Status:Closed    Asked:May 15, 2015 - 09:44 PM

Probability of duplicates if using multi-year data from the ACS?

I apologize if this seems like a pretty basic question, but I'm using ACS data from 2008 to 2013 in order to expand a specific subsample that is rather rare. My question is when researchers use multiple years of the survey to do statistical analysis and estimations, do they worry about households that may be a part of multiple surveys?

Do you have the same question? Follow this Question

Staff Answer




The short answer is not to worry about repeated households. Once a household has been selected for the ACS that household is removed from the sampling frame for at least 5 years. Since your sample includes 6 years, it is technically possible that a household could be repeated; however, the probability of a household being selected in the 1% national random sample for 2008 and then selected again in 2013 is only about 0.01%, which is much too small to affect your analysis. Even if there were a small number of repeated households in your sample, no variables exist to identify such households.

Hope this helps.


May 18, 2015 - 01:42 PM

Report it


Change in CPS EMPSTAT (A-LFSR Labor force status recode) in 2001?
I have an extract 2001 to 2015. Total number of people: 4 million, but in bir...
Why are there fewer counties in the 1950 1% sample compared to the 1940 1% sa...
Is the value of INCWKCOM included in the value of INCDISAB?
Trying to access CPS online analysis tool
ACS METRO variable construction
Login   |   Register

Recently Active Members

View More »

Share |