Codes and Frequencies
Description
For sample adults (and, prior to 2019, for all persons), MARSTLEG reports the legal marital status of the respondent. This variable differs from MARSTCUR by recoding responses that do not fall into a federally-recognized marital category. Prior to 1997, this includes same-sex couples who reported themselves as married. From 1997 forward, persons who responded that they were living with an unmarried partner were assigned to a category corresponding to their legal status.
NCHS collected information about spouse location (present, absent, or unknown) from 1973-2003. For samples from 2004 forward, IPUMS NHIS constructs these categories from data available in the variable MARSTCUR. To create a value of “married” that is comparable across all available samples, users should combine all codes that begin with the number 1 into a single category. Using sample year 2000 as an example:
11 (Married – spouse present): 41,138 +
12 (Married – spouse absent): 1,140 +
13 (Married – spouse unknown): 2 = 42,280
This variable replaces MARST and MARSTAT, which were deprecated in 2024.
Comparability
Along with universe changes, there have been changes in question wording and in the number of recognized categories for this variable over time.
The question corresponding to MARSTLEG remained essentially the same for 1968-1996.
At the start of this period, interviewers asked, "Is [person] now married, widowed, divorced, separated, or never married?"
At the end of this period, interviewers asked, "Is [person] now married, widowed, divorced, separated, or has [person] never been married?" The only categories recognized in these questions were those corresponding to legal marital status.
In contrast, beginning in 1997, interviewers initially asked, "[Are/Is] [you/subject's name] now married, widowed, divorced, separated, never married or living with a partner?"
The inclusion of "living with a partner" melds legal marital status with something more akin to union status (since a person living with a partner could be legally married, widowed, divorced, or separated).
Persons living with a partner were, however, asked a follow-up question about their legal marital status, and it is this supplementary information that is included in MARSTLEG. Thus, while the underlying question wording changed after 1997, MARSTLEG consistently reports legal marital status for the entire period.
In 1973, the distinction was first drawn between "Married - Spouse present" and "Married - Spouse absent."
The 1973 Field Representative's Manual offers the following explanation for the distinction: "Mark 'Married - spouse present' for each married household member whose spouse is also listed on the questionnaire. This includes Armed Forces members living at home as well as those whose spouses are temporarily absent." "Mark 'Married - spouse absent' for a married person who is not legally separated, and whose husband or wife is not a member of the same household. This includes Armed Forces members who are not living at home." Beginning in 1997, the presence or absence of a married person's spouse was ascertained via the question, "Is [subject's name] spouse living in the household?" and a new category, "Married - Spouse in household unknown" was added (presumably to accommodate cases where this question went unanswered).
Common-law marriages were first referred to in 1982.
The Field Representative's Manual which instructed the interviewer to include "as 'Married,' persons who state they have a common law marriage, or who are living together as husband and wife."
Beginning in 2004, all currently married persons were simply labeled "married," and additional information about the spouse's location (present in the household, not in the household, or unknown whether in the household) was dropped.
This change derived from the adoption by the National Center for Health Statistics (NCHS) of a single variable for reporting legal marital status in the public use files of the NHIS. This variable followed the "standards" for reporting legal marital status set by the Centers for Disease Control and Prevention (CDC).
The introduction of a substantially revised questionnaire in 2019 may also affect comparability with earlier years. The NHIS questionnaire was substantially redesigned in 2019 to introduce a different data collection structure and new content. For more information on changes in terminology, universes, and data collection methods beginning in 2019, please see the user note.
Universe
- 1963-1981: Persons age 17+.
- 1982-2018: Persons age 14+.
- 2019-2022: Sample adults age 18+.
Availability
- 1963-2022
Weights
- 1963-2018 : PERWEIGHT
- 2019-2022 : SAMPWEIGHT
