The unpublished results taboo

Late last year, I retweeted a university press release about some topical research on bees that hadn’t yet been published or, apparently, peer reviewed (I can’t find the paper online anywhere, so it looks like it is still yet to be published).

To be fair, the release stated upfront that the research was ‘preliminary findings’ and the source mentioned at the end was an upcoming conference presentation, not a journal article. But should it have been the subject of a press release in the first place?

The story subsequently got a fair bit of coverage in news media, most of which presented the results as fact. Although most stories clarified within the text that the results hadn’t been published yet, the headlines didn’t. This is a problem, given that many online news consumers skimread or just look at headlines. While subsequent media misrepresentation is not the researcher’s responsibility (journalists have their own code of conduct to follow), it is a known risk that they should take into account.

Similar arguments against talking about unpublished results come up in scicomm discussions. If research hasn’t yet been peer reviewed, many academics think the results shouldn’t be communicated at all.

But hang on a minute. A lot of standard science communication involves discussing unpublished or preliminary research…

Conferences: A lot of work presented at conferences is currently in review or hasn’t been written up yet. In fact, some people think the discussion of unpublished results is a key goal of science conferences. Live-tweeting of conferences, and even Twitter-only conferences, are increasingly promoted as an excellent way to engage with a wider audience. Although some people aren’t comfortable with their results being tweeted beyond the conference audience, we are mostly all comfortable with presenting those results at the conference in the first place.

Preprints: The preprints juggernaut and post-publication peer review are being pushed as the future of scientific publishing. Publishing research openly online before it’s been peer reviewed isn’t any different to telling your institution’s press office about it before it’s been peer reviewed.

Academic blogs: Many academics who blog discuss their current research on their blog. Out of the ones I follow regularly, most seem to post details of results after they’ve been published. Brian Roman lists some good reasons for taking this approach, some of which I think also apply to preprints. I also wait until my papers are officially online before I write about them on this blog…but not because I don’t think it’s okay, I’ve just never really thought about it until now. I do sometimes post natural history observations here instead of publishing them in a peer reviewed journal – mostly because very few journals see the value in natural history observations. But also see Simon Leather’s posts on some of his unpublished data and ideas he never followed up.

Teaching: A lot of lecturers discuss their current research with students if it’s relevant to the course topic. As an undergrad, I remember quite a few of my lecturers using their preliminary or unpublished data to teach us theory or analysis techniques.

Citations: Put up your hand if you’ve cited a “personal communication” or “unpublished data” in one of your published papers.

So does it matter if researchers talk about unpublished results in the public domain?

I think it depends a lot on the context. Some data are less risky than others. For example, natural history observations can’t be ‘wrong’, but the potential for it to be a rare or context-dependent observation should be acknowledged, in case people get too excited (like the carnivorous kangaroo sighting). Interesting species-habitat correlations may have less of an impact if misrepresented as fact than a prediction or cause-effect correlation.

If you’re not sure, steer clear of your institution’s media office. A press release, by nature, is an official communication from an organisation. Many modern news media rely on press releases for a large proportion of their news. Interestingly, I’ve started to notice more and more universities releasing press releases about a researcher’s upcoming presentation at a conference…is this a new engagement strategy?

But in general, I don’t think there’s anything wrong with talking about new results in public…as long as the research is clearly presented as unpublished or not yet peer reviewed, and the narrative focuses on modal verbs indicating likelihood not certainty, e.g. possibly, could, may, might etc. This also provides a good opportunity to discuss the scientific process, not just the research itself, which is an important (and overlooked) aspect of science communication.

I’m keen to know what others think about this!

© Manu Saunders 2017

4 thoughts on “The unpublished results taboo

  1. sleather2012 May 18, 2017 / 1:57 AM

    Great post – Personally, I think it is fine to talk about unpublished results at conferences and at Departmental seminars, although some don’t, presumably for fear of someone stealing their idea. This tends to make for less interesting boring conferences, especially as some people do seem to make a habit of recycling the same talk at different conferences 🙂

    Liked by 1 person

  2. U Träger May 19, 2017 / 11:20 PM

    Totally agree. Great post. New results should be communicated. To other researchers at conferences to avoid doubling efforts/ hear if others have seen similar thing. But also to the public- BUT scientists need to be very careful about this. It needs to be clear that it is preliminary work! No press releases, they are official. In worst case patients get their hopes up, similar to press releases overstating results. Both to be avoided!

    Liked by 1 person

What do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s