John Lott emails and comments that we “probably should correct” this post where we wrote:

In the first section of his article, Lott claims to be examining news coverage of quarterbacks, and it was overall bias that Rush had alleged. But what subset of his data did Lott use when presenting his simple comparison of averages? The one that showed articles about black QBs to be 67% positive, compared to 61% for whites? Lott only used weeks in which these QBs played. Should that matter? Big time.

Lott’s total dataset includes 1346 articles, 1013 about white QBs, 333* about black QBs. The “simplest measure” shows that white QBs receive positive comments in 57% of all articles, while black QBs’ positive articles only amount to 53%. Though again a small difference, the result is exactly the opposite of what Lott (and Limbaugh) claimed. [Lott’s selection excludes 50 articles about white QBs that are 70% positive, and 41 articles about black QBs, 30 of which are negative. What a convenient feature of the data this turned out to be.]

Lott included results of the “average” for news coverage when QBs did not play. Lott comments (on his web site🙂

During the weeks when quarterbacks played about 67 percent of the news coverage for black quarterbacks was positive and about 61 percent of the coverage for whites was positive.

Progress has been achieved at last — now that we know how Lott calculated his average (how we came to ours should be pretty clear from the quote above.)

When running his regression analysis, Lott argued that one had to take into account “differences in media coverage.” But when looking at averages, such considerations are apparently irrelevant. What Lott did was take the average of positive stories about each QB every week, and averaged these weekly ratios. The problem, we think, is quite simple. Let’s take Donovan McNabb: in week 4, there were 11 positive stories, and 5 negative ones, i.e. 69% of stories were positive. In week 2, there were 3 positive stories, and 29 negative ones, i.e. 9% of the stories were positive.

Lott’s average would say that 39 % [69 + 9 /2] of the news coverage for Donovan McNabb was positive. We would say that 29% of said coverage was positive [11 + 3 / 16 + 32.]

Lott should at least have made the method behind the calculation of his “average” clear — as it is when he writes:

During the weeks when quarterbacks played about 67 percent of the news coverage for black quarterbacks was positive and about 61 percent of the coverage for whites was positive.

We think he is (deliberately or not) giving a very misleading picture of what ‘average’ news coverage is. We assumed, and think it reasonable to have done so, that Lott looked at aggregate numbers. As it is, we think most people would be surprised to learn that for weeks 1 and 4, Kurt Warner’s coverage was, according to “Lott-average,” 22% positive: 0 out of 2 one week (0%) and 16 out of 36 the other week (44%.) The Sadly, No! Average? would find that Warner’s coverage was 42% positive.

Roger Ailes has additional comments here, as does Tim Lambert there.

* Typo corrected, read 303 rather than 333.