NHISPID is a 16-character string variable.
Description
NHISPID is an IPUMS NHIS-constructed value that is an identifying number unique to each person in a given survey year. Using NHISPID, analysts can link IPUMS NHIS data with additional data elements, from the NHIS source public use data files, that are not currently available in IPUMS. For example, researchers can use NHISPID to link variables in IPUMS NHIS with other variables from the core NHIS files variables. Alternatively, users may use NHISPID to link variables from IPUMS with variables from NHIS supplements that are not yet part of IPUMS NHIS. NHISPID is a single variable constructed by concatenating multiple data elements from the NHIS source data files.
Analysts wishing to link household-level NHIS data should use the household-level linking key NHISHID.
See the user note on LINKING for instructions on how to create links between IPUMS NHIS and NHIS source data.
Comparability
The variables concatenated to construct NHISPID differ by periods. There are five discrete periods, namely: 1) 1969-1991 and 1993-1994; 2) 1992; 3) 1995-1996; 4) 1997-2003; 5) 2004-2018; and 6) 2019-forward. While the particular combination of variables used to construct NHISPID differs between these periods, the substantive meaning of the variable is consistent across years. In the discussion that follows, the elements of NHISPID are identified via the variable names used in the original NHIS public use data source files.
The sequential elements in NHISPID before 1994 (except for 1992) are processing year (YEAR), processing quarter (QUARTER), PSU number (PSUNUMR), census week (WEEKCEN), segment number (SEGNUM), household number (HHNUM), and person number (PNUM).
The sequential elements of NHISPID for 1992 are processing year (YEAR), processing quarter (QUARTER), PSU number (PSUNUMR), census week (WEEKCEN), segment number (SEGNUM), household number (HHNUM), and person number (PNUM).
The sequential elements of NHISPID for 1995 and 1996 are processing year (YEAR), household ID (HHID) and person number (PNUM).
From 1997 through 2003, the sequential elements of NHISPID are processing year (YEAR), household serial number (HHX), family serial number (FMX), and person number (PX).
For 2004-2018, the sequential elements of NHISPID are processing year (YEAR), household serial number (HHX), family serial number (FMX), and person number (FPX).
For 2019, the sequential elements of NHISPID are processing year (YEAR), household serial number (HHX), and record type (RECTYPE). Because a maximum of one sample adult and one sample child are pulled from each household, RECTYPE, which identifies a sample adult or sample child, can uniquely identify a person within a household.
The NHIS questionnaire was substantially redesigned in 2019 to introduce a different data collection structure and new content. Beginning in 2019, HHX, a component of NHISPID, leads with the character "H" followed by six numeric digits. Prior to 2019, HHX consisted only of six numeric digits. This does not change the fact that NHISPID is unique within sample. For more information on changes in terminology, universes, and data collection methods beginning in 2019, please see the user note.
Universe
- 1963-2018: All persons (NHIS samples).
- 2019-2022: Sample adults age 18+ and sample children age 0-17.
Availability
- 1963-2022
Weights
- 1963-2018 : PERWEIGHT
- 2019-2022 : SAMPWEIGHT
