August 22, 2007

This could explain a lot

Among the most important critters in the ecology of science journalism are the "public relations" or "communications" specialists who write the press releases for universities, corporations, scientific societies and so forth. As far as I can tell, nearly all the science stories that make the news are picked from among the press releases that flow through conduits like EurekAlert. And in most cases, what's published or broadcast features the general spin and also the specific details -- quotes and numbers -- provided by the press releases.

The people who write the PR materials have a hard job. They need to take complicated results, full of background assumptions and layers of caveats, and present them in a form that non-specialist journalists will understand, and will find interesting enough to choose from out of the flood of competing alternatives.

The PR people are not specialists themselves, and in many cases may not have a lot of scientific or mathematical background. But I was still surprised to see this, on Dan Santow's excellent Word Wise blog ("Writing tips for public relations professionals"):

I just came across an online tool that could help writers (who are not known for their mathematical skills - or is it just me?) that’s really worthwhile (especially if you work in financial communications). It’s called the percent change calculator and all you do is enter a number, then the number it changed to and it tells you the percent change. No long division, no embarrassing yourself by asking colleagues if they know how to figure out the percent change from 745 to 13 (98.3 percent decrease), no feeling really old and decrepit. It’s divine.

I gather from this that there are educated, intelligent and otherwise skilled adults who are not sure how to turn two numbers into a percentage change, and that some of them are working as public relations professionals. I'm not trying to be snarky, I'm just truly and sincerely surprised.

[Update -- Ed Keer is surprised that I'm surprised:

Let's see if I can help Mark understand. First, in my experience most writers in PR, advertising, or communications have humanities educations. Remember back in college there were all those people taking English courses and their parents would say, "What are you going to do with that?" Well, that's what they're doing with that. And second, if the last time you were asked to figure out a percentage change was 20 years ago, when you were more focused on the cute potential mate in the seat across the way, you might be a little rusty.

But, but, percentages are taught in the 5th grade, more or less, well before most students are distracted by mating opportunities. And they play a role in a number of everyday activities, like tipping and sales discounts and the like (though I admit that those can easily be faked). So this seems unfair and disrepectful to English majors, who are generally smart people. I've known quite a few English majors who could calculate percentages better than I can.

And Holly Cordner, who a college student "studying English" (well, and also "information systems") agrees:

If you can’t even do elementary math, how did you get through your formal education? How did you get a college degree? No one expects journalists to be the next Isaac Newton or anything, but they should at least be math- and science-literate enough to be able to recognize blatant errors in either calculations or reasoning.

But Zeno at Halfway There has some hair-raising anecdotes about people in positions of authority who demonstate complete failure to understand elementary percentages and related concepts. And Dick Margulis blogged some serious innumeracy in the 1/22/2007 issue of the New Yorker (though percentages are not crucially involved, just an apparently inability to detect a factor-of-a-thousand error in the sum of a set of numbers given as components of total energy expenditures in the U.S.).

When you add this to Bill Poser's anecdote about a case in which working teachers were felt to need instruction in how to calculate the percentage error in students' in test results, I guess that I have to stop being surprised.. There's a prima facie case that there are some high-functioning adults who are seriously percentage-impaired.

If this is true -- and it would be nice to have some real evidence about how many people are affected, not just suggestive stories -- then we've got a major disability on our hands that has somehow escaped the process of identity politics. Where are the lobbyists, counselors and support groups for the percentage-impaired? Where are the stories about the effect on our GNP of the financial-communications professions who worry about "embarrassing yourself by asking colleagues if they know how to figure out [a] percent change"? Seriously, I worry that people with this problem are embarrassed to admit it and to seek help. In some cases, they may not even realize that they need help. But they do. ]

Posted by Mark Liberman at August 22, 2007 10:19 AM