NPR's Morning Edition produced a five-minute radio piece on the U.S. Airforce Academy's attempt at improving learning results by modifying the ability-grouping of their cadets.

Shankar Vedantam

According to the piece, reported by Shankar Vedantam, based on research by Dartmouth researcher Bruce Sacerdote and colleagues:

  • Weaker students did better when in squadrons with stronger students (but note caveats below).
  • However, when researchers intentionally created squadrons with only the strongest and weakest students (that is, the middle students were removed), the weaker students did worse than they otherwise would have. The researchers argue that this was caused by the splintering of the squadron into groups of strong students and groups of weak students.
  • Middle students did better when they didn't have weaker and stronger students in their squadrons.
  • It appears that the middle students acted as a glue in the mixed-ability squadrons–and specifically, they helped the squadron to avoid splitting into groups.

Of course, one study should not be taken without some skepticism. Indeed, there is a long history of research on academic ability grouping. For example see the review article:

Schofield, J. W. (2010). International evidence on ability grouping with curriculum differentiation and the achievement gap in secondary schools. Teachers College Record, 112(5), 1492-1528.

As Schofield reports:

International research supports the conclusion that having high-ability/high-achieving schoolmates/classmates is associated with increased achievement. It also suggests that ability grouping with curriculum differentiation increases the achievement gap. For example, attending a high-tier school in a tiered system is linked with increased achievement, whereas attending a low-tier school is linked with decreased achievement, controlling for initial achievement. Furthermore, there is a stronger link between students’ social backgrounds and their achievement in educational systems with more curriculum differentiation and in those with earlier placement in differentiated educational programs as compared with others.

But she also warns:

However, numerous methodological issues remain in this research, which suggests both the need for caution in interpreting such relationships and the value of additional research on mechanisms that may account for such relationships.

In addition, social effects are probably not the only effects in play. For example, the research tells us that learners do better when they are presented with information and given instructional supports targeted specifically to their cognitive needs. So for example, this could be why the middle-ability students did better when they were grouped together.

Also interesting is that neither the NPR piece or Shofield's abstract reports specifically on how the mixed groupings affect the stronger learners.

Indeed, other researchers have advocated that gifted students should not be so ignored. See for example the following review article:

Subotnik, R. F., Olszewski-Kubilius, P., & Worrell, F. C. (2012). A proposed direction forward for gifted education based on psychological science. Gifted Child Quarterly, 56(4), 176-188.

Here's what these authors recommend:

In spite of concerns for the future of innovation in the United States, the education research and policy communities have been generally resistant to addressing academic giftedness in research, policy, and practice. The resistance is derived from the assumption that academically gifted children will be successful no matter what educational environment they are placed in, and because their families are believed to be more highly educated and hold above-average access to human capital wealth. These arguments run counter to psychological science indicating the need for all students to be challenged in their schoolwork and that effort and appropriate educational programing, training and support are required to develop a student’s talents and abilities.