There was a terrific article a few weeks back in the New York Times about the demographic imbalance in the gifted and talented (G&T) programs within the public school system in New York City. The school system serves primarily students of color. The G&T programs serve a disproportionate percentage of white students. The only classrooms in many schools that have a majority of white students are those that host G&T programs.

The piece included a brief and helpful summary of G&T programs:

THE idea of gifted education has drifted in and out of vogue in American schools. It was elevated in the 1950s, when educators and lawmakers pushed gifted programs in math and science amid fears about communism’s rise. It waned in the 1960s but re-emerged with a White House task force on giftedness and the signing of several federal bills in the 1970s that recognized gifted children’s needs.

Urban districts were seen as using the programs to help prevent white flight from the schools, in essence offering a system within the system that was white-majority and focused on achievement. “There have been claims that gifted education resegregates the public schools,” Dr. Borland said.

Reading the article, I wondered if Denver looks any different. It’s not too hard to get a quick read on the data — the Colorado Department of Education lists four categories of G&T: Language Arts, Math, Both Language and Math, and Other (these categories are exclusive). For simplicity, I combined all four into a single category of G&T, and ran the numbers for DPS.

First some quick context. The National Association for Gifted Children estimates that about 6% of the total population is academically gifted. Now note first that the 6% estimate is for a national population, which is significantly different from an urban school district like DPS, where roughly 72% of students qualify for free and reduced meals (FARM) – a basic indication of poverty.

So what did the DPS G&T population look like in 2012 (data from CDE’s Data Lab)? Out of a total population of 43,638 kids in grades 3-10:

• Fully 19.1% meet some sort of G&T classification — this is over three times the 6% national estimate of gifted kids, in a more challenging demographic population.

• By income: 13% of FARM eligible (i.e. low-income) kids are G&T, while 35% of Non-FARM kids are G&T. In other words, more than one in every three kids not in poverty in DPS is classified as G&T.

• By race: Looking at just black, hispanic, and white students (93% of the sample), the G&T classification includes 10% of all black students, 14% of all hispanic students, and 40% of all white students. White students are thus 4 times as likely as black students to be classified as G&T, and almost 3 times as likely as hispanic students to be classified as G&T.

• By both race and income: for FARM students, approximate* G&T percentages are 9% of black kids, 13% of hispanic kids, and 21% of white kids. For kids who do not receive meal assistance, approximate G&T percentages are 15% of black kids, 25% of hispanic kids, and 45% of white kids. Low-income white kids are roughly twice as likely as low-income children of color to be classified as gifted and talented.

This is somewhat rough data, and clearly there is a lot more that should be done in a complete analysis. It would be interesting to look at the kids classified as highly gifted and talented (HGT) compared to simply “regular” G&T — but frankly I think the division into high and low G&T populations is itself compelling evidence of a system that segregates kids from the general population for reasons beyond sheer intellectual promise.

It’s my guess that these numbers are directionally correct, and moreover that aggregating the data across the district probably lessens an even sharper discrepancy at many specific schools. I’ve written before about the segregation in selective admissions schools, but I think the disparity within G&T programs is far more pronounced. With nearly 1 in 5 of all DPS students somehow classified as G&T it seems pretty obvious that other criteria are at play. Unfortunately it probably also means that the roughly six percent of kids who are truly G&T — the true focus of G&T programs — are probably not being served as well as they should be.

But the G&T system is imbalanced. Currently, just short of half of non-FARM white students meet some G&T classification. And almost unbelievably, there is a higher percentage of low-income white kids (21%) who are G&T than the percentage of non low-income black kids (15%), so there is more going on here than just a correlation between poverty and G&T admissions.

The NY Times story notes that the evidence seems to suggest that many G&T programs, whatever their intentions, are now both geared towards and predominantly serving white middle-class families, and that this is an institutional and deeply rooted issue within the public school system in many cities. It’s easy to read this claim about other places, but this story seems as much a part of the Denver landscape as anywhere else.

Many people may ascribe all sorts of malicious tendencies here — it’s not clear to me that this is true. But I do believe that, particularly over time, most benefits accrue to people who have resources to advocate for them, and these tend to reinforce themselves. This trend can become far worse in a large centralized system which is hard to change.

But whatever the origin or rationale, these numbers tell me that there needs to be a far deeper discussion about the purpose and process for G&T programs in Denver. That dialogue will make lots of people uncomfortable — but hopefully less so than a hard look at the mirror in the data above, which is deeply unsettling.

—
Notes:

*At this level of specificity, there are some G&T categories where the number of students is less than 16, and thus not recorded in CDE data, so the percentages may be slightly underrepresented. These categories are: black FARM, white FARM, and black Non-Farm.

I use math N count data; there are occasional differences between the N counts for math, reading and writing, but these are unlikely to make any changes in the percentages.