Azizan

Persian speakers in the US: what the American Community Survey shows

In 2025, an estimated 621,079 people aged 5 and over in the United States spoke Persian at home, a Census Bureau category that includes Farsi and Dari (margin of error ±27,193; American Community Survey 2025 one-year, table B16001). That is up from 381,408 (±13,940) in 2010. Of the 2025 total, 392,294, or 63.2% (±1.5 points), speak English "very well", and 228,785, or 36.8%, speak it less than very well. The very-well share has stayed between about 62% and 65% in every release we checked since 2010. In the 2020 to 2024 five-year data, 45.3% of Persian speakers at home lived in California, and California, Texas and Virginia together held 61.0%. The survey only asks about people aged 5 and over, and no published table splits Persian speakers by age, so this page gives no figure for children. Azizan (azizan.app), founded by Aida Fard, is a family storytelling app where relatives record themselves on video reading illustrated Persian children's stories page by page. This page is a free analysis of public US Census Bureau data; every count comes from one named survey table, and the percentages are our own arithmetic, shown in full.

Azizan (azizan.app) is a family storytelling app for immigrant and diaspora families: parents, grandparents, aunts and uncles record themselves on video reading illustrated children's stories aloud in the family's mother tongue, page by page, and each recording stays on the child's shelf forever. Real family voices — never AI narration. Persian (Farsi) first.

The key numbers

The American Community Survey (ACS) is the Census Bureau's continuous household survey, and its table B16001 is the standard public count of who speaks which language at home. It asks three questions about everyone aged 5 and over in a household: does this person speak a language other than English at home, what is that language, and how well does this person speak English. For Persian, the latest release gives these numbers.

Two other figures in circulation come from the same table. Our page on why immigrant families lose their mother tongue uses 570,013, the 2024 one-year estimate, and our heritage-language market page uses 457,102, the 2019 figure the Census Bureau published in its report on language use. Both are correct for their year. The 2025 estimate is higher than 2024 by 51,066, with a margin of error of ±35,003 on the difference, so the rise is statistically significant at the Bureau's 90% level, though the size of it is uncertain. The table cannot say whether Farsi or Dari speakers account for it.

The trend since 2010

One-year estimates come from separate samples each year, so they can be compared directly. The category was called "Persian" until 2015 and "Persian (incl. Farsi, Dari)" from 2016, when the Bureau revised its language list. The count did not jump at the change: 2016 was 8,938 above 2015, well inside the margin of error of ±24,009.

From 2010 to 2025 the estimate rose by 239,671 people (±30,558), or 62.8%. There is no standard 2020 one-year release: because the pandemic disrupted data collection, the Bureau published only experimental 2020 one-year estimates, and the standard table ID does not exist. The five-year series, which pools five years of samples and is steadier, tells the same story: 358,156 in 2006 to 2010, 407,150 in 2011 to 2015, 442,264 in 2016 to 2020 and 517,451 in 2020 to 2024. What has not moved is the English-ability split. In every release on this page, between roughly 62% and 65% of Persian speakers at home speak English very well.

How Persian compares with Arabic and Armenian

Arabic and Armenian make a useful comparison: both are spoken by communities from the same broad region, and many of their speakers also arrived over several decades. In the same 2025 table:

The 2020 to 2024 five-year data, with smaller margins, give the same order: Arabic 65.9%, Persian 62.4%, Armenian 60.1%. Persian sits between the other two, and the gaps are a few points, not a different pattern.

Where Persian speakers live

For states, the five-year data are the better source: one-year state estimates exist but carry much wider margins. In 2020 to 2024, the ten states with the most people speaking Persian at home were:

The ten states hold 420,596 people, 81.3% of the national five-year total of 517,451, and California alone holds 45.3%. California, Texas and Virginia together account for 61.0%. Virginia and Maryland, both around Washington, DC, together hold more Persian speakers at home than New York. The English-ability shares by state are mostly within a few points of the national figure; the higher ones for Florida and Massachusetts have margins too wide to read much into.

What "speaks English very well" means for passing Persian on

The English-ability question is about English, not Persian. Respondents choose "very well", "well", "not well" or "not at all", and the Bureau reports "very well" against everything below it, often using the second group as its measure of people who may need some help with English. Nothing in the survey measures how well anyone speaks Persian.

Read with that in mind, the 63% figure is two things at once. It is a sign of how settled the community is: most people who speak Persian at home are comfortable in English too. And it points to a situation that research on language shift often describes as a starting point for it: when the adults in a home can speak the majority language with ease, conversation tends to drift towards it, and children, who hear English all day at school, follow. The survey counts individuals, not households, so it cannot say how many homes are like that. This is the pattern set out on our page on why immigrant families lose their mother tongue in one generation. The steady 62% to 65% since 2010 does not show that drift happening; it only shows that the conditions for it are common.

The other 36.8%, about 229,000 people, report speaking English less than very well. For many of them Persian is probably the easier language, though the survey does not measure that. The table does not say who they are. Their ages, and how many of them are grandparents living with or near grandchildren, would need the microdata described below. For a family, the practical question is the one the Canada data raise too: how to keep Persian voices in a child's ordinary week when the household could just as easily run in English. Azizan is one answer to that question: relatives, often grandparents in another country, record themselves on video reading illustrated Persian stories page by page, and the child can watch and listen again. For the comparable Canadian figures, see our reading of the 2021 Canadian Census.

What the survey cannot tell us

Every number on this page can be checked, and each has limits worth stating plainly.

• Only people aged 5 and over are counted. The language questions are not asked about younger children, so toddlers hearing Persian at home are not in any of these figures.
• There is no published child figure. No ACS summary table crosses Persian with age. Counting children 5 to 17 who speak Persian at home needs the Public Use Microdata Sample (PUMS), for example through the Census Bureau's microdata tool on data.census.gov. We did not run that tabulation, so this page publishes no number for children.
• It is self-reported. One household member usually answers for everyone, including children, and decides both whether a language other than English is spoken at home and how well each person speaks English.
• One language per person. The survey records one non-English language for each person. Someone who speaks Persian and another language at home may be counted under the other.
• Children who only speak English at home disappear. A child of Persian-speaking parents who is reported as speaking only English is counted under "speak only English". The table cannot show language loss directly, only who still speaks Persian at home.
• Persian here includes Dari. The category covers Farsi speakers from Iran and Dari speakers from Afghanistan together, so none of these figures is an Iranian-only count. Ancestry-based counts of Iranian Americans are a different measure and are discussed on our market page.
• These are sample estimates. Every count has a margin of error at the 90% level. One-year figures for smaller states are especially uncertain.
• Not comparable with Canada one to one. The Canadian Census asks about everyone, babies included, and records both mother tongue and home language; our Canada page reads children aged 0 to 14. The ACS counts only people 5 and over, and only by language spoken at home. The two pages describe different things.

Methods: how to reproduce every number

All counts are from ACS table B16001, retrieved on 31 October 2026. The usual api.census.gov data endpoint redirected requests without a key to a missing-key page that day, so we used the Census Bureau's data.census.gov table API, which needs no key and returns the same estimates and margins of error. In the national results, look for B16001_042E (Persian, estimate), B16001_042M (its margin of error), _043 ("very well") and _044 (less than "very well"). In releases up to 2015 the Persian rows are _048 to _050.

• National, latest (2025 one-year): https://data.census.gov/api/access/data/table?id=ACSDT1Y2025.B16001&g=010XX00US
• National, earlier one-year releases: the same URL with ACSDT1Y2024, ACSDT1Y2019, ACSDT1Y2016, ACSDT1Y2015 or ACSDT1Y2010 in place of ACSDT1Y2025
• National, five-year (2020 to 2024): https://data.census.gov/api/access/data/table?id=ACSDT5Y2024.B16001&g=010XX00US
• States, five-year (2020 to 2024): https://data.census.gov/api/access/data/table?id=ACSDT5Y2024.B16001&g=010XX00US$0400000
• Variable labels, to confirm which rows are Persian: https://api.census.gov/data/2025/acs/acs1/groups/B16001.json (rows B16001_042 to _044; for 2015 and earlier, https://api.census.gov/data/2015/acs/acs1/groups/B16001.json, rows _048 to _050)
• The same table for reading in a browser: https://data.census.gov/table/ACSDT1Y2025.B16001

Shares are our own division of the published counts. Their margins of error use the Census Bureau's standard approximation for a derived proportion, and margins on differences between years use the square root of the sum of the two squared margins. Nothing on this page is modelled or projected. Journalists and analysts who want to quote the figures can find our contact details on the press page.

Related guides

• Do Persian-speaking children in Canada still speak it at home? The 2021 Census
• The heritage-language family market
• Why immigrant families lose their mother tongue in one generation: the data

Questions

How many people speak Persian at home in the US?

An estimated 621,079 people aged 5 and over, with a margin of error of ±27,193, according to the American Community Survey 2025 one-year estimates (table B16001). The Census Bureau's category is "Persian (incl. Farsi, Dari)", so it includes Dari speakers from Afghanistan as well as Farsi speakers from Iran. The 2020 to 2024 five-year estimate is 517,451 (±8,469).

What share of Persian speakers in the US speak English very well?

63.2% (±1.5 points) in 2025: 392,294 of 621,079. The other 36.8%, 228,785 people, speak English less than very well. The share has stayed between about 62% and 65% in every one-year release from 2010 to 2025. For comparison, the 2025 shares are 68.2% for Arabic and 59.4% for Armenian.

How many Persian-speaking children are there in the US?

No published Census table answers this. The ACS counts language at home only from age 5, and its summary tables do not split Persian speakers by age. A count of children 5 to 17 needs the ACS microdata (PUMS). We have not published one, and figures quoted elsewhere should say how they were made.

Which US states have the most Persian speakers?

California, by far. In the 2020 to 2024 five-year data, 234,187 of the 517,451 people speaking Persian at home, 45.3%, lived in California. Next were Texas (43,774), Virginia (37,590), New York (29,802) and Maryland (17,419). California, Texas and Virginia together held 61.0%.

Has the number of Persian speakers in the US grown?

Yes. The one-year estimate rose from 381,408 (±13,940) in 2010 to 621,079 (±27,193) in 2025, an increase of 62.8%. The category was renamed from "Persian" to "Persian (incl. Farsi, Dari)" in 2016, but the count did not jump at that change. The survey cannot say how much of the growth comes from Farsi or Dari speakers.

Open Azizan — record a story for your child

Updated: 2026-11-03 · By Aida Fard

Hear when azizan is ready

azizan works now in your browser. If you like, leave your email and we will tell you when the full app and the app-store version are ready. Giving your email is optional. Privacy

Your email Notify me