Read the report

SECOND LANGUAGE PROJECT • RESEARCH EDITION 1.1 · PRINT PAGE 1

Languages of the Bay Area

Home, learning, work and community access

Nine counties. Many language repertoires. Different needs in different settings.

Prepared for the Second Language Project
22 October 2026
English edition • 40 pages • public aggregate data

2020-2024 American Community Survey
2025-26 California English learner records
District documents and selected research literature

01 / READING THE EVIDENCE · PRINT PAGE 2

Start with the question

About 3.18 million Bay Area residents age 5 and older report a language other than English at home: 43.9% of that population. This includes people with many different levels of English ability. [C01] [C02] [C03]

The most useful headline is linguistic diversity. A ranking alone cannot decide which language a person prefers, which language they learned first, or what support they need for a particular conversation.

01 / READING THE EVIDENCE · PRINT PAGE 3

Three perspectives, three denominators

[C01] [C07] [D02]

School counts are attached to enrollment records and school locations. ACS counts concern residents. A child may cross a district boundary to attend school. Neither population can be substituted for the other.

Workplace language use would require a defined population of workers, jobs and locations. Bilingual position requirements show institutional demand; they do not count everyone who uses a language at work.

02 / HOME AND PRIMARY LANGUAGE · PRINT PAGE 4

What does “primary language” mean?

[C04] [AL03] [AL04]

These answers can differ within one person. This report uses “primary home language” only for an instrument that actually asks about most-frequent home use. It allows ties rather than forcing one language.

Research on bilingual profiles treats history, current use and proficiency as separate dimensions. Earliest acquisition does not establish current dominance. The same person can be comfortable speaking a language and less comfortable reading it. [AL03] [AL04] [AL06]

02 / HOME AND PRIMARY LANGUAGE · PRINT PAGE 5

What the ACS actually asks

The ACS asks whether a person sometimes or always speaks a language other than English at home, which language, and how well the person speaks English. If several non-English languages are used, instructions select the one used most often; if that cannot be determined, the first learned. This does not establish whether that language is used more often than English. [C04]

Standard coding retains one non-English language. Research on multiple write-ins demonstrates why the published field cannot describe a person’s entire repertoire. A Census conference poster explores this limitation; it is research, not a peer-reviewed regional estimate. [C06]

“English only” means only English reported at home. It does not mean the person knows no other language. “Non-English at home” does not mean English is absent or that the person needs an interpreter. [C04] [C05]

03 / LANGUAGES AT HOME · PRINT PAGE 6

A regional snapshot

[C01] [C02] [C03]

The region is Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano and Sonoma counties. We reconstruct it from 62 whole 2020 Public Use Microdata Areas (PUMAs), all verified to nest within those counties. County results in this guide are derived from the same PUMAs. [C03]

The five-year estimate describes a period, not the population on one day in 2024 or 2026. Counts are survey estimates, not a complete enumeration. The printed precision supports reproducibility; it should not imply exact head counts.

03 / LANGUAGES AT HOME · PRINT PAGE 7

Top 20 reported categories

[C01] [C02] [C03]

03 / LANGUAGES AT HOME · PRINT PAGE 8

What a ranking leaves out

Chinese combines Mandarin, Cantonese and other Chinese varieties. Persian includes Farsi and Dari. Several categories combine distinct languages, such as Ilocano, Samoan, Hawaiian and other Austronesian languages. Their combined rank cannot be assigned to any one member. [C02]

The dataset separates Hindi, Telugu, Tamil and Punjabi, but combines Nepali and Marathi with other Indic languages, and Malayalam and Kannada with other Dravidian languages. Category design changes the apparent league table. Preserve these labels when comparing sources. [C02]

A lower regional count does not imply a less important access need. A language may be concentrated in one neighborhood, one school or one service audience. A small language community can be missed by a top-20 cutoff.

ACS does not provide a separate count of American Sign Language users in this table. It also does not enumerate languages known but unused at home. Absence from this ranking is not evidence of absence from the community. [C05]

03 / LANGUAGES AT HOME · PRINT PAGE 9

The regional average hides local patterns

[C01] [C03]

The share ranges from about 24% in Marin to about 55% in Santa Clara. These are shares of each county’s age-5+ population, not each county’s contribution to the region.

Language-access planning should examine local concentrations as well as regional totals. The explorer offers all nine county aggregates and 62 PUMAs. PUMAs are statistical areas; they are not school districts or necessarily city boundaries.

03 / LANGUAGES AT HOME · PRINT PAGE 10

Three useful local comparisons

[C01] [C02] [C03]

In San Francisco, the Chinese category exceeds Spanish among reported non-English home languages. Santa Clara accounts for about 69% of the region’s Vietnamese category. These geographic patterns do not establish an individual’s preferred communication language.

Chinese is a combined category here. District records can distinguish Cantonese and Mandarin, but their school populations differ from the resident ACS population. More specific labels do not make the two datasets directly interchangeable.

03 / LANGUAGES AT HOME · PRINT PAGE 11

English ability is a separate measure

Among residents reporting a non-English home language, 1,224,228 report speaking English less than “very well”: 38.5% of that group, or 16.9% of all residents age 5+. This is a self- or household-reported survey category. [C01] [C04]

The published split separates “very well” from the combined “well,” “not well,” and “not at all” responses. It concerns speaking English, not reading, writing, translation skill or a tested level of proficiency. [C04]

For reproducibility, the data store names the field english_less_than_very_well. The reader labels its denominator explicitly. We do not convert it into a count of people who always require an interpreter.

Nor does this split measure “English as a second language.” ACS does not establish language acquisition order. Someone reporting English very well may still be learning another language. [C04] [AL03]

04 / STUDENTS AND FAMILIES · PRINT PAGE 12

Schools ask different questions

School language evidence can describe a student’s background, English learner status, instructional needs or a family’s communication preferences. These are related but distinct fields. Home Language Surveys help identify students for English-language assessment; they do not by themselves test proficiency. [D01] [D02]

CALPADS primary language is a stable administrative designation informed by early and home language information. It is not an annually updated measure of current dominance. Castro Valley and San Leandro surveys distinguish first spoken language, most frequent child home language, adult-child language and adults’ home language. [D04] [D01] [D05]

[D01]

A district-wide ranking needs all acquisition statuses, the coding rule and enrollment denominator. Family communication preferences need separate fields.

04 / STUDENTS AND FAMILIES · PRINT PAGE 13

Castro Valley: read the labels carefully

The 2021-2024 district plan’s language analysis makes the importance of home-language evidence clear. Its 2020-21 acquisition-status chart lists English Only 68.12%, EL 9.66%, initially fluent 6.18% and reclassified fluent 16.05%. [D01]

A nearby “Ten Most Commonly Spoken Languages in CVUSD” chart lists Cantonese 917, Spanish 686 and Mandarin 312. An adjacent chart refers to English learners but includes English. The population labels are not sufficiently clear to treat those figures as a current all-student or EL ranking. They are preserved as historical evidence, with this ambiguity. [D01]

[D02]

04 / STUDENTS AND FAMILIES · PRINT PAGE 14

San Leandro: an EL population profile

[D02]

The CDE file yields 1,992 English learners in San Leandro Unified. Spanish accounts for 1,274 of these records (about 64%). Cantonese, Mandarin, Arabic and Vietnamese also appear prominently.

“Uncoded languages” is a published residual category. It is not one language. Do not distribute it among named languages without further evidence, or assume its students share the same communication needs.

SLUSD’s plan describes Spanish and Mandarin dual-language immersion for English learners and English-dominant students. This demonstrates a learning pathway, not a count of all additional-language speakers. [D08]

04 / STUDENTS AND FAMILIES · PRINT PAGE 15

A regional school-language lens

[D02] [D03]

The nine-county aggregation totals 166,614 English learners across 172 district entries. Spanish represents about 72.3% of EL records. These proportions use EL enrollment, not total enrollment or the resident population.

Language codes are more detailed than ACS in some places and less detailed in others. For example, the school file separates Mandarin and Cantonese, and also includes “Philippine languages” and other residual categories. Keep the published categories intact.

04 / STUDENTS AND FAMILIES · PRINT PAGE 16

Let families answer the access question

Proposed practice: record the student’s language background separately from each caregiver’s communication preferences. One household may contain people who need different languages or formats.

SLUSD’s 2022-23 LCAP planned parent preferred-language settings in Aeries. This is historical evidence of a separate communication field, not a current preference distribution. Offer oral and accessible responses; keep nonresponse distinct from “English preferred.” [D09]

Preserve year, respondent, grade range, population and collection purpose. CDE updated reclassification guidance in September 2026; historical district procedures should not be treated as current rules. [D10]

05 / ADDITIONAL LANGUAGES · PRINT PAGE 17

“Second language” is only a starting point

For this guide, an additional language is one a person learns or develops alongside their earliest language or languages. This editorial term includes people learning a third or fourth language. It does not impose a single acquisition sequence on everyone.

The CEFR Companion Volume describes an interconnected repertoire with uneven and changing strengths. A learner can have different levels of listening, speaking, reading, writing and interaction. A single fluent/not-fluent label can hide useful capabilities. [AL02]

LEAP-Q research examines language history and self-reported experience alongside proficiency. Its original validation involved particular adult samples; a local abbreviated survey does not automatically inherit that validation. [AL03] [AL06]

SLP can separately describe existing language resources, desired learning, and access needs. An interest in learning English does not mean a person can already participate fully in an English-only service.

05 / ADDITIONAL LANGUAGES · PRINT PAGE 18

Recognize what learners already bring

Heritage-language development connects learning with family and intergenerational life. UCLA’s National Heritage Language Resource Center emphasizes existing learner resources and goals such as literacy and formal-register development. Heritage learners are not necessarily beginners. [AL05]

These profiles are illustrations, not measured Bay Area population groups. The Bilingual Language Profile distinguishes history, use, proficiency and attitudes, including different settings and skills. This supports asking about actual activities rather than assuming ability from identity. [AL04]

05 / ADDITIONAL LANGUAGES · PRINT PAGE 19

From population evidence to learning choices

Spanish is the largest reported non-English home-language category across the nine counties. Practical Spanish is therefore a reasonable program-development priority to investigate. Population size alone does not establish enrollment demand, teacher availability or the best learning format. [C01]

Vietnamese, Tagalog/Filipino, Japanese, Korean, Hindi and Persian/Farsi are all visible in the regional evidence. Chinese varieties and several South Asian languages also deserve attention. The exact list should remain open to local voices and staffing capacity.

06 / WORKPLACE LANGUAGE · PRINT PAGE 20

Language at work needs its own evidence

A resident can use one language at home, another with coworkers, and another with customers. Work may also involve reading, writing, interpretation or signing. The ACS home-language measure does not answer these workplace questions. [C04]

This review did not establish a representative nine-county ranking of languages used at work. Job advertisements, bilingual pay assignments and service requests are useful institutional indicators, but none counts all workers’ language use.

Statistics Canada provides a useful design comparison: it distinguishes languages used most often at work from other regular use and allows ties. Its definitions concern work tasks, not merely the physical location of a person working from home. These are methodological examples, not Bay Area statistics. [AL07] [AL08]

06 / WORKPLACE LANGUAGE · PRINT PAGE 21

A concrete workplace case: San Francisco

The San Francisco Budget and Legislative Analyst’s April 2025 report lists 1,157 positions with language special conditions in Exhibit 6. Spanish and Chinese account for most of these designated positions. The report’s staffing discussion concerns FY2022-23. [C07]

[C07]

This is a case study of institutional staffing demand. A position is not a person, a filled job, a service encounter or an estimate of all workplace speakers. The report also discusses a broader certified bilingual employee pool; do not mix those denominators.

07 / COMMUNITY LANGUAGE ACCESS · PRINT PAGE 22

Language access begins with participation

For a listening session, school meeting or neighborhood program, the planning question is practical: can people understand, express themselves, ask questions and follow up in a way that works for them?

Recommendations should be tested with actual participants. Translated publicity may improve awareness, but it does not by itself make the event accessible. Interpretation, written translation and accessible communication formats address different parts of participation.

The MTC’s 2023 Title VI report provides a regional transportation planning example. Its audience, service context and period differ from this guide’s resident ranking. It belongs in the source library as institutional context, not as a substitute population estimate. [C10]

07 / COMMUNITY LANGUAGE ACCESS · PRINT PAGE 23

Include language, modality and context

Language assistance and language learning serve different purposes. A class develops skills over time. Interpretation or translation supports participation now. One should not be treated as proof that the other is unnecessary.

Effective communication can involve sign-language interpretation, captioning, accessible electronic text, large print or other aids. DOJ’s ADA guidance emphasizes the person’s usual method of communication and the nature, length and complexity of the interaction. Do not assume every Deaf person uses ASL or that all communication access is spoken-language translation. [C08]

For complex or consequential exchanges, the guide recommends qualified support appropriate to the task. Casual conversational ability does not demonstrate professional interpretation or translation competence. [C08]

08 / PROPOSED RESEARCH DESIGN · PRINT PAGE 24

A better local survey: home and preference

The following are original proposed fields, not a validated questionnaire. Pilot wording with community reviewers and make participation optional. Offer oral or accessible completion as well as written forms.

Use a consistent person-level unit. If asking about a household, explicitly allow differences between members. Record “not answered” separately from “none” or “English only.”

08 / PROPOSED RESEARCH DESIGN · PRINT PAGE 25

Add skills, work and learning goals

The proposed dimensions draw on research instruments that separate language history, use and proficiency. Their inclusion here is a design synthesis. We are not reproducing or validating the LEAP-Q or Bilingual Language Profile. [AL03] [AL04] [AL06]

A convenience sample can describe its respondents’ interests. It cannot establish region-wide prevalence without a defensible sample design and weighting. Report who was invited, who responded, and whose voices may be missing.

08 / PROPOSED RESEARCH DESIGN · PRINT PAGE 26

Rules for responsible analysis

For multiple-response survey questions, percentages may sum above 100%. Say so explicitly. For ACS B16001, the reported categories are mutually exclusive and sum to the age-5+ total. These are different data structures.

This publication uses only public aggregate data. A future local study should establish consent, access controls and retention practices before collecting personal information. It should avoid unnecessary names, exact employers or detailed demographic combinations.

09 / METHODS AND REPRODUCIBILITY · PRINT PAGE 27

How the regional estimates were built

We use the Census Bureau’s table-based 2020-2024 ACS five-year B16001 file and its variable dictionary. Published PUMA rows were selected using the official 2020 PUMA-to-county-subdivision relationship file. No population allocation or proportional splitting was applied. [C01] [C02] [C03]

All 62 selected PUMAs fall entirely within one of the nine target counties. The script verifies complete inclusion and absence of outside-county areas, then sums the same disjoint areas for each county and the region.

09 / METHODS AND REPRODUCIBILITY · PRINT PAGE 28

Uncertainty is part of the finding

For a sum of geographic estimates, this guide approximates the 90% count margin of error as the square root of the sum of squared component margins of error. This conventional approximation does not model covariance between estimates. [C12]

The explorer retains count MOEs for each category. Percentages use the corresponding age-5+ point-estimate total. We do not compute share confidence intervals, significance tests or rank confidence intervals. Do not use count MOEs as if they were percentage-point MOEs.

Adjacent ranks can be uncertain. “Rank 10” means tenth by the point estimate, not statistically greater than every category below it. Small-area estimates can be particularly imprecise.

The five-year period improves coverage but combines observations across years. A change comparison needs compatible categories and geographies, and should account for overlapping survey periods. This edition does not estimate growth rates.

09 / METHODS AND REPRODUCIBILITY · PRINT PAGE 29

How school records were aggregated

The CDE 2025-26 school-by-language EL file contains 59,111 statewide rows. Filtering the nine counties selects 15,630 rows, covering 1,764 school or office codes and 172 district entries. These records aggregate to 166,614 EL students. [D02] [D03]

The validation checks nonnegative integer values, duplicate school-language keys, grade sums against TOTAL_EL, and agreement across county, district and language totals. No duplicate school-language rows or arithmetic failures were found.

The source includes 44 rows explicitly labeled District Office with a zero school code, totaling 166 EL students. We retain them as published entries; they are small enrollment records rather than district-wide subtotals. These records are kept distinct from school-level entries in the analysis.

For the current all-student language picture, obtain a district export across acquisition statuses and documentation of its Home Language Survey coding. Preserve school year and the district’s enrollment population.

09 / METHODS AND REPRODUCIBILITY · PRINT PAGE 30

What is supported, and at what level?

[C01] [D02] [C07]

This is a targeted evidence review, not a systematic review. Sources were selected for direct relevance, primary authority and accessible definitions. We prioritized official data, original research and institutional methods over secondary rankings.

10 / NEXT RESEARCH · PRINT PAGE 31

What would strengthen the next edition?

First, obtain current district-wide language and caregiver-preference data. Castro Valley’s historical chart needs a clarified denominator; current EL records alone cannot answer the primary-language question. An October 2026 announcement records approval of a new multilingual plan, which we could not retrieve. [D01] [D07]

Second, conduct a carefully piloted local needs assessment. Measure most-used home languages, full repertoires, communication preferences and learning interests separately. Document recruitment so readers can judge how far the findings generalize.

Third, collect comparable institutional service indicators: requests, completed assistance, unmet requests and participant feedback, using the same period and audience. Staffing counts can then be interpreted alongside actual service demand.

Independent methodological review and community feedback can refine terminology, source interpretation and proposed survey questions. Check the edition date when comparing findings, since later evidence may change the interpretation.

10 / APPLYING THE GUIDE · PRINT PAGE 32

Using this evidence in practice

Use this guide to identify questions, plan outreach and start conversations with residents. Regional patterns can help you decide where to look more closely; local experience helps explain what support people want.

Until more local evidence is available, avoid treating regional rankings as a service plan or a home-language category as an individual preference. Later analyses may refine the picture; document the population, period and uncertainty behind each decision.

11 / SOURCE LIBRARY • 1 OF 4 · PRINT PAGE 33

Sources and reading notes

The links below lead to original public sources. SLP’s prepared data exports require a key; these source references remain openly accessible.

{
  "id": "C01",
  "title": "2020-2024 ACS B16001 table-based summary file",
  "organization": "U.S. Census Bureau",
  "date": "2026-01-29 file listing",
  "url": "https://www2.census.gov/programs-surveys/acs/summary_file/2024/table-based-SF/data/5YRData/acsdt5y2024-b16001.dat",
  "locator": "B16001 E001-E128 and M001-M128; 795P200 PUMA rows",
  "type": "official statistics",
  "local": "data/raw/acsdt5y2024-b16001.dat",
  "use": "Regional, county-derived and PUMA home-language estimates; not primary language or L2 totals",
  "accessed": "2026-10-22"
}

{
  "id": "C02",
  "title": "B16001 variable definitions",
  "organization": "U.S. Census Bureau",
  "date": "2024 dataset",
  "url": "https://api.census.gov/data/2024/acs/acs5/groups/B16001.html",
  "locator": "Labels for estimates and margins of error",
  "type": "data dictionary",
  "local": "data/raw/variables.json",
  "use": "Preserve exact grouped language definitions",
  "accessed": "2026-10-22"
}

{
  "id": "C03",
  "title": "2020 PUMA to county subdivision relationship file",
  "organization": "U.S. Census Bureau",
  "date": "2020 geography",
  "url": "https://www2.census.gov/geo/docs/maps-data/data/rel2020/puma520/tab20_puma520_cousub20_natl.txt",
  "locator": "California PUMA and county-subdivision GEOIDs",
  "type": "geographic crosswalk",
  "local": "data/raw/puma-county.txt",
  "use": "62 whole PUMAs nest within exactly nine counties",
  "accessed": "2026-10-22"
}

{
  "id": "C04",
  "title": "2024 ACS Subject Definitions",
  "organization": "U.S. Census Bureau",
  "date": "2024",
  "url": "https://www2.census.gov/programs-surveys/acs/tech_docs/subject_definitions/2024_ACSSubjectDefinitions.pdf",
  "locator": "Language Spoken at Home and Ability to Speak English",
  "type": "official methodology",
  "local": "research/raw/acs-definitions.pdf",
  "use": "Definition and limitations of language concepts",
  "accessed": "2026-10-22"
}

{
  "id": "C05",
  "title": "Frequently Asked Questions About Language Use",
  "organization": "U.S. Census Bureau",
  "date": "2021-12-16 page revision",
  "url": "https://www.census.gov/topics/population/language-use/about/faqs.html",
  "locator": "Language questions; ASL; household language",
  "type": "official methodology",
  "local": "research/raw/census-faq.html",
  "use": "English-only, proxy response and ASL coverage limits",
  "accessed": "2026-10-22"
}

{
  "id": "C06",
  "title": "Examining Multilingualism in the United States Using ACS Language Write-ins",
  "organization": "Daniela Mejía, U.S. Census Bureau",
  "date": "2024-04",
  "url": "https://www.census.gov/content/dam/Census/newsroom/press-kits/2024/paa/paa2024-poster-multilingualism-acs-language-write-ins.pdf",
  "locator": "Background and data; methods; research disclaimer",
  "type": "research conference poster; not peer-reviewed article",
  "local": "research/raw/multilingual-writeins.pdf",
  "use": "Standard coding cannot describe full language repertoires",
  "accessed": "2026-10-22"
}

{
  "id": "C12",
  "title": "Calculating Margins of Error the ACS Way",
  "organization": "U.S. Census Bureau",
  "date": "2020",
  "url": "https://www.census.gov/data/academy/webinars/2020/calculating-margins-of-error-acs.html",
  "locator": "Derived estimates and uncertainty",
  "type": "official methodology",
  "use": "Approximate count MOE; point ranks do not establish statistically significant differences",
  "accessed": "2026-10-22"
}

11 / SOURCE LIBRARY • 2 OF 4 · PRINT PAGE 34

Sources and reading notes

{
  "id": "C07",
  "title": "Language Access Services",
  "organization": "San Francisco Budget and Legislative Analyst",
  "date": "2025-04-14",
  "url": "https://sfbos.archive.sf.gov/sites/default/files/BLA.Language_Access.041425.pdf",
  "locator": "Exhibit 6; staffing and language-access analysis",
  "type": "local government analysis",
  "local": "research/raw/sf-bla.pdf",
  "use": "Institutional language staffing is distinct from all workplace use",
  "accessed": "2026-10-22"
}

{
  "id": "C08",
  "title": "ADA Requirements: Effective Communication",
  "organization": "U.S. Department of Justice",
  "date": "2020-02-28 update",
  "url": "https://www.ada.gov/resources/effective-communication/",
  "locator": "Auxiliary aids; qualified interpreters; context and communication methods",
  "type": "official guidance",
  "local": "research/raw/ada.html",
  "use": "Include signing, captioning and accessible formats; context matters",
  "accessed": "2026-10-22"
}

{
  "id": "C09",
  "title": "Limited English Proficiency notice",
  "organization": "U.S. Department of Justice",
  "date": "2025-07-17 notice",
  "url": "https://www.lep.gov/language-access-plans",
  "locator": "EO14224 and suspended operations notice",
  "type": "current policy-status notice",
  "local": "research/raw/lep.html",
  "use": "Do not present older federal language-access guidance as unchanged",
  "accessed": "2026-10-22"
}

{
  "id": "C10",
  "title": "MTC Title VI Triennial Report 2023",
  "organization": "Metropolitan Transportation Commission",
  "date": "2023",
  "url": "https://mtc.ca.gov/sites/default/files/documents/2023-11/Title_VI_Triennial_Report_2023_MTC_FTA.pdf",
  "locator": "Language assistance plan and county language context",
  "type": "regional administrative report",
  "local": "research/raw/mtc.pdf",
  "use": "Transportation language-access planning context; not 2024 estimates",
  "accessed": "2026-10-22"
}

{
  "id": "C11",
  "title": "DataQuest",
  "organization": "California Department of Education",
  "date": "live data portal",
  "url": "https://www.cde.ca.gov/Ds/ad/dataquest.asp",
  "locator": "English learner and enrollment reports",
  "type": "official education data portal",
  "use": "District statistics require school year and exact population definition",
  "accessed": "2026-10-22"
}

{
  "id": "D01",
  "publisher": "Castro Valley Unified School District",
  "title": "2021–2024 Plan for English Language Learners",
  "url": "https://files.smartsites.parentsquare.com/8941/2021-2024_plan_for_english_language_learners.pdf",
  "period": "2021–2024 plan; chart observations 2020–21",
  "accessed": "2026-10-22",
  "locator": "Printed pp.9–11 charts (PDF pp.10–12); printed pp.19–20 identification and HLS (PDF pp.20–21)",
  "type": "district plan",
  "status": "downloaded and chart pages visually inspected",
  "localFile": "cvusd-2021-2024-plan.pdf",
  "supports": [
    "Historical English language acquisition status percentages",
    "Four-dimensional home language survey"
  ],
  "limitations": "Printed p11 language-count list has an ambiguous denominator alongside chart labeled EL that includes English; do not reuse as a verified EL/all-student distribution. Superseded plan; not current procedural authority.",
  "organization": "Castro Valley Unified School District",
  "date": "2021–2024 plan; chart observations 2020–21"
}

{
  "id": "D02",
  "publisher": "California Department of Education",
  "title": "English Learners by Grade & Language — elsch25",
  "url": "https://dq.cde.ca.gov/dataquest/dlfile/dlfile.aspx?cLevel=School&cYear=2025-26&cCat=EL&cPage=fileselsch",
  "landingUrl": "https://www.cde.ca.gov/ds/ad/fileselsch.asp",
  "published": "2026-04-16",
  "period": "2025–26",
  "accessed": "2026-10-22",
  "type": "administrative TSV dataset",
  "status": "downloaded and computed",
  "localFile": "elsch25.txt",
  "sha256": "656fa33260bcecaf8e2f0713b1aebd22ae67ae3e5250b9052789f05473b29e40",
  "unit": "students classified EL, by school and reported language",
  "geography": "school/district county attribution; nine selected counties",
  "supports": [
    "Regional, county and district EL counts by language"
  ],
  "limitations": "EL-only; excludes IFEP/RFEP, general English-only and parent language preferences; 44 District Office rows with 166 EL records retained; seven TK records in selected data; no sampling margins of error.",
  "organization": "California Department of Education",
  "date": "2025–26"
}

11 / SOURCE LIBRARY • 3 OF 4 · PRINT PAGE 35

Sources and reading notes

{
  "id": "AL02",
  "authors": [
    "Council of Europe"
  ],
  "title": "Common European Framework of Reference for Languages: Learning, teaching, assessment – Companion volume",
  "date": "2020-04",
  "type": "official educational framework",
  "url": "https://rm.coe.int/cefr-companion-volume-with-new-descriptors-2020/16809ea0d4",
  "locator": "Section 2.3, printed p.30; Section 2.7, printed pp.38–40",
  "access_status": "full PDF available; relevant sections read",
  "applicability": "Uneven and changing repertoires; proficiency profiles",
  "limitations": "No prevalence estimates; not an ACS scale crosswalk.",
  "accessed": "2026-10-22",
  "organization": "Council of Europe"
}

{
  "id": "AL03",
  "authors": [
    "Viorica Marian",
    "Henrike K. Blumenfeld",
    "Margarita Kaushanskaya"
  ],
  "title": "The Language Experience and Proficiency Questionnaire (LEAP-Q): Assessing Language Profiles in Bilinguals and Multilinguals",
  "date": "2007-08",
  "journal": "Journal of Speech, Language, and Hearing Research",
  "volume": "50",
  "issue": "4",
  "pages": "940–967",
  "doi": "10.1044/1092-4388(2007/067)",
  "type": "peer-reviewed original validation study",
  "url": "https://bilingualism.northwestern.edu/wp-content/uploads/2012/02/MarianBlumenfeldKaushanskaya.pdf",
  "locator": "Abstract, printed p.940; Questionnaire appendix",
  "access_status": "author-hosted full PDF; abstract, measurement sections read",
  "applicability": "Language history and self-report",
  "limitations": "Validation sample is not representative of all ages or literacy backgrounds.",
  "accessed": "2026-10-22",
  "organization": "Viorica Marian, Henrike K. Blumenfeld, Margarita Kaushanskaya"
}

{
  "id": "AL04",
  "authors": [
    "David Birdsong",
    "Libby M. Gertken",
    "Mark Amengual"
  ],
  "title": "Bilingual Language Profile: An Easy-to-Use Instrument to Assess Bilingualism",
  "date": "2012",
  "publisher": "COERLL, University of Texas at Austin",
  "type": "research instrument; creator documentation",
  "url": "https://blp.coerll.utexas.edu/",
  "supporting_urls": [
    "https://blp.coerll.utexas.edu/blp-components/"
  ],
  "locator": "Overview; BLP Components modules 1–4",
  "access_status": "full overview and components pages read",
  "license": "CC BY-NC-SA, as stated on site",
  "applicability": "Multiple dimensions of language dominance",
  "limitations": "Instrument documentation, not regional survey findings.",
  "accessed": "2026-10-22",
  "organization": "David Birdsong, Libby M. Gertken, Mark Amengual"
}

{
  "id": "AL05",
  "authors": [
    "National Heritage Language Resource Center, UCLA"
  ],
  "title": "About the Center",
  "date": null,
  "date_type": "undated",
  "type": "research-center institutional framework",
  "url": "https://nhlrc.ucla.edu/nhlrc/about",
  "locator": "Mission",
  "access_status": "full page read",
  "applicability": "Heritage-language development",
  "limitations": "Institutional summary; broad statements about benefits are not used as findings.",
  "accessed": "2026-10-22",
  "organization": "National Heritage Language Resource Center, UCLA"
}

{
  "id": "AL06",
  "authors": [
    "Margarita Kaushanskaya",
    "Henrike K. Blumenfeld",
    "Viorica Marian"
  ],
  "title": "The Language Experience and Proficiency Questionnaire (LEAP-Q): Ten years later",
  "date": "2020-11",
  "online_publication_date": "2019-04-15",
  "journal": "Bilingualism: Language and Cognition",
  "volume": "23",
  "issue": "5",
  "pages": "945–950",
  "doi": "10.1017/S1366728919000038",
  "type": "peer-reviewed methodological reflection by instrument authors",
  "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7899192/",
  "locator": "Abstract; Best-practice discussion",
  "access_status": "open author manuscript read",
  "applicability": "Self-report interpretation and validation limits",
  "limitations": "A locally shortened instrument does not inherit validation.",
  "accessed": "2026-10-22",
  "organization": "Margarita Kaushanskaya, Henrike K. Blumenfeld, Viorica Marian"
}

{
  "id": "AL07",
  "authors": [
    "Statistics Canada"
  ],
  "title": "Language used most often at work of person 15 years or over",
  "date": "2022-08-16",
  "date_type": "page modified",
  "standard_approved": "2009-04-20",
  "type": "official statistical standard",
  "url": "https://www23.statcan.gc.ca/imdb/p3Var.pl?Function=DEC&Id=34028",
  "locator": "Definition; Usage",
  "access_status": "full page read",
  "applicability": "Work-language measurement comparator",
  "limitations": "Canadian standard; no Bay Area population evidence.",
  "accessed": "2026-10-22",
  "organization": "Statistics Canada"
}

{
  "id": "AL08",
  "authors": [
    "Statistics Canada"
  ],
  "title": "Dictionary, Census of Population, 2021",
  "date": "2023-11-15",
  "catalogue": "98-301-X2021001",
  "type": "official methodology",
  "url": "https://www12.statcan.gc.ca/census-recensement/2021/ref/dict/98-301-x2021001-eng.pdf",
  "locator": "Language spoken most often at home: printed pp.309–310; PDF indexes322–323; Language used most often at work: printed pp.310–312; PDF indexes323–325; Mother tongue: printed pp.312–313; PDF indexes325–326",
  "access_status": "full PDF available; specified sections read",
  "applicability": "Distinct operational home, work, childhood-language definitions",
  "limitations": "Design comparator only; cannot relabel U.S. data.",
  "accessed": "2026-10-22",
  "organization": "Statistics Canada"
}

11 / SOURCE LIBRARY • 4 OF 4 · PRINT PAGE 36

Sources and reading notes

{
  "id": "D03",
  "publisher": "California Department of Education",
  "title": "File Structure: ELs by Grade & Language",
  "url": "https://www.cde.ca.gov/ds/ad/fselsch.asp",
  "updated": "2026-04-10",
  "accessed": "2026-10-22",
  "locator": "File structure table",
  "type": "data dictionary",
  "status": "read via search-indexed primary text; direct retrieval challenge",
  "localFile": "cde-fselsch.html",
  "supports": [
    "TOTAL_EL and grade fields definitions",
    "CDS and language code fields"
  ],
  "limitations": "Local HTML is a Radware challenge, not dictionary. Dictionary says LC width 2; actual 2025–26 data uses 3-character codes, so parse tabs rather than fixed width.",
  "organization": "California Department of Education",
  "date": "2026-04-10"
}

{
  "id": "D04",
  "publisher": "CALPADS",
  "title": "Primary Language — CALPADS User Manual",
  "url": "https://documentation.calpads.org/Glossary/EnglishLanguageAcquisitionData/PrimaryLanguage/",
  "accessed": "2026-10-22",
  "locator": "Primary Language definition, three paragraphs",
  "type": "official documentation",
  "status": "read",
  "supports": [
    "Primary language recorded from early/home language information",
    "Administrative field intended to remain stable during school career"
  ],
  "limitations": "Does not measure current dominant language, second-language ability or parent communication preference.",
  "organization": "CALPADS",
  "date": "accessed 2026-10-22"
}

{
  "id": "D05",
  "publisher": "San Leandro Unified School District",
  "title": "EL Master Plan: Chapter 1 — Initial Identification, Assessment & Placement",
  "url": "https://www.slusd.us/english-learner-master-plan/el-master-plan-chapters/el-master-plan-chapter-1/",
  "accessed": "2026-10-22",
  "locator": "Identification: four HLS questions; Assessment: initial ELPAC, IFEP, EL, RFEP",
  "type": "district plan",
  "status": "read",
  "supports": [
    "Home language and English learner classification are separate stages",
    "First language, most frequent home language, caregiver-child language and adult home language are separate questions"
  ],
  "limitations": "Undated current webpage; procedural details should be checked against September 2026 CDE change.",
  "organization": "San Leandro Unified School District",
  "date": "accessed 2026-10-22"
}

{
  "id": "D06",
  "publisher": "Castro Valley Unified School District",
  "title": "CVUSD Community Connection Newsletter — August 15, 2026",
  "url": "https://cv.k12.ca.us/index.php?articleID=83855428&pageID=smartSiteFeed&psqFeed=true",
  "published": "2026-08-15",
  "observationDate": "2026-08-05",
  "accessed": "2026-10-22",
  "locator": "From the Board Meeting / opening-school update",
  "type": "district newsletter",
  "status": "downloaded HTML and verified date",
  "localFile": "cvusd-aug2026.html",
  "supports": [
    "Approximately 9,750 students representing 60 home languages as of August 5, 2026"
  ],
  "limitations": "Rounded district-wide diversity statement; no distribution table or denominator for particular languages.",
  "organization": "Castro Valley Unified School District",
  "date": "2026-08-15"
}

{
  "id": "D07",
  "publisher": "Castro Valley Unified School District",
  "title": "CVUSD Community Connection Newsletter — October 16, 2026",
  "url": "https://www.cv.k12.ca.us/index.php?articleID=92793353&pageID=smartSiteFeed&psqFeed=true",
  "published": "2026-10-16",
  "accessed": "2026-10-22",
  "locator": "Strengthening Support for Multilingual Learners",
  "type": "district newsletter",
  "status": "read",
  "supports": [
    "Approval of updated Comprehensive Plan for Multilingual Learners",
    "Approximately 12 percent multilingual learners in district program context"
  ],
  "limitations": "Underlying newly approved plan not recovered; no explicit category denominator in newsletter.",
  "organization": "Castro Valley Unified School District",
  "date": "2026-10-16"
}

{
  "id": "D08",
  "publisher": "San Leandro Unified School District",
  "title": "EL Master Plan: Chapter 2",
  "url": "https://www.slusd.us/english-learner-master-plan/el-master-plan-chapters/el-master-plan-chapter-2/",
  "accessed": "2026-10-22",
  "locator": "Dual Language Immersion program section",
  "type": "district plan",
  "status": "read",
  "supports": [
    "Spanish and Mandarin DLI pathways serve English learners and English-dominant students"
  ],
  "limitations": "Program description does not measure number of additional-language speakers.",
  "organization": "San Leandro Unified School District",
  "date": "accessed 2026-10-22"
}

{
  "id": "D09",
  "publisher": "San Leandro Unified School District",
  "title": "LCAP and LCFF 2022–23",
  "url": "https://www.slusd.us/wp-content/uploads/2022/09/LCAP-and-LCFF-2022-23.pdf",
  "period": "2022–23",
  "accessed": "2026-10-22",
  "locator": "Action 2.20 Multilingual Family Engagement Webinars and Workshops",
  "type": "district plan",
  "status": "read indexed primary text",
  "supports": [
    "Parent preferred communication language settings are separately recorded in Aeries"
  ],
  "limitations": "Historical planned practice, not current quantitative parent-language distribution.",
  "organization": "San Leandro Unified School District",
  "date": "2022–23"
}

{
  "id": "D10",
  "publisher": "California Department of Education",
  "title": "Survey, Assessments, Programs, & Reclassification",
  "url": "https://www.cde.ca.gov/SP/ML/parentinfo.asp",
  "updated": "2026-09-27",
  "accessed": "2026-10-22",
  "locator": "Update notice; Home Language Survey section",
  "type": "official guidance",
  "status": "read via indexed primary-source text",
  "supports": [
    "HLS is screening for possible English proficiency assessment",
    "Reclassification guidance changed September 2026"
  ],
  "limitations": "Do not repeat older four-criterion procedures as current.",
  "organization": "California Department of Education",
  "date": "2026-09-27"
}

APPENDIX / 1 OF 2 · PRINT PAGE 37

All language categories

[C01] [C02] [C03]

APPENDIX / 2 OF 2 · PRINT PAGE 38

All language categories

[C01] [C02] [C03]

APPENDIX / DATA DICTIONARY · PRINT PAGE 39

Inside the data store

Formats: JSON for the reader, SQLite for queries, and CSV exports for reuse. Source identifiers, retrieval dates and file hashes document the evidence trail. Dataset labels and null values are preserved.

SECOND LANGUAGE PROJECT · PRINT PAGE 40

Use the evidence. Keep the distinctions.

Suggested citation: Second Language Project. (2026). Languages of the Bay Area: Home, learning, work and community access. Research edition 1.1, 22 October 2026.

This edition was prepared through computer-assisted research, source extraction, analysis and publication production. It has not undergone independent academic peer review. Quantitative tables are reproducible from public sources; interpretations and recommendations remain open to community and subject-matter review.

The public reader includes the full text, searchable sources, geography filters, school and workplace evidence, and downloadable tables. Source references and methodological notes help readers examine how the findings were produced.