The Public Use Microdata Sample (PUMS) files contain records for a subsample of the housing units and persons of the American Community Survey (ACS) annual sample. A weighting process was introduced for the 2009 PUMS that expanded the raking matrix to include more demographic controls and family equalization with the goal of forcing more consistency between the PUMS and the ACS full sample estimates. This paper discusses the preliminary research, the trade-offs of doing the weighting at the state versus Public Use Microdata Area (PUMA) levels, and some of the impact on estimates of the new weighting procedure.