New York City charter schools replace students who leave between kindergarten and third grade, according to new data from the Independent Budget Office, though seats that open up in older grades sometimes go unfilled.

A new report from the city’s education-data watchdog offers the clearest look yet at how the charter schools “backfill” their seats, an issue that has become a focus of debates about whether those schools serve the city’s neediest students. It also includes a trove of other statistics about charter schools at a moment of rapid growth for the charter sector, whose enrollment jumped 364 percent between 2007 and 2014.

The report reaffirms that New York City charter schools still lag behind district schools in serving both English language learners and students with disabilities, although they continue to serve larger shares of black and Hispanic students. It also includes rankings of charter school networks based on their state test scores for each tested grade in 2014 in math and English.

The numbers will add new detail to debates about where and how students are educated in the nation’s largest public school system. But they are unlikely to settle long-simmering disagreements about the role that charter schools — which are privately managed but receive public funds — play in the city, where most students still attend traditional district schools.

Less than 7 percent of the city’s 1,084,955 students attended a charter school in 2013-14, according to the report. Mayor Bill de Blasio often refers to those numbers when responding to questions about charter schools, saying he is focusing attention and resources on schools serving the vast majority of city students.

But the IBO numbers show that charter schools are much more dominant in some parts of the city. Thirty-seven percent of students attending school anywhere in Harlem’s District 5 were in a charter school in 2013-14. In another four districts — 4, 7, 16, and 23 — at least one in five students attended a charter school.

Charter schools continue to serve a smaller share of English language learners, especially in elementary schools, where 17.5 percent of district school students are English learners, compared to less than 7 percent of charter school students. Charter schools are also less likely to serve students born in another country or who speak another language at home.

The gap is narrower when it comes to special education, although charter schools still serve fewer students with disabilities overall. That includes students with learning disabilities and who are emotionally disturbed, and with a couple of notable exceptions, charter schools don’t serve students with autism. The city’s few charter high schools serve a slightly higher percentage of special-education students, though.

Debates about who charter schools serve have recently focused on what happens when a student leaves a charter school. Charter schools aren’t required to replace that student with another, and some choose not to — a practice that some observers, including the city teachers union, say is a reason why their student populations are less needy and why some charter schools achieve better academic results than nearby district schools. (Charter schools that don’t backfill say it avoids academic and social disruption, and ensures that all of their students can keep up.)

The report tracks more than 3,000 students who started kindergarten at a charter school in 2008 through their fifth grade year, an analysis that included 53 charter elementary schools. The schools included those belonging to the city’s largest networks, including Success Academy, Uncommon Schools, and Achievement First.

The data shows that the schools filled all of the seats that opened up in their early grades, and even added extra students during those years. But their average backfill rate fell to 80 percent after third grade, the first year that standardized state testing begins for math and English.

After third grade, 23 of the 53 schools backfilled all or almost all of their open seats. Twelve schools filled fewer than one in three open seats, and five admitted just one or zero new students to their fourth and fifth grades.

Ray Domanico, the IBO’s director of education research, cautioned against drawing broad conclusions from the analysis, noting that the group of schools he looked at was relatively small and did not count the 137 schools that have opened since 2008.

“As time passes it will get easier to study this,” Domanico said in an email.

The report did not look at backfill in middle schools, though some schools — most prominently, the Success Academy network — have come under fire for not adding students who didn’t start at the network in younger grades. The data shows one outcome of Success’ approach: just 32 students took the eighth-grade exams in 2014, part of the network’s first cohort of students, and their average raw score was higher than every other charter network’s in reading and math.

The report is a first for the IBO, which is required by state law to research and report on city education data. The IBO has released three comprehensive reports on the city’s traditional public schools, but had not published similar data on the demographics and performance of charter schools.

The report also compared charter networks by their students’ average raw scale scores on the state tests in English and math. Such comparisons are rarer than comparisons of proficiency rates, which don’t show whether most students just missed — or just surpassed — the state’s proficiency bar.

Success Academy, the city’s largest and fastest-growing network, had the highest average test scores in 2014 for both English and math in every tested grade. The Icahn, Uncommon Schools, and Public Preparatory networks had the number-two spots for various grades and subjects.

Bronx-based Lighthouse Academies, Kunskapsskolan Education, which manages a school that closed this year, Harlem Village Academies, and National Heritage Academies were among the lowest-scoring networks.

A spokesman for the United Federation of Teachers highlighted the finding that district schools continued to serve more high-needs students.

“The IBO report makes clear many of the inconvenient facts that charter cheerleaders keep managing to ignore — charters enroll fewer than half as many English Language Learners as regular public schools and far fewer high-needs special ed students, even as many charters leave seats empty when children leave,” the spokesman, Dick Riley, said.

Charter school supporters, however, said the sector’s growth was evidence of charter schools’ popularity, while the backfill rates showed that most charters were replacing most students who leave.

“This independent report confirms that charter schools are an increasingly popular choice among New York City parents and why they should be supported in their continued growth,” said James Merriman, head of the New York City Charter School Center.

Read the complete report here: