Equity: who the changes affect

The case for a different solution for immersion families

The argument in brief

The plan relies on moving immersion programs. Scenario A would move 1,586 students in dual language immersion (DLI) programs out of their schools. They are 29% of the students who would have to change schools because of a closure, a program move or the César Chávez change (§1).
The programs PPS would move have plenty of families. Atkinson gets 121 kindergarten applications a year for about 24 seats. Beach gets 96 for 23 (§2).
The burdens fall on students of color, in different ways. In Scenario A, students in the moving programs are 45% Latino and 13% Asian, against 19% and 5% of PPS elementary and middle school students. Black students are only 2% of immersion students. The plan reaches them mainly in two ways: closures, and moving César Chávez’s neighborhood program out so the school can be immersion-only. That program is 34% Black. Schools that lose an immersion program are 51% historically underserved and 44% low-income; the schools closing outright are 61% white (§3).
Immersion families would travel farther. Their schools would move farther (1.7 vs 1.2 miles). The share of school-age Spanish speakers who live within a mile of a K-5 Spanish site would fall from 52% to 36% (§4).
Moving a program loses the families who would have joined it. In 2023, PPS moved Bridger’s Spanish program to Lent. Lent now enrolls 31% fewer immersion kindergartners a year than the two schools averaged per grade before. Measured the same way, the district fell 14%. The 2023 plan had projected 527 students at Lent; it has 256. PPS’s own school projections now expect only 80–84% of students in a moved program to follow it. The rest would stay at neighborhood schools. By PPS’s own numbers, the moves shrink immersion (§5).
The last round of changes also moved neighborhood enrollment. After the 2023 Southeast changes, fewer Southeast elementary children attend their own neighborhood school. The share fell from 71% to 66%, while it rose elsewhere in PPS. Public data can’t separate the part of that drop that was built into the changes from the part that was families’ choice (§6).
PPS’s equity check can’t see any of this. It reports one move rate for each student group. Even so, multilingual learners are already the group PPS moves most often (§7).
Southeast Spanish immersion is too small for its demand. Richmond alone enrolls 89 Japanese kindergartners a year. That is more than Southeast’s two Spanish programs combined (70). Atkinson puts about two applicants on its waitlist for every one it offers a seat. Lent’s problem is not a shortage of families who want Spanish. Its problem is that it no longer serves its own neighborhood, and the plan would do the same to Rigler (§8).

1. The plan relies on moving immersion programs

PPS’s own documents put immersion near the center of the plan. Its scenario memo names “underenrolled dual language immersion classrooms” among the problems the modeling responds to. It calls immersion consolidation a “primary driver” of the changes in the Roosevelt/Jefferson region (PPS, Rightsizing Update: Scenario Release board memo, October 5, 2026, pp. 1–2 (BoardBook)). Consolidation means combining programs at fewer schools. Both scenarios would cut elementary Spanish immersion sites from nine to four and move two Mandarin programs. Scenario A would also move Vietnamese immersion.

In Scenario A, 1,586 students are in immersion programs that would move. Another 3,763 attend schools that would close. And 143 students in César Chávez’s neighborhood program would move out, so that César Chávez can become an immersion-only school (§3). We call all three groups “directly displaced”: they would have to change schools because a school closes or because PPS moves their program. Immersion students are 29% of everyone directly displaced. Yet they are only 13% of PPS enrollment. Boundary changes move many more students, but none of these groups counts them. Public data can’t show who those students are.

2. “Underenrolled” describes seats, not demand

Table 1

Program move	Language	Kindergarten lottery, average per year
Demand at the programs being moved (Scenario A)
Program move	Language	Applications	Offers	Waitlisted	Applications per offer
Atkinson → Lent	Spanish	121	24	49	4.9
Beach → César Chávez	Spanish	96	23	46	4.1
MLK Jr. → Boise-Eliot/Humboldt	Mandarin	59	46	5	1.3
Scott → Rigler	Spanish	40	29	8	1.4
Rose City Park → Vestal	Vietnamese	33	29	1	1.1
James John → César Chávez	Spanish	27	18	4	1.5
Sitton → César Chávez	Spanish	25	19	4	1.3
Clark → Woodstock	Mandarin	16	7	0	2.4
Average of the 2022-23 to 2026-27 lotteries (fewer years where a program wasn't in the lottery). Counts are applications, not children: a family can apply to several programs.
Sources: PPS Enrollment & Transfer, Elementary School / Kindergarten Lottery Results Summary, lotteries for 2022-23 through 2026-27 (prior lottery results); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown. Method details: Sources & methods.

Atkinson and Beach, the two programs with the most demand, have far more families than seats. They are the two Spanish programs with the most applicants in the district. Each gets four to five applications for every kindergarten seat, and dozens of families are waitlisted every year. Their classrooms are small because PPS offers few seats, not because families aren’t asking for them.

Other moves are different. James John and Sitton draw little more than one application per seat. Clark has had no waitlist, and its applicants are often placed at programs they ranked higher. The case for consolidating those programs is stronger. A plan that treated these programs differently would take less from immersion families than one that moves them all. The Demand & scorecard page has more on demand. It also explains why Lent’s own lottery numbers understate its draw.

District totals hide where the demand is. PPS’s October 13 slides count 278 Spanish kindergarten lottery seats for 2026-27 and 250 Spanish kindergartners enrolled. That looks like spare room. The 2026-27 lottery shows where families actually applied. Spanish programs put 151 applications on waitlists. A family on two waitlists counts twice, so this counts places on waitlists, not children. Most were at Atkinson (47), Ainsworth (41) and Beach (36). Scott, James John and Sitton waitlisted only 1 to 4 each. Atkinson and Beach are two of the programs PPS would move. (The slide doesn’t say what its counts include. In October 2025, PPS enrolled 306 Spanish immersion kindergartners.)

Sources: PPS, PPS Dual Language Immersion (DLI) Programs, presentation to the Board’s Teaching, Learning, and Enrollment Committee, October 13, 2026 (BoardBook); PPS Enrollment & Transfer, 2026-27 Kindergarten Lottery Results Summary (prior lottery results); PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting). Method details: Sources & methods.

3. Who would have to move, by race and ethnicity

The plan’s burdens fall differently on each racial and ethnic group. The moving immersion programs are mostly Latino, with many Asian students in the Mandarin and Vietnamese programs. Few Black students are in immersion, so the plan reaches Black students mainly through closures and the César Chávez change.

We can see this because PPS’s Enrollment Summary by Race and Ethnicity (October 2025) gives each immersion program its own row. The table reports each group separately. PPS’s combined “historically underserved” share counts Latino, Black, Native American, Pacific Islander and multiracial students together, and it would hide these differences.

Figure 1: Share of each racial and ethnic group among students who would have to change schools in Scenario A, compared with all PPS elementary and middle school students (hollow gray ring). October 2025 enrollment. An open marker means fewer than 5 students, drawn near zero. Hover for counts; the table below has every number and Scenario B.

Sources: PPS, Enrollment Summary by Race and Ethnicity, October 2025 (separate rows for each school’s immersion and neighborhood programs; PDF), retrieved October 11, 2026; PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; school closures and receiving schools: PPS scenario maps and memo (October 2026) as transcribed in pps-explorer; PPS, How Could Your School Be Impacted?, César Chávez guide (October 8, 2026), page 3, “What is being proposed?” (page); PPS, How Could Your School Be Impacted?, César Chávez guide (October 8, 2026), page 4, “Who would the school serve?”, with no changes and under Options A and B (page); PPS, PPS Dual Language Immersion (DLI) Programs, presentation to the Board’s Teaching, Learning, and Enrollment Committee, October 13, 2026 (BoardBook). Method details: Sources & methods.

Table 2

	Students	Latino	Black	Asian	Native American	Pacific Islander	White	Multiracial		Historically underserved (PPS)
Race and ethnicity of the students who would have to change schools
October 2025 enrollment
	Students	Latino	Black	Asian	Native American	Pacific Islander	White	Asian and White	Other	Historically underserved (PPS)
Scenario A
Students in immersion programs that move	1,586	45%	4%	13%	<1%	<1%	31%	4%	4%	52%
Students at closing schools (outside immersion)	3,763	15%	7%	4%	1%	1%	59%	6%	8%	31%
Scenario B
Students in immersion programs that move	1,379	51%	4%	6%	<1%	<1%	32%	3%	4%	59%
Students at closing schools (outside immersion)	3,013	15%	8%	4%	1%	1%	57%	5%	9%	33%
Scenarios A and B
César Chávez neighborhood program, moved out so the school can be immersion-only	143	33%	34%	5%	<4%	<4%	15%	<4%	10%	80%
PPS today
All PPS elementary and middle schools	26,430	19%	8%	5%	1%	1%	53%	6%	7%	35%
All students in PPS K-8 immersion programs	4,457	39%	2%	11%	<1%	<1%	35%	11%	3%	44%
To protect privacy, groups with fewer than 5 students show as <5 or <X% (the share 5 students would be).
Latino includes students of any race. The other groups count only students who are not Latino, and the single-race groups (Black, Asian, Native American, Pacific Islander, White) count only students who list one race. So Native American and Pacific Islander students who are also Latino or multiracial appear under Latino or multiracial. PPS defines historically underserved students as Latino, Black, Native American, Pacific Islander and multiracial (other ancestries); Asian and Asian/White students are not in its definition. Immersion moves use each program's row; César Chávez's grade 6-8 move uses its K-8 program's mix. Closing schools count students outside immersion (their immersion programs are counted as moves). Every move and closure is in grades K-8, so the comparison is PPS's elementary and middle schools.
Sources: PPS, Enrollment Summary by Race and Ethnicity, October 2025 (separate rows for each school’s immersion and neighborhood programs; PDF), retrieved October 11, 2026; PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; school closures and receiving schools: PPS scenario maps and memo (October 2026) as transcribed in pps-explorer; PPS, How Could Your School Be Impacted?, César Chávez guide (October 8, 2026), page 3, “What is being proposed?” (page). Method details: Sources & methods.

Latino students would carry most of the program moves

In Scenario A, 45% of students in the moving programs are Latino (51% in Scenario B). Latino students are 15% of students at the closing schools and 19% of PPS elementary and middle school students. The Spanish programs that would move are 32% to 79% Latino. Of all Latino students who would have to change schools in Scenario A, 54% are in a moving immersion program.

Black students are rarely in immersion, and the plan reaches them in other ways

In Scenario A, about 351 Black students would have to change schools. Few of them are in immersion. Most attend a school that would close or César Chávez’s neighborhood program.

PPS’s own history explains why burdens on Black students need a close look. PPS describes Oregon’s 1844 Black exclusion law and the “whites only” clause in the state’s 1857 constitution. It says the district “had already adopted racist practices that disadvantage students of color – especially our Black and Native students.” In lower Albina, in North and Northeast Portland, urban renewal and Interstate 5 “forcibly displaced thousands of primarily Black residents and businesses.” PPS adds that “students bore the cost. Forced busing sent many Black students to schools outside their neighborhood” (PPS, Center for Black Student Excellence, Historical Context (page), retrieved October 11, 2026). The City of Portland describes how “red lines drawn on a city map would dictate where people of color could live” (City of Portland, Portland Housing Bureau, N/NE Neighborhood Housing Strategy: Historical Overview (page), retrieved October 11, 2026).

In 2011 the school board adopted a Racial Educational Equity Policy. In it, PPS “committed to significantly changing its practice in order to achieve and maintain racial equity in education” (PPS, Racial Educational Equity Policy (2.10.010-P, approved June 13, 2011; page), retrieved October 11, 2026).

Black students are largely missing from immersion. They are 2% of students in PPS’s K-8 immersion programs (91 of 4,457), compared with 8% of elementary and middle school students. Nearly half of them, 43, are in one program: MLK Jr.’s Mandarin program. Both scenarios would move it to Boise-Eliot/Humboldt. That school is 41% Black today.

The Spanish programs that would move enroll almost no Black students: each has fewer than 5. The neighborhood programs in the same buildings are more often Black: Beach 23%, Ockley Green 20%, Sitton 19%, Scott 18%, James John 11%, Beaumont 4%, Atkinson <3%. So in Scenario A, the program moves would move 56 Black students, 43 of them from MLK Jr.

Figure 2: Each moving immersion program (blue circle) compared with the neighborhood program in the same building (hollow gray ring): share of Black students (left) and Latino students (right), October 2025. César Chávez compares its whole K-8 Spanish program with its neighborhood program, which would also move. An open blue circle means fewer than 5 students, drawn near zero.

The plan reaches more Black students in two other ways:

Closures. The schools that would close enroll 247 Black students outside immersion in Scenario A (226 in Scenario B).
César Chávez becomes immersion-only. To make César Chávez a K-5 school with only Spanish immersion, PPS would move its neighborhood program out. Its K-5 students would go to Rosa Parks and its grade 6-8 students to George (PPS, How Could Your School Be Impacted?, César Chávez guide (October 8, 2026), page 3, “What is being proposed?” (page)). That program has 143 students. 34% are Black, 4.2 times the share in PPS elementary and middle schools. 80% are historically underserved. This repeats the pattern of Lent’s 2023 conversion. The district’s equity audit found that the students displaced then “were disproportionately Black and Asian Lent neighborhood families” (Green, T. L., & Hanson, H. (2026, June 29). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing initiative: School-by-school analysis (PPS SEGC documents), p. 47; see whole-school vs. strand research). We count them in §1 as directly displaced.
César Chávez after the change. PPS’s own guide projects how the school would change. Its share of students from underserved racial groups would fall from 82% to 69%. Its share of students in poverty would fall from 66% to 48%, and its share in special education from 21% to 14%. PPS’s October 13 slides on César Chávez show only today’s mix.

Of the 351 Black students who would have to change schools in Scenario A, 70% would move because their school closes.

Figure 3: Black students who would have to change schools in Scenario A, by school or program and by reason, October 2025. Schools and programs with fewer than 5 Black students are combined into one bar for each reason, to protect privacy; hover to see which ones.

Table 3

Program move	Language	Grades	Scenarios	Students moving	Immersion program				Neighborhood program at the same school
Each moving program, and the neighborhood program it shares a building with
October 2025
Program move	Language	Grades	Scenarios	Students moving	Latino	Black	Asian	White	Black	Historically underserved
Scott → Rigler	Spanish	K-5	A and B	222	47%	<3%	<3%	43%	18%	59%
Rose City Park → Vestal	Vietnamese	K-5	A	207	<3%	<3%	63%	24%	4%	29%
MLK Jr. → Boise-Eliot/Humboldt	Mandarin	K-5	A and B	193	13%	22%	7%	34%	45%	88%
Atkinson → Lent	Spanish	K-5	A and B	178	70%	<3%	<3%	28%	<3%	19%
Beach → César Chávez	Spanish	K-5	A and B	155	47%	<4%	<4%	46%	23%	45%
Sitton → César Chávez	Spanish	K-5	A and B	138	70%	<4%	<4%	30%	19%	58%
James John → César Chávez	Spanish	K-5	A and B	133	75%	<4%	<4%	21%	11%	56%
César Chávez → George	Spanish	6-8	A and B	105	79%	<2%	<2%	16%	34%	80%
Beaumont → Roseway Heights	Spanish	6-8	A and B	101	73%	<5%	<5%	22%	4%	19%
Clark → Woodstock	Mandarin	K-5	A and B	91	<6%	<6%	68%	18%	23%	66%
Ockley Green → George	Spanish	6-8	A and B	63	32%	<8%	<8%	60%	20%	50%
To protect privacy, groups with fewer than 5 students show as <5 or <X% (the share 5 students would be).
César Chávez's grade 6-8 move uses the mix of its whole K-8 Spanish program. Its neighborhood program would not stay: PPS would move it to Rosa Parks and George. Beach and James John close in both scenarios, and Rose City Park in Scenario A, so their neighborhood programs move too; they are counted with closures above.
Sources: PPS, Enrollment Summary by Race and Ethnicity, October 2025 (separate rows for each school’s immersion and neighborhood programs; PDF), retrieved October 11, 2026; PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown. Method details: Sources & methods.

Asian, Native American and Pacific Islander students

Asian students are 13% of students in Scenario A’s moving programs, compared with 5% of elementary and middle school students. Rose City Park’s Vietnamese program and Clark’s Mandarin program are mostly Asian. PPS’s definition of historically underserved students leaves out Asian students. So PPS’s own equity measures, and the “historically underserved” column above, don’t show this burden on Vietnamese and Chinese families. Scenario B keeps the Vietnamese program in place (6% Asian).

Native American and Pacific Islander students are rarely in immersion, at least as PPS counts them. PPS’s K-8 immersion programs enroll fewer than 5 Native American students and fewer than 5 Pacific Islander students. (To protect privacy, we don’t show counts under 5.) These counts include only students who list one race and are not Latino. PPS counts Native American and Pacific Islander students who are also Latino as Latino. It counts students who list more than one race as multiracial (PPS, Enrollment Summary by Race and Ethnicity, October 2025 (separate rows for each school’s immersion and neighborhood programs; PDF), retrieved October 11, 2026, page 6 footnote). So the true number of Native students in immersion is likely higher.

The schools that would close enroll 30 Native American and 27 Pacific Islander students outside immersion in Scenario A. These numbers are too small for program-by-program comparisons. Even so, they suggest that immersion reaches few students in either group.

Some schools that lose a program would become whiter

Immersion programs bring many of the Latino and Asian students to the schools that host them. At Atkinson, the Spanish program has 124 of the school’s 139 Latino students. The neighborhood program that stays is 76% white. At Beaumont Middle School, the Spanish program has 74 of 99 Latino students. Its neighborhood program is 70% white.

PPS’s own school guides show the result. They project that Atkinson’s share of students from underserved racial groups would fall from 44% to 21%, and Beaumont’s from 33% to 17%. Scott, MLK Jr. and César Chávez would also fall. Clark and Ockley Green would rise. The guides’ projections include boundary changes too, so not all of each change comes from the program move. The measure also leaves out Asian students: Vestal’s share falls because the Vietnamese program it would receive is mostly Asian.

The School changes page follows every school the plan touches, including the schools that take in students. It also covers Title I and SUN schools, and whether the moves would separate students by race.

Figure 4: PPS’s projected share of students from underserved racial groups at each school that would lose or receive an immersion program: if nothing changes (hollow gray ring) and in Scenario A (filled marker). Both are PPS projections from its school guides. The projections include boundary changes as well as program moves.

Sources: PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 4, “Who would the school serve?”, column “Underserved racial groups”, with no changes and under each option (page); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown. Method details: Sources & methods.

Table 4

School	No changes	Scenario A	Change (points)
PPS's projected share of students from underserved racial groups
Schools that lose or receive an immersion program, Scenario A
Loses an immersion program
Atkinson	44%	21%	−23
Beaumont	33%	17%	−16
MLK Jr.	63%	47%	−16
Scott	56%	42%	−14
César Chávez	82%	69%	−12
Sitton	63%	56%	−7
Ockley Green	48%	58%	+10
Clark	52%	64%	+12
Receives an immersion program
Vestal	46%	30%	−16
Boise-Eliot/Humboldt	74%	64%	−10
Rigler	67%	61%	−6
George	77%	72%	−5
Woodstock	16%	14%	−2
Lent	72%	71%	0
Roseway Heights	42%	47%	+6
Beach, James John and Rose City Park close in Scenario A and are not shown. Scenario B differs only at Scott (no changes 56%, Scenario B 54%) and Vestal, which receives no program in B. The guides' projections include boundary changes as well as program moves.
Sources: PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 4, “Who would the school serve?”, column “Underserved racial groups”, with no changes and under each option (page); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown. Method details: Sources & methods.

Whole schools: income and home language

PPS publishes income (direct certification) only for whole schools. The table below shows both race and income for the schools involved.

Table 5

	Students	Historically underserved	Low-income (direct certification)	Latino	White	English learners
Who attends the schools involved (Scenario A)
Whole-school composition, October 2025, weighted by enrollment
Schools losing an immersion program	3,895	51%	44%	32%	40%	21%
Schools closing (no immersion program)	3,188	29%	28%	14%	61%	7%
Schools receiving an immersion program	2,419	54%	49%	33%	34%	24%
All PPS schools	43,479	35%	31%	18%	53%	10%
PPS defines historically underserved students as Latino, African American, Native American, Pacific Islander and multi-racial (other ancestries). Direct certification: students the state has identified as eligible for free meals. Each school is counted once: Beach, James John and Rose City Park, which close and lose an immersion program, are counted with schools losing an immersion program. Clark is not printed in PPS's report and is left out. These are whole-school figures; for race by program, see the first table in this section.
Sources: PPS, School/Neighborhood Enrollment by Ethnicity & Program, October 2025 (School rows; PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; school closures and receiving schools: PPS scenario maps and memo (October 2026) as transcribed in pps-explorer. Method details: Sources & methods.

The schools whose immersion programs would move are 51% historically underserved and 44% low-income. “Historically underserved” is PPS’s term for students of several races and ethnicities, listed in the note under the table. “Low-income” means the state has identified the student as eligible for free meals. The district averages are 35% and 31%. The schools that would close outright are 61% white, 29% underserved and 28% low-income. Scenario B shows the same pattern (56% vs 31% underserved).

These race and income figures describe whole schools, including students in each school’s neighborhood program, who don’t move. PPS doesn’t publish income by program. We can also describe the moving students by home language and English-learner status:

Figure 5: Share of students whose home language is Spanish, or any language other than English: students in immersion programs that move, students at schools that close (excluding their immersion programs), and all PPS students. October 2025 enrollment.

Sources: PPS, Enrollment by Language and School, October 2024 and October 2025 (PPS Data & Reporting); PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; school closures and receiving schools: PPS scenario maps and memo (October 2026) as transcribed in pps-explorer. Method details: Sources & methods.

In Scenario A, 29% of students in the moving programs speak Spanish at home, and 40% speak a language other than English. At the closing schools, the figures are 3% and 7%. 36% of students in the moving programs are multilingual learners still receiving English language services. Of the Spanish-speaking students directly displaced in Scenario A, 82% are immersion students.

Two-way immersion depends on these students. It needs native speakers of the program language in every classroom, alongside English speakers.

4. Immersion programs move farther from families

How far the school moves (measured). Immersion programs would move farther than closing schools’ students. Students at closing schools mostly move to the next school over, 1.2 miles away on average. Immersion programs would move 1.7 miles on average. Atkinson, Beach, James John and Sitton would each move 1.6–2.5 miles. Scott → Rigler, at 0.7 miles, is the exception.

Table 6

	Scenario A	Scenario B
How far the school moves, miles (straight line)
Average over affected students
Immersion program moves	1.7	1.8
School closures	1.2	1.2
Distance from the current school to the new school, averaged over each change's enrolled students (October 2025). For closures split among several receiving schools, weighted by the share of the closing school's attendance area each receives.
Sources: PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS, Enrollment by Language and School, October 2024 and October 2025 (PPS Data & Reporting). Method details: Sources & methods.

Access (estimate). Today, 52% of school-age Spanish speakers live within a mile of a K-5 Spanish immersion site. Under either scenario, 36% would. These counts come from a Census survey, which has a margin of error. Allowing for that error, the drop is 11–20 percentage points.

Distance from home (model estimate). PPS doesn’t publish where students live, so we estimate it with a model. The model places children where the 2020 Census counted them. For immersion programs, it assumes families live in the area nearest their program’s current site. On that basis, the trip from home to school grows by about 1.0 mile for immersion families. It grows by 0.5 mile for families at closing schools.

Figure 6: Model estimate, not measured: average home-to-school distance before and after each change in Scenario A, for children living in the closing school’s attendance area, or in the area nearest the immersion program’s current school (an assumption about where immersion families live). Hover for details.

Sources: PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; US Census Bureau, 2020 Census redistricting data (P.L. 94-171), tables P1 and P3, census blocks. Method details: Sources & methods.

PPS’s report of where each program’s students live offers a check. 53% of students in the moving programs live in their school’s own attendance area. PPS reports the rest only by high-school area. Placing them there gives an increase of 0.8 to 1.1 miles, close to the model. But the model overstates how far families live from school today: 1.8 miles, against 1.2 miles in the check. The details and sensitivity checks are at the end of this page.

Buses (PPS’s October 13 plan). PPS doesn’t bus every immersion student today. Its October 13 slides say kindergarten routes go first to native and heritage speakers. Any other student can ask to ride an existing route as a “hitcher”. For the North Portland Spanish moves to César Chávez, PPS proposes 3 to 4 bus routes just for immersion students. They would have 10 to 14 stops, at schools and community centers, and take about 20 minutes on average. Draft routes are due in spring 2027. The slides show no bus plan for the Southeast (Atkinson to Lent) or the Northeast (Scott to Rigler). They also show none for the neighborhood students who would have to leave César Chávez.

Sources: PPS, PPS Dual Language Immersion (DLI) Programs, presentation to the Board’s Teaching, Learning, and Enrollment Committee, October 13, 2026 (BoardBook). Method details: Sources & methods.

5. Moving a program loses the families who would have joined it

When a program moves farther away, some families choose their neighborhood school instead. PPS staff described this at a board committee meeting in September: “when we move, for example, a program … we did not see students necessarily leave the district, but they may not have moved with the program” (PPS Board of Education Teaching, Learning, and Enrollment Committee, September 8, 2026, at 33:22 (video; automated transcript via Save PDX Schools)). PPS’s equity audit of the 2023 Southeast changes found that students in moved programs who stayed in PPS generally followed their program (Green, T. L., & Hanson, H. (2026, June 10). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing, pp. 11, 48–49). Because enrolled students mostly followed, the larger loss is likely among families who haven’t joined yet. These are kindergarten families deciding whether to choose a program that is now farther away.

The audit is less clear about who left the moved programs. It says leavers “more often identified as white” (p. 11). The audit reports Latino/a/x students as a separate group alongside White, Asian and Black students. So this most likely means non-Latino white students. But the report does not say how it counted students who are both white and Latino. On home language, the audit contradicts itself. It says leavers “more often speak a language other than English at home” (p. 11). It also says “families who left were disproportionately English-speaking” (p. 49).

The clearest test is the last time PPS consolidated Spanish immersion. In 2023-24, PPS moved Bridger’s Spanish program to Lent (PPS, October 2025 Enrollment Summary, 10-year detail, footnote 6 (“Bridger Spanish Immersion moved to Lent”)). Before the move, the two schools had about 55 immersion students per grade. Since then, Lent has enrolled 32 and 43 immersion kindergartners. That is 31% fewer. Measured the same way, district-wide enrollment fell 14%. So the program lost about one in five families beyond the district trend (details). The audit also found that the Southeast projections overestimated enrollment at 12 of 20 schools. Of the 7 schools still well short of target, 3 had a program relocation or grade change (Green, T. L., & Hanson, H. (2026, June 10). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing, p. 6).

The 2023 plan expected Lent to grow, not shrink. It turned Lent into a whole-school immersion school, where every class is an immersion class. A strand, by contrast, is an immersion program that shares a school with a neighborhood program. The audit’s school-by-school appendix summarizes PPS’s Phase 2 impact analysis. That analysis projected that converting Lent to whole-school Spanish immersion “by adding two strands of Spanish immersion from Bridger” would take it “from 277 students in 2021–22 to an estimated 527 students in 2025–26”. It projected building use rising “from 43.8% to 87.5%” (Green, T. L., & Hanson, H. (2026, June 29). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing initiative: School-by-school analysis (PPS SEGC documents), p. 34).

In October 2025, Lent enrolled 256 students, about half that projection. The appendix doesn’t say whether the 527 includes pre-kindergarten. Even counting PPS’s pre-kindergarten students, Lent is far short. Of the 265 students at Lent the spring before the conversion, neighborhood and immersion alike, 49 (18.5%) had left PPS within two years (p. 45).

Enrollment forecasts by Portland State University (PSU) did not predict the shortfall either. PSU’s June 2023 forecast for PPS built in the Bridger → Lent move. Its notes read “Bridger Spanish DLI (grades K-5) moved to Lent ES”. The forecast kept almost all of the combined program’s 319 students: 314 for Lent in 2023-24. Lent enrolled 273. For three years running, the forecast for Lent was 14–16% too high, while its district-wide forecast was within 1%.

Table 7

	2023-24			2024-25			2025-26
Lent's enrollment fell well short of PPS's forecast
	Forecast	Actual	Error	Forecast	Actual	Error	Forecast	Actual	Error
Bridger + Lent Spanish: forecast made before the move (June 2022)	333	273	+22%	335	256	+31%	340	256	+33%
Lent: forecast made with the move built in (June 2023)	314	273	+15%	298	256	+16%	291	256	+14%
District total: same June 2023 forecast	44,288	44,005	+1%	43,826	43,375	+1%	42,950	42,622	+1%
Forecasts and October actual enrollment (kindergarten through grade 12) from the PSU Population Research Center forecasts for PPS. Error = (forecast − actual) ÷ actual. Other schools also miss their forecasts, in both directions; Lent's miss is unusually large and lasting for a school whose program was moved.
Sources: Portland State University Population Research Center, Portland Public Schools Enrollment Forecasts, 2022-23, 2023-24 and 2026-27 editions (Appendix C / Table 5.5; archive). Method details: Sources & methods.

This time, PPS’s projections expect some families not to follow. PPS’s school guides for the scenarios project each school’s enrollment with no changes and under the options (PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page)). Their “no changes” figures match PSU’s 2026-27 forecast exactly. So we can compare a moved program’s arrival with PSU’s forecast for that program. In the three moves where nothing else changes at the receiving school, PPS projects that 80–84% of the program’s students arrive. Where the guides show the sending school, the rest stay there as neighborhood students:

Table 8

Year	Program size (PSU forecast)	Arrive at new school	Share arriving	Stay at old school	Share staying
PPS's projections have about one in five students not following a moved program
Atkinson → Lent (Spanish, Scenario A)
2027-28	182	145	80%	–	–
2031-32	183	147	80%	–	–
Scott → Rigler (Spanish, Scenario B)
2027-28	223	181	81%	49	22%
2031-32	223	184	83%	54	24%
Clark → Woodstock (Mandarin, Scenario A)
2027-28	100	84	84%	16	16%
2031-32	87	81	93%	16	18%
Arrive = receiving school's projected enrollment under the options minus its projection with no changes. Stay = sending school's projection under the options minus PSU's forecast of its other programs (shown only where the sending school has no other change: Scott in Scenario B, Clark). Program size = PSU's 2026-27 forecast of the moving program in that year. Other moves (to César Chávez, Sitton and Vestal) come with boundary or grade changes and can't be separated. The guides don't state a follow rate; these shares are implied by their numbers.
Sources: PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page); Portland State University Population Research Center, Portland Public Schools Enrollment Forecasts, 2022-23, 2023-24 and 2026-27 editions (Appendix C / Table 5.5; archive). Method details: Sources & methods.

That is close to the Bridger → Lent precedent. It is far from the roughly 98% that PSU assumed for that move in 2023. It also means the loss is part of the plan: PPS expects the consolidated programs to be smaller than the programs they replace.

Table 9

If this share of families doesn't follow the program	Share	Students in moving programs	Students not following	Kindergartners not joining, per year
Families who might choose a neighborhood school instead (Scenario A)
Estimate: current students and yearly kindergarten entry in the moving programs, under three assumptions
Everyone follows	0%	1,481	0	0
7% (figure cited in public comment)	7%	1,481	104	16
Bridger to Lent precedent	20%	1,481	293	46
Students: October 2025 enrollment in each moving program, in the grades that move (program moves only). Kindergartners: October 2025 kindergarten immersion enrollment at each sending K-5 program. The 7% row uses a figure a public commenter attributed to PPS at the October 6, 2026 board meeting (PPS's estimate of families who would leave the district); we could not find it in PPS documents. The precedent row applies the excess loss measured after the Bridger to Lent move.
Sources: PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; PPS, Enrollment Details for Language Immersion Schools (Language Immersion Track Profiles), October 2021–October 2025 (current, archive); PPS, Class Size Detail, 2021-22 through 2025-26 (archive, current); PPS, October 2025 Enrollment Summary, 10-year detail, footnote 6 (“Bridger Spanish Immersion moved to Lent”). Method details: Sources & methods.

If Scenario A loses families at the Bridger → Lent rate, about 46 kindergartners a year who would have joined these programs would go elsewhere. At 7%, it is 16. A public commenter attributed that figure to PPS (see the table note). Over six years of kindergarten classes, that adds up to roughly 279 children. We would expect most to enroll at their neighborhood school instead.

PPS’s guides place these students at neighborhood schools. They flag whether each school meets its size and space goals, but they give no capacity figures. PPS publishes each building’s current capacity separately, in its School Profiles dashboard (§8). Either way, the students lost would come from programs at schools that are about half historically underserved (§3).

What staff and families at Lent and Bridger described

The audit’s interviews and surveys describe what the 2023 move was like for the schools involved. The appendix identifies speakers only as participants or staff at each school.

Families who pushed for the move didn’t all stay. A Lent participant: “A lot of the parents who pushed from Bridger for their kids to come to the school didn’t end up staying…they felt like it was out of their way [and that] their students didn’t feel supported” (Green, T. L., & Hanson, H. (2026, June 29). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing initiative: School-by-school analysis (PPS SEGC documents), p. 39).
Promises made to families weren’t kept. A Lent participant: “the fourth grade parents were really upset… they were promised it would be a three strand school…and the parents [started] bringing the letters that they were promised” (p. 41). In the main report, families described transition staffing or funding that was “reduced, delayed, or pulled back because of budget constraints” (Green, T. L., & Hanson, H. (2026, June 10). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing, p. 83).
The transition was hard on staff and the neighborhood. “Those years were really terrible, those first few years…There were a lot of tears during the transition” (appendix, p. 36). After 2022-23, 63% of Lent staff left the school, against a 32% average across the affected schools (p. 39). Staff described hostility from neighborhood families who lost access: “We have been cussed out. We have been yelled at…by parents who moved into the neighborhood that do not want their students to speak Spanish, and that feel that it’s unfair that their students can’t go to their neighborhood school” (p. 41). They also described losing a school children could walk to: “We had a walking school bus every day…We don’t have that anymore, because our entire population gets on a bus to go home at the end of the day” (p. 43).
Families who moved with the program paid in travel. “Families who moved from Bridger to Lent described losing walkable access to school, longer commutes, early bus times, transportation uncertainty, and a loss of neighborhood connection” (main report, p. 83).
Bridger changed too. Participants “experienced the Bridger student body as becoming more white and more affluent, while also losing the additional staffing and Title I support that had previously accompanied a higher-poverty enrollment” (appendix, p. 17).
Marysville participants questioned whom the consolidation served. They “noted that the DLI strand was concentrated in a school farther from many Latino/a/x families, while Atkinson, located in a whiter and more affluent neighborhood, retained a single-strand DLI program”. They also observed “that some immigrant families may have been hesitant to enroll their children in full Spanish immersion in the broader political climate” (p. 61).

Parents summed up what the move needed: “Whole school DLI is great. It needs to be funded, and it needs to be prioritized” (main report, p. 84).

6. Fewer Southeast children now attend their neighborhood school

The scenario projections place most students at the schools that new boundaries and program sites assign them to. PPS tracks how often that happens with a measure called the capture rate. The capture rate is the share of children living in a school’s attendance area (the neighborhood it serves) who attend that school. District-wide, it has held near 68% for years. So about a third of PPS students already attend a school other than their neighborhood school.

Figure 7: Neighborhood capture rate: share of K-12 students living in each elementary attendance area who attend that area’s school, October counts. Southeast = the 16 elementary areas around the 2023 Southeast changes, excluding Lent. Hover for counts.

Sources: PPS, Neighborhood Capture Rate Metrics (Enrollment Summary by K-12 Students’ Neighborhood and Type of School Attended), October 2019 to October 2025 (current, archive). Method details: Sources & methods.

After the 2023 Southeast changes, capture in the Southeast elementary areas fell from 71% to 65%. It has stayed there (66% in 2025-26). Over the same years, capture in the rest of PPS rose from 72% to 73%. In the Southeast, the share attending PPS alternative and focus schools rose from 19% to 22%.

Some of this drop was built into the changes and doesn’t reflect families’ choices. For example, children allowed to finish at their old school after a boundary moved count as attending “another neighborhood school”. Also, Creative Science, a focus school, moved into the Bridger building. That put an alternative school inside the Southeast. PPS’s reports don’t break the drop down, so we can’t measure the share due to families’ choices. If the drop were mostly students finishing at their old schools, it should fade as they move up. Three years on, Southeast capture has recovered by only 0.6 percentage points.

The Lent area shows what a whole-school immersion site does to its neighborhood. Capture there fell from 63% to 30% when Lent converted. That drop is mostly by design, not a sign of families leaving. PPS assigned Lent’s non-immersion neighborhood children to Marysville, and its report counts them as attending “another neighborhood school”. Today, 80 of Lent’s 256 students live in its area, so most of its families travel from elsewhere. In both scenarios, three of the four K-5 Spanish sites would be whole-school immersion sites like this: Lent, Rigler and César Chávez.

7. PPS’s equity check misses these patterns

PPS’s equity check is one slide. It shows the share of kindergarten–8th grade students in each group who would change schools.

Table 10

Student group	Scenario A		Scenario B
Student group	Students	Share	Students	Share
All Students	7,981	27%	6,988	24%
Students Experiencing Poverty	2,648	27%	2,450	25%
Students Receiving Special Education Services	1,545	27%	1,379	24%
Multilingual Learners	1,006	31%	909	28%
Historically Underserved by Race	2,888	28%	2,731	26%
As printed by PPS. The slide gives no year and does not define "move" (for example, whether a student whose immersion program relocates counts). Groups overlap.
Sources: PPS Board of Education, 10/06/26 PowerPoint, slide 44, “Balance and Impact: Kindergarten – 8th Grade Students Who Move” (BoardBook). Method details: Sources & methods.

Even this table shows multilingual learners as the most-moved group (31% in A, 28% in B, against 27% and 24% overall). But it still understates the imbalance in three ways:

It averages each group over all students in kindergarten through 8th grade, and boundary changes move most of them. That hides what this page shows. Closures fall mostly on whiter, higher-income schools. Program moves fall on schools that are about half underserved, and on students far more likely to speak Spanish at home.
It counts a move to the next school over the same as a program moved across town.
It has no row for immersion students, Latino students or home language. It does not measure distance. It does not account for families who stop following a program.

8. Southeast Spanish immersion is too small for its demand

In Southeast Portland, PPS offers more kindergarten seats in Japanese than in Spanish. Richmond, a whole-school Japanese immersion school, enrolled 89 kindergartners in October 2025 (512 students in all). Southeast’s two Spanish programs, Lent and Atkinson, enrolled 70 between them. Le Monde, a French immersion charter school, enrolled 48. That is more than any Spanish program in the district except Rigler. District-wide, Spanish still has the most immersion seats. The gap is in the Southeast, home to the district’s most oversubscribed Spanish program.

The only other Spanish immersion option nearby is a private school. The International School of Portland is in South Waterfront, across the river from inner Southeast. It runs Spanish, Chinese and Japanese immersion tracks and charges tuition. It enrolled 376 students from preschool through 5th grade in 2023-24, including 72 kindergartners. It doesn’t publish enrollment by language (National Center for Education Statistics, Private School Universe Survey, 2023-24 (International School of Portland); grade counts as shown by US News; languages from the school’s website, accessed October 8, 2026).

Table 11

	Kindergartners, Oct 2025	Program students, Oct 2025	Kindergarten lottery, average per year
Southeast immersion: kindergarten seats and demand
	Kindergartners, Oct 2025	Program students, Oct 2025	Applications	Offers	Waitlisted	Applications per offer
Richmond (Japanese)	89	512	138	102	19	1.4
Le Monde (French, charter)	48	385	–	–	–	–
Lent (Spanish)	43	256	38	19	9	2.0
Atkinson (Spanish)	27	178	121	24	49	4.9
Lottery: 2022-23 to 2026-27 average. Le Monde is a charter school and runs its own admissions, so it isn't in the PPS lottery reports. Lent's lottery counts only families from outside its neighborhood (see the Demand & scorecard page). Applications are not children: a family can apply to several programs.
Sources: PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS Enrollment & Transfer, Elementary School / Kindergarten Lottery Results Summary, lotteries for 2022-23 through 2026-27 (prior lottery results). Method details: Sources & methods.

Richmond admits most of the families who apply. Atkinson admits about one in five. Atkinson’s waitlist alone, about 49 applications a year, is larger than its kindergarten class. Families can apply to several programs, so this doesn’t count children exactly. Still, it suggests that the waitlist alone could likely fill a second Atkinson class each year. PPS’s plan moves this program to Lent instead. Its school guide gives the reason: “Moving Atkinson DLI to Lent brings Lent’s student numbers above steady target levels (400+ students)” (PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page)).

Lent’s problem is its neighborhood, not its distance from the center. Lent sits about 0.3 miles from the district’s edge. Families describe students crossing “205, 92nd and 82nd Ave” to reach Marysville, the new home of Lent’s old neighborhood program (Green, T. L., & Hanson, H. (2026, June 29). Equity audit evaluation of Portland Public Schools’ Southeast Enrollment and Program Balancing initiative: School-by-school analysis (PPS SEGC documents), p. 46). But Rigler is also about 0.3 miles from the edge. It works as a whole-school Spanish immersion school that serves its own neighborhood: 215 of its 289 students live in its attendance area. At Lent, after PPS moved its neighborhood program out, 80 of 256 do.

Lent’s surroundings have plenty of Spanish speakers too. About 393 school-age Spanish speakers live within 1.5 miles of Lent, against 311 near Atkinson. What Lent lacks is the English-speaking families who fill Atkinson’s waitlist. They live closer to Atkinson.

Table 12

	Miles to district edge	Spanish immersion students	Live in school's own area	Share	School-age Spanish speakers within 1.5 mi
Two whole-school Spanish sites at the district's edge
October 2025; Atkinson for comparison
Rigler	0.3	289	215	74%	622
Lent	0.3	256	80	31%	393
Atkinson	1.8	178	60	34%	311
Distances are straight lines from the school's map location to the PPS boundary. School-age Spanish speakers are ACS estimates spread over census blocks (see methods).
Sources: PPS, School Enrollment by Neighborhood of Residence, October 2025 (PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; US Census Bureau, American Community Survey 2020–2024 5-year estimates, tables C16001 and B16007, census tracts in Multnomah, Washington and Clackamas counties; US Census Bureau, 2020 Census redistricting data (P.L. 94-171), tables P1 and P3, census blocks. Method details: Sources & methods.

The plan would make Rigler more like Lent. Rigler’s guide (p. 3) says it “would remain open as a focused regional Spanish DLI K-5 school”. It also says that “Rigler’s non-DLI neighborhood boundary would shift to Scott K-5” (PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page)). On the scenario maps, Rigler has its own attendance area today and none in either scenario (PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown). Lent went through the same change in 2023. It ended up with about half the enrollment PPS projected (527 projected, 256 in 2025).

PPS’s projection for Rigler doesn’t show that risk. It has Rigler growing from 306 students with no changes to 487 with Scott’s program, and still 505 in 2031-32. That is Rigler’s current enrollment plus the arriving program. The projection treats Rigler as if its neighborhood families keep choosing it after PPS assigns new kindergartners from its area to Scott. The guide also projects Rigler’s share of multilingual-learner students falling from 50% to 42%. It projects its share of students in poverty falling from 54% to 46%. The guides don’t say whether families in Rigler’s area would keep any priority to enroll there.

The buildings have room. Functional capacity is how many students a building can hold. The 2021 Long-Range Facility Plan defines it as a building’s classrooms times PPS’s standard class size, minus rooms set aside for other uses. Utilization is the share of that capacity in use. The table shows PPS’s current figures from its School Profiles dashboard, including 2025-26 utilization. It shows the older plan’s figures for comparison.

Table 13

	Functional capacity		October 2025		PPS projection, 2027-28 (Scenario A)
Building capacity at the Southeast Spanish sites
	2021 plan	2025-26	Students	Share of capacity	Students	Share of capacity
Atkinson	567	519	366	71%	395	76%
Creston	558	429	213	50%	closes	closes
Lent	707	798	256	32%	401	50%
Rigler	589	621	289	47%	487	78%
Shares use PPS's current (2025-26) capacity. PPS's space-use goal is 70–100% of available room (school guides). Creston's 2021 figure counted its annex. Enrollment is October 2025, kindergarten through grade 5, all programs; it matches the dashboard's enrollment for every school.
Sources: PPS, School Profiles, 2025-26 (Power BI dashboard: functional capacity, utilization, enrollment, direct certification, capture rate; page), retrieved October 9, 2026; PPS, Long-Range Facility Plan 2021, Volume 1, p. 53 (elementary) and p. 57 (K-8), functional capacity (PDF); compiled in meub/pps-data by Alex Meub (MIT license); PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page). Method details: Sources & methods.

Atkinson uses about 71% of its capacity. That leaves room for roughly 150 more students, about 25 per grade. That would fit a second Spanish class in every grade without moving anyone. It would not fit a full doubling of today’s program, which has 178 Spanish immersion students in K-5. (PPS’s 2021 plan gave Atkinson more room: 567 seats.) A whole-school Spanish program would not fit unless Atkinson’s 188 neighborhood-program students moved out. That is what went wrong at Lent.

Lent is the opposite case. Even after Atkinson’s program moves in, PPS projects Lent at 401 students, 50% of its capacity. That is still short of PPS’s own space-use goal, and Lent’s guide marks the goal as not met.

That points to two other ways to serve Southeast families, both closer to where the demand is:

Add a second Spanish class at Atkinson, filled from its waitlist, and keep its neighborhood program. This fits in the current building, as long as PPS doesn’t also send Creston’s neighborhood children there.
Make Creston’s building a Southeast Spanish immersion school. Creston closes in both scenarios, and PPS already plans to send its neighborhood children to Atkinson (PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page)). Moving Atkinson’s immersion program into Creston instead of Lent would move it 0.8 miles rather than 2.4. The building has room for about 72 students per grade at PPS’s current capacity of 429 (PPS, School Profiles, 2025-26 (Power BI dashboard: functional capacity, utilization, enrollment, direct certification, capture rate; page), retrieved October 9, 2026). That is more than twice Atkinson’s Spanish program today. Creston would need work first. It is an unreinforced masonry building that PPS has targeted for seismic retrofit (strengthening against earthquakes), according to the PPS facility data compiled in meub/pps-data.

Either way, Lent could get back a neighborhood program serving its own families, many of them Spanish speakers. One question remains. Two-way immersion needs native speakers. About 213 school-age Spanish speakers live within 1.5 miles of Creston, fewer than near Atkinson or Lent. So a Creston site would need to recruit them.

PPS’s case for whole-school sites is that teachers need same-grade partners. A second class per grade at Atkinson would give them one. See whole-school vs. strand research for what the evidence says about whole-school and strand programs.

9. Other options would take less from immersion families

The data point to options that would take less from immersion families. PPS should evaluate each of these before it decides:

Keep high-demand programs where the families are. Atkinson and Beach have more applicants than seats every year. Consolidate only where demand is low (James John, Sitton, Clark).
In the Southeast, grow Spanish immersion where the demand is. Add a second class at Atkinson, or open a Southeast immersion school in Creston’s building, 0.8 miles away instead of 2.4 to Lent. Restore Lent’s neighborhood program (§8).
Keep Rigler’s neighborhood boundary, the feature that makes it work today. At least guarantee its neighborhood families a seat.
Move the neighborhood program instead of the immersion strand. The plan already does this at César Chávez. There, non-immersion students would move to Rosa Parks so the school can become an immersion hub.
Prefer two-class strands to whole-school conversions where the goal is teacher collaboration. A two-class strand has two immersion classes in each grade. No study we found shows whole-school programs outperform strands (whole-school vs. strand research).
Fill “underenrolled” immersion classrooms from the waitlists instead of moving them.
Phase changes in and protect current families. Let enrolled students finish where they are. Guarantee transportation to any relocated program.
Do the equity analysis PPS hasn’t published. Report every change by immersion enrollment, home language, English-learner status, race and income. Include home-to-school distance before and after. Publish the share of families the projections assume will follow each moved program. Judging by the school guides, it is about 82%. Then track how many actually do. PPS has the student-level data to do this.

10. Some moves would send students to schools beside freeways

Both scenarios move Atkinson’s Spanish immersion program into Lent. Lent’s school grounds are about 37 meters from I-205. The US Environmental Protection Agency (EPA) says pollution from traffic is highest within about 150 meters (500 feet) of a busy road. It falls to background levels by about 600 meters (2,000 feet). In Scenario A, the plan would move groups from schools farther away into schools within 150 meters of a freeway. These groups enroll about 878 students today. In Scenario B, those groups enroll about 597 students.

A few specific moves drive this. In Scenario A, 46% of the students in these groups are from underserved racial groups. That compares with 36% of all students the plan moves and 35% of all K-8 students. In Scenario B, the share is 60%, against 39% of all students it moves. The pattern comes from which programs PPS chose to move, not from a district-wide trend:

Atkinson’s Spanish immersion program to Lent (both scenarios): 71% from underserved groups. Lent is 37 m from I-205.
Faubion’s grades 6-8 to Ockley Green (both): 70%. Ockley Green is 120 m from I-5.
Vernon’s grades 6-8 to Harriet Tubman (both): 40%. Tubman is 10 m from I-5.
Stephenson’s students to Markham and Capitol Hill (Scenario A only): 17%. Both schools are within 78 m of I-5.

K-8 schools near freeways already enroll somewhat more students from underserved groups. Today, 38% of the K-8 students at schools within 150 m of a freeway are from underserved racial groups. That compares with 35% of all K-8 students. The plan would widen that gap. After Scenario A, 42% of the students at schools within 150 m would be from underserved groups (44% after Scenario B). One move goes the other way: Odyssey would leave a building 55 m from US 26. Mandarin immersion would move from MLK Jr. to Boise-Eliot/Humboldt. Its grounds are 51 m from an I-5 ramp and 315 m from the freeway itself.

Today, 2,933 students (11%) attend schools within 150 m of a freeway. After Scenario A, schools that close to a freeway would enroll about 3,531 (13%). After Scenario B, they would enroll about 3,250 (12%). These counts use today’s K-8 students, like the rest of this page.

With high school students. Benson and Lincoln high schools are also within 150 m of a freeway. The plan doesn’t move any high school students to or from them. Counting their students, 5,605 students attend schools within 150 m today (6,203 after Scenario A, 5,922 after B). Of them, 35% are from underserved racial groups, about the same as all PPS students (34%).

Table 14: K-8, middle and alternative schools whose grounds are within 400 m of a freeway: distance, what each scenario does, and the share of students from underserved racial groups today (PPS’s guides). Student counts are October 2025 enrollment, a measure of the size of each change. Ramp distance is shown when a ramp is much closer than the mainline.

School	Grounds to freeway	Students today	Scenario A	Scenario B	Underserved today (PPS)
Harriet Tubman	10 m (I-5)	402	Takes in students (598 after)	Takes in students (598 after)	50%
Capitol Hill	23 m (I-5)	325	Takes in students (409 after)	No change	23%
West Sylvan	24 m (US 26)	683	No change	No change	24%
Lent	37 m (I-205)	256	Takes in students (434 after)	Takes in students (434 after)	72%
Odyssey	55 m (US 26)	217	Program moves to another building (0 after)	Program moves to another building (0 after)	12%
Markham	78 m (I-5)	339	Takes in students (536 after)	No change	40%
Ockley Green	120 m (I-5)	460	Takes in students and sends some away (620 after)	Takes in students and sends some away (620 after)	48%
Sunrise	130 m (I-205)	251	No change	No change	44%
Laurelhurst	222 m (I-84)	610	Sends some students away (375 after)	Sends some students away (375 after)	16%
Beverly Cleary	252 m (I-84)	480	Takes in students and sends some away (535 after)	Takes in students and sends some away (535 after)	16%
Jackson	276 m (I-5)	689	Boundary or feeder change only (689 after)	Boundary or feeder change only (689 after)	26%
Clark	279 m (I-205)	364	Sends some students away (273 after)	Sends some students away (273 after)	52%
Ainsworth	314 m (US 26)	560	No change	No change	36%
Boise-Eliot/Humboldt	315 m (I-5); ramp 51 m	294	Takes in students (487 after)	Takes in students (487 after)	74%
Kelly	318 m (I-205)	411	Boundary or feeder change only (411 after)	Boundary or feeder change only (411 after)	42%
Sources: freeway distances: OpenStreetMap contributors, freeways (highway=motorway and motorway_link) and school grounds (amenity=school), snapshot of October 13, 2026 via the Overpass API (ODbL); computed in data-raw/freeway_distance.R; where each school’s students would go: PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 3, “What is being proposed?” (page); every sentence used is in data-raw/plan_flows.csv; PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 4, “Who would the school serve?”, column “Underserved racial groups”, with no changes and under each option (page). Method details: Sources & methods.

Table 15: Race and ethnicity of the groups the plan would move into schools within 150 m of a freeway (from schools farther away), compared with all groups the plan moves and all K-8 students today. October 2025 enrollment; counts by program for immersion moves.

	Scenario A		Scenario B		All K-8 students today
	Moved within 150 m	All moved	Moved within 150 m	All moved	All K-8 students today
Latino	27%	21%	36%	23%	19%
Black	10%	7%	13%	7%	8%
Asian	2%	6%	1%	4%	5%
Native American	1%	1%	<1%	1%	1%
Pacific Islander	1%	1%	1%	1%	1%
White	47%	53%	37%	52%	53%
Multiracial: Asian and White	5%	5%	2%	5%	6%
Multiracial: other	8%	7%	10%	8%	7%
Underserved racial groups	46%	36%	60%	39%	35%
To protect privacy, groups with fewer than 5 students show as <5 or <X% (the share 5 students would be).
Sources: PPS, Enrollment Summary by Race and Ethnicity, October 2025 (separate rows for each school’s immersion and neighborhood programs; PDF), retrieved October 11, 2026; where each school’s students would go: PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 3, “What is being proposed?” (page); every sentence used is in data-raw/plan_flows.csv; freeway distances: OpenStreetMap contributors, freeways (highway=motorway and motorway_link) and school grounds (amenity=school), snapshot of October 13, 2026 via the Overpass API (ODbL); computed in data-raw/freeway_distance.R. Method details: Sources & methods.

Figure 8: Moves that start or end within 600 m of a freeway: distance from the current school’s grounds (open circle) to the new school’s grounds (diamond). The shaded band is within 150 m. Moves in both scenarios unless the hover says A only. Hover for students and distances.

Sources: freeway distances: OpenStreetMap contributors, freeways (highway=motorway and motorway_link) and school grounds (amenity=school), snapshot of October 13, 2026 via the Overpass API (ODbL); computed in data-raw/freeway_distance.R; where each school’s students would go: PPS, How Could Your School Be Impacted? school guides (per-school PDFs dated October 8, 2026), page 3, “What is being proposed?” (page); every sentence used is in data-raw/plan_flows.csv; US Environmental Protection Agency. (2015, November). Best practices for reducing near-road pollution exposure at schools, PDF pp. 3, 10–12, 16 (PDF). Method details: Sources & methods.

What we know about the air at Lent and Tubman

No one has measured the air at Lent the way Portland State University measured it at Harriet Tubman Middle School. That 2018 study found traffic pollution falling with distance from I-5, “reaching background levels about 200-300 feet from the freeway.” The study recommended that “Student outdoor activities be limited at HTMS, especially during high traffic periods,” and filters rated at least MERV 16.

For Lent, what exists is advocacy and neighborhood sampling, not school measurements:

In 2017, the Lents nonprofit Green Lents wrote to a state Senate committee: “There are several elementary schools (Lent and Kelly Elementary) within a few short blocks of I-205.”
In 2019, Neighbors for Clean Air told the Legislature that I-5 and I-205 flank “Tubman and Lent Elementary among others.” It added: “At Lent Elementary, parents are worried about their kids staying sick longer than normal, and having breathing problems.”
In 2016-17, a community project in Lents measured black carbon, a sign of diesel exhaust, with a handheld monitor. A partner said readings “far exceed the benchmarks that Oregon has set.” The news report names no school as a sampling site. It says the partner blamed old diesel engines, not the freeway.

PPS’s school guides don’t mention freeways or outdoor air. At the October 6 board meeting, Director Virginia La Forte said of Tubman: “we know that there are significant air quality concerns. That’s a fact. We have the tests, we have the studies, we know that that’s real.” We found no mention of air quality at Lent in the board meetings we reviewed.

Sources: Gall, E. T., George, L. A., Cal, R. B., & Laguerre, A. (2018, April 18). Indoor and outdoor air quality at Harriet Tubman Middle School and the design of mitigation measures: Phase I report. Portland State University, for PPS (PDF); Green Lents (Adam Brunelle, program director), letter to the Oregon Senate Rules Committee on SB 1008, 2017 (PDF); Neighbors for Clean Air (Tori Cole), testimony on HB 2007, Oregon Legislature, 2019 (PDF); Ashton, D. F. (2017, January). Diesel pollution studied in Lents. East Portland News (article); PPS Board of Education, regular meeting, October 6, 2026, Director Virginia La Forte at 4:01:33 (video; automated transcript via Save PDX Schools); PPS, How Could Your School Be Impacted? school guides (“PPS Family-Friendly School Guide”, per-school PDFs dated October 8, 2026), page 4, projected enrollment 2027-28 to 2031-32 (page). Method details: Sources & methods.

What schools near freeways can do

The EPA suggests steps for schools near busy roads:

Filters. Use “the highest MERV-rated filters that the HVAC system can handle.” Filters rated MERV 11 to 16 can cut particles indoors by 74% to 97% compared with outdoors.
Timing. “Plan strenuous outdoor activities outside of rush hour and drop-off/pick-up times.”
Layout. Put playgrounds, fields and classrooms farther from the road, and parking or storage closest to it.

We don’t know what filters or schedules Lent, Ockley Green, Markham or Capitol Hill use now.

Sources: US Environmental Protection Agency. (2015, November). Best practices for reducing near-road pollution exposure at schools, PDF pp. 3, 10–12, 16 (PDF). Method details: Sources & methods.

Limits

Grounds, not playgrounds. We measure from the edge of each school’s grounds as mapped by OpenStreetMap volunteers. Mapped grounds may include fields or parks. Lent’s building is about 138 m from I-205, farther than its grounds.
Distance only. We don’t model wind, hills, sound walls, traffic volume or trucks.
Freeways only. Busy streets such as 82nd Avenue, Powell, Lombard and Columbia Boulevard also carry heavy traffic. We don’t count them.
Today’s students. The groups are today’s students moved all at once, as on the rest of this page. The changes would start in 2027-28 or later.

Details: distance estimates and every change

Estimated home-to-school distance

Figure 6 in §4 shows the estimate for each change. The model places children by 2020 census block, a small area the Census reports on. For closures, it uses the children living in the closing school’s attendance area. Their new school is the one the scenario map assigns. For immersion programs, it uses the children nearest the program’s current site for that language. This is an assumption. Their new school is the program’s new site.

Table 16

Version of the estimate	Today	After	Increase
Estimated home-to-school distance, miles: how sensitive is it?
Model estimate; averages over affected students
Scenario A
Immersion moves: all (main estimate)	1.8	2.8	1.0
Immersion moves: excluding Vietnamese (one site, whole-district catchment)	1.5	2.5	1.1
Immersion moves: Spanish only	1.3	2.5	1.2
Immersion moves: Spanish only, weighted by Spanish-speaking children	1.1	2.2	1.0
Closures: all (main estimate)	0.5	1.1	0.5
Closures: weighted by Spanish-speaking children	0.7	1.0	0.4
Scenario B
Immersion moves: all (main estimate)	1.5	2.5	1.1
Immersion moves: excluding Vietnamese (one site, whole-district catchment)	1.5	2.5	1.1
Immersion moves: Spanish only	1.3	2.5	1.2
Immersion moves: Spanish only, weighted by Spanish-speaking children	1.1	2.2	1.0
Closures: all (main estimate)	0.6	1.0	0.5
Closures: weighted by Spanish-speaking children	0.7	1.0	0.3
All versions assume immersion families live in the area nearest their program's current site of that language; that assumption cannot be tested without student addresses. "Weighted by Spanish-speaking children" places children by ACS school-age Spanish speakers rather than all children.
Sources: PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; US Census Bureau, 2020 Census redistricting data (P.L. 94-171), tables P1 and P3, census blocks; US Census Bureau, American Community Survey 2020–2024 5-year estimates, tables C16001 and B16007, census tracts in Multnomah, Washington and Clackamas counties; PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting). Method details: Sources & methods.

Checked against where students live. PPS’s School Enrollment by Neighborhood of Residence report counts how many of each program’s students live in the school’s own attendance area. For the rest, it gives the high-school area they live in. We place those students on census blocks within that area in two ways: spread over all the area’s children, or over the half living closest to the program’s current school.

Table 17

Move	Students	Live in school's own area	From where students live (PPS report)			Model
Estimated home-to-school distance: model vs where students live (Scenario A, miles)
Move	Students	Live in school's own area	Today	Increase (area average)	Increase (nearer half)	Today	Increase
Scott → Rigler (Spanish, K-5)	222	82%	0.7	0.3	0.3	1.0	0.4
Rose City Park → Vestal (Vietnamese, K-5)	207	33%	1.7	0.4	0.7	4.2	0.5
MLK Jr. → Boise-Eliot/Humboldt (Mandarin, K-5)	193	21%	1.6	0.4	0.5	2.7	0.0
Atkinson → Lent (Spanish, K-5)	178	34%	1.5	1.0	1.6	1.7	1.4
Beach → César Chávez (Spanish, K-5)	155	47%	1.4	1.4	1.6	1.3	1.8
Sitton → César Chávez (Spanish, K-5)	138	86%	0.7	1.1	1.5	0.9	1.7
James John → César Chávez (Spanish, K-5)	133	72%	1.1	0.3	0.9	1.8	0.7
César Chávez → George (Spanish, 6-8)	300	38%	1.1	0.3	0.7	0.7	0.7
Beaumont → Roseway Heights (Spanish, 6-8)	101	73%	0.9	1.0	1.1	1.2	1.1
Clark → Woodstock (Mandarin, K-5)	91	78%	1.0	1.5	2.0	1.7	1.5
Ockley Green → George (Spanish, 6-8)	63	83%	1.1	2.3	2.7	1.5	2.7
Students: all students in the sending program, October 2025 (César Chávez's row covers grades K-8). Students living outside their school's area are known only by high-school area; Jefferson's three sub-areas are approximated by which of the Grant, McDaniel and Roosevelt areas is nearest. Out-of-district and undetermined addresses are left out. Straight-line distances.
Sources: PPS, School Enrollment by Neighborhood of Residence, October 2025 (PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown; US Census Bureau, 2020 Census redistricting data (P.L. 94-171), tables P1 and P3, census blocks. Method details: Sources & methods.

Every closure and immersion move

Table 18

Change	Grades	Students	Spanish at home	Non-English at home	School moves (mi)
Every closure and immersion move, Scenario A
Schools that close
Woodmere	K-5	227	9%	22%	0.5
James John	K-5	232	9%	17%	0.7
Marysville	K-5	307	5%	16%	0.6
Creston	K-5	213	4%	13%	0.8
Peninsula	K-5	239	3%	5%	0.7
Irvington	K-5	259	2%	4%	1.0
Buckman	K-5	283	2%	4%	1.0
Beach	K-5	133	2%	8%	1.1
Sabin	K-5	269	1%	4%	0.7
Maplewood	K-5	287	1%	4%	0.9
Rose City Park	K-5	210	1%	2%	1.2
Lewis	K-5	259	1%	3%	0.7
Stephenson	K-5	281	1%	5%	1.5
Sellwood	6-8	564	1%	3%	2.8
Mandarin immersion moves
MLK Jr. → Boise-Eliot/Humboldt	K-5	193	3%	8%	1.0
Clark → Woodstock	K-5	91	1%	63%	3.0
Spanish immersion moves
César Chávez → George	6-8	105	62%	64%	1.1
Beaumont → Roseway Heights	6-8	101	54%	57%	1.7
Sitton → César Chávez	K-5	138	54%	55%	2.2
James John → César Chávez	K-5	133	52%	52%	1.6
Atkinson → Lent	K-5	178	40%	41%	2.4
Scott → Rigler	K-5	222	36%	39%	0.7
Beach → César Chávez	K-5	155	21%	21%	2.5
Ockley Green → George	6-8	63	16%	17%	3.4
Vietnamese immersion moves
Rose City Park → Vestal	K-5	207	0%	41%	1.6
Students: October 2025 enrollment; for immersion moves, students in the moving language program in the grades that move; for closures, the school's non-immersion students (its immersion students are counted under the program move). Home language: share of those students. School moves: straight-line distance from the current school to the new one; for closures split among several receiving schools, weighted by how much of the closing school's attendance area each receives.
Sources: PPS, Enrollment by Grade and Program Type, October 2024 and October 2025 (PPS Data & Reporting); PPS, Enrollment by Language and School, October 2024 and October 2025 (PPS Data & Reporting); PPS board packet for October 6, 2026: scenario maps and Rightsizing Update: Scenario Release memo, as digitized and transcribed in pps-explorer by Jason Brown. Method details: Sources & methods.

Limits of this analysis

Lottery figures count applications, not children.
Race and income are whole-school figures; PPS doesn’t publish them by program.
Home-to-school distances and the number of families who don’t follow are estimates.
This page does not count students moved by boundary changes.
This is an independent analysis, not a PPS publication.

See Sources & methods for full definitions.