Is it appropriate to compare “unweighted” response rates for the 2011 National Household Survey and the 2011 Census with the “net undercoverage estimates from the technical reports (net undercoverage being a reflection of both undercoverage and overcoverage link) for 1996-2006?
We never compare the Census response rates and the net undercoverage rates estimates. These are two different concepts. Census response rates are calculated at the dwelling level based on the estimated number of occupied dwellings while undercoverage is estimated at the person level. Net undercoverage includes people missed in enumerated households (dwellings) and will also take into consideration the overcoverage. They are very different.