I'm using the Current Population Survey and studying immigration. There's a variable: Year of Entry into the United States for the Foreign Born.

The survey is administered to a representative sample of the United States population, regardless of whether the people are citizens or not. It doesn't ask about whether the person is an illegal immigrant, but still, people probably don't want to share that info or hint at it.

I figure there has to be inherent measurement error in this variable because people might not tell the truth if it hints at anything that reveals that their visas have expired or something.

What can I do to confirm that there is measurement error? Or to account for the error?

  • 2
    $\begingroup$ Almost all legal US immigration is based on having a related family member already be in the US or skill based. Perhaps you could figure out proxies for those attributes and see if they are systemically related to immigration year. $\endgroup$ – BKay Mar 31 '16 at 21:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy