Black kids in math on the latest NAEP!


Storyline battles statistics: In a new, intriguing Kevin Drum mentions a conversation, then offers a set of statistics about public school test scores in the various states.

According to Drum, the conversation went like this. There's something we might be able to learn from right here:

DRUM (4/10/23): Last night the conversation at dinner somehow turned to the subject of education and the South. No conclusions were reached except for a vague agreement that Southern states didn't spend a lot on education and had generally poor outcomes.

But is that true? Naturally I decided to check...

It's possible that Kevin is being diplomatic. Within our blue tribe, we tend to assume the worst about public schools in those Southern states. Within our blue tribe comment threads, we're sometimes less than fully polite.

In this instance, the dinner companions formed a vague agreement that Southern states "had generally poor outcomes" in public education—but uh-oh! When Kevin checked the latest NAEP scores, this is what he found:

DRUM (continuing directly): My metric is 8th grade reading scores on the NAEP because (a) 12th grade scores are old, sparse, and cover only a few states, and (b) reading scores are a better reflection of general education than math scores.


The perhaps surprising result of all this is that, relative to other states, six Southern states are above average when it comes to Black scores. Only three are above average when it comes to white scores. I'm not quite sure what to conclude from this.

In any case, when you look at the results for all students it turns out that Southern states are unexceptional. As a group, they're slightly below average (256 vs. 259, a gap of about one-third of a grade level), but nothing more than that. Among Black children, they're slightly above average (241 vs. 240).

In all honesty, that last paragraph is a bit strange. It compares apples to fruit baskets in a way we won't stop to explain.

In his fuller post, Drum mentions "the usual statistical caveats." Also, he supplies the average reading scores for black eighth-graders and white eighth-graders in the nation's various states. 

Drum's main takeaway is this:

"The perhaps surprising result of all this is that, relative to other states, six Southern states are above average when it comes to Black scores." 

Also this:

Concerning the region as a whole, Southern black kids scored "slightly above average (241 vs. 240)." For the record, that statistical difference is so small that it barely matters.

One caveat of our own would be this—this was the first NAEP test administration after the public school dislocations brought on by Covid. Just as a guess, we'll guess that a few years will have to pass before we can return to statistical comparisons in which we can have significant confidence.

That said, we blue tribe members might want to check our stereotypes concerning certain regional issues. We love, love, love to assume the worst about the yahoos down South. In this instance, Drum's numbers might almost seem to diverge from the standard preferred tribal narrative.

We decided to take a quick look at Grade 8 math scores. Here are some of the numbers we found:

Average scores, Black kids only 
Grade 8 math, 2022 NAEP

Defense Department schools: 272.2
Texas: 259.9
Massachusetts: 259.6 
Georgia: 256.3
Florida: 254.9
North Carolina: 254.5 
United States: 252.1 
New Jerey: 250.9
New York: 250.6
Connecticut: 248.9
California: 248.0
Wisconsin: 237.2

(We included Wisconsin just to be cruel. Permit your gaze to linger on the number from those Defense Department schools.)

Those are average scores for just one year. As noted, it was an unusual year.

That said, black kids in six Southern states exceeded the national average for black kids in Grade 8 math. As you can see, some of our finest blue tribe states were lagging a bit behind.

Could statistical tricks be lurking therein? Everything is possible! Also, various socioeconomic factors haven't been factored into those average scores.

That said, you'll never see such questions discussed, assessed, resolved. Performative virtue to the side, no one actually cares about any of this, and no one ever will.

It's Storyline pretty much all the way down! We humans love our tribal tales. That includes us the occasionally self-impressed people here in our own blue tribe.

For all NAEP data, start here.


  1. Drum said he disregarded private schools since there appeared to be the same percentage of kids in them everywhere. However, who attends public school and who leaves may differ across regions.

    Note that the percentage of kids in private school is likely NOT the same for the Defense Dept schools.


  2. What's with the dumb race-mongering, dear Bob? Aren't you supposed to be a smart liberal?

    ...every time you relapse into this "white children"/"black children" gibberish, it becomes obvious that you are, alas, just a garden variety brain-dead lib...

  3. The concern with private schools is that large numbers of kids are taken out of the public school sample. The reason why kids leave public school matters because those who leave to find enrichment will be better prepared and their leaving will tend to lower the average. Those who leave to attend special education may increase the average by leaving because they would not be expected to do as well on the NAEP. Those who leave for racial or social class issues may decrease the average left behind if they are wealthy (with educated professional parents) or increase the average left behind if they are religious or immigrant children (going to Catholic schools) or decrease the average left behind if they are leaving to attend prep school, as occurs in New England and large urban areas in the East. So it does matter why children are in private school, not simply how many of them have left publis school.

    While the percentage of kids in private schools may remain constant around the nation, the percentage of kids leaving in different racial groups differs, and the percentage of minority kids in different kinds of private schools also varies. Kevin Drum didn't explore that and Somerby doesn't mention it.

    "African Americans accounted for 6.9 percent of the students in Montessori schools and 18.5 percent of students at schools that served special education students. Blacks were 12.4 percent of the private school students in the South but only 4.7 percent of the private school students in the West. African Americans were 12 percent of all students who attended small private schools that had less than 50 students. But Blacks were only 6 percent of all students who attended larger private schools with 750 or more students."

    That makes Drum's analysis kind of meaningless because there are unexamined confounds that make it difficult to draw conclusions about what the data mean. Drum is good at making graphs and charts, but less good at analyzing them. He tends to like showing that an expected social phenomenon is not justified by the data. In this case, he is trying to say that, but he really cannot because of the possibility that the college prep white students are drained off in the urban areas, reducing the average NAEP scores of those kids remaining in public schools. That makes the Southern schools look better by comparison. A higher percentage of black kids are attending private schools in the South, but they are also more likely to be attending a special ed school. That means the low black performers in the South may be going to private school, which will make the public school averages for black students higher, because the worse performers have been siphoned off into private schools.

    Patterns reported in studies of private school attendance by minority students suggest that Drum's gloss of private school attendance may be unwarranted. He didn't examine this and his assurance that it doesn't matter is based on nothing except that it would be difficult to analyze.

    It always bothers me when Drum speaks with assurance about a superficial analysis of results that he thinks contradict everyone else in an area of study. His preferred narrative is that other people's work is bunk, which he confidently asserts with no qualifications whatsoever in that field.

    Drum is not an expert in education. That bothers Somerby when the writer is a journalist, but not when Drum (whose degree is in journalism) says something. I think that is because Somerby likes Drum's preferred narratives, not because Drum has qualifications or is making sense.

  4. Why would Somerby show us the math NAEP scores instead of the reading ones that Drum analyzed?

    Reading scores correlate more strongly with tests of IQ or general intelligence than math scores do.

    Reading scores are more difficult to raise because they depend on early childhood experiences that disadvantaged kids may not have had, which makes reading comprehension more difficult even when decoding is taught. Impoverished early learning environments affect reading and the deficits at entrance compound, causing children to fall behind in later grades. This makes reading scores more intransigent than math scores.

    Reading scores are more strongly correlated with parental income and level of education, thus appear to be less related to teaching efforts, quality of school, etc.

    The symbol system used by math is easier to acquire for non-native speakers of English or children with verbal deficits. It is also more modular, encapsulated, less affected by other aspects of learning and thus can be acquired despite difficulties in other processing (such as reading, although math benefits from reading skills).

    Math skills are less correlated with early childhood experiences, parental education and income, and IQ. They can be taught without reference to outside experiences that a child might not have experienced, although a relatable context helps kids acquire math skills.

    What experiences might a child of a military family have that other kids would not? For one, the chance to live in a variety of places and have enriched experiences of different places and people in their lives. They also have a system set up to incorporate children who float into and out of the learning space, with teachers who have assessment tools and skills to bring new kids up to speed quickly despite different prior learning experiences. Or it could be something as straightforward as a consistent curriculum across all of the Defense Department schools, regardless of location.

    Somerby is so eager to write-off any impact of racism in the South that he doesn't consider why Defense Department schools might have higher average scores. It could be higher per pupil spending. It could be the inclusion of all kids, without any going to private schools. As usual, Somerby doesn't answer any questions himself. It is enough for him to just hint at his preferred narrative while assuring us it cannot be race because Drum said so.

  5. I left a message for Somerby in the Drum comment section, asking him to clear the spam out of the comments here. Since he has done nothing about it, I think that proves he doesn't care about discussion here at his blog.

    1. It's mostly just one sad troll here in comments.

  6. Can we get some statistics on liberals who
    spend a lot of time thinking or talking about
    the test scores of students in Southern States?
    They don’t seem to have a lot to talk about
    at Kevin’s house….

  7. The second amendment is evil.

  8. “We love, love, love to assume the worst about the yahoos down South.”

    If “yahoos” = “Republicans”, then I would urge Somerby to broaden his view. We liberals (including the millions of us down South) don’t love to assume the worst about Republicans, but they keep being the worst. See: Tennessee, recent events in state legislature of. It is what it is.

    Also, my experience as a southerner is that it is (or used to be) northerners, liberal AND conservative, who believe the south has worse schools than the North, whether you’re talking about black, white, Asian, or anything else.

    On the other hand, I don’t believe Somerby or Drum have proven anything at all.

    1. Missouri just voted to defund its public library system.

  9. It’s not that the yahoos in the south love to reminisce about the confederacy, but … well, the yahoos in the south DO love to reminisce about the confederacy. I believe it is now confederate appreciation month somewhere in these southern United States, where we celebrate the defense of slavery and treason to the United States. Hope they don’t run afoul of the stop woke act. Good times!

    Arkansas’ previous governor managed to get rid of the celebration of Robert E Lee’s birthday, which (coincidentally/not coincidentally) was celebrated on Martin Luther King‘s birthday. We cannot predict the future with current governor Lord Huckabee sanders, however.

    1. Bob southern snowflakery is annoying,
      but he retreats to it when the big
      old ugly world is closing in. I
      wish the Drum Family would
      find something to talk about that
      wouldn’t throw it raw meat.

    2. IMO one should distinguish between things that affect blacks merely symbolically and things that make a real difference in their lives. I sympathize with blacks who are offended by Civil War memorials. However, statues of Robert E Lee on a horse don't affect their lives like failing inner city schools, or watered-down black studies courses, or increased rates of being murdered or robbed, or type of employment opportunities in their neighborhoods.

    3. Why should this be either/or? Why should black people be asked to tolerate any of these unacceptable things? An in-your-face symbol of past slavery is not trivial and it should not be happening anywhere.

    4. @8:56 It should not be either-or, but in practice it is. Liberals are doing away with confederate statues, That’s fine. BUT liberals are opposing school choice, which would improve education. And liberals are opposing stronger police protection, which would black potential victims.

      Instead liberals just call conservatives “racists”.

    5. Above comment from David in Cal.

    6. David,
      School choice is as smart as Defense choice. Parents should be the ones making the choice of how to best defend their families (children).
      Every parent should get a voucher for defense. Big government elitists think the Pentagon and Department of Defense know more than parents about what's best for their children.
      Besides, watching Raytheon marketing anti-missile defense systems to parents, will be totally worth it.

  10. Strike a match department: I’ve never been a
    Jon Stewart fan but his podcast with a Defamation Law Professor on the Dominion case and the surrounding issues was really excellent, just the cure for Bob’s embarrassing
    Inability to deal with said issues in
    any way.