CASEID is the unique identifier for each woman in a given sample year. Using CASEID and CLUSTERNO, analysts can link IPUMS-DHS data to the original DHS datasets to attach variables that are not currently available in IPUMS-DHS. For example, researchers can use CASEID and CLUSTERNO to link variables from the IPUMS-DHS with other variables from the Woman's Recode file to merge country-specific variables, or to the Household or Couples Recode file to attach variables that are not yet part of IPUMS-DHS.
In most samples, users can identify women who are in the same household by using the first 12 characters of CASEID (and CLUSTERNO (V001)), as this indicates a unique household identifier.
CASEID is generally created by The DHS Program using the variables CLUSTERNO (V001), HHNUM (V002), and LINENO (V003), with the last three characters indicating the respondent's line number in the household file.
For instructions on how to create links between IPUMS-DHS data and DHS source data, see the User Note on "Merging IDHS and DHS Data." [URL omitted from DDI.]
Identifiers Variables -- TOPICS
Imputation and derivation
CASEID is a 18-character string variable.