PolitiFact Wisconsin gives us a fantastic example of that flawed method with its fact check of Republican presidential candidate Ben Carson. Carson highlighted problems with advancement in the black community by saying there are more blacks involved with the criminal justice system than with higher education.
PolitiFact Wisconsin decided to evaluate claim and rated it "False." But of course there's a problem with the rating. PolitiFact Wisconsin found it had poor data with which to work:
(T)here is only one solid figure -- 75,000 black males ages 18 to 24 in prison. We’re not aware of any recent counts of the black males in that age group who were arrested, in jail, or on probation or parole at a particular time.PolitiFact Wisconsin emphasized that relatively low solid figure in its summary paragraph:
Carson did not provide evidence that backs his claim. The latest federal figures we found show 75,000 black males in that age group who were in prison in 2013 and in the range of 690,000 to 779,000 who were in college. We are not aware of any recent figures for the number of black males ages 18 to 24 arrested, in jail, or on probation or parole at any particular time.Perhaps it makes sense if the number of blacks ages 18 to 24 in college outnumber those involved with the criminal justice system 779,000 to 75,000. But that number comparison is rigged against Carson. PolitiFact Wisconsin acknowledges Carson's claim on unknown numbers of young blacks arrested, in jail, on probation or on parole.
If figures do surface, we’ll re-evaluate this item, but we rate Carson’s claim False.
And that's the Achilles' heel of PolitiFact Wisconsin's fact check. It collected enough information to enable rough estimates of those categories.
Was Carson's claim plausible?
We start our estimate by noting the percentage of blacks ages 18 to 24 in the federal prison population was fairly high: PolitiFact said the number of about 14 percent of the total black population.
We aim to create a conservative estimate, erring on the side of caution, so we'll assume that just 10 percent of blacks in the other categories fall in the 18 to 24 age range.
PolitiFact Wisconsin noted that one individual might be arrested more than once. Still, the fact checkers gave the number 3 million for arrests of adult blacks in 2012. Ten percent of 3 million gives us a figure of 300,000, but to help account for multiple arrests we'll cut that number in half and use 150,000.
PolitiFact Wisconsin gave a figure of 261,500 for jail inmates in mid-2013. Ten percent of that figure gives us about 26,000.
On Probation or Parole
PolitiFact Wisconsin said 4.75 million people of all races were on probation, parole or other supervision in 2013. The Bureau of Justice Statistics, PolitiFact's source, says blacks account for 30 percent of that number. That gives us 1.4 million, and 10 percent of 1.4 million comes to 140,000.
Combined with the 75,000 prison population PolitiFact Wisconsin used, our conservative estimate comes to 391,000--about half of PolitiFact Wisconsin's peak figure for black college enrollment. Based on our estimate, we think it's very unlikely Carson's claim exaggerates the truth by more than 100 percent, probably exaggerates it by substantially less than 100 percent and perhaps doesn't exaggerate at all.
For comparison, PolitiFact Georgia recently gave a "Mostly False" rating to a claim exaggerated by over 200 percent.
Do these numbers potentially support Carson's underlying point about the upward mobility of young black males? For some reason, PolitiFact Wisconsin did not deem that point worth considering.
Shall fact checkers rate claims "False" if they are difficult to settle? We think that's the wrong method. We also think fact checkers err by selectively ignoring politicians' underlying arguments. Either consider the underlying argument every time or never consider the underlying argument. Fairness demands it.
Post a Comment
Thanks to commenters who refuse to honor various requests from the blog administrators, all comments are now moderated. Pseudonymous commenters who do not choose distinctive pseudonyms will not be published, period. No "Anonymous." No "Unknown." Etc.