Caught with their stats down
Be sure to give a look to the Public Editor column in today's NYT: Can a 15-year-old be a 'woman without a spouse'? Not to spoil the ending or anything, but Barney finds both the reporter and the editors playing fast and loose with the data on a 1A trender from last month -- the one that proclaimed that "51% of women are now living without spouse."
The picture he paints is dismaying because it's so familiar. The data in the analysis aren't quite the same ones reported in the graphic and story (but the reporter didn't think they made any difference, so ...). A few common-sense questions about the makeup of the trend -- say, "is it a good idea to count 15-year-olds as spouseless women?" -- got brushed aside in the interest of being out-front with "this tipping point." The reporter ends up defending the story because its "essence" remains accurate:
about half — maybe a little bit more, maybe a little bit less, depending on the age group — of American women are living without a spouse at any given time
even though the whole point is that the magic 51% mark has been reached.
It's a relief, sort of, to know that somebody got the glass offices' attention: Boss editor Bill Keller "has decided to meet with staffers with expertise in statistics and demographics to create a 'vetting network to help with the editing of articles dealing with those subjects.'" That's nice. But the problem in this sort of case isn't usually a lack of experts; it's the unholy lust that makes newsrooms see only what they want a story to say, not what it says in the cold light of day.
Expertise in statistics? Nah. You could chisel an introductory stats text onto stone tablets and haul 'em from Mount Sinai to the Times' 1A budget meeting and it wouldn't make any difference.
The picture he paints is dismaying because it's so familiar. The data in the analysis aren't quite the same ones reported in the graphic and story (but the reporter didn't think they made any difference, so ...). A few common-sense questions about the makeup of the trend -- say, "is it a good idea to count 15-year-olds as spouseless women?" -- got brushed aside in the interest of being out-front with "this tipping point." The reporter ends up defending the story because its "essence" remains accurate:
about half — maybe a little bit more, maybe a little bit less, depending on the age group — of American women are living without a spouse at any given time
even though the whole point is that the magic 51% mark has been reached.
It's a relief, sort of, to know that somebody got the glass offices' attention: Boss editor Bill Keller "has decided to meet with staffers with expertise in statistics and demographics to create a 'vetting network to help with the editing of articles dealing with those subjects.'" That's nice. But the problem in this sort of case isn't usually a lack of experts; it's the unholy lust that makes newsrooms see only what they want a story to say, not what it says in the cold light of day.
Expertise in statistics? Nah. You could chisel an introductory stats text onto stone tablets and haul 'em from Mount Sinai to the Times' 1A budget meeting and it wouldn't make any difference.
posted by fev at 7:29 PM
0 Comments:
Post a Comment
<< Home