Tuesday, February 23, 2016

Dispelling the Ludicrous Conspiracy Theory About ADHD/LD Diagnoses at UNC

I am not going to waste much time on this blog entry, because the issue I'm addressing is indeed nothing but a ludicrous conspiracy theory generated by the crazed ABC crowd.

That crowd's new hero, BlueDevilicious (whose real name is Ted Tatos), has been obsessively tweeting highlighted portions of emails that are being released in an ongoing, massive document dump from UNC. From these emails, he has discovered that an abnormally high percentage of UNC athletes were diagnosed with ADHD and/or learning disabilities (LD). He also discovered an email thread from independent physicians stating that the methods used for diagnosing ADHD/LD were inadequate.

The conspiracy theory that has thus emerged from BlueDevilicious's monomaniacal tweeting is that UNC overdiagnosed athletes in order to secure for them academic accommodations that would give them an advantage in the classroom.

This conspiracy theory is so ludicrous I'm annoyed that I'm even addressing it.

Yes, the rates of diagnoses were alarmingly high. Yes, the methods of diagnosis were in some ways inadequate. Is that evidence of a scandal?

Absolutely not.

Two facts, which BlueDevilicious and his groupies conveniently overlook, undermine his entire campaign: (1) No one from UNC was doing the diagnosing. (2) The independent psychologist who was doing the majority of the diagnosing throughout the 2000s was none other than Mary Willingham's research partner.

Let that sink in for a moment.

Oh, and while you are, also keep in mind that the assessments leading to those high rates of diagnoses were the basis of the data from which Willingham made her CNN-famous claims about athletes' reading levels.

Therefore, ironically, when BlueDevilicious cites the emails from the independent physicians who criticize the methods Willingham's partner used to diagnose athletes, he is providing further reason to doubt Willingham's claims.

Some history will help clarify the situation even more. Considering the fact that many DI football (FB) and basketball (BB) players come from urban schools, with few resources, and low socioeconomic backgrounds, UNC, like most other DI schools, has new FB and BB players assessed for potential academic difficulties soon after they arrive on campus. Sometime around the early 2000s, UNC began contracting Lyn Johnson, Willingham's eventual research partner, to conduct those assessments. Johnson conducted her assessments completely independently. She did not even communicate with coaches or athletics officials. Academic counselors and learning specialists coordinated the assessments with her. UNC continued to contract Johnson because she and Willingham were conducting their research together, and UNC trusted Willingham at that time.

Not long after I arrived in 2011, I began having concerns about the rates of ADHD/LD diagnoses. One of my research interests is the linguistic patterns of African American youth, and research suggests that some African American youth are diagnosed with learning disabilities when the issue is actually just a matter of linguistic difference. I began advocating to use a new psychologist, one who was more familiar with the challenges of assessing African American students of low socioeconomic status.

Eventually, we stopped contracting Johnson, after we had concerns that she and Willingham had not conducted their research properly. Yet before we switched psychologists, we sought the input of outside experts, the independent physicians BlueDevilicious cites. I want to repeat that: seeking the outside opinions of those physicians was our idea. And they confirmed what I had suspected, that the methods Johnson was using were in some ways inadequate.

Again, no one had questioned Johnson, because she was conducting research with Willingham, and people trusted the two of them at that time.

To be clear, I am not accusing Johnson of intentionally inflating the numbers. I believe she simply wasn't familiar with the latest research on assessing diverse students.

BlueDevilicious also likes to point to some spreadsheets that list student-athletes and have checkboxes for ADHD medication. He believes this is evidence that a high number of athletes were taking ADHD medication. What he doesn't know is that those checkboxes meant only that the psychologist recommended the student-athlete discuss medication with a physician. Those checkboxes did not indicate who was actually taking medication. In fact, most of the athletes I worked with didn't take medication even when the psychologist recommended it. Most of them either tried the medication and didn't like the effect, or they declined to try because they worried about the effect.

Furthermore, the other academic accommodations that students diagnosed with ADHD/LD receive do not provide much advantage without the students doing considerable work. The primary accommodation is a notetaker. But if a student is not active in class and taking their own notes, simply looking over someone else's notes will not suffice to learn the material and pass the exams. Similarly, if a student gets extra time on exams but does not study, that extra time does little good.

In sum: UNC was not behind the high rates of ADHD/LD diagnoses. The person behind those rates was Willingham's research partner, conducting her assessments independently. When UNC learned that the psychologists' methods were, to an extent, inadequate, UNC stopped working with her. Furthermore, the accommodations that students diagnosed with ADHD/LD receive do not provide any unfair advantages anyway.

I've talked with three journalists who were interested in doing a story based on BlueDevilicious's tweets, but, after talking with me, they began to question BlueDevilicious's conspiracy theory. He and his followers are fanatics who make invalid inferences from incomplete information. Their conspiracy theory is ludicrous, and no journalist acting with integrity is going to present it as legitimate.

Again, I am annoyed I even addressed this buffoon BlueDevilcious, and I hope this will be the last time I do. I suggest reasonable people ignore him.

Update: There are two points I forgot to mention that further explain why no one at UNC was especially concerned about the high rates of diagnoses. First, research suggests that students of low socioeconomic status (SES) are diagnosed at a higher rate than the general population is. Therefore, considering that the percentage of FB and BB players who come from low SES backgrounds is significantly higher than among the general student body, we should expect the rates of diagnoses to be higher, too. The second point to consider, which BlueDevilicious conveniently ignores in his "probability" analysis, is that 100% of the FB and BB players were assessed, whereas nowhere near 100% of the general population is assessed. Therefore, again, we should not be surprised that the rates were higher. In other words, we should expect a subpopulation assessed at 100% to have higher rates of diagnoses than the general population. In conclusion, although the rates of diagnoses were high, there were legitimate reasons to expect the rates to be high.