Topic

Students from new homes

How many public school students does a new home bring, and what should the forecast assume?

It depends mostly on bedrooms. A studio or one-bedroom brings almost none, a two-bedroom about 0.22 and a three-bedroom about 0.37, whether it is a house, a condominium or an apartment. Across the mix of homes Brookline has been building, that comes to about 0.24 to 0.27 a home, so the forecast keeps its 0.244.

Every estimate of what a new home costs the schools rests on one number: how many public school students the home brings. This page sets out where the forecast's figure, 0.244 students a home, comes from, checks it against four other sources, and shows what different kinds of building would bring. How the forecast counts the homes themselves is on Counting Brookline's homes, and how enrollment is forecast is on Enrollment: three forecasts compared.

Definitions

Condominium is a form of ownership, not a kind of building: a three-family house split into condominiums and a 50-unit condominium building are both condominiums here. That is one reason this page leans on bedrooms rather than kind of home.

Bedrooms matter more than the kind of home

The Census Bureau's American Community Survey asks each household about its home (bedrooms, building size, owned or rented, condominium fee) and about everyone living there, including whether each child attends public school. Its records for individual households are published only for areas of at least 100,000 people; the smallest that includes Brookline also covers Dedham and Milton, and Brookline is about half of it. The 2020 to 2024 records cover about 2,400 homes.

At the same number of bedrooms, the kinds of home differ by less than their ranges, with two exceptions. A two-bedroom condominium holds fewer students (0.15) than a two-bedroom house or apartment (0.26 and 0.23). And four-bedroom homes that are owned hold far more than the few that are rented.

Owned and rented. Up to three bedrooms, rented homes hold as many students as owned ones, or a few more:

Only at four or more bedrooms do owners hold more, and only about one rented home in twenty is that large.

Brookline's own counts

The survey covers three towns. Two sources cover Brookline alone, and both agree with it.

Neighborhoods. The 2010 and 2020 Censuses count children on every block. Brookline's 44 block groups (neighborhoods of about 1,500 people) differ widely in children per home, and the share of their homes with three or more bedrooms explains most of the difference: the correlation is 0.89 in 2010 and 0.90 in 2020. The share of homes that are owned tracks nearly as closely (0.77 to 0.83), because the two go together in Brookline's neighborhoods, so the block groups alone cannot say which one matters. The survey above can, because it compares owned and rented homes of the same size.

Matching each block's homes in the assessors' records to its Census count gives children per home for each kind (all children 5 to 17, public and private):

Houses hold the most because they have the most bedrooms. Condominiums with three or more bedrooms held about 0.34 children in 2020, about 40% more than two-bedroom condominiums (0.24).

New developments. The Town and the Schools counted the students enrolled in fall 2025 in ten developments of eight or more homes built since 2015, not counting age-restricted and Housing Authority buildings. They range from 0.03 to 0.56 students a home and average 0.21. The eight apartment developments average 0.16 at an average of 1.56 bedrooms, close to what the survey schedule gives for a typical rental mix (0.16, below). The two townhouse developments, at three bedrooms, hold 0.25 and 0.56, around the schedule's 0.37.

How the forecast's 0.244 fits

The forecast's figure was not measured by dividing students by homes. It is the per-bedroom schedule in RKG Associates' fiscal study for the Town (none for a studio or one-bedroom, 0.25 for two bedrooms, 0.35 for three) applied to the bedrooms of the 34 homes Brookline built from FY2018 to FY2025, which average 2.31. Students who live outside Brookline, through METCO or by paying tuition, never enter it. The steps from the Census's count to the forecast's figure:

Brookline's existing homes and its recent new ones come out close because their bedrooms are close, about 2.4 and 2.31 on average. The survey schedule runs a little higher than RKG's at three bedrooms and more, but about 82% of the three towns' children attend public school against 75% of Brookline's, and correcting for that brings it back to about 0.245. The forecast's 0.244 holds up.

What different buildings would bring

Applying the survey schedule to a building's mix of homes:

The first two and the family-sized mix are illustrations, not real buildings. The range is six-fold, and nearly all of it comes from the share of homes with three or more bedrooms. A building's average bedrooms is not enough on its own, because students rise faster than bedrooms above two: a building of all two-bedrooms gives 0.22, while one that is half one-bedrooms and half three-bedrooms, the same average, gives 0.19. For a single project, the unit mix answers most of the question.

How the forecast uses it

The enrollment forecast and the students from new homes are counted separately, so that no family is counted twice.

Families moving in, at the recent pace, are in the enrollment forecast. NESDEC's method turns births into kindergartners and moves each class up a grade by ratios drawn from Brookline's recent history, and those ratios already include the families who moved into new homes in those years, as well as those who moved into older ones. The forecast assumes new homes keep coming at the recent pace, about about 105 a year, and adds no students for them beyond what the enrollment forecast already holds.

Faster building adds students on top. Homes built beyond that pace add 0.244 students each, and the 384 homes proposed at Chestnut Hill, mostly studios and one-bedrooms, add 0.126 each.

A check on the recent pace. NESDEC's ratios grow a class about 2% from kindergarten to grade 12. That sits between the years before COVID, when classes shrank about 7%, and the years since, when they grew about 6%, so it leans on the recent years. Those were busy years for building: permits issued in 2019 to 2022, finished as those classes moved up, came to about 155 homes a year not counting age-restricted ones, against about 43 a year in 2012 to 2018. The forecast's pace of about 105 a year includes age-restricted homes, about 77 a year without them. So NESDEC's ratios probably carry somewhat more children from new homes than the forecast's pace of building will bring, perhaps 5 to 20 students a year, or roughly 60 to 250 by FY2040, 1% to 5% of enrollment. If anything, that puts the base case's enrollment a little high, not low. The ratios also include families returning after COVID, which this check cannot separate.

What could change it

• Families may keep filling homes faster. Children per home rose from 0.27 to 0.33 between 2010 and 2020, and none of the rise came from building different kinds of homes: houses, condominiums and apartments each held more children. A schedule measured in 2020 to 2024 could run high or low by the 2030s. This is the largest uncertainty, and it works through the enrollment forecast's rate of families moving in more than through new homes. The Forecast Explorer's Families moving in setting tests it: each class 0.35 point a year larger, the rate since COVID, brings +200 students by FY2040, and 1 point a year larger brings +580 (Assumptions register). As Brookline's owners age, more family-sized homes should change hands, which points toward the first of these (Aging homeowners and family-sized homes).
• Brookline is not the three towns. Brookline's single-family houses hold more children than the survey's three-town figure for houses (0.72 against 0.61 children 5 to 17), its condominiums fewer (0.18 against 0.27), and its families send a smaller share to public school. These pull in different directions, which is why the schedule is checked above against Brookline's own counts.
• New buildings fill slowly. The ten developments were counted once, and several are only a few years old. Families with school-age children may arrive as tenants turn over.
• The ten developments are averaged as equals. A building of 8 homes and one of 250 count the same in their 0.21. Weighting them by homes, and comparing each with what the schedule predicts for its own mix of homes, needs each building's bedrooms, which are being gathered from building permits and plans.

Sources and method

• Census Bureau, American Community Survey public-use microdata, 2020 to 2024, Public Use Microdata Area 00902 (Brookline, Dedham and Milton), households and people, with the survey's replicate weights for the ranges.
• Census Bureau, 2010 Census Summary File 1 and 2020 Census Demographic and Housing Characteristics File, Massachusetts, every block; block outlines from TIGERweb.
• Brookline Assessors' records for FY2010 and FY2020, placed on Census blocks by parcel; children per home for each kind of home are fitted across blocks, with ranges from resampling block groups.
• Town and Schools count of enrolled students in developments built since 2015, ERSC Final Report, p. 568, read from its chart.
• RKG Associates' per-bedroom schedule and the 34 homes it is applied to, as used in the forecast.
• Massachusetts Department of Elementary and Secondary Education, enrollment, October 2019.

The code and data behind every number on this page are on GitHub.

Go deeper

• Counting Brookline's homes: How many homes has Brookline been adding, of what kind, and what does the forecast assume?
• Enrollment: three forecasts compared: Why do the enrollment forecasts disagree, and which one does the base case use?
• Why families settle elsewhere: Why does Brookline draw fewer young families than towns like Lexington?
• Aging homeowners and family-sized homes: As Brookline's homeowners age, will more family-sized homes pass to families with children?
• New homes and Brookline's budget: What does a new home cost the Town, now that enrollment is falling?
• Growth: when a new home pays for itself: When does a new home, or new commercial space, add more tax than it costs the Town?
• Assumptions register: What does the forecast assume, and why?

Last updated 2026-10-28. What has changed.