A few days ago, Education Week released a survey of 1,284 teachers gifted education teachers, coordinators, and other education personnel. The full survey can be downloaded here, and it has some interesting tidbits. Here are my reactions:
Who is gifted?
According to educators, an average of 12% of students are identified as gifted. It’s a plausible value, but I would not take this number as gospel truth. A better metric would be calculating this number from state data by dividing the number of formally identified gifted children by total school enrollment. When I did this with data in Texas, I found that the number hovered at around 9% from 1999 to 2013 (Warne & Juntune, 2009; Warne & Price, 2016).
The most common types of giftedness mentioned in legal definitions are intellectual giftedness (79%), academic giftedness (76%), creative/artistic giftedness (48%), and giftedness in specific academic areas (43%). Good. It is the job of schools to education children and build their academic skills. A definition of giftedness should align with this. Other types of giftedness (e.g., athletic, leadership) are much less important for schools.
The most common methods of identifying gifted children are non-IQ tests (79%), nominations/referrals (71%), IQ scores (66%), multiple criteria model (64%), and the vaguely worded “range of approved assessments” (50%).
I do not think that one method of identifying children for gifted programs works for all districts and all programs. Best practice is to align the identification method with the services that will be provided (Peters, Matthews, McBee, & McCoach, 2014). For example, if a writing program will require high performance in language arts and writing, then the identification process should be based on high performance on those skills.
But nominations are almost always a bad idea. They introduce subjectivity into the system, and add an extra–often unnecessary–barrier–for identification. Nominations make it harder for kids without an adult advocate (who often come from low-income families or diverse backgrounds) to be labeled as gifted. Nominations also greatly increase false negatives in the gifted identification process. Screening every child is best practice (McBee, Peters, & Miller, 2016). With 72% of respondents saying that they screen for giftedness after a teacher referral and 67% after a parent referral, it is clear that most school districts are ignoring this advice.
As for specific tests, the Cognitive Abilities Test was most popular (54%), with the Wechsler Intelligence Scale for Children (40%), and the Naglieri Nonverbal Ability Test (34%) rounding out the top 3. Again, I usually care less about the specific test and more about whether the test aligns with the program services. However, there is one test that I found very unsettling, as did a colleague:
The Stanford-Binet L-M was created in 1972 from items cannibalized from the 1937 Stanford-Binet. It is incredibly outdated and has no place in modern assessment. One prominent person in gifted education advocated for using the L-M in the 1990s (years after it had been replaced by the Stanford-Binet IV), and I suspect that the lingering online pieces she wrote at the time are prolonging this the L-M’s life.
Another disappointment is the 10% of respondents who said that a district-created assessment is used to identify children for gifted services. I am highly skeptical that these tests are as technically sophisticated or developed with the same ethical standards that professional tests are created with.
There has long been a tension in gifted education about the underrepresentation of Black and Hispanic students in gifted programs and the overrepresentation of White and Asian students. Therefore, I’m not surprised when most respondents stated that Black (61%), Hispanic (57%), and Native American (53%) students were underrepresented in their programs and Asian (24%) and White (46%) students were overrepresented. Similarly unsurprising results are found when asking about family income: 58% say that students who live in poverty are underrepresented in gifted programs, and 51% say that students from high-income families are overrepresented. Students with disabilities (43%) and English language learners (63%) are also underrepresented.
This is despite the fact that 88% of respondents who say that at least one group is underrepresented in their program are making an effort to correct it. Yet, the underrepresentation still happens. Why? Because the cause of under- or overrepresentation in gifted programs is not confined to gifted education. It is a symptom of the larger achievement gap:
So, the tactics that survey respondents say they use to correct underrepresentation are ineffective or partially effective. The most popular ones are adopting a new or additional assessment (30%), universal screening (23% Yay! Helpful, but doesn’t fully fix the problem), combining multiple criteria (20%, does not help with diversity), targeting disadvantaged populations in the screening process (20%), and training teachers and staff (17%). None of these fix the achievement gap, so none will fix the problem of underrepresentation of some groups in gifted programs. This seems to be one of the hardest lessons for gifted education to learn.
Characteristics of Gifted Programs
Late elementary and middle school grades are the most common times for gifted programs to serve children. Over 90% of respondents say that their programs serve Grades 3-6, and 88% say they serve Grades 7-8. I am not surprised, though I wish the percentage were higher in first (66%) and second (78%) grade. The low-ish percentages for high school (66-69%) don’t bother me much because there are often other ways of serving advanced learners that don’t require a full blown gifted program (e.g., Advanced Placement, online high schools, dual enrollment).
What is more disappointing is the services provided. Pull-out programs (86%) are the most common. This is when a child is removed from the classroom for a short period each week (usually a total of 1-5 hours) for gifted services. It’s popular, but it’s a stupid gifted service because it tries to fix a problem that exists all day, every day (the mismatch between the child and the typical curriculum) with a solution that is a few hours per week. Pull-out enrichment programs should be the last resort of gifted services and only implemented when other services are not financially or politically feasible.
Other programs are inconsistent in quality: the most common options are honors/advanced work (82% 👍), regular classroom services (52% 👎), dual enrollment (47% 👍), cluster grouping within a classroom (39% 👎), and self-contained classrooms for gifted children (32% 👍).
The survey is not surprising to a gifted education scholar like myself. The results indicate a mix of practices supported by evidence and others that are not. It is nice to have some up-to-date numbers, but it is important to remember that these are the perceptions of the respondents. How well these align with reality is not clear. Still, they are knowledgeable informants, and I think that the survey is an excellent contribution to the conversation surrounding gifted education.
McBee, M. T., Peters, S. J., & Miller, E. M. (2016). The impact of the nomination stage on gifted program identification: A comprehensive psychometric analysis. Gifted Child Quarterly, 60, 258-278. doi: 10.1177/0016986216656256
Peters, S. J., Matthews, M. S., McBee, M. T., & McCoach, D. B. (2014). Beyond gifted education: Designing and implementing advanced academic programs. Waco, TX: Prufrock Press.
Warne, R. T., & Juntune, J. (2009). Recent trends in gifted identification in Texas. Tempo, 29(4), 22-26. (link to article)
Warne, R. T., & Price, C. J. (2016). A single case study of the impact of policy changes on gifted programs. Journal for the Education of the Gifted, 39, 49-61. doi:10.1177/0162353215624159