Constructing a household panel



I’m trying to see if it’s possible to construct a household panel having US census data 1990-2000-(2005)-2010.

As far as I understand the extracts are random samples for each waves, which would only give me a repeated cross-section. Is there a way to select a random sample from one of the waves and follow it over time?

Thank you



The Public Use Microdata Series is composed of a random sample of respondents from each year. Because there is no effort made to include the same respondents in multiple samples (with the ACS actively avoiding including the same households in consecutive samples), constructing a panel of respondents or households over modern samples is not possible (though historical samples have been linked).

I hope this helps.