States and localities have discretion over gifted programs, but surprisingly little large-scale research compares the education environments of students in gifted programs to high-achieving, non-gifted students or investigates how these learning environments vary across districts. In this guest blog, Ben Backes, James Cowan, and Dan Goldhaber discuss their IES-funded exploration study, where they use administrative and survey data to describe the relationship between gifted participation and access to educational resources across nearly 300 school districts in Washington State.
Gifted Access and Participation in Washington
The underrepresentation of low-income and minority students in gifted programs has attracted attention because identification procedures often include nomination or referral processes requiring subjective evaluation of student ability. Nationally, low-income and non-White students are significantly less likely to participate in gifted programs. To better understand who is in these gifted programs in Washington State, we are investigating participation in gifted programs by student race/ethnicity and socioeconomic status in grades 4–12. Consistent with prior studies, relative to White students, we observe Asian students being more likely to be found in gifted programs, while Black, Hispanic, and free and reduced-price lunch students are less likely to receive gifted services. Washington districts frequently use universal screening policies, and the Black-White and Hispanic-White gifted gaps disappear once statistical adjustments for prior test scores are used. We find little association between use of modifications for underrepresented minorities or low-income students—as reported by district coordinators—and gifted participation.
In sum, we find consistent evidence of disparities in access to gifted programs conditional on student achievement in Washington for low-income students, but less consistent evidence of disparities by student race/ethnicity. However, we only observe data on student academic aptitude beginning in third grade, and many classification decisions are made before this time. There may be disparities in initial gifted classification decisions for younger students.
Unsurprisingly, participation in gifted programs does affect student learning environments. Gifted students are much more likely to sit in classrooms with other high-achieving students and in more homogenous classrooms. These differences persist even after limiting the sample to high achievers. These patterns are most pronounced in elementary school. Gifted students are taught by more qualified teachers in elementary and middle school, as measured by experience, licensure test scores, and educational attainment. However, these effects are very small.
Differences Across Districts and Program Types
We find that although gifted students do tend to take more advanced courses with higher-achieving peers, there is considerable variation in the design of gifted programming across school districts.
- Although school districts tend to assign gifted students to more advanced academic tracks, we find that these effects are mostly concentrated in large urban and suburban districts. The estimated gifted effects on access to more advanced courses are typically much smaller in the western and eastern school districts in smaller cities and rural areas of the state.
- Larger, higher income districts in cities and suburbs operate gifted programs that provide more significant changes in learning environments. Students in these programs are more likely to share classrooms with other gifted students and with high-achieving students, and—in the case of large districts—sit in smaller classrooms with more qualified teachers.
- The structure of gifted programming also influences the type of instructional approaches districts employ. Self-contained gifted programs—where students are assigned to specialized classrooms for most of their instruction—report using a broad array of acceleration strategies. However, about one third of gifted students participate in programs offered through services in regular classrooms, where independent study, supplemental instruction, and flexible ability grouping appear to be important strategies.
- Well under half of districts have established gifted curricula for math or ELA. About 20% of gifted students are districts that report having a districtwide math curriculum and 25% are in districts that report having districtwide ELA curriculum. This finding is consistent with another study that surveyed districts in three states.
What's Next?
There is a growing body of empirical literature providing causal estimates of the effect of gifted participation on student achievement which generally uses administrative data from a single school district. The results from this study of gifted programs across an entire state suggest that district-specific gifted programming effects are likely to vary substantially as the nature of the programs vary substantially across districts. This implies both that we should be cautious about generalizing based on district-level studies and that the variation in findings across studies may be indicative of true variation in program effectiveness. In the next stage of this project, we plan to investigate the extent to which this heterogeneity generates differences in the relationship between gifted participation and student achievement.
Ben Backes is a Senior Economist with CALDER at the American Institutes for Research.
James Cowan is a Senior Researcher with CALDER at the American Institutes for Research.
Dan Goldhaber is the Director of CALDER at the American Institutes for Research and CEDR at the University of Washington.
