Thursday, July 30, 2009

Research tips for writers who hate numbers

I'm not sure why, but many reporters I've worked with over the years, either colleagues or outside journalists calling for information, seem to have a strong aversion to crunching numbers. (I'm excluding, of course, financial reporters.) They have no problem quoting "experts" who offer this or that estimate--on anything from the growth of inflation to the exchange rate of euros to the percentage change in some corporation's quarterly operating income. But they rarely seek to develop their own estimates, let alone extract signifcance from a column of numerical data.

What I hope to do with this part of the blog is to provide some simple steps for such writers or editors to get the numerical information they need, without trepidation, and without having to chase after some academician or security analyst. Some of the tips will be super-simple, such as going to a particular website and following the prompts. Others will entail a few more steps. But whatever the process, I hope to present it as easily as possible.

First challenge: measuring inflation, or translating dollars from years past to today. Suppose you're doing a story that includes somebody's salary in the year 1975. You find out, through an interview, that he or she earned $9,000. How much is that in today's currency?

Do this: go to the website of the US Bureau of Labor Statistics' inflation calculator. The caculator will appear. In the top box, type in $9000. In the second box, type in the year 1979. In the third box, type in 2009. Hit "calculate." The answer appears as $26,738.80; you can write something like "roughly $26,700."

Why "roughly?" Because the Bureau measures general consumer inflation, which is what most people are interested in. But that measure doesn't mean much to, say, a hospital or construction company or university, where costs tend to rise far more rapidly than basic consumer goods, like a basket of food. Later, we'll talk about how to measure inflation in those areas. For most purposes, though, the BLS estimate is appropriate: you don't need to call an economist to figure out what yesterday's prices mean in today's prices, or vice versa.