4

The ACS Survey provides the most up-to-date census estimates for US population groups. The 5 year survey is the most comprehensive. For example, the most recently available 5 year survey as of Sept 2015 is the 2009-2013 ACS 5 Year Survey.

I'm trying to calculate population changes since that survey was taken with a number of methodologies, but what I don't know is when the effective estimate date is supposed to be. For the 2009-2013 survey, is it January 1, 2013? Is it December 31, 2013? Is it some date in 2012? I'm assuming it is January 1, 2013, but have no way of knowing this other than an educated guess.

Thanks

Improve this question

2 Answers 2

Reset to default
2

As a student of demography myself, my professor's often treat the effective date as the midpoint of the survey period. So for a 1-year survey period, the estimate would be said to be July 1st of that year. When you start getting into 5-year survey periods, the flaws in thinking of the survey as a point-in-time estimate become more apparent. As fdonnelly has suggested, the best way to think of ACS estimates is that they are the average population or proportion estimate across 12 (or 60) months.

If you are still keen on using the point-in-time estimate approach, then I would say the midpoint date for the 2013 5-Year ACS would be July 1st, 2011.

Improve this answer
0
5

The ACS is a rolling annual sample. They take the sample for the entire year (about 3.5 million addresses in 2013) and divide it into 12 equal, monthly samples, and they mail out forms every month. So unlike the 10 year census (which is on April 1st) there is no effective date for the ACS. The analogy they use is that the 10-year census is like a snapshot (one day in time) while the ACS is more like a moving image (constantly collected each month and tallied for 1, 3, or 5-year time periods).

Technical docs that mention this are here: http://www.census.gov/programs-surveys/acs/methodology/sample-size-and-data-quality/sample-size-definitions.html

Improve this answer
1
  • So the middle of 2013 is a safe assumption? Or is it the end? I guess it would be the end? I need something to plug into program regardless of whether or not it is the absolute truth or not
    – boulder_ruby
    Dec 2, 2015 at 22:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.