Featured Resources
October 2022
Inequity in Access:
An Examination of the Test-Based Admissions System
To the City’s Gifted & Talented Program
Summary
For years, entry into the city’s public gifted and talented program was based on scores on an admissions test—largely taken by 4-year-olds. The test-based system had been widely criticized for contributing to segregated classes; White and Asian students have been heavily overrepresented in the gifted program. Last spring, Mayor Adams expanded the program and announced changes to its admissions process. Eligibility is now based on universal screening by pre-K teachers for kindergarten and on grades for older-elementary students. Like the process in the test-based system, eligible students are then invited to apply to the program at various schools with seats awarded by lottery.
In this brief, IBO examines the demographic differences at each step of the admissions process for students entering kindergarten in the 2018-2019 school year, when the test was still in place. We found demographic disparities at nearly every part of the process. Among our findings:
While eligibility for the program no longer begins with a test, this research provides insights into the admissions process as whole, as well as a baseline for comparing outcomes under the new system.
History of Admission to the City’s Gifted & Talented Program
In April 2022, the Adams administration announced changes to the Department of Education’s (DOE) gifted and talented program, which offers specialized instruction to students in kindergarten through fifth grade. The city increased the number of seats in the program and made changes to its admissions process, which for about a dozen years, was based solely on standardized exams. The test-based admissions system had been widely criticized for leading to a gifted program that overrepresents White, Asian, and non-low income students. Although the gifted and talented program has undergone several changes over time, it has been a fixture of academic enrichment in New York City public education for nearly 50 years.
The test-based admissions system to the city’s gifted and talented program was first introduced by the Bloomberg administration, and lasted from the 2008-2009 school year through the 2020-2021 school year. During the pandemic, the admissions test was first suspended temporarily and then the de Blasio administration announced plans to phase the test out after 2021-2022 school year. However, the Panel for Educational Policy, a DOE oversight group, rejected the the city’s contract to administer the test for that year, and admissions for 2021-2022 were based on teacher recommendations.
During the last months of his administration in 2021, then-Mayor de Blasio announced a complete system wide phase-out of the gifted and talented program. His move was aligned with a 2019 report published by the City’s School Diversity Advisory Group.1 The report included recommendations for improving diversity in the city’s schools by modifying or eliminating the use of the city’s academic screening tests. It called for phasing out the gifted and talented program and instead recommended that the city offer school-wide enrichment programs in all schools in an effort to expand opportunities for accelerated learning to more students.
The Adams administration’s changes reversed this phase out, although it did not reinstate use of the standardized test. Admissions now are based on either teacher recommendations (for kindergarten admissions) or grades in core subjects (for admissions in all other grades). According to the DOE, using grades in the four core subject areas allows for a more well-rounded assessment of students’ skills as opposed to relying on any single subject or test, and that situating the screen at the school level can foster a program composition that is more representative of the district’s population. Families are considered for placement at all of their application choices and offers made based on district and sibling priorities, as well as seat availability.
The recent policy change instituted by the Adams administration more closely resembles the pre-2008 policy. Before 2008, the city’s 32 school districts each created their own criteria for admission to their gifted and talented program which included evaluations from teachers and classroom observations, in addition to partly relying on tests, and by comparing the results of students from within a district.2 The number of children entering gifted programs immediately dropped by half under the testing admissions policy introduced in 2008.3
While admissions based on teacher recommendations yielded larger and more diverse gifted and talented kindergarten classes during 2021-2022, in other contexts, researchers have expressed concerns about racial biases inherent in relying solely on teacher recommendations for identifying gifted students in the selection process.4,5
Research Questions
In this brief IBO examines admission into the city’s gifted and talented program in the 2018-2019 school year when the exam was still in place. Our study focuses on the students who took the screening test in pre-K in 2017-2018 for entry into kindergarten in the 2018-2019 school year. We consider students’ gender, low-income status, disability status, self-reported race/ethnicity, home language, and neighborhood socio-economic status to understand the degree to which access to the program varies.6
Our research questions include:
It is important to note that our sample is limited to those students who were in the public school system in pre-K in 2017-2018. When looking at the kindergarten class, the brief also includes any students from non-public schools who joined the public school system for kindergarten. We do not have demographic information on students from private schools who might apply to the city’s gifted program, but ultimately do not enroll in a public school.
Gifted & Talented Program Changes Made by the Adams Administration Under the changes made by the Adams administration there are two types of gifted and talented programs—K to 5th grade programs and a new 3rd to 5th grade program—with separate admissions criteria for each. For the K to 5th grade programs, admission in kindergarten is based on universal screening of all students currently attending a public preschool, including district schools, charters and Community Based Organizations (CBOs). These students are evaluated by their preschool teachers who then nominate eligible students for gifted and talented admissions.7 Admissions to K to 5th grade programs in grades 1, 2, and 3 (as space allows) are now based on student grades. For students in DOE public schools, their core course grades need to be in the top 10 percent at their school, while for students in private, parochial or charter schools, the DOE determines eligibility based on submitted school grades. For the newly-established 3rd to 5th grade gifted and talented programs, of which there is at least one in each community school district, eligibility is based on grades in the four core subject areas of science, English Language Arts, math, and social studies, with 2nd grade students in the top 10 percent at their school invited to apply. The city has also added 100 seats beginning this year in kindergarten for the K to 5th grade gifted and talented program and 1,000 seats beginning this year in 3rd grade for 3rd to 5th grade gifted and talented programs, based on research that showed that identifying gifted behavior in later grades may provide a more accurate assessment of students’ ability.8 |
Overview and Eligibility Under the Testing Admissions System
In October of the year before kindergarten, parents registered online with the DOE to select a location and time for their child to take the gifted and talented test that would determine their eligibility to apply to the program. In January or February, the student took the test and in April, the child’s results were mailed to parents. Eligible students could then tour schools and apply to program(s) online in April. Final placement letters were sent to parents in May and June. Curriculum design for each gifted and talented program was and still is at the discretion of the school; the only requirements are that they must meet Common Core learning standards for math and English Language Arts and meet state learning standards for other subjects. Some schools may offer certain classes not found in others.
There are two types of gifted programs in New York City: district programs and citywide programs. Under the testing admissions system, students who scored at the 90th percentile or above on the gifted and talented test were eligible for a district program, and those that scored at least at the 97th percentile were eligible for a citywide program. All but four of the city’s public school districts (Districts 7 and 12 in the Bronx and Districts 16 and 23 in Brooklyn) have at least one district kindergarten program.9 In general, district gifted programs are classrooms contained within neighborhood public schools that include majority general education classes. Conversely, all classrooms in schools housing citywide programs are reserved for gifted program students.
Once eligibility was determined, students then applied to programs. Students who were eligible for the citywide programs could apply to any of the city’s five citywide schools, which had a total of 300 seats for kindergarteners in 2018-2019, or to any of the district programs, which had almost 1,600 seats. The citywide programs accepted students from all boroughs and there was no geographic priority. There were three citywide programs with 200 offered seats in Manhattan; one program in Brooklyn with 50 offered seats; and one in Queens with 50 offered seats. Students scoring at 97th percentile or above applying to a program where a sibling already enrolled were placed first. After all eligible siblings were placed, non-sibling applicants were placed by lottery based on their overall score.
For the district programs, eligible students could apply to any program in any district across the city, but applicants who lived in the same district as the school received first priority, beginning with those who had a sibling at the school. Families with siblings applying at the same time needed to submit a separate application for each child, with each sibling treated as an individual applicant. However, twins and other multiples had unique priority: they were placed together in the same program if each child was eligible for it. After all eligible siblings were placed, non-sibling applicants were placed by their overall score.
Each eligible applicant had priority for one or more district programs based on the district where they were zoned for elementary school. However, in some cases, such as when a district did not offer a gifted program, applicants could have a priority for one or more program options in neighboring districts. Students zoned for or currently attending a school with a gifted program did not have any additional priority to that school over other students in the district.
While Brooklyn and Queens had the largest shares of district gifted and talented program seats, roughly equal to their share of all kindergarteners, Manhattan had a disproportionately higher share of seats, and the Bronx had a disproportionately lower share of seats. The share of Manhattan gifted and talented students (24.3 percent) was double their respective share of all kindergarteners in the 2018-2019 school year. The share of Bronx gifted program students (5.0 percent) was one-fourth of their respective share of all kindergarteners. These borough-level differences are reflective of differences in the student populations that opt to take the test, become eligible, apply and ultimately accept offers.
Demographic Profile of Pre-K Class of 2017-2018
There were 73,700 students attending public pre-K in New York City in the 2017-2018 school year, constituting the pool of potential public school applicants to the city’s gifted programs. This includes students enrolled in DOE programs offered at Community Based Organizations (CBOs). In addition to public pre-K, students from private schools could also take the exam and, if eligible, apply for a spot. In 2017-2018, a total of 14,500 students took the gifted and talented test of which 10,800 (almost three quarters) were enrolled in a public preschool and 3,800 (about a quarter) were either in a private preschool or were not attending any preschool.
Looking at student demographic groups among the public pre-K population, there was a near even split of female and male students. In terms of student race/ethnicity, Hispanic students represented the largest share of students, 37.5 percent. Black and White students each represented approximately one-fifth of students, 22.1 and 20.3 percent, respectively, while Asian students accounted for 16.9 percent. Multiracial and Native American students accounted for the smallest shares of the pre-K class at 2.0 percent and 1.2 percent, respectively. Students with a disability accounted for just over 10 percent of the population. Low-income students accounted for nearly two-thirds (63.2 percent). Students who do not speak English at home represented just over one-third.
When examining students within each race who did not speak English at home, a majority of Asian students (61.2 percent) did not speak English at home, compared with 46.9 percent of Hispanic students. About 20 percent of White students and just 5.8 percent of Black students did not speak English at home.
IBO categorized the census tracts (which we refer to as neighborhoods) in which students reside using three dimensions—income level, violent crime rate, and educational attainment. We categorized a neighborhood as low-income if the census tract had a median household income below the Mayor’s Office for Economic Opportunity’s (formerly known as the Center for Economic Opportunity) threshold of $33,562 in 2017.10 We defined a neighborhood as having a high violent crime rate if it fell in the top quartile of the distribution of violent crime rates, or 3.5 violent felonies or more per 1,000 residents in 2017-2018—a figure we calculate based on the number of crimes that occurred from 2013 through 2017 as recorded by the New York Police Department.11 Finally, we defined a neighborhood as one with low educational attainment if it fell in the bottom quartile of educational attainment (adults attaining a bachelor’s degree or higher) based on census data among New York City’s census tracts: 20.2 percent or less of adult residents with a college or higher degree in 2017-2018.12
In terms of students’ neighborhood socioeconomic status, 21.9 percent came from low-income neighborhoods, 28.1 percent came from neighborhoods with high rates of violent crime, and close to one-third of students (30.7 percent) came from neighborhoods with low levels of educational attainment.
When examining students’ neighborhood socioeconomic status within race, approximately 31.0 percent of Black students and Hispanic students came from low-income neighborhoods, compared with 9.2 and 6.4 percent of Asian and White students, respectively. Similarly, larger shares of Black and Hispanic students, 43.1 and 33.0 percent, came from neighborhoods with high crime rates, compared with 15.6 and 14.0 percent of White and Asian students. There was a relatively more even distribution of students from neighborhoods with low rates of educational attainment across three of the four major racial groups: 39.8 percent of Hispanic students, 35.6 percent of Black students, and 27.6 percent of Asian students. Only 13.1 percent of White students came from neighborhoods with low educational attainment.
Who Took the Gifted & Talented Test?
IBO compared the racial composition of pre-K test-takers with the racial composition of the entire public pre-K population. Overall, we found that disproportionately larger shares of Asian, Multiracial, and White students took the test and disproportionately smaller shares of Black and particularly Hispanic students took the test. Among the student demographic and neighborhood factors we considered, we found that students who speak English at home were more likely to take the test, while low-income students, and those who live in boroughs other than Manhattan were less likely to take the test.
Descriptive Statistics. The public pre-K sample totaled 73,700 public school students, of which 10,800 (14.6 percent) took the gifted and talented test—this excludes students who took the test but did not attend public pre-K. Manhattan had a disproportionately large share of test-takers (17.6 percent) compared with the borough’s share of public pre-K students (11.9 percent). Conversely, the Bronx accounted for 10.5 percent of test-takers despite accounting for 20.6 percent of public pre-K students. The districts that had a greater share of test takers than their relative proportion of the pre-K population were Districts 1, 2, 3 (Manhattan), 13, 15, 20, 22 (Brooklyn), and 25, 26, 28, and 30 (Queens).
Among those who took the gifted and talented test, the share of Asian students, 30.7 percent, was 1.8 times their share of the pre-K population, 16.9 percent. Similarly, Multiracial students were more than twice as likely to take the test compared with their share in the population. White test takers represented 1.5 times their share of the population. By contrast, only 17.8 percent of gifted and talented test takers were Hispanic, while they made up 37.5 percent of the public pre-K population. The discrepancy among Black students was smaller; they represented 15.0 percent of test takers, but comprised 22.1 percent of the pre-K population. The share of test takers that were Native American, 0.9 percent, was also smaller than their share of the public preschool population, 1.2 percent.
Regression Analysis: Who Is Likely to Take the Gifted & Talented Test? In addition to student race, IBO investigated what other factors were associated with students’ likelihood of taking the gifted and talented test. We used a multivariate logistic regression framework to analyze how students’ demographics, neighborhood characteristics, and borough of residence were associated with test taking among all public pre-K students. Student demographics included gender, race, low-income status, disability status, and whether students spoke English at home. Neighborhood characteristics included whether the student came from a low-income census tract, whether the student came from a census tract with low educational attainment, and whether the tract had a high rate of violent crime. Our analysis follows the literature in using a logistic regression because our outcome variables are binary—meaning they have discrete values of either 0 (the student did not take the test) or 1 (the student did take the test).
Male students had a slightly lower propensity to take the test; they were 92 percent as likely compared with female students. (An odds ratio of less than 1 means that the associated variable makes the outcome less likely, and vice versa—for example, an odds ratio of 0.92 for male students makes them only 92 percent as likely as female students to take the gifted and talented test.) Asian students were much more likely to take the test—nearly twice as likely as White students—as were students who speak English at home, who were 1.6 times as likely compared with students who do not speak English at home. On the other hand, Black and Hispanic students took the test at significantly lower rates compared with White students: they were only 58 percent and 45 percent as likely, respectively. Students from less affluent families were only 36 percent as likely to take the test compared with their relatively more affluent peers.
Students with a disability had a lower rate of participation in the gifted and talented test compared with students without a disability (55 percent as likely to take the test). Similarly, students who live in boroughs other than Manhattan took the test at a rate just over half of that of students residing in Manhattan. Students from neighborhoods with low levels of educational attainment and those whose neighborhoods are low income were 88 percent and 71 percent as likely to take the test, respectively, compared with students from neighborhoods with higher educational attainment and neighborhoods that are not low income.
Who Was Eligible for Gifted & Talented Programs?
Among the 10,800 public pre-K students who took the gifted and talented test, approximately 2,400 students (22.6 percent) scored high enough to be eligible for the program. IBO found that 9.2 percent of test takers (1,000 students) were eligible for a citywide program and 13.5 percent of test takers (1,400) were eligible for a district program.
Looking at eligibility by student race, IBO found that there were stark differences: Multiracial, White, and Asian test-takers were more likely to be eligible for both the citywide and district programs compared with Black, Hispanic, and Native American test-takers. In terms of other demographics, low-income students and students from boroughs other than Manhattan were also less likely to be eligible for both programs. We also found that among Asian and Hispanic students, those who speak English at home were more often eligible for both citywide and district programs, while among White, Black and particularly Multiracial students, those who do not speak English at home were more likely to be eligible for the citywide program.
Descriptive Statistics. The shares of Multiracial and White test-takers who were eligible for a citywide program, 15.6 percent and 12.5 percent, respectively, were three to four times as high as those of Black and Hispanic test-takers—the pattern was similar with respect to eligibility for district programs. The shares of Asian test-takers eligible for either citywide or district programs were also much higher than those of their Black and Hispanic peers. Overall, Multiracial test-takers had the highest rate of gifted and talented eligibility, 36.4 percent, followed by White and Asian test-takers. For Black test-takers the rate was 10.3 percent and was only marginally higher for Hispanic test-takers (11.5 percent).
Regression Analysis: Who Is Likely to Be Eligible for the Gifted and Talented Program? We again used a regression framework to understand the factors (student demographics, neighborhood characteristics and borough) that predict eligibility among students who took the gifted and talented test—specifically whether the student was eligible for a gifted and talented program (scoring at or above the 90th percentile), and whether they were eligible for a citywide gifted and talented program (scoring at or above the 97th percentile).
Compared with White students, Asian students were not any more or less likely to be eligible for gifted programs —this is also true when looking at eligibility for citywide gifted programs. But Black and Hispanic students were significantly less likely to be eligible for gifted and talented programs, being 37 percent and 39 percent as likely as White students, respectively. These patterns also hold true when looking at eligibility for citywide programs. Multiracial test-takers were more likely (1.4 times as likely) to be eligible for gifted programs compared with White students. There was no difference for citywide eligibility.
Although we previously found that male students were less likely than female students to take the test, we also found that when they did take the test, male test-takers were more likely than female test-takers to be eligible for gifted and talented programs. There was no such difference for the eligibility for the citywide program.
Low-income students had lower rates of eligibility for gifted and talented programs, both overall and citywide, compared with peers from higher-income families. Students with disabilities were about 70 percent as likely to be eligible compared with students without disabilities. Students who speak English at home were not significantly more likely to be eligible for gifted and talented programs, despite the fact that we found that they were more likely to take the test. This was true for eligibility for citywide gifted programs too.
In terms of where students live, IBO found that students who live in the boroughs other than Manhattan significantly lagged behind in terms of eligibility; they were 68 percent as likely to be eligible for any gifted programs and 62 percent as likely for citywide programs. Residing in a low-income neighborhood is associated with a lower rate of overall eligibility, but is not significantly correlated with eligibility for citywide programs. Neither of our neighborhood measures on resident education level nor violent crime rate had a significant affect on gifted eligibility (at either the overall level or for citywide programs).
Regression Analysis: Who Was Likely to Take the Test and Become Eligible Among Students Who Speak English at Home? Because we found that students who speak English at home took the gifted and talented test at a significantly higher rate, but did not have a similar advantage in eligibility for programs at any level, we explored this in more detail. We interacted the student race variable with whether the student speaks English at home in our regression model. This exercise highlights differences across students who speak English at home and those who do not among the various racial groups, with respect to students’ likelihood to take the test and receive a score high enough to be eligible for a placement.
In terms of who is more likely to take the test, we found Hispanic students who speak English at home were more than twice as likely (2.4 times as likely) to take the test compared with their non-English speaking counterparts. The same is true for Black students, although the strength of the association was weaker: 1.6 times as likely. White students who speak English at home were more likely (1.2 times) to take the test compared with their non-English speaking counterparts. For test-takers, there was no statistically significant difference among Asian students in terms of speaking English at home.
As for eligibility, Hispanic and Asian students who speak English at home were eligible for both district and citywide programs at higher rates than their same-race peers who do not speak English at home. For district programs, Hispanic students from homes where English was spoken were 2.3 times as likely, and Asian students from such homes were 1.5 times as likely to be eligible compared with their counterparts who did not speak English at home. The differences were larger with respect to citywide programs, with English-speaking Asian and Hispanic students being almost three times as likely to be eligible compared with their same-race peers who do not speak English at home.
What this also demonstrates is that the similar likelihood of gifted and talented eligibility for Asian students compared with White students is driven by the increased likelihood of eligibility of those Asian students who speak English at home, compared with Asian students who do not speak English at home. In the full pre-school sample, more than 60 percent of Asian students do not speak English at home.
Who Applied for Gifted & Talented Programs?
Next, IBO looked at the distribution of students, who after being eligible, actually applied for a spot in a gifted and talented program. Among eligible students, 85.5 percent (2,100 students) applied for programs—the share of citywide eligible students who applied (93.9 percent) was higher than the share of district eligible students who applied (79.8 percent). Looking across racial groups, the share of eligible students who applied to a citywide program ranged from 86.1 percent for Hispanic students to 97.7 percent for Asian students. These differences across races were smaller among students who were eligible for district programs, the exception being Asian students who applied at a higher rate of 87.2 percent relative to the other racial/ethnic groups, which ranged from 77.7 percent for Black students to 75.0 percent for White students. While White students eligible for citywide programs applied at a high rate of 93.5 percent, White students eligible for district programs had the lowest rate of application among all racial groups.
Who Received Offers: District and Citywide Programs
Offers for gifted and talented programs were given based on students’ percentile ranking on the test and the choices that applicants made, as well as sibling priority and geographic priority (for the district programs). Beginning with the highest-scoring students (those in the 99th percentile), students were randomly assigned a lottery number and offers were assigned sequentially to each student based on their choices. If an offer was not available in any of the programs a student chose, then that student did not receive an offer. The process continued sequentially among students within the same percentile ranking on the test (for the 98th percentile, 97th percentile, and so on). The number of offers was typically more than the number of seats as the DOE assumed that a share of offers would not be accepted.
Among eligible students who applied for the 2018-2019 school year, about 2,300 students received offers to enroll in one of the city’s gifted and talented programs, with almost 70 percent of offers (1,600 students) going to students who attended public pre-K the prior year. It is important to note that the 700 students who received offers and did not attend public pre-K the prior year includes two groups of students. About 300 students enrolled in the public school system the following year but roughly 400 students did not. While 70.4 percent of offers to district programs went to students who attended public pre-K the year before, public pre-K students accounted for a disproportionately smaller share (61.3 percent) of offers to citywide programs.
Applicants who were eligible for a citywide program received offers at a higher rate compared to applicants who were only eligible for a district program, as the former can apply for both programs. However, the vast majority of the students eligible for a citywide program received an offer to a district program. The majority of applicants eligible for the citywide program (86.0 percent) received an offer, with 20.1 percent (200 students) receiving a citywide offer and the remaining 65.9 percent (about 600 students) receiving a district offer, a reflection of the limited number of seats available in citywide programs for eligible students. Among applicants who were eligible for a district program, 69.5 percent of students (800 students) received an offer.
Looking at seat offers among applicants by race, offer rates for citywide programs were relatively even. There was a similar pattern by race among citywide-eligible applicants who received a district offer, ranging from 25.6 percent of Hispanic students to 32.4 percent of Multiracial students. The starkest differences in offer rates across racial groups were among students who were district eligible: more than half of Black students, 53.3 percent, received a district offer, compared with only 35.2 percent of White students. White students were also most likely not to receive any offer: over a quarter of White applicants did not receive an offer, compared with only 8 percent of Black applicants, a function of the number of seats available in the districts to which students applied.
Who Accepted Offers?
The overwhelming majority of students, 91.3 percent, who were offered a citywide program accepted, and 63.7 percent of those offered a district program accepted. Overall, over two-thirds of students, 66.9 percent, who received offers accepted. Asian and Multiracial students were most likely to accept an offer (70.2 percent each), while Black and Hispanic students were least likely (61.9 percent and 63.0 percent). When looking at the type of program offered, among students who received an offer to a citywide program, Asian and White students were most likely to accept an offer (97.1 percent and 89.5 percent) while Black and Multiracial students were least likely (80.0percent and 84.6 percent). Among students who received an offer to a district program, Multiracial students were most likely to accept an offer (68.3 percent) while Black and Hispanic students were least likely (60.3 percent each).
The Resulting Demographics of 2018-2019 Gifted & Talented Kindergarten Class
IBO examined the demographic composition of the kindergarten gifted and talented program in 2018-2019 to see how the disparities within the admissions process impacted the overall degree of representativeness within the program. Overall, there were about 1,900 kindergarteners enrolled in gifted and talented programs, 2.4 percent of the 78,600 kindergarteners in 2018-2019 in traditional public schools (districts 1-32 and 75, the citywide special education programs) and charter schools.
The program was heavily overrepresented by Asian students and White students compared with Black and Hispanic students, and their shares were disproportionate to those in the kindergarten population. White and Asian students in the gifted programs were more than double their share of kindergarteners, while Hispanic and Black students in these programs account for one-third and one-quarter, respectively, of their share of the kindergarten population. As our analysis demonstrates, this demographic breakdown is largely the result of fewer Black and Hispanic pre-K students taking the gifted and talented test, fewer being found eligible, and fewer accepting offers among those who received one.
Conclusions
IBO found that 15 percent of 73,700 public school pre-K students took the gifted and talented test in the 2017-2018 school year. Black students and particularly Hispanic students took the test at lower rates compared with Asian and White students, a crucial factor in the mismatch of the makeup in the gifted and talented program compared with the rest of the public school system. Students with a low family income and students with disabilities had significantly lower rates of taking the test. Students’ home neighborhood rates of educational attainment, income, and violent crime were also a significant factor in terms of test taking.
Less than one-quarter of test-takers were eligible for a program. Coming from a low-income family was associated with a lower chance of being eligible for the program. Compared with White students, Black and Hispanic students were less than half as likely to score high enough to be eligible for both of the citywide and district programs, compounding the racial mismatch seen at the test-taking level. We also found that students who speak English at home had a higher likelihood of taking the test and were more likely to be eligible for the program within their racial group, particularly among Asian and Hispanic students.
The overwhelming majority of eligible students applied for programs. More than three-quarters of those eligible received offers, the vast majority of which were for district programs.
In terms of citywide offers by race among eligible students, there were no significant racial gaps, though there were starker racial gaps in district offers. Among district eligible students, White students had the lowest offer rate at 35.2 percent, compared with 53.3 percent of eligible Black students. Asian, White and Multiracial students accepted offers at the highest rate.
IBO found that racial disparities are compounded across multiple levels of access to the gifted and talented program: fewer Black and Hispanic students take the test and are eligible, which result in fewer Black and Hispanic students receiving and accepting offers. When they get offers of a gifted and talented kindergarten seat, Black and Hispanic students also accept these offers at a lower rate than Asian and White students. Additionally, differences within racial groups in terms of speaking English at home factors into both the likelihood of taking the test and the likelihood of being eligible for the program. Should testing become a part of the eligibility determination again in the future, this study invites future research on students’ access to testing and academic enrichment programs more broadly across the city.
Prepared by Stephanie Kranes, Joydeep Roy,
Vanessa Witenko, Jeannie Kim
New York City Independent Budget Office
Endnotes
1The Advisory group was comprised of education researchers, advocates, and city employees at the Department of Education (DOE)—both in schools and in the central administration.
2Baker, Al. “In One School, Students Are Divided by Gifted Label — and Race.” The New York Times 12 Jan. 2013. NYTimes.com. Web. 1 May 2014.
3Gootman, Elissa, and Robert Gebeloff. “Fewer Children Entering Gifted Programs.” The New York Times 30 Oct. 2008. NYTimes.com. Web. 2 May 2014
4Elsen-Rooney, Michael. “Black and Hispanic Admissions to ‘Gifted and Talented’ Program Double After NYC Scraps Entrance Exam.” New York Daily News. Web. 14 April 2022.
5Grissom, Jason A. and Christopher Redding (2016). Discretion and Disproportionality: Explaining the Underrepresentation of High-Achieving Students of Color in Gifted Programs, AERA Open January-March 2016, Vol. 2, No. 1, pp. 1–25.
6Race is self-reported by students and/or their families to the DOE. Data on race that IBO receives from the DOE includes one of the following classifications for each student: Asian, Black, Hispanic, Missing, Multiracial, Native American, and White.
7For current pre-K students attending private or parochial schools, or not attending a school as of yet, their families can apply to the Department of Education who sets up interviews - early childhood education experts conduct these interviews and nominate eligible applicants for G&T admissions. See https://www.schools.nyc.gov/enrollment/enroll-grade-by-grade/gifted-talented for more details.
8Mayor Eric Adams and Chancellor David Banks press conference, April 14, 2022: Mayor Adams, Chancellor Banks Announce Expansion of Gifted and Talented Programs Citywide | City of New York (nyc.gov).
9In 2018-2019, an additional three districts (District 5 in Manhattan, District 8 in Bronx, and District 18 in Brooklyn) had schools that offered seats to students, but had no students enrolled in Gifted and Talented programs in that district. Furthermore, Districts 7, 12, 16, and 23 have Gifted and Talented programs beginning in 3rd grade.
10A detailed explanation of the motivation for IBO’s creation of a student neighborhood poverty indicator can be found in this 2015 report. IBO obtained five-year estimates of median household income for each census tract from the 2017 American Community Survey (ACS), covering data from 2013 through 2017. The most recent poverty report from the Mayor’s Office of Economic Opportunity is for 2019.
11Data source for violent felonies: NYC OpenData: NYC crime | NYC Open Data (cityofnewyork.us).
12IBO obtained five-year estimates of the share of residents ages 25 and older that have attained a bachelor’s degree or higher (master’s, doctorate, or professional degree) from the 2017 American Community Survey (ACS), covering data from 2013 through 2017.