No slPressReleases please, a scientist begs you.

September 6, 2017 11:03 AM

Can we not link to university (or corporate) press releases for scientific discoveries. If we must do so, can we especially avoid single links?

Press releases from the discovering institution are almost invariably AWFUL. They are biased breathless accounts. They are often garbage scientifically, ignoring nuance, ignoring other advances or the history of the field, overstating their claims, and the importance of their claims.

They are often NOT written by scientists - or even scientifically literate people. They are barely better than clickbait (which is what the press office is often judged on!)

They are the beginning of the dumb science news cycle.

Their many problems have been described here, here, here and here, to name the first four places I could find.

University Press releases -especially slUniversityPress releases! - are NOT the best of the web. Without context written by someone more neutral and informed, they are actively harmful.

posted by lalochezia to Etiquette/Policy at 11:03 AM (28 comments total)

Seconding this. I'm a publicist who works in the field.

Press releases from discovering institutions routinely misrepresent research findings -- primarily because they tend to be written by communications majors and researchers are not always accurate assessors of their own findings. There's nothing wrong with being a communications professional, but if they don't have a technical writer/science/medical/pharma background there may be an emphasis on hyperbole over accuracy in reporting.

Press releases are not unbiased news sources. They're supposed to be the foundation upon which reporting is built. A journalist's job is to take their content, vet it and turn it into an unbiased story.

PLEASE link to original source material when possible. As well as coverage of the discovery from a reputable media outlet or news journal which analyzes both submissions and industry news and reports on them, (like Nature.) If your only source is a press release, please pay attention to whether your post is offering unbiased, accurate information or promotional content.
posted by zarq at 11:20 AM on September 6, 2017 [12 favorites]

While we're at it, maybe don't use links that open Print dialog boxes?
posted by Etrigan at 11:30 AM on September 6, 2017 [5 favorites]

I disagree with this callout completely.

The press release in question is out of Stanford, a highly reputable institution, and refers to a study which is being published in a very prestigious and peer-reviewed journal:

...Stanford researchers report August 22 in the Proceedings of the National Academy of Sciences ....

posted by jamjam at 12:06 PM on September 6, 2017 [2 favorites]

If they're posting a press release for an article in a journal, why not also include the link to the article (even if it's paywalled (should be noted))? I see this a lot in general when academic research is discussed - they cite a report or paper but often don't link back to the actual research or bury it.

I think this callout is fine and something people should consider when making FPPs.
posted by kendrak at 12:25 PM on September 6, 2017 [15 favorites]

> The press release in question is out of Stanford, a highly reputable institution

The purpose of a press release is to say "look how great we are!" And that may be true! But it is not an unbiased discussion about the subject of the press release. A press release is not journalism. It shouldn't be the sole basis for an FPP, especially a science-related one. If all that exists about the subject outside a paywall is the press release, I don't think it makes for a good post.
posted by rtha at 12:49 PM on September 6, 2017 [30 favorites]

Okay but how much of this is, some people (working scientists, publicists) have the mental model of wanting to preserve the traditional filter/way of disseminating information: source -› journalists –› public, versus in a post-postmodern 2.0 crowdsourced world, regular people can actually directly access and interface with information including that which comes from institutions?

This is also a matter of empowerment and separating the responsibilities / ethical obligations of the middle-person, and seeing how this is also about structure and roles which change as technology and society changes. Don't assume people don't know or can't tell the difference between ads / PR / marketing versus substantial science, and casual vs serious discussion, etc.

I think a great reason to not have university press releases is because they are newsfilter and that's by definition generally not what's best of the internet. The fact that they are not vetted or are neither primary nor tertiary but rather a kind of specialized, industry-specific, see-how-the-sausage-is-being-made information matters less. Like, do you really want to be making the argument that because the reader is not a professional scientist, they should not be exposed to this material? I'd rather think that distinguishing information types/tenors and learning how to distinguish sources is good, because learning is good. But that's a process of understanding and a conversation, and having a rule "don't do a thing" isn't the best way to teach.

But I'll write my thoughts into a corner and also say that I basically agree with a goal of less reliance on press releases. So a workable alternative is, if someone sees neat info that they would like to share, that's still workable—just give it a couple more days or even weeks when a) famous scientist bloggers are writing cool things about it, b) various science news and long-form journalists can write up good stories. It does mean having to tamper the initial excitement, which really is what science advocacy is all about in a sense, but the benefit is discussion quality can be higher with more information (= the no single links idea) to work with.
posted by polymodus at 1:08 PM on September 6, 2017 [1 favorite]

I have worked in communications adjacent roles for several "highly reputable institutions," and I'm inclined to agree with this callout with a few caveats. University communications offices generally work very hard to accurately represent work coming out of their respective schools, but the people writing these releases are typically not the researchers involved. They are also generally working under a tight deadline with limited information and lack the background or resources to present the data with a lot of nuance. For better or for worse, press releases are generally intended as Cliff's Notes for the layman. This isn't to say they are always or even usually wrong, but it would be a much better practice to also link to the actual paper in question whenever possible and give any interested parties the opportunity to do a deeper dive.
posted by Diagonalize at 1:27 PM on September 6, 2017 [1 favorite]

Why don't these kinds of press releases cite the underlying paper in a way that (modulo paywalls) those who want to dive deeper can?
posted by the antecedent of that pronoun at 1:30 PM on September 6, 2017 [2 favorites]

My suspicion is because the conventions for how to cite the underlying paper can be dramatically different from field to field. In some fields, it might be trivially easy to link to the paper. In physics you could just pop in a link to the ArXiv and that's all you need, but other fields might only have their most recent findings in a print journal or book published by a press outside of the university's control. And in other cases, there might not even be a paper to link to yet because the initial findings are so fresh. It's not ideal, obviously, but every office handles it differently.
posted by Diagonalize at 1:40 PM on September 6, 2017

It doesn't really matter which field it is, the abstract for an academic paper is often more informative and less misleading than a university press release. Press releases are designed to get people who don't know anything about the field excited about the research. Waiting for an article from someone who is or has talked with a researcher in the field who isn't an author is probably the best thing, but just providing the abstract is not too bad.
posted by demiurge at 2:14 PM on September 6, 2017 [6 favorites]

I don't disagree with this callout, but I will say that at the research institutions I've worked at, the press offices take GREAT care to ensure that the findings are appropriately represented. My former institution just put one out after going back and forth with me (maybe 5+ times) to ensure that everything was accurate and appropriately tempered. They also asked me for a copy of the paper so they could send it out to anyone who requests it, and linked to the abstract in the release. It seems like there's no excuse for not doing that, although I understand that not all institutions have these resources, and not all researchers will be responsive to these kinds of clarification requests (or are appropriately humble about the implications of their findings). I agree with the recommendation in the BMJ article linked through one of the articles above that institutions that are the worst offenders on this should be publicly shamed and those who put out high-quality releases should be recognized.

In case folks aren't aware, there are many ways to find free PDFs of research papers. Any NIH-funded research should be available for free on PubMed, although there may be an embargo. Unpaywall is a browser plug-in that finds free PDFs. Sci-Hub is a legally murkier option. If none of those work for you, you can always email the first author (or a friendly scientist with university access to research articles). Publicly posting the paper can get researchers into hot water with publishers, but we are allowed to send them to people on request, and most researchers I know are totally thrilled to increase readership of their work in this way.
posted by quiet coyote at 3:25 PM on September 6, 2017 [11 favorites]

I wonder if folks would accept the University of West Florida's press kit about the Tristán de Luna colony as an example of a university press release done fairly well: contextualized with a separate first-hand account, historical timeline, photos, etc. Personally, I think it would have held up as a single-link post, although that isn't what I did with it: "April, 1561: Florida man leaves 'priceless' artifacts in empty lot." My feeling is that there's reasonable variation both in how FPP creators feel about supporting links and in the quality of particular university media relations efforts.
posted by Wobbuffet at 3:30 PM on September 6, 2017

Adding to quiet coyote's options - A legal alternative to academic publishing paywalls - a few more options/systems below the fold and in the comments.
posted by Wordshore at 3:42 PM on September 6, 2017 [1 favorite]

Totally agree. There's a temptation to link to a press release if the original publication is paywalled or overly technical, but it's an unacceptable substitute.
posted by All Out of Lulz at 4:32 PM on September 6, 2017 [1 favorite]

Worth noting: The intended audience of a press release is not the public but rather media professionals.
posted by zarq at 4:42 PM on September 6, 2017 [4 favorites]

I've posted a few university press releases, some better, some worse. They often have interviews with the authors that give context that's not there in the abstract or article, which is nice. They're often about something which no journalist is ever going to investigate in depth; if a news story ever comes, it'll be the press release with - if you're lucky - a few words rearranged.

They're self-promotional, yes, but so are a lot of the interesting links that get posted. Cute puppies hit the front page... don't really need an academic paper to go with it, nor a hard-hitting news story to present All Sides. Scientist finds a bunch of weird DNA in human blood... same thing. It's interesting; nobody knows what the implications are; we make jokes about it; it's all good.

Yes, somebody somewhere is trying to figure out how to turn this research into a bullshit product that they can market on Goop and Infowars, and that's Bad for Science, but they'll be doing it whether or not we have a fluff-level conversation about it on Metafilter. Nothing wrong with the occasional bit of scientific fluff.
posted by clawsoon at 6:54 PM on September 6, 2017 [2 favorites]

One university press release was so unusually good I had to double-check that I was reading a university press release. Now that's how it's done.
posted by a snickering nuthatch at 9:24 PM on September 6, 2017 [1 favorite]

Honestly, a press release is someone put out into the world what they want to have said about a thing that is happening or has happened. It isn't news, it's a press release. We don't call news articles press releases. (Although regrettably the difference is becoming more difficult to discern.)
posted by hippybear at 10:09 PM on September 6, 2017

Press releases are designed to get people who don't know anything about the field excited about the research.

Are blue posts....not....supposed to do that? I mean, I suppose the ideal science discovery post would be a brief precis that linked to both a nuanced, contextualised, objective report for laypeople along with links to the actual paper/works themselves. But a lot of posts are less than ideal. If the paper's paywalled and the press hasn't covered it, you might be stuck with the flack spew. But the purpose of a blue post is to spark discussion, no? And the flack spew's fine for that.
posted by Diablevert at 6:55 AM on September 7, 2017 [5 favorites]

And the flack spew's fine for that.

It's not, though.

We live in an age where research findings and accepted scientific theories are incessantly questioned by skeptics, who often have personal and.or political agendas for doing so. The doubt those questions cast have lead (in the worst case scenarios with real-life repercussions,) to things like climate change denial and the introduction in multiple school districts across the US of religion-based, so-called "Intelligent Design" being taught alongside evolution.

In such an environment, accuracy in science reporting matters a great deal. Each false report about the nature of research findings, theories or discoveries chips away at their credibility until the average layperson feels justified in saying that a storm which drops 3 feet of snow on New Jersey disproves evidence of anthopogenic climate change.

As someone who has written and worked with medical and science journalists on many complex stories, I feel comfortable saying that it really is impossible to definitively determine what will or won't be covered in depth, now or in the future. Studies that didn't make much of an impact when their related journal articles were published have a tendency to form the foundation of later research hypotheses and results.

There's no downside to greater FPP accuracy.
posted by zarq at 8:48 AM on September 7, 2017 [3 favorites]

My feeling is that the blue is not Mt. Sinai, a post is not a stone tablet. The purpose of a post is to generate chit-chat, not to stand in eternal lines to time, and we are not the internet bullshittters of record. If a post is thinly sourced, people usually complain about it in the comments. Complaining about it in the comments seems a robust enough method of fact checking for chit chat.

The downside of waiting around until when/if one can source a post perfectly is that the post doesn't get made, and the discussion never happens, and the site is more boring. I think we have a far greater obligation as mefites to make the site interesting than to act as guardians of the abstract public interest; metafilter is a niche site. We ain't Walter Cronkite. No one is taking a blue post as the final word from on high about anything.
posted by Diablevert at 9:08 AM on September 7, 2017 [5 favorites]

No one is asking for perfection. What we are asking for is a Filter - with Standards.
posted by lalochezia at 10:46 AM on September 7, 2017 [2 favorites]

> It's not, though.

It is, though. You're reacting as you are because you feel a personal stake in it. I understand; I grind my teeth whenever I see some idiotic piece on language linked. But we both have to remember that MeFi is neither Nature nor Linguistics; it is a general-interest site where all sorts of crap gets posted, and if people have problems with it they can shoot it down in the comments. Nobody is getting their full education exclusively from posts on the blue.

> The downside of waiting around until when/if one can source a post perfectly is that the post doesn't get made, and the discussion never happens, and the site is more boring. I think we have a far greater obligation as mefites to make the site interesting than to act as guardians of the abstract public interest; metafilter is a niche site. We ain't Walter Cronkite. No one is taking a blue post as the final word from on high about anything.

Exactly.
posted by languagehat at 10:50 AM on September 7, 2017 [4 favorites]

Thank you. These are all good things to think about as a non-scientist who is interested in posting about science from time to time.

My comfort zone around posting FPPs about scientific researchconsists of: 1.) at least linking to the abstract of the journal article if a non-paywalled version is unavailable, and 2.) Finding longer form interviews with one of the authors (if available) where they explain their work in more details and get into the caveats/limitations of their findings - so definitely going beyond a SL post.

If nothing else, a paywalled journal link typically provides the abstract - which helps to clarify what the findings actually are.

The info about finding non-paywalled versions here is great, though - thanks for mentioning it.

I totally get the caveats around press releases from institutions, but some of them seem to be fairly good - case bay case, if you will. I used this one in a post, and it also handily pointed me to the paper itself right away - a it wasn't paywalled, so that was a bonus.
posted by mandolin conspiracy at 10:50 AM on September 7, 2017 [2 favorites]

I endorse this callout.

It's not asking you to get all of science to bless your post. Just think twice before posting single links to institutional propaganda.

Y'all smarter than that.

Love,
Chief explainer of science to my mom
posted by zennie at 2:58 PM on September 7, 2017 [5 favorites]

The purpose of a post is to generate chit-chat

All this time I thought the purpose of a post was to share a thing with people and the chit-chat part was beneficial but secondary.
posted by hippybear at 8:38 PM on September 7, 2017 [3 favorites]

a study which is being published in a very prestigious and peer-reviewed journal

...Stanford researchers report August 22 in the Proceedings of the National Academy of Sciences ....

Let's talk about what peer review means for a moment. I'm a genomics researcher, and peer-review here usually means an impartial handling editor will find three to four anonymous and independent expert reviewers and decide whether the paper should be published and whether any aspects of the paper need to be changed in response to reviewer comments.

This paper was published through the "Contributed Track" at Proceedings of the National Academy of Sciences (PNAS), which, famously uses a different process. Use of the Contributed Track is a privilege given to National Academy members, generally the most prominent established scientists. In the Contributed Track, the contributing author effectively serves as their own handling editor—the contributing author decides who the reviewers are and influences to what extent any adjustments need to be made. The two reviewers here are well-known prominent scientists and if the paper were published over grave objections from them we might expect to hear about it. But on the other hand, someone who suspected that they might not find the paper acceptable might just decline to review in the first place, given that they don't have the usual protections of anonymous peer review.

The Contributed Track once served a useful purpose in allowing scientists of considerable stature to publish new research even when it would otherwise be blocked by gatekeepers due to controversy. Now, of course, they can just post their new research to the internet so it's less clear why we need this. Regardless, I wouldn't assign it the same value as traditional peer review.

This is the sort of thing you don't expect to be revealed in a university press release. Better to get your commentary on new research from an independent source.
posted by grouse at 9:30 PM on September 7, 2017 [7 favorites]

Contributed track FAQ.

There have been arguments that CT papers are a cliquey way of getting not-impressive research published in a high status, high impact factor journal. The arguments are not entirely without merit.
posted by lalochezia at 5:23 AM on September 8, 2017 [1 favorite]