Religion forms buffer against work stress - how to overstate a survey

[drmatthewl]

The Telegraph and the Daily Mail (among many others) recently reported that religious people are less stressed, less anxious and take less sick days. This article is a great example of how not to report on a study and why more effort needs to be taken to make sure study details are reported with conclusions. Both the Telegraph and the Mail reported the findings of the survey with out supplying information on the study to put any of it in context. The result is a bias article that reports how religion is beneficial to a working, modern life.

A quick google search is all it takes to find a bit more information that helps put the articles in perspective. The survey was carried out by Dr Roxane Gervais from the Health and Safety Laboratory in Stockport. It turns out the results of the survey are based on 34 full time employees, all from the Caribbean. This already casts doubt on the validity of the findings as the results are drawn from a very, very small and isolated population. How do we know that some other factor about the religious people in this group isn't responsible for the reduced stress? We don't, because no controls are in place to account for other lifestyle factors.

A sentence as simple as, 'Dr Roxane Gervais surveyed 34 full time employees from the Caribbean', would have allowed the reader to look more critically at the findings.

Links:

telegraph.co.uk/news/religi...

dailymail.co.uk/health/arti...

[Hidden]

There are two other clues that this isn't very robust evidence for this news.

First, the Daily Mail story is entirely based on The Daily Telegraph's (to be fair to the Mail, they acknowledge this and link to the Telegraph's page).

Second, this is based on a report from a conference. News stories based on conference talks are almost always based on research that hasn't been peer-reviewed, and are very difficult to check the facts of - finding even an abstract for such research can be incredibly difficult.

In this case at least, the abstract is publicly available (here: bps.org.uk/system/files/use... - see page 245) and comparatively detailed.

[JossS in reply to Hidden]

It would be interesting to see how this came into the Telegraph's hands - in what form and by what route. Was it a press release? Was it picked up by some other correspondent at the conference? How was the original report or release worded? And so on.

Leaving out the Carribean link is very important, too, as religion plays a huge part in society there whereas in the UK it has a relatively smaller role. The same applies to any research that would be carried out in the US - relating to this country would be misleading. I can think of some companies I have worked for where adding religion in any form would have added to tension!

[Hidden in reply to JossS]

Hi JossS,

It could well be via a press release (the PR machine goes into overdrive when an organisation is holding a conference), but it could well be simply diligent journalists scouring the abstract books for quirky or controversial news angles. I spent many a long hour poring over gastroenterology conference abstract books for a scoop (to little avail) in my time.

[JossS in reply to Hidden]

Yes, I know the PR machine all too well. I also know how dangerous it can be in the wrong hands. Many a misleading story has had its source in the words of some well-meaning scientist or health professional who doesn't know how to get their idea across clearly and unambiguously. (And why should they? That is not their profession)

[archimboldo]

Hard to believe that they thought this research was worth national attention. And what a simple generalising headline!

[Hidden in reply to archimboldo]

Clearly the Mail will believe it merited attention as it has had >3,500 shares on social media and 399 comments.