Description

MARRNO indicates whether ever-married persons had been married more than once.

Comparability

The universe shifted considerably over time. For comparability with 1910 (U.S. and Puerto Rican Censuses), the universe must be limited to currently-married persons; for comparability with the 1940 full count, the universe must be limited to ever-married sample-line women. A fully comparable universe of ever-married, sample-line females can be created using the SEX, MARST, and SLREC variables. The 1940 1% sample further restricts the 1940 full count universe to women age 14+. For comparability with 1950, the universe must be limited to 1-in-6 ever-married females using the AGE, SEX, MARST, SLREC, and LINE variables (Note: MARRNO is not available in Alaska or Hawaii for the 1950 full count). For 1960 through 1970 samples, the universe expanded to include all ever-married persons age 14+ and from 1980-onward, the universe consists of ever-married person, age 15+, excluding Puerto Rico.

In all years, respondents were simply to indicate whether they had been married more than once. In the 1910 U.S. Census, many enumerators recorded the number of marriages for persons married more than twice, and the 1910 U.S. Census sample preserves this information for more than 4,000 individuals (this is not applicable to the 1910 Puerto Rican Census sample). Because enumerators were recording information beyond their instructions, users should not assume that these instances of 3+ marriages are representative. In the 2008 ACS and 2008 PRCS, though, "Three or more times" is an actual category.

Universe

Availability

• 2023: All samples
• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: ACS; ACS 3yr
• 2010: ACS; ACS 3yr
• 2009: ACS
• 2008: ACS
• 2007: --
• 2006: --
• 2005: --
• 2004: --
• 2003: --
• 2002: --
• 2001: --
• 2000: --
• 1990: --
• 1980: All samples
• 1970: 1% state fm1; 1% metro fm1; 1% neigh fm1
• 1960: All samples
• 1950: All samples
• 1940: All samples
• 1930: --
• 1920: --
• 1910: 1%; 1.4% ovrsmp
• 1900: --
• 1880: --
• 1870: --
• 1860: --
• 1850: --

• 2023: All samples
• 2022: All samples
• 2021: All samples
• 2020: All samples
• 2019: All samples
• 2018: All samples
• 2017: All samples
• 2016: All samples
• 2015: All samples
• 2014: All samples
• 2013: All samples
• 2012: All samples
• 2011: PRCS; PRCS 3yr
• 2010: PRCS; PRCS 3yr
• 2009: PRCS
• 2008: PRCS
• 2007: --
• 2006: --
• 2005: --
• 2000: --
• 1990: --
• 1980: --
• 1970: --
• 1930: --
• 1920: --
• 1910: All samples

Flags

Editing Procedure

MARRNO (Times married) YRMARR (Year married)
ACS Years: 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015
ACS editing procedure:
This question asks a person how many times they were married and the year in which they last got married. The questions are asked only to ever-married people 15 and older.
If an ever-married (MARST) person is younger than 20 reports being "Married twice," MARRNO will be replaced with "Married once". If an ever-married (MARST) person is younger than 25 reports being "Married three times or more," MARRNO will be allocated from another person with a similar SEX, AGE, and marital status (MARST). The flag variable (QMARRNO) will indicate if the value is edited or allocated.
If an ever-married person leaves MARRNO or YRMARR blank, MARRNO and YRMARR will be allocated from another person with a similar SEX, AGE, and marital status (MARST). The flag variables (QMARRNO QYRMARR) will indicate if the value is allocated.
For 2008 to 2013, if an ever-married person (MARST) reports a value for YRMARR that is prior to 1900 or after the survey year, YRMARR will be allocated from another person with a similar SEX, AGE, and marital status (MARST). In 2014 and later, this same allocation occurs if an ever-married person reports a year of marriage that is over 100 years before the survey year or after the survey year.
If a divorced person (MARST) reports they were married in the survey year, YRMARR will be replaced with the survey year minus one. The flag variable (QYRMARR) will indicate if the value is allocated.
If an ever-married person reports their last marriage occurred when they were less than 15 years old, YRMARR will be allocated from another person with a similar SEX, AGE, and marital status (MARST). The flag variable (QYRMARR) will indicate if the value is allocated.
If the allocated value is either prior to when the person was 15 or after the survey year, it will be replaced with the year midway between the survey year and the year the person turned 15. The flag variable (QYRMARR) will indicate if the value is allocated.
After the above edits have been made for all household members, a further set of edits are then performed.
This set of edits applies if the household contains male-female married couples in the same subfamily with different years of last marriage. If the female's value of YRMARR was not allocated, the male's value of YRMARR will be replaced with the female's value. If the female's value of YRMARR was allocated and is less than the male's value of YRMARR, the female's YRMARR will be replaced with the male's value. If the female's value of YRMARR was allocated and is greater than the male's value of YRMARR, the male's YRMARR will be replaced with the female's value. The flag variable (QYRMARR) will indicate if the value is edited.
Starting in 2014, another set of edits applies to married same-sex couple with different values for the year of last marriage. If the householder's marriage year was not allocated, and is a valid value for the spouse's year of marriage (e.g., the year of marriage occurred when the spouse was 15 or older), the spouse's YRMARR value will be replaced with the householder's. If the householder's YRMARR is not valid for the spouse, the householder's value of YRMARR will be replaced with the spouse's. If the householder's value for year of marriage was allocated, and the spouse's value for YRMARR is valid for the householder, the householder's value will be replaced with the spouse's. If the spouse's value is not valid for the householder, the spouse's value will be replaced with the householder's.
Internal ACS variable : MARHT MARHY