By: John Besl
In my last blog entry, I wrote about a measure of neighborhood residential mobility that’s derived from the American Community Survey. Since the measure is based on responses from a small sample of the population, there’s some degree of uncertainty associated with the tract-level estimates. Besides that limitation, the ACS-based mobility measure represents a short time frame, a single year. In this post, I take a longer view – 10 years – and develop a synthetic estimate of child migration derived from the last two decennial census enumerations. The new measure relies on the ironclad assurance that children who were counted in the 2000 Census will be 10 years older when enumerated in the 2010 Census. We would expect that a residentially stable community (i.e., where migration is low) would see little change between the 2000 Census count of children in the preschool age range of 0 to 4 and the 2010 Census count at ages 10 to 14.