USC 2010 10 percent has no one over age 95


I was using the 2010 Census 10 percent extract and noticed that the maximum age in this dataset is 95. I thought it was a bit odd that zero 95+ year olds (0.1 percent of the population) would enter a 10 percent random sample. On the variable description for AGE, there is no availability for ages over 96 in USC 2010. I’m not sure what that means.

It would be great if I could confirm whether the age values are actual ages and that random sampling lead to no 95+ year olds in your 10 percent data.

Best, Shiro Kuriwaki

This is a good question. The Census Bureau actually topcodes AGE values by state. Appendix E of this document shows how the topcode is defined for each state. So, for individuals who have an age at or above the state-specific topcode their age is replaced by the average of the top-coded values within their state. This procedure is done in order to maintain confidentiality of individuals included in the census.