The Volokh Conspiracy - Bogus Statistics, Courtesy of the <i>L.A. Times</i> Editorial Board?

Late last month, an L.A. Times editorial reported that "In our America, 60 million people survive on $7 a day" — which is to say that 20% of the population survives on $2555 or less a year.

That's obviously way wrong: As of 2006, according to the Census, 12.3% of the population lived at or below the poverty level, which was $10,294 for a single person and $24,382 for the average family of five ($23,691 if one assumes only one adult and four children). Thus, even if all the poor people in the country were in families of one adult and four children (which I suspect substantially overestimates the average family size), that would mean 12.3% of the population surviving on $4600 or less, not 20% on $2555 or less. Naturally, I wouldn't want to live even on $10,294 per year, but the Times made a specific assertion about a particular number. It's pretty clearly a false assertion.

Annie Jacobsen (Pajamas Media) tries to track down the source of the error, in considerable detail; I encourage you to read Jacobsen's piece, which among other things suggests the Times is too ready to rely on advocacy sites rather than tracking down the facts itself. Among other things, Jacobsen reports that she questioned the Editorial Page Editor about this, in enough detail, I think, to put him on notice that there might be a problem there; my quick search, though, suggests that the Times hasn't published a correction.

Jacobsen's story also shows how careless or insufficiently skeptical editors can fall prey to a journalistic broken telephone: Once one tracks down the source of the $7 a day statistic, one finds it in this N.Y. Times story, which says -- with various sensible qualifiers -- that, according to IRS data, "the poorest 60 million Americans reported [to the IRS] average incomes of less than $7 a day each" (emphasis and bracketed material mine). "The I.R.S. data does not include the value of government benefits like food stamps, the earned-income tax credit for working families and subsidized medical care. It also excludes unreported income, which the Treasury Department and the I.R.S. have said is a major and growing problem among the highest-income Americans, especially those who own businesses, invest in stocks and have overseas financial interests." There's a lively debate in the comments to Jacobson's post about the N.Y. Times story, and the debate includes posts from the author to the story. But it's pretty clear that even if the N.Y. Times story is right, the L.A. Times' indirect echo of that story is way wrong.

Thanks to OpinionJournal's Best of the Web for the pointer.

	[Eugene Volokh, January 14, 2008 at 7:03pm] Trackbacks Bogus Statistics, Courtesy of the L.A. Times Editorial Board? Late last month, an L.A. Times editorial reported that "In our America, 60 million people survive on $7 a day" — which is to say that 20% of the population survives on $2555 or less a year. That's obviously way wrong: As of 2006, according to the Census, 12.3% of the population lived at or below the poverty level, which was $10,294 for a single person and $24,382 for the average family of five ($23,691 if one assumes only one adult and four children). Thus, even if all the poor people in the country were in families of one adult and four children (which I suspect substantially overestimates the average family size), that would mean 12.3% of the population surviving on $4600 or less, not 20% on $2555 or less. Naturally, I wouldn't want to live even on $10,294 per year, but the Times made a specific assertion about a particular number. It's pretty clearly a false assertion. Annie Jacobsen (Pajamas Media) tries to track down the source of the error, in considerable detail; I encourage you to read Jacobsen's piece, which among other things suggests the Times is too ready to rely on advocacy sites rather than tracking down the facts itself. Among other things, Jacobsen reports that she questioned the Editorial Page Editor about this, in enough detail, I think, to put him on notice that there might be a problem there; my quick search, though, suggests that the Times hasn't published a correction. Jacobsen's story also shows how careless or insufficiently skeptical editors can fall prey to a journalistic broken telephone: Once one tracks down the source of the $7 a day statistic, one finds it in this N.Y. Times story, which says -- with various sensible qualifiers -- that, according to IRS data, "the poorest 60 million Americans reported [to the IRS] average incomes of less than $7 a day each" (emphasis and bracketed material mine). "The I.R.S. data does not include the value of government benefits like food stamps, the earned-income tax credit for working families and subsidized medical care. It also excludes unreported income, which the Treasury Department and the I.R.S. have said is a major and growing problem among the highest-income Americans, especially those who own businesses, invest in stocks and have overseas financial interests." There's a lively debate in the comments to Jacobson's post about the N.Y. Times story, and the debate includes posts from the author to the story. But it's pretty clear that even if the N.Y. Times story is right, the L.A. Times' indirect echo of that story is way wrong. Thanks to OpinionJournal's Best of the Web for the pointer.