CPS-ASEC: Topcodes, Swapvalues, or both?


I am using the INCWAGE variable from CPS-ASEC and want to conduct analysis using the most accurate version of this data for the period 1988-2013. Despite reading the paper by Larrimore, both pages of IPUMS documentation, and lots of questions here in the forum, it is still not clear to me whether I should update my extracts using the files and documentation for the Topcodes and/or the Swapvalues?

Any suggestions are greatly appreciated,


The US Census has implemented a number of techniques that preserve respondent confidentiality at high levels of income. These procedures are not consistent over time and so the methods discussed on this page allow for a more consistent treatment of top-coded values. Given that your analysis spans multiple years and across various techniques used by the Census Bureau for handling top-coded values, the added consistency may be beneficial for your analysis. Ultimately, however, whether or not you perform these adjustments is a decision that is based on your research question and intended audience.