At one elementary school in rural Appalachia, most of the children are white and poor; 90 percent qualify for free or reduced priced lunch. Guess how many of the 800 students are gifted? The answer: three. At least, that’s the determination of a widely-used national intelligence test, on which few students living in poverty score highly.
School administrators wanted to boost the number of gifted students and invited a team of researchers to come up with another way to find them. The researchers asked 16 teachers to rate their students to indicate which ones were far above average in their classrooms, if not the nation, and could benefit from advanced instruction.
When the research team tallied up the teacher ratings for all 282 students in this 2021 experiment, they were startled. Different methods of creaming off the top 10 percent produced entirely different groups of students who would be identified as gifted with almost no overlap. The top 10 percent in each classroom yielded one group of gifted students. The top 10 percent school-wide yielded another. Only six kids were in both groups.
“It was inconsistent from classroom to classroom,” said Karen Rambo-Hernandez, an associate professor of education at Texas A&M University, who presented her unpublished findings at the annual meeting of the American Educational Research Association in April 2022. “Teachers may be making different judgment calls.”
Related: Gifted programs provide little to no academic boost, new study says
Despite the training that teachers received on assessing students by answering a list of 37 questions, some teachers were inclined to rate their students more generously than others. The definition of who is gifted appeared to change as you walked across the hallway.
This experiment is important because many school systems around the country rely on these sorts of teacher checklists, often called “scales” or “instruments” in the field of education, to identify who is gifted. New momentum is building to lean even more on these teacher ratings as school systems wrestle with how to address the underrepresentation of Black and Hispanic students in gifted education. Only 10 percent of the nation’s gifted students were Black, far less than their 15 percent share of the school population in the federal government’s most recent data. The gap is even larger for Hispanic students, who made up only 18 percent of gifted students but over 25 percent of the school population.
In April 2022, New York City permanently eliminated a gifted test for four-year-olds that resulted in a terribly lopsided allocation of only 16 percent of the gifted and talented seats to Black and Hispanic children, who make up 63 percent of the city’s kindergarten population. The city is replacing the test with teacher evaluations of students, which will involve judgments of traits, such as perseverance and curiosity.
Related: Gifted classes may not help talented students move ahead faster
The advantage of teacher ratings is that they can assess important aspects of giftedness that tests cannot measure. A newer HOPE scale of teacher ratings was expressly designed to improve racial equity in the selection of gifted students, and includes questions on social behaviors, such as whether a student shows compassion for others.
The teacher rating scale used in the Appalachian experiment was developed by education psychologist Joseph Renzulli. He theorized that a combination of creativity, motivation and ability indicate a high potential for inventiveness and productivity that can be nurtured even if a student doesn’t score high on an intelligence test.
Among the evaluation questions used at the Appalachian school were how often a student demonstrates “imaginative thinking ability,” “the ability to concentrate intently on a topic for a long period of time,” “curiosity about scientific processes,” and “is eager to solve challenging math problems.” A sense of humor is an important indicator of intelligence on Renzulli scales too.
All of these questions involve subjective judgment calls. To some, five minutes is a long period of concentration. For others, it’s a half hour. Some teachers might see exceptional curiosity when a child asks questions. Others might see questions as normal child behavior.
In the Appalachian school, math and science were emphasized in the questions about each student because the school wants to create a gifted program in computational thinking and computer coding for students. Teachers’ math ratings were more consistent from classroom to classroom, but science marks were much higher in some classrooms than others. Across all 16 classes, teachers tended to think that girls were more creative than boys.
Rambo-Hernandez, the Texas A&M professor who conducted this experiment, fears that teacher ratings of giftedness may ultimately benefit children from wealthier families with more educated parents who tend to be more verbal. Their imagination, curiosity and tenacity may be more visible to a teacher. Quiet students could be overlooked.
The field is in a pickle. Intelligence tests disadvantage children in poverty. Efforts to shift the test score for giftedness school by school, giving poorer schools lower cutoffs, haven’t moved the needle as much as many had hoped and don’t improve racial balance in more integrated schools. Even lotteries for children above a certain threshold will end up advantaging demographic groups that do better on the test. Now this experiment shows that teacher ratings of gifted indicators aren’t a clear solution either.
Related: What research tells us about gifted education
Joni Lakin, an associate professor of educational research at the University of Alabama who has developed tests to identify gifted children, praised the study. “I think we’re too fixated on identification,” Lakin Said. “I’ve lost my faith in fixing gifted’s equity problem by fixing how we identify students.”
Lakin and Rambo-Hernandez both want the field of gifted education to focus more on improving the services offered to gifted children first. They point out that most schools have one sort of gifted program that doesn’t necessarily help many children who are in them.
“Children are diverse in their characteristics,” said Lakin. “Some are creative. Some are rigid but have stick-to-itiveness. If you put them in the same services, they’re not going to be served well.”
The Appalachian research team is going back to the drawing board on how to select which students will get the extra instruction in computational thinking next year. They are considering using other assessments that the school is already giving children. It’s a work in progress.
This story about gifted identification was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for the Hechinger newsletter.