Skip to main content

Census Workshop

A research guide designed to be paired with the UVA Library's Research Data Services workshop on the Census. Learn the basics of Census data, and how and why you want to use it. https://tinyurl.com/UVAcensusWorkshop

Getting Started

Welcome!  This guide was designed for the "Census Basics" Workshop at UVA Library, Fall 2019. 

These are your learning outcomes:

  • Understand the available Census surveys, and their strengths and weaknesses,
  • Understand Census geographies,
  • Understand where to find easy-to-download data, 
  • Recognize Standard Error in the ACS and how to compare estimates,
  • Recognize advanced download options and other specialized Census products. 

Do you still have questions?  I invite you to schedule a one-on-one consultation!

Most Common Census Surveys: Decennial Census and American Community Survey (ACS)

The two most commonly used Census surveys are the Decennial Census and the American Community Survey.  

The Decennial Census is conducted every 10 years; the most recent one was in 2010.  The Census attempts to count the entire US population (including households and group quarters).  You can get only the most general population characteristics from the decennial Census.  The Census aids in apportioning congressional districts and determining federal funding for local programs.  The main purpose is a relatively simple count of population. 

The American Community Survey (ACS) is conducted on a rolling basis.  It has a lot more in common with survey estimates rather than a complete population count.  Only some people are surveyed at any given time -- about 1% of the US is surveyed each year.  There are 1-year and 5-year estimates available. They have strengths and weaknesses in regards to geography size, timelineness, statistical error, and ease of interpretation.  

Census survey comparisons
  Decennial Census ACS 1-Year (incl. Supplemental Tables) ACS 5-Year
Data Collection Point-in-time (April 1 of years ending in 0) 12 month continuous survey 60 month continuous survey
Example 2010 Decennial Census 2017 ACS 1-year estimate 2013-2017 ACS 5-year estimate
Available Geographies Census block and larger

Geographies with populations >65,000

Note that Supplemental, "high-level" tables are available for populations with >20,000 people (e.g., City of Charlottesville) starting in 2014.

Census tract and larger
Design Counts every person Smaller sample size Larger sample size
Margin of Error Smallest margin of error, i.e. very reliable estimates Highest margin of error, i.e., least reliable estimates Margin of error smaller than 1-year estimate, i.e., more reliable than 1-year estimates
Best used when When you are only interested in general population characteristics (count; age, sex, race) Currency is more important than precision; when you are working solely with large population geographies; when you need more-easily interpreted population data Precision is more important than currency; When you are working with less populous geographies (small-sized geographies or rural areas)
Drawbacks Only available every 10 years (data age quickly); only most general population characteristics Not available for less-populated geographies; high margin of error 5 year estimate is harder to interpret; margin of error can still be high for small geographies

Which survey should you use?  Ask yourself these questions:

  • What population characteristics am I interested in? 
    • If you only want the most general population counts and characteristics (age, sex, race), your best bet is the Decennial Census (or official Population Estimates for county geographies and larger for intercensal years). 
    • If you want more descriptive population characteristics (e.g., education, income, nativity), you have to work with the ACS. 
  • What type of geography am I interested in? 
    • If you are interested in larger geographies (states, larger cities/MSAs, or populous counties), the ACS 1-year is a good choice. 
    • If you are interested in anything but the largest geographies, you will have to choose the ACS 5-year estimates.  For example, if you want smaller geographies like census tracts or all counties in Virginia (some of which are rural and have smaller populations), you will have to work with the 5-year estimates. 

Understanding and Using American Community Survey Data: What All Data Users Need to Know

This handbook from the Census is an excellent reference for understanding the ACS: