Hi there! I am kind of a IPUMS and CPS newbie and find it hard to understand the different values for the quality flags. I already read the information stated here (https://www2.census.gov/programs-surveys/cps/methodology/PublicUseDocumentation_final.pdf) and here (IPUMS USA) but could you please explain to me the differences between these three types:
- xy to value
- xy to allocated value
- xy to longitudinal value
Thanks a lot and apologies in advance if this question was already answered somewhere else!
Thanks for your question. I’ve provided a short explanation of each of these three data quality methods below based on the information provided in the Edits and Imputations section of the CPS technical paper (pg. 133). Note these edits are made by the Census Bureau and not by IPUMS. I hope this helps. Please let me know if you have any additional questions.
xy to value is referred to as a “relational imputation” in the CPS technical paper. These edits refer to values that have been assigned based on other characteristics about the individual or about others in their household. For example, if marital status is missing but relationship to the householder is “spouse,” the missing marital status value will likely be reassigned to “married.”
xy to longitudinal value is referred to as a “longitudinal edit” in the CPS technical paper. These edits refer to values that have been assigned based on information available for that person from the previous month if the person is in their second or later interview month. This is most often used for the labor force items.
xy to allocated value is referred to as a “‘hot deck’ imputation” in the CPS technical paper. These edits refer to values that have been assigned from a donor record that shares similar characteristics with the record that has missing information. Using this process, a donor is selected, and their information is used to fill in the missing information for the person with the missing value.
1 Like
