In the infamous presidential election of 2000, for example, many of the crucial issues were statistical in nature, but the commentators were almost always lawyers and journalists. Regression analysis of the Buchanan vote in the various Florida counties would have clearly demonstrated how much of an outlier Palm Beach County was. An examination of the tiny difference between Gore and Bush in the official vote totals, especially given the crude Florida election apparatus, would have shown it to be statistically meaningless.

Paulos also mentions Alan Dershowitz’s bad probability arguments made about the O. J. Simpson trial (during which, David Blatner observed, there was also some chicanery about the number pi). The Australian writer Russell Blackford and I analyzed some claims about the Essjay incident, a briefly scandalous affair regarding Wikipedia, and found the numerical accuracy of media reporting to be astonishingly bad.

In short, on subjects where a scientific talking head should be televised, none is in evidence.

Of course, even coverage of clearly scientific topics shows abominable gaps in its fact-checking. Exhibit A is the uncritical reporting about Louann Brizendine’s book The Female Brain.

It would be useful for non-journalist science people (and anyone else interested) to understand how information flows through and what tends to be done to it at various stages of the journalistic and publishing process.

After all, as Kevin Padian states in his e-mail: “we don’t get trained in this”..!

journalists don’t write their own headlines!

As a reader, I long ago learned to assume the headline was written by someone who didn’t read beyond the first paragraph.

Dr Padian’s advice is good. I would just add a couple of things that we tell people in media training:

1. Respect deadlines, but don’t get bounced into responding off the cuff. You can always say you’re busy right now and ask the reporter to call back in fifteen minutes. That gives you time to write out the two or three key points you want to get across. Use everyday analogies, real-world examples, and figures to convey a point.

2. We don’t advise trying to go “off the record.” If you don’t want to say it, don’t say it. Good journalists are information sponges, so always assume that anything you say to a reporter in any situation could find its way into print.

Contact your press office and find out what services they offer; if you think your work is going to get into the news at some point, build a relationship with the reporter at your local paper who covers that — that could be science, medicine, higher ed or business/technology, depending on the size of the paper.

Some tips on what to do “When a reporter calls” here.

There was a story in the UK about a serious childhood disease that could be communicated in various ways from parent to child. A reporter asked the researcher whether it was possible for the disease to be caught through a kiss. The researcher replied that it was theoretically possible. The headline was something along the lines of – ‘Parents could kill with a kiss’.

Of course, the real risk of passing on the disease through a kiss was negligible. You could say that the problem here was the scientist’s use of technical terms to describe the level of risk involved. If he had been trained to find other terms to describe risk levels when talking to the media – perhaps saying “there is no real risk” rather than “there is a theoretical risk” – then the reporter wold not have picked that line.

Sometimes it is just a question of learning each other’s language.

The “stick to the message” and “don’t answer the question asked” can have you coming off with that sort of smug condescention some folks in the Bush administration are famous for.

Personally, if anything will make me hostile, it’s this approach–the epitome of PR bullshit merchants.

I’d suggest that if you are so threatened by talking to a reporter that you can’t imagine doing it without following guidelines like these, then you shouldn’t talk to reporters. But you SHOULD acknowledge you’re kind of neurotic.

One of my father’s jobs in the newsroom was to lay out pages: decide which articles go where, choose the points to break the front-page articles which are continued deeper in the paper, and so forth. While working on the page layout, he would put in headlines, just to see how they worked with the space. He always used nonsense strings of letters (“Xyzzy xyx”) to make absolutely clear that these weren’t real headlines. Some people, he said, used silly phrases about brown foxes jumping over lazy dogs and such, but there are risks to doing that.

For example, he said, one editor (Kirk Scharfenberg) stuck a joke title on an editorial about President Carter. He thought it would get a chuckle out of his Boston Globe coworkers, but somehow, the paper got printed and, oops, Carter’s speech on inflation was forevermore known as “mush from the wimp.”

This has nothing in particular to do with science journalism; I just think it’s a great piece of lore.

Definitely a good point to bring up about headlines! Since paleontology doesn’t result in too many “scare stories”, instead we get the following trite headline elements repeated again and again (perhaps because the subeditor thinks they are being original…):

If it has to do with something related to bird origins, then “ruffles feathers” will be in the headline.

If it has to do with some new phylogenetic arrangement of taxa, than “overturns evolutionary theory” or (even worse “challenges Darwinian evolution” might show up.

And, of course, the old-fashioned superlatives: oldest, youngest, largest, smallest, fastest, meanest, rootinest tootinest, etc. (Okay, these I get: the press does need to have a hook to get the reader to actually want to read that story, and superlatives are an easy way to do it).

I wholeheartedly agree with Kevin’s comments about handling the interview from the interviewee point of view (short quotes, deliminate what is on and off the record, credit where credit is due, etc.)

And he also has good points about misperceptions on the part of the press (and general public, actually) about the usefulness of a media presence in academic careers. Granting agencies, promotion-and-tenure committees, etc., really don’t care too much about this. It might work towards your professional advantage if your home institution regards “public outreach” as a component of your job, but that’s about it.

Something I find useful (time prior to an interview permitting) is to summarize to myself what I think the main points of a study would be, and then make sure that I can articulate all of them as I might to a 100-level undergraduate class or a public audience lecture. Have a list of bullet points of key concepts handy (metally, if not on paper or computer monitor!), and see that you get all of them out. They will NOT all be used; perhaps only one (or part of one). But at least you’ll know that you did your job: conveying your information and interpretation to the professional press.

There is a curious breed of person called the ‘subeditor’ whose job it is to summarise someone else’s story in a few pithy words. I have seen many a well-written and balanced story reduced to a scare story through the simple use of an overblown headline. This matters even more in the online world, where scanning (and not full reading) is the order of the day).

Obviously, this doesn’t apply to misquoting or similar transgressions, but it does apply to many examples like Carl’s iPod incident.