The shortage of top-quality schools in certain Boston neighborhoods has undercut the city’s bold efforts to provide access to good schools close to home, according to a new study led by a Northeastern University-based research center.

Four years ago, Boston began implementing a school assignment system that uses an algorithm to produce individualized school options for families, with the goal of increasing students’ access to high-quality schools while reducing the distance they must travel to get to school. Schools were ranked into one of four tiers based on their most recent MCAS scores and historical trajectory, with tier 1 being the highest.

The researchers concluded that the new assignment system failed to counteract the city’s longstanding geographic, racial, and socioeconomic disparities, noting that in some ways it further diminished geographic and racial integration across the district.

The study focused on evaluating the system’s effectiveness over three years for kindergarten and sixth grade. Boston Public Schools began implementing its new assignment system with kindergarten and sixth grade during the 2014-15 academic year, and has been sequentially phasing in other grades since then.

One of the researchers’ primary findings is that geography largely determines access to quality schools. “If you don’t have quality throughout the city, you can’t offer equal access to quality,” said Daniel T. O’Brien, an associate professor in Northeastern’s School of Public Policy and Urban Affairs and School of Criminology and Criminal Justice. O’Brien, a lead author on the report, co-directs the Boston Area Research Initiative at Northeastern, which conducted the evaluation.

The researchers cited the lack of top-tier schools near predominantly minority neighborhoods in the southern part of the city, including South Boston, Roxbury, Dorchester, and Mattapan. Therefore, those students had fewer top schools from which to choose, had greater competition for seats in those schools, were less likely to attend them, and had to travel longer distances when they did attend them.

If you don’t have quality throughout the city, you can’t offer equal access to quality.

Daniel T. O’Brien, Co-director of the Boston Area Research Initiative

The researchers found that pre-existing disparities in the number of students living in different parts of the city led to greater competition for seats for black and Latino students, compared to white and Asian students. As a result, they underscored the need for a system that defines access to quality in terms of competition for seats, rather than the number of school options.

“Compounding the problem of having fewer tier 1 seats available in these communities, there are more students vying for those seats,” said Nancy E. Hill, a co-author on the report and the Charles Bigelow Professor at the Harvard Graduate School of Education. “Because more children were vying for fewer seats in these communities, fewer children in these communities received their first choice. And in a cascading fashion, more of these students were assigned to tier 3 and tier 4 schools or were administratively assigned.”

The researchers presented their report to the Boston School Committee on Monday night.

Other top findings:

• The new system achieved one of its main goals: for students to travel shorter distances and for shorter periods of time to and from school. In particular, the longest 25 percent of commutes among kindergarten students were reduced by a half-mile and two-and-a-half minutes each way.
• The system didn’t succeed in creating “neighborhood schools”—in which neighborhoods concentrate their students at fewer, more local schools.

“The home-based school assignment system was designed in such a way that failed to address many existing inequities, and, in some cases, exacerbated them,” O’Brien said.

The previous school assignment system divided the city into three zones. Under the new system—the algorithm for which was generated by researchers at the Massachusetts Institute of Technology—each family had the opportunity to choose from at least six top-tiered schools, including the two closest tier 1 schools.

The researchers analyzed administrative records, including school choice options under both the current and previous assignment systems as well as enrollment data for both systems. They also examined data generated by the public schools’ lottery process, including the choices submitted by families and the steps by which they were assigned to schools.

Another finding, Hill said, was a “surprising and avoidable” error in how the system was implemented for sixth grade. Families were supposed to have access to the closest two tier 1 schools for children in sixth grade, but the algorithm was set to determine school quality based on the closest tier 1 kindergarten schools. As a result, schools that don’t offer sixth grade were removed and pathway schools were added.

“Removing the schools without sixth grade, in many cases, meant that no tier 1 schools remained,” Hill added. “The result was that many communities had no access to tier 1 schools with sixth grade.”

Increasing the number of good schools in the district is the most direct way to improve equity and access to high quality education for all, the researchers said.

“We cannot lose sight of the fact that black and Latino children, who are already disadvantaged in other ways, face greater competition to get into tier 1 schools,” Hill said. “The deck is already stacked against them in this society, and this policy has made it harder for them to get the educational foundation they need to succeed.”