When the abstract or conclusions aren't accurate or enough

By Grant Jacobs 08/08/2011

(Particularly for those interested in science writing.)

Some time ago I wrote about a local study surveying errors of fact and logic in media reports about immunisation. Also some time ago, attracting a little fuss on the same topic, was a study in PLoS One, arguing that misrepresentations in the conclusions and summaries of ADHD research papers might be one contribution to the gap between what is current best understanding in neurobiology and what is stated in the media. (Ironically given the topic, whereas the authors write ‘neurobiological facts’, I prefer ‘current best understanding’ to ‘fact’.) As their Background statement reads:

There is often a huge gap between neurobiological facts and firm conclusions stated by the media. Data misrepresentation in the conclusions and summaries of neuroscience articles might contribute to this gap.

Their paper is open-access, I encourage readers to read it for themselves.

I’d like to briefly widen their thoughts to science communication in general. Since I started on post I’ve learnt that other bloggers have examined this paper, among them:

I’m fighting a losing battle with the to-do list encourages me so forgive me for limiting myself to two quick anecdotal thoughts. For sake of what I write let’s accept that their conclusions are justified (but see Jacqueline‘s criticisms).

Lack of critical reporting
It’s not good if abstracts or conclusions in a research paper are misleading but, having accepted that, if a journalist is more-or-less parroting an abstract or conclusion are they doing much better than uncritically echoing the content of a press release? (The latter practice is regarded as poor science journalism.)

Research papers are generally best regarded as arguments for a case. The data may be the data, but the interpretation is open. With this in mind, shouldn’t writers bring at least some basic critical judgement to their subject matter?

Forward-looking statements
It is fairly common to see forward-looking statements in the end of discussion sections (e.g. at the very end of some research papers).

These are not conclusions from the work but are hopes for the longer-term direction or the (prospective) context in which the work sits.

Readers would note that for two of the examples ‘over-statements’ in Table 3 of the paper the over-statement is qualified with ‘may’. Should these be read as forward-looking tentative prospects, i.e. for future research to address, not conclusions?

The examples are open to interpretation (I’m not a specialist in that area) but I think the point is worth raising nevertheless.

I’ve seen media reports pick up clear examples of forward-looking statements and report them as if they were conclusions. I sometimes wonder if this is in part because these throw-away statements often sound more significant–long-term goals tend to be–and are simpler to read than the detailed body of the paper as they often use less jargon.

In the interests of better science communication perhaps there these forward-looking statements should be more clearly identified as such, and also for journalists to be more aware of this issue – ?

These are more readily seen for what they are if the reader is familiar with the field. Read this way, it would be yet another pointer to a need to use writers with domain knowledge.

For all the debates I’ve read about who should write about science, I don’t think that there is any getting around that those with appropriate domain knowledge are better placed to write about a subject with accuracy and nuance.


This is rescued from the Drafts pile as I’m too busy to write today. I had hoped to make this more substantial, but I hope it’s still worthy someone’s time.



Gonon, F., Bezard, E., & Boraud, T. (2011). Misrepresentation of Neuroscience Data Might Give Rise to Misleading Conclusions in the Media: The Case of Attention Deficit Hyperactivity Disorder PLoS ONE, 6 (1) DOI: 10.1371/journal.pone.0014618

Other articles in Code for life:

Fact or fallacy, a survey of immunisation statements in the print media

Media thought: Ask what is known, not the expert’s opinion

On alternatives to academic careers and “letting go”

Accessing digital legacies (experimental ones, too)

0 Responses to “When the abstract or conclusions aren't accurate or enough”