I am trying to reconstruct the median household income by metropolitan area statistics in this report https://www.census.gov/content/dam/Ce…
I am using the variable met2013 and HHINCOME and not matching their statistics. I am using survey weighting.
There are two reasons why your estimates may differ slightly. First, and most importantly, is the inexact correspondence of geographical boundary definitions with official definitions. As is stated on the Description Tab of the MET2013 variable: “Since 1990, the only sub-state-level geographic information available in census PUMS data is for PUMAs, areas which occasionally straddle official metro area boundaries. Given this limitation, MET2013 cannot identify the exact set of households residing in each metro area. The protocol used by MET2013 is to identify the metro area in which the majority of each PUMA’s population resided. If MET2013 identifies a metro area for a given household, it indicates that, for the PUMA in which the household resided, a majority of the PUMA’s 2010 population resided in the identified metro area.” Second, we generally don’t expect to exactly replicate official statistics with public use data. This is because official statistics use restricted access data, which are not subject to confidentiality requirements. More information on this second point is noted in this recent blog post.