscribe to is that none of the three "problems" they cite are actually problems:
- This is 1) plagiarized from Wikipedia, 2) quotes ChatGPT as evidence (???), 3) appears to equate success with being predatory, 4) appears to equate fast with being predatory (I wonder if the author realizes it's painfully easy to slow down - just sit on the paper for six months before working on it. If they genuinely think that's a good idea I dare them to print that on their blog.), and 5) alleges that MDPI articles cite themselves more, which is dealt with in the next bullet point.
- This is the source for the claim that MDPI articles cite other MDPI articles more. The claim comes from an article with some serious issues (some of them are mentioned in the comments here), and the article is also under an expression of concern from the journal it's published in, which the blog authors are either unaware of or willfully ignoring.
- This is the only source that is potentially problematic. However, there are several reasons to not be overly concerned. First, editors can accept a paper even if reviewers recommend reject. Second, MDPI publishes several hundred thousand articles a year, implying there are more than a million reviews. With this number of reviews, we should expect peer review failures, so a single incident is not enough to be concerning. I would be interested in statistics if whether MDPI experiences more peer review failures than other publishers, but it's not in the blog, and the only other source I'm aware of (this, linked in my previous answer) indicates that MDPI doesn't have a higher retraction rate than its peers. Third, I have handled several articles where some reviewers recommended minor revisions and we rejected the article anyway, including those which are charging full APCs (i.e. APC is not waived), and I didn't experience any kind of pressure to accept those papers. Finally, I handled an article recently that received 2 minor revision and 2 reject reviews. The guest editor wanted to accept regardless (it led to this question), and I saw firsthand that MDPI's editorial management system stops me from accepting the article until the second editorial board member has given a decision. The setup is such that there can be no accident. One must get a second editorial board member to approve the decision before a paper with outstanding reject reviews can be accepted.
- One thing that can be said is that after a paper is accepted, the editorial board member or guest editor who made the decision is usually known. You can find them in the citation on the bottom left of the first page if you're viewing the PDF file, or on the left if you're viewing the webpage:
For this paper, Lorenzo Iorio is the person who made the accept decision. If Sergey Gromov (the person who alleged the peer review failure in the bullet point above) talks to the listed Academic Editor and finds that the Academic Editor did not actually make an accept decision, he has a significantly more compelling case. However, I think it's more likely he'll get an explanation of why the paper was accepted anyway, invalidating his argument. (It's even possible the academic editor pre-empts the question by getting the journal staff to email the reviewer with their reasons to accept - happened to my paper.)
Another recent controversy was two of MDPI's biggest journals getting delisted by Clarivate. If you read the article, the reason was apparently the journal publishing content that is not related to its aims & scope. My reaction to this one was amusement, because it's a stupid way to get delisted and indicates an issue with MDPI's internal key performance indicators. However, publishing out-of-scope content does not indicate it wasn't peer reviewed properly. There is no evidence in the article for that. The news article quotes a lot of people who had doubts, but no evidence. If they actually come up with evidence that 1) there are a lot of papers published without proper peer review and 2) this happens at MDPI at a rate that is significantly higher than at other publishers (including non-OA ones), I'll reconsider.
(Incidentally MDPI is in the process of a major internal investigation in response to Clarivate's delisting. As I wrote above, there are a lot of advanced degree holders doing repair work each time something goes wrong.)
Finally one thing I want to address is Prof. Santa Claus's claim that MDPI plagiarized his colleague's call for papers. Based on what I've seen, this is a believable claim, and it implies gross incompetence by the person who did it (more examples I'm aware of below). However, it's still not sufficient for me to be concerned. There are in the vicinity of 100,000 MDPI special issues. A single plagiarized call for papers is not on its own a sign of deep problems, any more than a single student caught cheating in an exam does not indicate there is mass cheating going on (or the entire cohort are cheaters, which is apparently how some people view MDPI).
Other: Sometimes MDPI editors will do painfully bad things. I saw one email chain where a reviewer answered the "did you detect plagiarism?" question in the review form with "yes", but didn't elaborate. The editor handling the paper wrote to the author with "the reviewer detected plagiarism, now you have to say where you plagiarized". Really?! Then there was another case where the guidelines on MDPI's website says "[research articles have] a suggested minimum word count of 4000", and the paper had less than 2000 words. The editor wanted to tell a reviewer that the minimum word count is 4000, and enlist their help to find things for the author to add. Finally, I witnessed one person arguing that X is not a suitable editorial board member to handle Y paper, because Y is written by Russian authors and X has a Russian name (even though he's currently based in the UK). I don't want to know how X would react if he ever learns of that reason.
I don't know what to make of these incidents. They are of the gross incompetence kind that indicates the editors have no clue about academic norms. On the other hand, I have also had a frank discussion with an editor about a guest editor's decision. The paper in question had one reject review and one minor revision review, and the guest editor chose minor revision with some rather controversial reasoning based on the identity of the author. My colleague was troubled, but agreed with me that if we didn't trust the guest editor's judgment, we ought not to have invited them to guest edit in the first place. I take that to mean my colleague does care about doing the right thing.
I don't see any evidence MDPI has compromised its peer review process. Some guidelines are annoying, some are unsettling, but (aside from the soft guidelines for who is an appropriate reviewer) they're all defensible. However, these processes place a lot of stress on individual editors, and we simply do not have enough skilled editors. This is exacerbated by the fact that there is a fairly high turnover rate (presumably because of the stress - many of my colleagues have been in publishing for less than 2 years). My colleagues are well-intentioned people, but sometimes they just lack the necessary expertise. This is, I think, the root cause of most of the "disreputableness" of MDPI.
This suggests the solution is "hire better editors", which would be great but for one thing: it's not easy to find people with the skill set. Management tries (c.f. the advertisement linked in my previous answer), but it's one thing to advertise for PhD holders and another to actually get them. There are very few people with PhDs and editorial experience considering a job in publishing.
This means that if you hold an advanced degree, are interested in a job in publishing, and are based in (or are willing to relocate to) a city in which MDPI has an office, odds are MDPI will be happy to hire you. Editorial experience is an advantage but not required. If anyone reading this is interested, you can let me know and I will refer you.
I think that there is no definitive answer to this question.
I have recently published in a MDPI Journal (Materials) and my personal experiences with the journal were okay: The review times were short and the reviewers' comments were in no case different from other journals I previously published in. Communication with the editorial office went smoothly and without further complications. The editors of the special issue I published in are were well known in my field and reputable.
Unfortunately, since my paper was published, I am bombarded with spam mails. I don't want to speculate if this has to do with MDPI or could be a general problem of open-access.
What I have heard from others and read on the internet, however, differs greatly from my personal experience. From all of this, I personally got the feeling, that it is more about the individual journals and less about MDPI itself.
A lot of papers in my field are published in MDPI journals (higher ranking ones like Sensors and Materials). Most editors of those journals are respected and renowned experts.
I have published there myself, and as other have stated, the process was smooth and in many aspects very much like with other, more "reputable" journals. There were several rounds of peer review with competent and constructive feedback from the reviewers. The whole process is just noticeably faster than with most other journals I have submitted to, which - considering that the review process is not of bad quality - is in my eyes a good thing.
And yes, you do receive quite some (unsolicited) email from them, but in contrast to emails I receive from what I would definitely call disreputable journals,
- the emails are always correctly adressed to me, my full name and correct title
- they are always from journals within the MDPI range that fit my expertise
- they often include an offer to publish for less than the normal open access fees and at times even an offer to completely waive the open access fee for a speciall issue or similar
So, from my experience, and the experience of others in my field, MDPI is not a disreputable or predatory publisher, although some of their actions might rub some people the wrong way.