This data biography includes information about who created this data, and how, where, when, and why it was collected. We (the Assessment, Policy Development, and Evaluation Unit at Public Health – Seattle & King County) created it to help you understand where the data comes from and its strengths and limitations. If you have any further questions, please don’t hesitate to reach out to us at data.request@kingcounty.gov!

Who collects the data?

The American Community Survey (ACS) is a survey of households. Data are collected by the U.S. Census Bureau, primarily online and through the mail, with some phone and in-person interviews.

Who owns the data?

The data are owned by the U.S. Census Bureau. They are available at data.census.gov.

How is the data collected?   

Each month, 295,000 random residential addresses are surveyed. Residential addresses receive a packet in the mail with information about the survey and how to participate online. If the household doesn’t complete the survey, a paper version will be mailed in a few weeks. If no response is received online or by mail after multiple reminders the Census Bureau will contact the household for a phone interview. Finally, a sample of households who don’t respond are visited by Census Bureau field staff for an in-person interview. 

The print surveys are distributed in English, along with instructions for requesting language support services via telephone in Spanish, Russian, Chinese, Korean, and Vietnamese.  Interview guides, which include a translation of the survey questionnaire, are available in Chinese and Korean. Full surveys are available in Spanish in print and online. 

A sample of people living in group facilities, such as college dormitories, nursing homes, and prisons, are interviewed in person to ensure coverage of everyone in the country. College dorms are only sampled in non-summer months.  

Who is included and excluded from the data?   

Each year, 3.54 million households and group quarters (such as college dorms and correctional facilities) across the US are surveyed. The Census Bureau maintains a list of addresses of housing units, group quarters (such as dorms), and non-residential units across the US. The American Communities Survey uses this record, plus records of new addresses (such as newly built housing units), to identify possible participants. Because of this, the ACS is likely to exclude individuals without a residential address. 

Where is the data collected?  

ACS data are collected from households across the 50 U.S. states, the District of Columbia, and Puerto Rico.  

Why is the data collected?  

The ACS is designed to describe the social, economic, housing, and demographic characteristics of the U.S. population. It is used by federal government agencies to inform policy, distribute funding, and assess programs. It is also used by state and local agencies for planning services such as roads, hospitals, schools, and more. The ACS is covered by the U.S. Constitution and code that provide for the decennial census and allow for the Census Bureau to “make surveys and collect such preliminary and supplementary statistics related to the main topic of the census as are necessary.”  

How often is this data collected?  

The ACS is conducted throughout the year. The data are released annually in the form of 1-year and 5-year estimates. The 5-year estimates are meant to be representative of the nation and are the primary and complete version for analysis. One-year estimates are available for geographic areas with at least 20,000 people, but these are considered less reliable. 

What else is important to know about this data?  

Data collection was challenging in 2020 due to the pandemic, with lower response rates than in the past. The Census Bureau developed new techniques to account for this, and therefore considers 1-year ACS data to be experimental. They do not recommend comparing the 2020 ACS 1-year experimental estimates with standard ACS estimates or the decennial census, and you can read more about their methods here.   

The ACS was fully implemented in 2006 and replaced the long form version of the Census as the primary source of social and economic data in 2010. To ensure that the data they share better represents the populations described, the Census Bureau weights the ACS data. Weighting is a statistical process used to adjust any demographic (e.g. race/ethnicity, age, or gender) imbalances between survey respondents and the broader population that they represent (for example, households in King County). This way, results will be generalizable to all King County households, not just those who participate in ACS. This is a common approach used by nearly all large-scale population-based surveys. You can learn more about weighting here.  

Where can I learn more about this data?   

About the ACS (census.gov) 

Understanding and Using American Community Survey Data: What All Data Users Need to Know (census.gov)

Sample ACS & PRCS Forms and Instructions (census.gov) 

Design and Methodology Report (census.gov) 

View our data about King County here.

About data privacy and security   

The Assessment, Policy Development, and Evaluation Unit takes data security and the privacy of all individuals represented in our data very seriously. Data are stored in a secure environment. For data sets that are not publicly available, only authorized staff are able to access them. Each data set has privacy guidelines to prevent sharing any information that may be identifiable – for example, not sharing (sometimes called suppressing) numbers when they are very small.