The most common method for identifying students with dyslexia involves tests that detect a “discrepancy” between a student’s IQ and their reading performance.
It’s discredited, but still widely used, leaving large numbers of students without help they should be getting.
Up to around 20 percent of the U.S. population has dyslexia, a neurological condition that makes it difficult to decipher and spell written words. Someone with the disability might omit short words such as “and” and “the” while reading aloud, for example, or read “dog” as “god”—even if they speak normally in conversation. The condition impedes a person’s ability to process written information and can negatively impact their career and well-being. Yet only a fraction of affected students get a dyslexia diagnosis or the specialized assistance that can help them manage their difficulty reading.
One reason so many diagnoses are missed is that thousands of schools in the U.S. continue to use an iteration of the discrepancy model to test children for learning disabilities. Moreover, for a multitude of reasons, including biases in IQ tests, a disproportionate number of those diagnosed—and helped—have been white and middle- to upper-class. “It’s unfair, it’s discriminatory, and it disadvantages already economically disadvantaged kids,” says Jack Fletcher, co-founder of the Texas Center for Learning Disabilities in Houston and one of the first scientists to question the discrepancy model’s validity.
Why it matters: Compared to other fields, like medicine, education often struggles to stamp out discredited practices, even when their harms become well-documented.
Literacy and Humility
The movement to bring the science of reading into more U.S. classrooms runs headlong into the inertia of education practice.
In the final analysis, raising reading levels at scale—particularly among low-income children, minorities, and students who come to school speaking no English or who don’t grow up in language-rich homes with ample enrichment opportunities—requires a clear-eyed view of the complex nature of language proficiency. Phonics, while critical, is just the starting line. Reading comprehension is the long game. It requires patience and persistence, which are not in great abundance among either K–12 students or politicians. This makes it difficult to be sanguine that the latest reading reformation will stick this time, or that the bad-practice zombies will remain in their crypts. Columbia University, it must be noted, did not put Lucy Calkins out of business. She has formed a new company, the Mossflower Reading and Writing Project, and taken most of her Teachers College army of coaches and consultants—and her lucrative publishing contracts—with her. Her “Units of Study” remains among the most widely used reading programs in U.S. elementary schools.
If scores on standardized tests of reading comprehension don’t show quick results—and paradoxically, they won’t if schools are getting instruction right—there will be predictable calls to dismiss the science-of-reading movement as just another failed edu-fad. There will be no shortage of reading-war dead-enders eager to say, “Phonics? Oh, we tried that. It didn’t work.”
Why it matters: Top-down efforts to improve education are humblingly difficult. Their almost-inevitable failure leaves leaders disheartened, teachers cynically awaiting the next futile mandate, and smart observers betting early that the latest initiative will land among others in the graveyard of edu-fads.
This dynamic was expertly chronicled by Rick Hess in Spinning Wheels: The Politics of Urban School Reform. It doesn’t mean improvement is impossible. But it does suggest leaders of improvement efforts need to bring humility to the task.
Need to Know
A pandemic hangover isn’t the only factor driving the spike in chronic absenteeism in schools across America. There are deeper cultural shifts in how families approach daily school attendance. Attendance numbers in many places are getting worse, in part because more parents are keeping their kids home amid circumstances like mild illness.
Compared to similar workers in other professions, “teachers take the same, if not less, amount of absences” from the job.
OpenAI’s most advanced model, GPT-4, still pales in comparison to humans at abstract reasoning.
Fidget spinners and bouncy bands appear to hurt, not help, students’ academic performance.
School-based mental health services can reduce the risk of teen suicide.
The Last Word
Billions of dollars of federal aid have been sent to states to ameliorate learning loss, but the latest data show things are getting worse for students, not better.
These are not the result of a fragile American education system, but of one that is so resilient it survived the pandemic enriched. Indeed, it’s the product of a system that’s emboldened and self-interested instead of humbled, contrite, and open to improvement in the pandemic’s wake.