The snowstorm that hit much of the United States last week was one for the books. In Chicago, the 20.2 inches that fell was a “record.”

It wasn’t the most snow Chicago had ever received in one storm; that honor belongs to a 1967 storm that left twenty-three inches over two days in January; last week’s accumulation was only the third deepest. It wasn’t even the most snow Chicago had received in twenty-four hours; that was 18.6 inches in January 1999. (Because it snowed longer than twenty-four hours that year, that storm became the second-snowiest Chicago has seen, with a total snowfall of 21.6 inches.)

So just how did last week’s storm set a “record”?

Easy: Feb. 1, 2011, was the snowiest February day ever recorded in Chicago. (Cue the streamers, horns, and confetti!)

It’s easy to create a “record,” since nearly all of them are limited in some way. “The snowiest February day recorded in Chicago” limits the record to February days that have been recorded in Chicago. If it happened before someone started recording, well, it didn’t happen. And had this storm happened a day earlier, it would not have been a “record.”

But, among others, the Chicago Tribune really, really wants this storm to be memorable. It refers to it as “the near-record blizzard,” and its weather website brags that Chicago has had more than fifty inches of snow for four years in a row: “It’s the first time this has happened in Chicago since snowfall records began here 126 years ago.” Yay! Another “record”! A fairly obscure one, to be sure, but a record nonetheless. But is it worth reporting?

“Records” are popular in journalism because a “record” adds news value. (And they’re not “new records”; they’re just “records.”) But “records” are extremely easy to manipulate. A “record” crowd at a concert at a football stadium may be a “record” only for concerts, where the seating may differ from that at a football game. Even seating arrangements for different concerts can affect the capacity for crowds, and thus the potential for “records.” Rather than proclaim a “record” for events that may not be comparable, it would be more realistic—and honest—to simply give the statistics, making note of any previous statistics, without seeking to claim the crown.

Also, note what opportunities there were to break the previous “record.” If a two-year-old company posts “record” earnings for its second quarter, don’t buy into its hype. The company started at zero; its second-quarter earnings were a “record” last year, too.

Some “records,” of course, can be presented as such: “The most quarterback sacks in a Super Bowl game” is pretty straightforward. But beware when the qualifiers start adding up: “The most sacks of an AFC quarterback in a Super Bowl” might be OK, but “The most sacks of an AFC quarterback behind his own forty-yard-line in a Super Bowl game played north of the Mason-Dixon line” is just somebody’s idea of playing with stats.

Don’t get snowed under by what may be a blizzard of PR.

If you'd like to get email from CJR writers and editors, add your email address to our newsletter roll and we'll be in touch.

Merrill Perlman managed copy desks across the newsroom at The New York Times, where she worked for 25 years. Follow her on Twitter at @meperl.