Marriage & Family Review

ISSN: 0149-4929 (Print) 1540-9635 (Online) Journal homepage: http://www.tandfonline.com/loi/wmfr20

Premarital Cohabitation and Direct Marriage in

the United States: 1956–2015

Arielle Kuperberg

To cite this article: Arielle Kuperberg (2018): Premarital Cohabitation and Direct Marriage in the
United States: 1956–2015, Marriage & Family Review, DOI: 10.1080/01494929.2018.1518820

To link to this article: https://doi.org/10.1080/01494929.2018.1518820

Published online: 08 Oct 2018.

Submit your article to this journal

Article views: 16

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=wmfr20
MARRIAGE & FAMILY REVIEW
https://doi.org/10.1080/01494929.2018.1518820

Premarital Cohabitation and Direct Marriage in the

United States: 1956–2015
Arielle Kuperberg
Department of Sociology, The University of North Carolina at Greensboro, Greensboro, North
Carolina, USA

ABSTRACT KEYWORDS
Cohabitation rates and durations increased rapidly beginning cohabitation; demography;
in the late 1960s, and by 2011–2015, 70% of first marriages marriage; premarital
among women under age 36 began in premarital cohabitation relationships; social change;
socioeconomic status;
lasting an average of 32 months before marriage. The National United States;
Survey of Families and Households (n ¼ 3,594) and the
National Survey of Family Growth (n ¼ 9,420) are analyzed to
estimate selection into direct marriage and premarital cohabit-
ation from 1956–2015, and long- and short-term premarital
cohabitations from 1971–2015. Early premarital cohabitors
were more likely to be women of color and had the same
education as direct marriers. Later cohorts of premarital
cohabitors were less educated, from lower class backgrounds,
more likely to have experienced a parental divorce/separation,
less religious, and long-term premarital cohabitations were
more common among women of color.

Introduction
After being virtually nonexistent in the 1950s, rates of premarital cohabit-
ation rose dramatically in the U.S. since the mid-1960s, and the majority of
couples that have married in the 21st century lived together before
marriage (Kuperberg, 2014; Smock & Gupta, 2002). Past research has
examined cohabitors as a whole, whether premarital cohabitation is related
to divorce, and changing selection into any cohabitation and premarital
cohabitation in Northern European contexts (Blom, 1994; Bumpass & Lu,
2000; Kuperberg, 2014; Manning, 2010; Mooyaart & Liefbroer, 2016; Wiik,
2008). Less is known about the difference between premarital cohabitors
(those who marry after a period of premarital cohabitation) and direct mar-
riers (those who marry directly without first cohabiting with their spouse) in
the United States, differences between long-term premarital cohabitors and
short-term premarital cohabitors, or how those differences have changed
over time. The sociological and demographic roots of the rise of premarital

CONTACT Arielle Kuperberg [email protected] Department of Sociology, The University of North

Carolina at Greensboro, PO Box 26170, Greensboro, North Carolina 27402#6170, USA.
ß 2018 Taylor & Francis Group, LLC
2 A. KUPERBERG

cohabitation in the United States have also not been fully explored; past
research has focused on micro-aspects of cohabitation rather than overall
increases in cohabitation over the past 50 years (Kroeger & Smock, 2014).
Past research that examined selection into cohabitation as a whole
analyzed a group comprised of several “types”; some couples were cohabit-
ing as an “alternative to being single” or an “alternative to being married”
but may never have intended to marry their partner, and others intended
to eventually marry their partner, but were unwilling or unable to marry
without first undergoing a period of a “trial marriage” to resolve uncertain-
ties, or perhaps saving money, finishing education or paying down debt;
these couples tended to behave in substantially different ways from each
other, and examining them as one group can hide or exaggerate inequal-
ities (Kuperberg, 2012; Heuveline & Timberlake, 2004). Heuveline and
Timberlake, (2004) characterize cohabitation in the U.S. as most commonly
an “alternative to being single” with many engaging in short-term cohabita-
tions with a high dissolution rate, but also find a clear minority who form
long-term cohabitations as an alternative to marriage, and that among
cohabitations that had ended in the period studied, 48% of cohabitations
were preludes to marriage. This later group, those that married after
premarital cohabitation, and how they differ from those who marry
directly, is understudied. Even less is known about how these differences
may have changed over time. Factors which affect selection into premarital
cohabitation, and other factors which do not, may also impact the length
of time that couples cohabit before advancing to marriage, resulting in
further inequalities in family formation which may be obscured by examin-
ing selection into any cohabitation or premarital cohabitation alone.
This article examines the rise of premarital cohabitation rates among
women who married between 1946 and 2015 in the United States. Changing
duration of premarital cohabitation and selection into premarital cohabitation
and direct marriage between 1956 and 2015, and into short term (1 year or
less) and long term (5þ years) premarital cohabitation among women who
cohabited with their first husband before marriage between 1971 and 2015 is
also examined. Selection is examined by education, mother’s education (a
measure of class background), race, religiosity, prior cohabitation experiences,
and whether respondents lived with both biological parents at age 14;
duration models additionally examine whether cohabitors completed add-
itional education between moving in and marrying their first husband.

The rise of premarital cohabitation

In the 19th and early 20th century, the majority of U.S. states recognized
common law marriages; romantic relationships conforming to a pattern of
 MARRIAGE & FAMILY REVIEW 3

marital behavior, including living together, but never solemnized through a

legal ceremony (Bowman, 1996; Dubler, 1998). The acceptance of common
law marriages was widely debated in the U. S. court system throughout the
19th century, and between 1875 and 1917 began to lose legal standing; by
the mid 20th century, a majority of states no longer recognized common
law marriage (Bowman, 1996; Dubler, 1998). With common law marriages
no longer a legal possibility for most, a new relationship form rose to take
its place; cohabitation, or living together with a romantic partner either
before a legal marriage or outside of the legal marriage system entirely.
In March 1968, the New York Times reported on Barnard College
student Linda LeClair, who had circumvented university rules to illicitly
live off campus with her boyfriend, a Columbia University student. The
couple, who had spent the summer of 1967 in Haight–Ashbury during the
“summer of love,” sparked a widespread debate and several national news
articles about cohabitation; later referred to as the The LeClair Affair, this
incident created widespread public awareness of cohabitation as a viable
relationship, along with moral panic about that possibility (Danziger &
Greenwald, 1977; Pleck, 2012). By 1971 cohabitation had become “trendy”
among young celebrities and was discussed in every women’s magazine
(Pleck, 2012). Cohabitation rates subsequently increased rapidly in the U.S.,
and by 1987 one-third of women aged 19–44 had previously cohabited
with an unmarried partner, with rates rising to 58% in 2006–2008
(Wydick, 2007; Manning, 2010).

The social and demographic roots of cohabitation

Policy shifts away from recognizing common-law marriage, along with
other technological, legal, social and economic changes in the 1960s and
1970s and thereafter, led to conditions primed for this new relationship,
and may perhaps be traced back to the advent of the birth control pill,
which became widely available to single women in the late 1960s and early
1970s, affording women consistent control over the timing of pregnancy
and childbearing (Goldin & Katz, 2002). The 1973 Supreme Court decision
(Roe vs. Wade) that legalized abortion in the United States further
cemented this control, while the Civil Rights Act of 1964 that barred
discrimination on the basis of gender and The Pregnancy Discrimination
Act of 1978, (EEOC, 1978) opened up a new range of occupations to
women, and further reduced the potential costs of pregnancy. Women,
many of whom had gotten a “taste for work” during World War II, subse-
quently increasingly invested in their own careers and education, with
reduced fear of career derailment due to pregnancy (Coontz, 1992; Goldin
& Katz, 2002). Due to these changes, along with the more general
4 A. KUPERBERG

expansion of higher education in the 1950s and 1960s, women’s rates of

college and graduate school attendance and labor force participation rose
during the post 1960s era (Kuperberg, 2009; Rosenfeld, 2007; Snyder &
Dillow, 2012; Mosisa & Hipple, 2006). At the same time, American men’s
labor force participation declined from around 97% in 1950 through 1970
to 90% in 2005, while their wages stagnated and showed no growth since
the late 1960s; the typical man in the mid-2000s actually earned less than
in 1969 once wages were adjusted for inflation (Madrick & Papanikolaou,
2007; Mosisa & Hipple, 2006).
The availability of the birth control pill and access to abortion, along
with increasing economic opportunities for women and other legal changes,
contributed to the rise in cohabitation rates through multiple mechanisms.
First, women could engage in premarital sex with little fear of unwanted
pregnancies, corresponding with increased rates of premarital sex over this
time-period (Finer, 2007) and a dramatic liberalization of attitudes and
norms regarding sexuality outside of marriage that has been called the
“sexual revolution” (Bailey, 2011; Coontz, 1992). Second, attending college
and living apart from parents in young adulthood itself contributed to a
general rise in rates of young adults entering nontraditional relationships
such as cohabitation, because parents act as a socially conservative force
constraining the relationship patterns of their children, and because educa-
tion often leads to a delay in marriage (Oppenheimer, 1988; Rosenfeld,
2007). Rising rates of college attendance led to the creation of a normative
“independent life stage” among young adults who both did and did not
attend college, during which young adults increasingly left their parent’s
home and lived independently before forming marriages, contributing to a
greater diversity in family forms and a greatly increased median age at
marriage during this time period (Rosenfeld, 2007). As more women and
men delayed marriage, they increasingly cohabited prior to marriage.
Third, shifts in United States divorce laws during the 1970s that established
“no-fault” divorce, along with women’s increasing economic independence,
led to rising divorce rates in the United States in the 1970s and 1980s,
which have remained relatively high ever since (Lundberg & Pollak, 2007;
Nakonezny, Shull, & Rodgers, 1995). The high divorce rate may have
increased the reluctance of couples, many of whom experienced a parental
divorce, to enter marriage without first undergoing a “trial marriage” in the
form of cohabitation to ensure compatibility (Hoelter & Stauffer, 2002;
Kiernan, 2002), potentially leading to an association between premarital
cohabitation and higher parental divorce rates.
Finally, in part due to increases in women’s labor force participation,
along with men’s stagnating wages and declining labor force participation
rates, marriage as an institution shifted in its meaning and function for
 MARRIAGE & FAMILY REVIEW 5

married couples. Women’s rising economic independence and the rising

ratio of their earnings compared to men’s led to a reduction in the
economic necessity of marriage for women and marriage offered fewer
economic benefits for women than in earlier years (Hoelter & Stauffer,
2002; Lundberg, Pollak, & Stearns, 2016; Seltzer, 2000), perhaps resulting in
increasing cohabitation rates as women did not have to advance to
marriage in order to gain economic security. Men’s stagnating economic
circumstances also violated norms of masculinity tied to employment and
earnings, potentially leading to declines in those willing to marry these
men, as those earning below the poverty level are considered by many
women to be “unmarriageable,” and those earning less than their partners
violate traditional gender norms dictating that husbands should be primary
breadwinners (Edin & Kefalas, 2005; Harknett & McLanahan, 2004;
Lundberg et al., 2016).
As marriage became less of an economic necessity for women, there was
a seemingly paradoxical increase in the emphasis on reaching certain finan-
cial achievements before entering marriage (Edin & Kefalas, 2005; Smock,
Manning, & Porter, 2005). Marriage became a symbol of individual
achievement and prestige that one must build up to by living with a
partner beforehand, starting a career, paying off debt, paying for a wedding
party, and possibly by having children or purchasing a home (Cherlin,
2004; Cherlin, 2009; Edin & Kefalas, 2005). This shift in the meaning of
marriage repositioned marriage as a symbol of status to be achieved once
financial prerequisites were met, rather than an economic necessity to
ensure economic support and stability (Cherlin, 2004; Edin & Kefalas,
2005; Smock et al., 2005). Rising standards of consumption for the middle
class in the late 20th century (Frank, 1999; Schor, 1998) further “raised the
bar” in terms of the financial goals that couples felt they must achieve
before entering marriage. At the same time, rising costs of homeownership
and the rising costs of a college degree, both of which outpaced inflation,
increased the amount of money necessary to achieve the “white picket
fence” lifestyle that couples hoped to achieve before entering marriage
(Edin & Kefalas, 2005; Warren & Tyagi, 2011). As a result, the transition
to adulthood was increasingly delayed (Furstenberg, Kennedy, McLoyd,
Rumbaut, & Settersten, 2004), and rates and duration of premarital and
non-marital cohabitation increased among couples unable to achieve these
financial goals, or unable to achieve them without first undergoing a period
of saving. These factors also contributed to the rising age at marriage
discussed above.
Since the LeClair Affair, as cohabitation rates have increased, and
religious constraints on behavior weakened due to an increasingly secular
culture (Casper & Bianchi, 2011) attitudes regarding cohabitation and
6 A. KUPERBERG

related activities became more accepting (Thornton & Young-Demarco,

2001). By the late-1990s the majority of younger cohorts of Americans
found premarital cohabitation not only acceptable, but a preferable living
arrangement before entrance into marriage (Axinn & Thornton, 2000;
Daugherty & Copen, 2016; Thornton & Young-Demarco, 2001). As
cohabitation gained acceptance, the normative pressure to marry quickly
after cohabitation was also reduced, potentially explaining why more recent
cohorts lived together for longer durations before marrying (Mernitz,
2018). Serial cohabitation, or cohabiting with more than one partner, also
became more common, as fewer cohabitations ended in marriage
(Eickmeyer & Manning, 2018; Lichter & Qian, 2008; Mernitz, 2018).

Selection into premarital cohabitation and direct marriage over time

As cohabitation rates rose, and became more socially acceptable, the func-
tion and meaning of cohabitation may have shifted as well, leading to shifts
in selection into premarital cohabitation and direct marriage over time.
Kiernan (2002) described four historical stages of cohabitation based on an
examination of Sweden that suggest selection into premarital cohabitation
may change over time. In the first, cohabitation was rare and seen as
deviant. In the second, cohabitation gained acceptability as a “trial mar-
riage” but was usually a childless relationship until marriage. Next, cohabit-
ation became socially acceptable as an alternative to marriage, and
childbearing occurred within cohabitation, and finally, cohabitation and
marriage were socially and almost legally indistinguishable, as is currently
the case in Sweden. Pleck (2012) traces the growth of cohabitation in the
US and notes that in 1962–1967 cohabitation was viewed as low class, and
more common among minorities, interracial couples and bohemians, but
that after the1968 LeClair affair, cohabitation began to be viewed as part of
the counterculture and “‘new morality’ of the young”; she notes by the late
1970s cohabitation was seen as a legitimate family form, and by the 1990s
it was normative, to the point where couples that did not live together
were seen as the “new abnormals.”

Education, class and student status

An innovation-diffusion perspective would predict that early adopters of
cohabitation would be relatively well off and have high levels of education
while later adopters may not have an education advantage. The innovation-
diffusion perspective has been used to examine the diffusion of contracep-
tion use in populations that demonstrate a rapid increase in this use over
time. Early adopters of new contraceptive technology or “innovators”
tended to be more cosmopolitan, highly educated and of a higher social
 MARRIAGE & FAMILY REVIEW 7

status; other research on “trendsetters” finds they have more disposable

income and leisure time (Cleland, 2001; Suzuki & Best, 2003). Similarly,
Becker (1963) refers to “moral entrepreneurs” who are “rule creators,” and
notes that they are typically in the upper levels of the social strata. Those
with more resources may be more able to live according to their own
beliefs, with lower costs resulting from dissent, while those with fewer
economic resources may need to rely on their social and family connec-
tions to “make ends meet,” (Edin & Lein, 1997) reducing their ability to
break social norms. These early adopters have more social influence than
other groups, causing diffusion of their behavior to other less educated and
less cosmopolitan groups, when knowledge of their behavior is spread
through personal networks or other means (Cleland, 2001). It may be no
coincidence that the couple featured in The LeClair Affair were college
attendees at an elite Ivy League university in New York City, and that
cohabitation rates rapidly increased after knowledge of their behavior and
that of other celebrities was diffused through media reports. Early reports
of cohabitation tended to focus on economic elites and especially college
students involved in the counterculture; cohabitation among the less
educated was ignored by media reports (Pleck, 2012).
Empirical research has found mixed results on class background, educa-
tion and cohabitation. Sociologists in the 1950s noted that common law
marriages were common among the lower class (Bowman, 1996) but later
research on cohabitors in the 1970s found that men with fathers with
college degrees were twice as likely to cohabit as men whose fathers had
not completed high school (Clayton & Voss, 1977). However, the same
researchers in the 1970s found that respondents who had dropped out of
high school were 25% more likely to cohabit than those who had
completed high school or college: they concluded that prior research that
had characterized cohabitation as a lifestyle of affluent persons may be
unaware of the rates at which lower class members of society had engaged
in cohabitation (Clayton & Voss, 1977). Similarly, Bumpass, Sweet, and
Cherlin (1991) found that early cohabitors were not elites as would be
predicted by the innovation-diffusion perspective. It may be that cohabit-
ation was undertaken at a higher rate by “elites” in early decades compared
to later periods, but that elites did not dominate the group of early
premarital cohabitors due to selection by other factors. Early research also
found a large number of college students among cohabitors: a study of
1970s data (Glick & Norton, 1977) found that a greater proportion of
cohabiting couples included two partners who were college students
compared to married couples, and up to one fourth of cohabiting couples
had at least one partner enrolled in college. However Pleck (2012, 47) notes
that while cohabitation among college students provoked the most
8 A. KUPERBERG

shock among older generations, cohabitation among the less educated was
and is more common, and cohabitation has often been considered “poor
people’s marriage.”
Research on more recent cohorts has found cohabitors in the United
States were less advantaged and had lower education and income compared
to married couples (Bumpass & Lu, 2000; Sassler & Miller, 2017; Smock &
Gupta, 2002). Their findings were at odds with results from Norway that
found cohabitation in recent cohorts was more common among those with
highly educated parents (Mooyaart & Liefbroer, 2016), suggesting the
United States may demonstrate distinctive patterns of cohabitation
compared to Europe. Cohabiting couples who had higher levels of income
and education were also found to be more likely to have plans to marry in
the future and to marry within a few years of cohabiting, while the less
educated were less likely to marry and progressed to marriage at a slower
pace when they did (Brown & Booth, 1996; Mernitz, 2018; Sassler & Miller,
2017). However, less educated women were no more likely to approve of
cohabitation than highly educated women (Raley, 2000), indicating that
cohabitation was probably higher among the less educated because of the
negative effects of unemployment and underemployment on financial
stability, which can reduce the marriage prospects of the less educated
(Harknett & Kuperberg, 2011).
Research on changes over time suggests a possible crossover in the
relationship between education and direct marriage. Raley (2000) found the
proportion of first unions that began as cohabiting unions increased more
steeply among the less educated over time, and higher levels of education
were associated with lower levels of cohabitation. As financial stability
became an increasingly dominant prerequisite to marriage due to changing
norms regarding the social acceptability of cohabitation and premarital sex,
direct marriage may have increasingly been undertaken only by those who
were most financially stable, specifically the college educated.

Race
The ability to meet financial goals and norms of masculinity vary by race
as a result of racial inequalities in employment, income and wealth, leading
to potential racial differences in the rate and duration of cohabitation.
Racial minorities are the “canary in the coal mine” when it comes to being
on the forefront of social change because as marginalized populations they
are most vulnerable to underlying problems in society which may encour-
age adaptive strategies (Guinier & Torres, 2009). Cultural differences in
norms related to marriage may also vary by race.
Sociologists and anthropologists in the late 1940s and 1950s believed
common law marriages were more common among African Americans, a
 MARRIAGE & FAMILY REVIEW 9

belief used to criticize common law marriages (Bowman, 1996). A study of

men in the 1970s found that Black men were almost twice as likely as
White men to have previously cohabited for at least 6 months with a part-
ner (Clayton & Voss, 1977) and a study of men and women using 1975
data found the cohabitation rate of Black people was three times the rate of
White people (Glick & Spanier, 1980). More recent literature finds that
rates of cohabitation vary by race, although findings have been inconsistent.
Bumpass and Lu (2000) find that by 1995 there were no significant racial
differences in the percent of women who had ever cohabited. However
other research during this time period found that Black women were more
likely than White women to begin their first union in cohabitation rather
than marriage (Raley, 2000), and both Black women and White women
were more likely to cohabit before marriage than Mexican American
women (Phillips & Sweeney, 2005). Black couples did not tend to be more
approving of cohabitation than White couples, which may be indicative of
a difference in the ability to marry; research has found that African
American women have a lower rate of marriage than other groups because
they do not have enough “marriageable men” due to high rates of incarcer-
ation and mortality, and racial differences in unemployment and under-
employment (Harknett & McLanahan, 2004; Phillips & Sweeney, 2005).
Black women may therefore cohabit with men who are under or
unemployed and delay marriage until they have achieved greater levels of
financial security. Supporting this idea, Brown (2000) found that
Black couples were just as likely as White couples to report marriage plans
- about 70% for both - but that Black couples were less likely to formalize
these plans into actual marriage.

Religiosity
As cohabitation became more common, direct marriers may have increas-
ingly been comprised of young adults with strong religious beliefs and
practice. Many religions have strong religious beliefs against premarital sex
which may reduce entrance into premarital cohabitation, popularly
described as “living in sin.” Sex among more religious individuals is more
likely to occur within long-term relationships and develops more slowly
and after more commitment compared to less religious individuals, while
individuals with no religious affiliation are more likely to have had many
sex partners (Laumann, Gagnon, Michael, & Michaels, 1994; Burdette,
Ellison, Hill, & Glenn, 2009). Premarital cohabitors with strong religious
views may enter marriage more quickly so that they can legitimize their
relationship in their religious community. While respondents’ religious
attendance at time of survey is an imperfect measure of religiosity at time
of cohabitation because religiosity changes over time and can be reduced
10 A. KUPERBERG

by premarital cohabitation as a result of rejection by a religious commu-

nity, religious attendance tends to be correlated across the life course
(Stolzenberg, Blair-Loy, & Waite, 1995; Wink & Dillon, 2002). Therefore
I examine both religious attendance at the time of the survey, and growing
up religious.

Method
I analyzed data from the National Survey of Family Growth (NSFG) 1995,
2002, 2006–2010 and 2011–2015 waves, along with data from the 1988
wave of the National Survey of Families and Households (NSFH) to
describe changes in premarital cohabitation and direct marriage over time.
The NSFG is a survey that is nationally representative of women in the
United States age 15–44 in 1995, 2002, 2006–2010, and 2011–2015. The
2006–2010 and 2011–2015 waves also surveyed men, but only female data
is examined in this paper. The NSFH is a nationally representative survey
of adults aged 18 and over in the United States. In the NSFH data I exam-
ined women respondents only, in order for results to be consistent with
NSFG results. The oldest respondent in the survey was aged 44 in 1956,
leading to results comparable to NSFG data. Limited results are also
presented for 1946–1955 but these earliest years may overrepresent
marriages that formed at young ages, and respondents may underreport
cohabitation rates due to issues of memory and recall (Hayford & Morgan,
2008); however as the only available national-level data on premarital
cohabitations that occurred during that era, results are still of interest.
Results from 1956–1985 (and in Figure 1, results from 1946–1955) are
drawn from the NSFH data. Results from the 1986–2015 period are drawn
from the NSFG. All marriages examined are first marriages only, and
women had to have married at least once and to have indicated whether
they cohabited with their first husband before marriage to be included in
the sample. The sample is limited to those who married at age 35 or
younger to account for age truncation in the NSFG dataset. The NSFG
dataset is additionally limited to those who responded within 10 years of
the survey date to account for age truncation. The NSFH was not age trun-
cated, and although marriages are limited to those that occurred at age 35
or younger to be comparable to the NSFG data, it is not limited to those
who answered the survey within 10 years of the survey to allow for histor-
ical comparisons, as one of the only national-level available dataset on
cohabitation that includes relationships that formed prior to the 1980s.
This method means that results from early periods may be somewhat less
complete than those more recently due to issues surrounding memory and
recall (Hayford & Morgan, 2008); results prior to the 1970s should be taken
 MARRIAGE & FAMILY REVIEW 11

80
67.0 69.6
70 63.7
58.1
60 54.6
50 45.2
40.0
40
30.4
30
20 13.2
6.5
10 0.6 0.7 2.8 2.5
0

Year of Marriage
Figure 1. Percent of first marriages preceded by cohabitation with eventual spouse, by year
of marriage.

with great caution and may underrepresent rates of premarital cohabitation.

Results from very early periods may also undercount cohabitors if some
aspect of selection led to a higher death rate among this group. Those
missing data on any variable except mother’s education (discussed below)
were removed from the sample. This resulted in a N of 3,579 first
marriages in the NSFH data that occurred between 1956 and 1985
(n removed for missing data ¼ 375), 777 first marriages in the NSFH that
occurred between 1946 and 1955 (n removed for missing data ¼ 54), and
9,480 first marriages in the NSFG data that occurred between 1986 and
2015 (n removed for missing data ¼ 210). A total of 13,059 first marriages
that occurred between 1956 and 2015 are examined.
Premarital cohabitation is measured by examining variables asking
whether respondents lived with their first husband prior to marriage. Prior
cohabitation with partners other than husbands is estimated in a separate
measure examining whether respondents formed any cohabitations prior to
their first date of marriage with a partner that they did not marry. Social
class background is examined by whether respondent’s mother completed a
college degree. Past researchers find mother’s and father’s education have
similar effects on premarital cohabitation (Mooyaart & Liefbroer, 2016),
therefore only mother’s education is examined. Mother’s education was fre-
quently unknown; 516 respondents in the NFSH data and 113 in the NSFG
data who were otherwise not missing data did not know their mother’s
highest level of education. To retain these respondents in the data a dichot-
omous variable for “mother’s education- don’t know” was included in
models. Parental divorce or separation is measured by whether respondents
reported living with both biological parents at age 14. Race was dummy-
coded to include four groups; White, Black, Latina, and “Other Race”, a
category comprised of the small number of Native Americans, Asian
12 A. KUPERBERG

Americans, and those who selected “other” as a racial category. Religiosity

was measured by whether the respondent was raised with no religion, and
whether respondents attended religious services at the time of survey fre-
quently (1þ/month), occasionally (1–11/year), or never.
Respondent’s education was measured at the time of coresidence and
marriage and was estimated on the basis of dates in which the respondent
obtained their degrees. In the 2002 and first half of the 2006–2010 wave
(2006–2008), the date of bachelor degree completion was not collected.
Following methods used by Lehrer (2008), for these years estimates were
calculated by assuming that any education beyond high school reported at
the survey collection date was obtained continuously without interruption.
For respondents missing information on dates of both high school degree
and bachelor’s degree completion, education at coresidence was imputed as
education at the time of survey. For premarital cohabitors, an additional
measure examined whether respondents completed additional education
between entering premarital cohabitation and marriage. This was measured
by whether respondents gained at least one “level” of education between
cohabitation and marriage, with four levels of education measures as Less
than High School, High School, Some College, and Bachelor’s Degree or
more. This measure underestimates the degree to which respondents may
complete additional education between cohabitation and marriage, because
those who had some college at cohabitation and complete additional years
of college but do not complete a degree, or those who attended graduate
school between cohabitation and marriage, are not captured by
this measure.
Overall rates of cohabitation are presented for 1946–1955, but the small
number of cohabitors in that cohort (n ¼ 13) precludes further analysis of
that group. Further results are presented in 15-year marriage cohorts for
the years 1956–2015. Additionally, several figures present results by 5-year
marriage cohorts. In addition to descriptive statistics, I estimated logistic
regressions predicting premarital cohabitation or direct marriage, and
premarital cohabitations of short (1 year or less) or long (5 years or more)
durations to examine the correlation of premarital cohabitation and
duration of premarital cohabitation with race, respondent’s education,
mother’s education, religious attendance, growing up with no religion,
parental divorce, and prior cohabitation experience. Models predicting
duration additionally controlled for whether couples completed additional
education between cohabitation and marriage. For the earliest cohort,
1956–1970, almost all couples cohabited for less than 12 months, so selec-
tions into short term and long term cohabitation are not examined for that
cohort. All results are weighted to account for the complex survey sampling
designs.
 MARRIAGE & FAMILY REVIEW 13

Results
Descriptive statistics by marriage cohort are included in Table 1. Reflecting
overall changes among Americans during this time period, later cohorts of
women entering their first marriage had fewer White women and more
women of color, education rates for both respondents and their mothers
and percent raised without religion increased, and level of religious attend-
ance declined in comparison to the earliest cohorts. Prior cohabitations
that did not end in marriage were also more common for later cohorts.
Figure 1 presents the percent of first marriages that were preceded by
premarital cohabitation with that partner from 1946–2015. In the first
cohort of marriages that occurred between 1946 and 1950 (n ¼ 420) only
five respondents (0.6%) reported living together with their husband before
marriage. In early 1950s 0.7% cohabited before their first marriage (8 of
385 marriages), and rates of premarital cohabitation were still under 3% by
the early 1960s. By the late 1960s rates had more than doubled to 6.5%,
and rapidly increased to 30.4% in the late 1970s, with the highest growth
rate during the 1970s. By 2011–2015 over two thirds of first marriages –
69.6% – began with premarital cohabitation.
As premarital cohabitation became more common, the duration of
premarital cohabitation rose (see Figure 2). The average premarital cohabit-
ing woman cohabited just under 6 months before marriage in the late
1950s, and 9–11 months before marrying in the 1960s. During the period
of rapid growth in cohabitation in the 1970s, duration of premarital
cohabitation grew rapidly to almost 20 months in the early 1970s, before
declining somewhat during the late 1970s to 17 months, but has been
increasing steadily since then, and grew to over two and a half years
(31.6 months) by 2011–2015.
As the average duration of cohabitation increased, fewer couples married
after living together for short term durations, of 6 months or less, or
12 months or less, and an increasing number lived together for 1–5 years, 5
or more years, or even 8 or more years before marriage (See Figure 3). In
the earliest cohort marrying in the late-1950s, 86% of couples married after
less than a year of cohabitation and 77% cohabited 6 months or fewer
before marriage, the modal duration of premarital cohabitation; by the
2011–2015 cohort couples were more likely to live together for five or
more years before marrying (18.5%) than they were to marry after living
together for 6 months or fewer (12.8%), and less than one-third married
before they could celebrate their one-year anniversary of coresidence.
However, cohabitations longer than 8 years that ended in marriage were
still unusual; just 3.2% of couples marrying in 2011–2015 had lived
together for 8 or more years before marriage.
14 A. KUPERBERG

Table 1. Descriptive statistics by year of marriage to first husband, women aged <36
at marriage.
Year of Marriage 1956–1970 1971–1985 1986–2000 2001–2015
Race
White 83.0 80.8 69.7 65.6
Hispanic 7.4 8.0 14.4 16.6
Black 8.7 8.5 9.7 10.4
Other Race 0.9 2.7 6.2 7.5
Education
<HS at Coresidence 27.6 21.5 10.6 14.6
HS At Coresidence 43.9 37.2 31.5 21.0
Some College at Coresidence 17.4 23.6 27.6 33.7
BAþ at Coresidence 11.2 17.6 30.3 30.7
<HS at Marriage 27.6 20.3 8.6 11.8
HS at Marriage 43.9 37.6 31.8 21.9
Some College at Marriage 17.4 23.5 27.3 32.5
BA þ at Marriage 11.2 18.6 32.3 33.8
Mother’s Education
Less than BA 83.9 83.8 83.0 75.9
BAþ 5.7 10.1 16.7 23.3
Unknown 10.4 6.2 0.4 0.8
Lived with Both Biological Parents at age 14 80.1 81.4 70.6 67.3
Raised with No Religion 2.1 4.5 7.6 8.1
Religious Service Attendance
None 17.5 20.1 21.0 20.8
0 < x < 12 /year 21.4 23.2 27.1 25.9
1þ/month 61.0 56.7 52.0 53.3
Prior Cohabitating Partner 0.5 4.7 14.3 20.1
n 1,491 2,088 4,665 4,210
n Premarital Cohabitors 81 614 2,510 2,759

35 32.5 31.6
29.9
30 27.6

25 23.0
21.5
19.8 19.2
20 17.0

15
11.2
9.1
10
5.8
5

Year of Marriage
Figure 2. Average duration of premarital cohabitation with first husband among premarital
cohabitors (months).

Over this time period, as premarital cohabitation became more common

and longer lasting, both premarital cohabitors and direct marries were
increasingly likely to have previously cohabited with partners other than
their first husband, before entering marriage (See Figure 4). Rates of prior
 MARRIAGE & FAMILY REVIEW 15

100
90
80
70
60
50
40
30
20
10
0

Year of Marriage
6 Months or Fewer 12 months or fewer 1 Year < X < 5 years

5+ Years 8+ Years
Figure 3. Percent of premarital cohabitations of long or short duration.

35
31.0
30 28.3
23.7
25 21.2 20.4
18.9
20
15.7
15 12.1
8.6 8.1 7.9
10 6.7
5.8 6.1 5.9
3.3 3.6 4.1
5 1.2
0.4
0.0 0.3 0.3 0.7
0

Year of Marriage

Premarital Cohabitors Direct Marriers

Figure 4. Non-marital cohabitation with partners prior to first husband.

non-marital cohabitation (cohabitations that did not lead to marriage) were

consistently low among direct marriers; rates among this group were as
low as .3% in the late 1960s and reached a peak of 8.6% in the late 1990s
before declining to 5.9% in 2011–2015, reflecting some increasing selection
among this increasingly small group. Rates of prior cohabitation with other
partners among premarital cohabitors continued to increase over time,
reaching a peak of 31% in 2011–2015, indicating that just over two-thirds
of women who married after cohabitation had only cohabited with one
partner before marriage; their first husband. Logistic regressions in Table 2
indicate that after a large gap in prior cohabitation rates in the 1971–1985
cohort, where cohabitors were over 6.6 times more likely than direct
marriers to have cohabited with other partners prior to their first husband,
16 A. KUPERBERG

Table 2. Odds ratios, from Logistic regressions predicting premarital cohabitation (1) or direct
marriage (0).
Year of Marriage 1956–1970 1971–1985 1986–2000 2001–2015
Race
Black 2.17 1.46 1.34 1.27
Latina/o 2.23† 0.93 0.58 0.85
Other Race 0.00 1.17 0.53 0.36
Education
HS at coresidence 0.93 0.78 0.65 0.70†
Some College at coresidence 1.33 1.01 0.57 0.61
BA þ at coresidence 1.10 0.85 0.53 0.46
Mother has BA† 1.79 1.19 0.76 0.64
Mother’s Education Unknown 1.11 0.77 0.29 1.07
Lived with both biological parents at age 14 0.74 0.62 0.62 0.75
Raised with No Religion 1.68 1.03 1.40 1.53
Religious Attendance
None 1.54 1.11 0.99 1.03
1þ/month 0.72 0.47 0.39 0.33
Prior Cohabiting Partner 5.58 6.62 2.61 3.22
n 1,491 2,088 4,665 4,210
Note. Reference: White, Less than high school at coresidence, mother has no bachelor’s degree, attends religious
services 1–11/Year. †p < .10, p < .05, p < .01, p < .001.

the difference between premarital cohabitors and direct marriers in prior

cohabitations before their spouse shrank to a factor of 2.6 times in
1986–2000 before increasing slightly to 3.2 times in 2001–2015.

Selection into premarital cohabitation and direct marriage by education,

student status, and class background
Regression results presented in Table 2 indicate differential rates of selec-
tion into premarital cohabitation by education for different cohorts,
and Table 3 indicates selection into long-term or short-term cohabitations
among premarital cohabitors. Figure 5 presents differences in college
graduation rates between premarital cohabitors and direct marriers,
including premarital cohabitors’ college graduation rates at the time of
co-residence and at time of marriage. Figure 6 presents the rate of cohabit-
ation by education at coresidence and year of marriage, and Figure 7
presents the percent of cohabiters obtaining an additional level of education
between cohabitation and marriage.
The earliest cohort of premarital cohabitors in 1956–1970 did not dem-
onstrate significant differences in education, with cohabitors just as likely
to come from any education level. These early cohabitors were also not
very likely to be students; less than 1% of cohabitors in this cohort com-
pleted additional education between cohabitation and entrance into
marriage. As seen in Figure 6, after The Leclair Affair in 1968, those with
less than a high school degree and those with some college had the highest
rates of cohabitation, but education differences remained insignificant in
this cohort in terms of overall cohabitation rates, although some significant
Table 3. Selection into short term (0–12 month) and long term (5þ years) premarital cohabitation among premarital cohabitors, by year of marriage.
Year of Marriage 1971–1985 1986–2000 2000–2015
Length of Premarital Cohabitation: 1–12 Months 5þ Years 1–12 Months 5þ Years 1–12 Months 5þ Years
Race
Black 0.93 0.99 0.82 1.97 0.73† 1.60
Latina/o 0.74 2.17 0.92 1.87 0.81 2.12
Other Race 0.67 2.25 0.90 0.57 1.79 1.63
Education
HS at coresidence 1.56† 0.44† 0.75† 1.05 1.00 0.92
Some College at coresidence 0.92 0.26 0.73 0.62† 0.81 0.79
BA þ at coresidence 1.35 0.32 0.86 0.58† 0.92 0.33
Mother has BAþ 0.92 0.40 1.04 0.83 1.30 0.91
Mother’s Education Unknown 0.90 2.50† 0.66 0.92 1.26 2.9
Lived with both biological parents at age 14 1.23 0.90 1.17 1.18 0.97 1.18
Raised with No Religion 1.22 0.83 1.03 0.83 0.85 1.48
Religious Attendance
None 1.03 2.02 1.04 1.03 0.81 1.45
1þ/month 1.15 1.62 1.33 0.79 1.50 1.13
Prior Cohabiting Partner 0.65 1.95 0. 75 1. 08 0.86 0.74
Married after Additional Education 0. 22 7.74 0. 23 1. 96 0. 20 1.96
n 614 614 2,510 2,510 2,759 2,759
Note. Reference: White, less than high school at coresidence, mother has no bachelor’s degree, attends religious services 1–11/Year. Results are restricted to premarital cohabitors and
predict whether couples cohabit for 1 year or less or longer, and 5 or more years or shorter. †p < .10, p < .05, p < .01, p < .001.
MARRIAGE & FAMILY REVIEW
17
18 A. KUPERBERG

60
50
40
30
20
10
0

Year of Marriage

Direct Marriers Cohabitors: At Cohabitation

Cohabitors: At Marriage
Figure 5. Percent with bachelor’s degree.

90
80
70
60
50
40
30
20
10
0

Year of Marriage

Less than High School High School

Some College Bachelors Degree Plus
Figure 6. Cohabitation rate by education at coresidence.

effects were found in length of cohabitation, with those who obtained some
college less likely than those without a high school degree to cohabit for
long periods before marriage (See Table 3). The post-Leclair era of
1971–1975 saw a spike in the number of students cohabiting; using a
conservative measure of additional schooling, over 9% of premarital cohab-
itors during this period completed additional education between cohabit-
ation and marriages, rates which dropped subsequent to that 5-year period
and which were not seen again until the 2000s (See Figure 7). Figure 5 also
demonstrates that during this period, premarital cohabitors had somewhat
(although not significantly) lower levels of college completion at the time
of cohabitation compared to direct marriers at marriage, but were as likely
as direct marriers to have a college degree by the time they married.
 MARRIAGE & FAMILY REVIEW 19

12 10.9
9.9
10 9.3 9.2 9.4
8.6
8
6.6 6.3
6 4.9

2 0.9
0.0 0.0
0

Year of Marriage
Figure 7. Percent of cohabitors obtaining additional education between cohabitation
and marriage.

Beginning in the mid-1980s, premarital cohabitation increased more

rapidly among those with lower education levels (See Figure 6), and pre-
marital cohabitors had significantly lower levels of education than direct
marriers (See Figure 5) both at cohabitation and marriage, as direct
marriage became less common, and increasingly the province of the well-
educated. Table 2 demonstrates that in 1986–2015, education was
negatively associated with premarital cohabitation. By 2011–2015, 48.1% of
direct marriers had completed college when they first married versus only
28% of premarital cohabitors when they moved in together, and just 33.6%
at marriage. Those with a college degree were only 46% as likely as high
school dropouts to premaritally cohabit, and those with some college were
61% as likely. In the 2000s cohabitors with a college degree were only
one-third as likely as those who dropped out of high school to cohabit for
long periods of time before marriage. Reflecting the increasing difficulty in
establishing themselves financially, the percent of cohabitors obtaining
additional education between cohabiting and marriage increased between
the late 1970s and 2010s (See Figure 7). In all cohorts, obtaining additional
education between cohabitation and marriage was associated with a signifi-
cantly higher rate of cohabiting for 5þ years before marriage, and a signifi-
cantly lower rate of cohabiting less than 1 year.
Mother’s education rates shown in Figure 8 provides a measure of social
class background that, like respondents’ education, may be correlated with
the ability to marry directly. Similarly to measures of respondent’s educa-
tion, in the 1956–1985 cohorts, mother’s education was not significantly
related to cohabitation (See Table 2) and Figure 8 demonstrates that for
some early cohorts, premarital cohabitors were more likely to have a
20 A. KUPERBERG

40
35
30
25
20
15
10
5
0

Year of Marriage

Premarital Cohabitors Direct Marriers

Figure 8. Percent with mother with bachelor’s degree (known education only).

mother with a college degree. In the 1986–2015 cohorts, those who have a
mother with a college degree were significantly less likely to cohabit with
their spouse prior to marriage. Mother’s education was not related to the
duration of premarital cohabitations.

Selection by race and ethnicity, religiosity, and parental marital status

Table 2 demonstrates that the earliest premarital cohabitors - the innova-
tors - were significantly more likely to be women of color, with both Black
and Latina women 2.2 times as likely as White women to cohabit before
marriage. In this cohort, no women of other races reported cohabiting
before marriage. By the second cohort, as cohabitation spread after The
Leclair Affair, fewer racial differences were seen, as White women’s rates of
cohabitation rose to meet Latina’s rates of cohabitation, but Black women
still cohabited at a significantly higher rate. In the 1986–2000 cohort,
reflecting more rapid growth in their premarital cohabitation rates, White
women were significantly more likely than Latina or women of other races
to cohabit before marriage, but less likely than Black women to do so. In
the last cohort, 2001–2015, racial differences diminished further as Latina
women caught up with White and Black women in rates of premarital
cohabitation and the Black–White difference became non-significant, lead-
ing to an equal likelihood of premarital cohabitation among Black, Latina,
and White women. Other race women were now the outliers, only 36% as
likely as White women to cohabit with a spouse before marriage, a signifi-
cant difference. As inequality in premarital cohabitation by race subsided,
another type of inequality in premarital cohabitation experiences arose; in
1986–2015, both Black and Latina women were significantly more likely
than White women to cohabit for 5þ years before they married.
 MARRIAGE & FAMILY REVIEW 21

Examining premarital cohabitors and direct marriers’ religious attend-

ance in Table 2 demonstrates that in the first cohort these differences were
not significant, but in 1971–2015, those who attended religious services at
least once a month were significantly less likely to have cohabited
before marriage compared to those who attended less frequently, while
non-regular attendees (1–11/year) and those who never attended religious
services had similar rates of premarital cohabitation. In 1956–1985, being
raised with no religion did not significantly impact cohabitation rates, but
in 1986–2015 premarital cohabitation was more common among those
raised with no religion. By 2011–2015 those who directly married were
very religious; 73% attended religious services at least once a month, while
only 46% of premarital cohabitors who married during that period attended
religious services frequently; 29% of premarital cohabitors in that cohort
never attended religious services versus only 11% of direct marriers
(results not shown, available from author). Frequent religious service
attendance was also associated with faster entry into marriage in 1986–2015
(See Table 3).
The first cohort’s rate of cohabitation was not related to whether individ-
uals had been raised with married biological parents or whether they
were raised by a single parent or had witnessed their parents’ divorce or a
parent’s death by age 14, as shown in Table 2. After the popularization of
cohabitation by The LeClair Affair, premarital cohabitation became signifi-
cantly more common among those who did not live with both biological
parents at age 14 for all cohorts from 1971–2015, perhaps as these cohorts
used cohabitation as a trial marriage in response to their childhood experi-
ences with marriage. Parent’s marital status was not related to the duration
of premarital cohabitations.

Discussion
After almost no such relationships existing in the late 1940s and early
1950s, couples began to live together before marrying in the late 1950s and
premarital cohabitation exploded in popularity in the 1970s, remaining
popular ever since; nearly 70% of first marriages in 2011–2015 began in
premarital cohabitation, indicating this is now the normative pathway that
young adults take into marriage. This article contributes to past literature
by examining the roots of these changes, along with the increasing preva-
lence, duration, and changing selection into premarital cohabitation and
durations over this period. The lengthening of premarital cohabitation
during this period reflected the growing acceptance and normativity of this
relationship stage; when cohabitation was less socially acceptable, cohabi-
tors may have felt under increased pressure to transition to marriage
22 A. KUPERBERG

quickly. Cohabitations may also have lengthened as young adults took

longer to build up their financial standing before marriage.
The reasons for the rise in cohabitation rates over this time period are
complex, and many are tied to the increased control of childbearing and
subsequent increase in education, labor force participation, and age at
marriage among women. Premarital cohabitation gained popularity after a
similar relationship that was previously codified as being a type of marriage
- common law marriage - lost its legal standing as marriage. Groups who
had already been marginalized by marriage law and civil society, and which
historically may have had higher rates of common law marriage - specific-
ally Black and Latina women – continued their prior practice of moving in
with partners without a marriage license. Within a decade, this practice
had spread further to include more students and more White women.
Following a convergence of social forces that included the rising age at
marriage and rates of college education and subsequent “independent life
stage” (Rosenfeld, 2007), the technologically innovative birth control pill
and subsequent sexual revolution, and popularized by The Leclair Affair
and further media stories about celebrities cohabiting, cohabitation diffused
to the general population in the 1970s. The historical timing of this popu-
larization is consistent with an innovation-diffusion perspective.
Selection into cohabitation versus direct marriage changed dramatically
over the time period studied. Results indicate that premarital cohabitors
can be roughly divided into three patterns of behavior over this time
period. The first cohort, those married in 1956–1970, represents the group
of “innovators” who for the most part cohabited (and later married) before
The LeClair Affair brought cohabitation to the public milieu in 1968.
Consistent with an innovation-diffusion perspective, the vanguard of
cohabiting women who transgressed strong norms to live with their first
spouse before marriage in the late 1950s and 1960s was unusual in a num-
ber of ways. Compared to later cohorts of premarital cohabitors, these
pioneers were more likely to have mothers with high levels of education
(a measure of social class background) and were comparatively more
educated- or at least did not exhibit the sharper education and class divi-
sions between premarital cohabitors and direct marriers of later cohorts.
Demonstrating their role as the “canary in the coal mine” (Guinier &
Torres, 2009) women of color were also more likely to cohabit during this
period. Premarital cohabitors did not exhibit the lower rates of religiosity
and higher rates of parental divorce/separation that characterized later
cohort. They were also not as elite as would be predicted by an innovation-
diffusion perspective. This can be explained by the unique nature of pre-
marital cohabitation; selection into this particular behavior among the
financially insecure and historically marginalized was a strong selective
 MARRIAGE & FAMILY REVIEW 23

force throughout the history of cohabitation. In the earliest period exam-

ined, this type of selection combined with selection into cohabitation
among the more-privileged “innovators,” and likely led to these results, in
which selection forces “canceled each other out” to some degree, and
cohabitors and direct marriers seem to have few education or class
differences.
As cohabitation diffused from a select group into the more general popu-
lation in the 1970s, cohabitors and direct marriers began to converge in
terms of race. Rates remained similar by education. Adoption of cohabit-
ation among students spiked during this period, although remained the
minority of cohabitors, with only 9.3% of premarital cohabitors completing
additional education before marriage in 1971–1975. Cohabitations in the
1971–1975 post-LeClair period were also of unusually long durations for
their time. Religiosity and parental marital status also began to affect selec-
tion in this second cohort of premarital cohabitors as it spread to the more
general population.
Later, as premarital cohabitation diffused to the general population
between 1986 and 2015, differences by education arose, as forces related to
selection into innovation subsided, and forces related to financial security
and the ability of young adults, and young men in particular, to establish
themselves persisted and perhaps grew stronger. As cohabitation became
more widespread, those who were less financial secure increasingly utilized
it prior to marriage, while more educated women and those of the highest
class background who had greater financial stability were more able than
those with lesser financial means to select into direct marriage, and there-
fore increased their cohabitation rates at a slower pace. In contrast to early
periods, women who were less educated and had less educated mothers
now premaritally cohabited at higher rates, perhaps in order to save funds
and achieve the financial prerequisites they hoped to achieve before mar-
riage. Those without a college degree were also more likely to cohabit over
5 years before marriage. Racial differences in premarital cohabitation rates
disappeared between White, Black and Latina women, although Black and
Latina women were considerably more likely than White women to cohabit
for long periods of time (5þ years) before marriage, revealing inequalities
not fully captured by examining overall rates of premarital cohabitation.
Direct marriers were also increasingly comprised of very religious women
during this time period, as religious constraints on premarital cohabitation
remained one of the few barriers to entry into an increasingly dominant
prerequisite to marriage.
This study demonstrates that selection into a new additional relationship
stage of couples that eventually marry (premarital cohabitation) shifted in
accordance with an innovation-diffusion perspective, but that selection
24 A. KUPERBERG

forces related to financial insecurity were a strong and sometimes opposing

force present throughout the history of cohabitation. It adds to prior
literature by examining selection into premarital cohabitation over time
and in historical context and proposing explanations for these changes, by
updating results through 2015, and by examining selection into short-term
and long-term durations of premarital cohabitation. Several questions
remain that may be better understood with additional qualitative and quan-
titative data analyses beyond the scope of this paper. Why do some young
adults cohabit for long periods of time; are these couples moving in quickly
after dating and delaying marriage as they get to know each other better,
or delaying marriage due to economic or other factors? To what extent do
men’s factors or couple-level factors affect selection into premarital cohabit-
ation? How do short-term economic recessions impact short-term selection
into premarital cohabitation and direct marriage? Recent research suggests
the Great Recession that took place from 2007–2011 did not have strong
impacts on overall marriage or cohabitation rates, but that marriage rates
declined somewhat for women with low levels of education, and one study
found an unusual increase in the number of new cohabitations in 2009 and
2010 (Kreiger, 2010; Schneider & Hastings, 2015; Schneider, 2017).
Preliminary analysis (not shown, available from author) shows a dip in the
percent of marriages that began in premarital cohabitation among mar-
riages formed in 2009 compared to the years earlier and later, which may
result from greater selection into marriage by the more-financially-secure
direct marriers during that year, one of the worst of the Great Recession.
Future research should explore these trends more fully.
This study had some limitations. Data prior to the 1970s likely underre-
presents rates of premarital cohabitation due to problems of recall for rela-
tionships that formed in distant periods, since retrospective life-history data
were not collected until 1988; results from prior to 1970 are presented for
interest, as the only available data from this time period, but strong conclu-
sions can not be drawn from that period and results should be interpreted
with caution. Women that died at young ages are also underrepresented
for those periods. Although the number of marriages in very early periods
allows for calculation of an overall rate of premarital cohabitation, the sam-
ple of premarital cohabitors for the period prior to 1956 is too small to
allow for detailed examination, and even fewer cohabitations in these data
prior to 1971 lasted longer than one year, precluding examination of
cohabitations of certain durations during this period. Finally, the measure
of obtaining additional education is a conservative estimate that does not
account for those who completed additional education without completing
an additional degree, or those who attended graduate school
while cohabiting.
 MARRIAGE & FAMILY REVIEW 25

In addition to providing insight into the rise of premarital cohabitation,

these data demonstrate the consequences of social changes that led to its
widespread adoption and domination. Examining which types of couples
are more likely to select into short term or longer-term premarital cohabit-
ation also reveals societal forces impacting relationship formation. These
forces can include normative pressures to avoid premarital cohabitation or
move into marriage quickly for certain groups (such as the more religious),
along with forces which can delay entrance into marriage for couples that
cannot achieve certain financial goals or are working to complete education
before marrying. As young adults found it harder to achieve the markers of
adulthood and masculinity that are cultural signals of readiness to marry,
and as cohabitation became more socially acceptable and marriage less of
an economic necessity, premarital cohabitation was increasingly likely to be
undertaken by those who have delayed marriage until they can achieve the
increasingly elusive cultural markers of economic success, indicating some
of those couples may have directly married if financially secure. Policies
aimed at promoting economic opportunities for young adults and reducing
racial disparities may, therefore, lead to a stabilization or decline in rates
and durations of premarital cohabitation.

Acknowledgments
The author wishes to thank Jerry Jacobs, Kristen Harknett, Herbert Smith, Barbara Risman,
Wendy Manning, and Sharon Sassler for their helpful comments on earlier versions of this
paper. Earlier versions of this research were presented at the 2010 Eastern Sociological
Society conference and the 2012 National Survey of Family Growth conference.

Funding
This research was funded in part by a National Science Foundation Graduate
Research Fellowship.

References
Axinn, W. G., & Thornton, A. (2000). The transformation in the meaning of marriage. In
L. J. Waite, C. Bachrach, M. Hindin, E. Thomson, & A. Thornton (Eds.), The ties that
bind (pp. 147–165). New York: Aldine de Gruyter.
Bailey, B. (2011). Sexual revolution(s). In A. S. Skolnick & J. H. Skolnick (Eds.), Family in
transition, 16th ed. New York: Pearson.
Becker, H. S. (1963). Outsiders: Studies in the sociology of deviance. New York: The Free
Press.
Blom, S. (1994). Marriage and cohabitation in a changing society: Experience of Norwegian
men and women born in 1945 and 1960. European Journal of Population ¼ Revue
Europeenne de Demographie, 10 (2), 143–173.
26 A. KUPERBERG

Bowman, C. G. (1996). A feminist proposal to bring back common law marriage. Oregon
Law Review, 75 (3), 709–780.
Brown, S. L. (2000). Union transitions among cohabitors: The significance of relationship
assessments and expectations. Journal of Marriage and Family, 62 (3), 833–846.
Brown, S. L., & Booth, A. (1996). Cohabitation versus marriage: A comparison of relation-
ship quality. Journal of Marriage and the Family, 58 (3), 668–678.
Bumpass, L., & Lu, H. (2000). Trends in cohabitation and implications for children’s family
contexts in the United States. Population Studies, 54 (1), 29–41.
Bumpass, L. L., Sweet, J. A., & Cherlin, A. (1991). The role of cohabitation in declining
rates of marriage. Journal of Marriage and the Family, 53 (4), 913–927.
Burdette, A. M., Ellison, C. G., Hill, T. D., & Glenn, N. D. (2009). “Hooking up” at college:
Does religion make a difference? Journal for the Scientific Study of Religion, 48 (3),
535–551.
Casper, L. M., & Bianchi, S. M. (2011). Cohabitation. In A. S. Skolnick & J. H. Skolnick
(Eds.), Family in transition, 16th ed. New York: Pearson.
Cherlin, A. (2004). The deinstitutionalization of American marriage. Journal of Marriage
and Family, 66 (4), 848–861.
Cherlin, A. J. (2009). The marriage-go-round: The state of marriage and the family in
America today. New York: Knopf.
Clayton, R. R., & Voss, H. L. (1977). Shacking up: Cohabitation in the 1970s. Journal of
Marriage and the Family, 39 (2), 273–283.
Cleland, J. (2001). Potatoes and pills: An overview of innovation-diffusion contributions to
explanations of fertility decline. In J. B. Casterline (Ed.), Diffusion processes and fertility
ltransition: Selected perspectives (pp. 39–65). Washington, DC: National Academy Press.
Coontz, S. (1992). The way we never were: American families and the nostalgia trap.
Hachette UK: Basic Books.
Danziger, C., & Greenwald, M. (1977). An Overview of Unmarried Heterosexual
Cohabitation and Suggested Marketing Implications. Advances in Consumer Research, 4
(1), 330–334.
Daugherty, J., & Copen, C. (2016). Trends in attitudes about marriage, childbearing and
sexual behavior: United States, 2002, 2006–2010, and 2011–2013. National Health
Statistics Reports, (92), 1–11.
Dubler, A. R. (1998). Governing through contract: Common law marriage in the
Nineteenth century. The Yale Law Journal, 107 (6), 1885–1920.
Edin, K., & Lein, L. (1997). Making ends meet: How single mothers survive welfare and low-
wage work. New York: Russell Sage Foundation.
Edin, K., & Kefalas, M. (2005). Promises I can keep. Berkeley: University of California
Press.
EEOC. (1978). The Pregnancy Discrimination Act of 1978. Retrieved from http://www.eeoc.
gov/laws/statutes/pregnancy.cfm
Eickmeyer, K. J., & Manning, W. D. (2018). Serial cohabitation in young adulthood: baby
boomers to millennials. Journal of Marriage and Family, 80 (4), 826–840.
Finer, L. B. (2007). Trends in premarital sex in the United States, 1954–2003. Public Health
Reports (Washington, D.C.: 1974), 122 (1), 73–78.
Frank, R. H. (1999). Luxury fever: Why money fails to satisfy in an era of excess. New York:
Robert H. Frank Free Press.
Furstenberg, F. F., Kennedy, S., McLoyd, V. C., Rumbaut, R. G., & Settersten, R. A. (2004).
Growing up is harder to do. Contexts, 3 (3), 33–41.
 MARRIAGE & FAMILY REVIEW 27

Glick, P. C., & Norton, A. J. (1977). Marrying, divorcing, and living together in the U.S.
today. Population Bulletin, 32 (5), 1–41.
Glick, P. C., & Spanier, G. B. (1980). Married and unmarried cohabitation in the United
States. Journal of Marriage and the Family, 42 (1), 19–30.
Goldin, C., & Katz, L. F. (2002). The power of the pill: Oral contraceptives and women’s
career and marriage decisions. Journal of Political Economy, 110 (4), 730–770.
Guinier, L., & Torres, G. (2009). The miner’s canary: Enlisting race, resisting power, trans-
forming democracy. Cambridge: Harvard University Press.
Harknett, K., & Kuperberg, A. (2011). Education, labor markets, and the retreat from
marriage. Social Forces; a Scientific Medium of Social Study and Interpretation, 90 (1),
41–63.
Harknett, K., & McLanahan, S. (2004). Racial and ethnic differences in marriage after the
birth of a child. American Sociological Review, 69(6), 790–811.
Hayford, S. R., & Morgan, P. S. (2008). The quality of retrospective data on cohabitation.
Demography, 45 (1), 129–141.
Heuveline, P., & Timberlake, J. M. (2004). The role of cohabitation in family formation:
The United States in comparative perspective.
Hoelter, L. F., & Stauffer, D. E. (2002). “What does it mean to be ‘just living together’ in
the new millennium? An overview.” In A. Booth & A. C. Crouter (Eds.), Just living
together: Implications of cohabitation on families, children, and social policy. Mahwah, NJ:
Lawrence Erlbaum Associates Inc.
Kiernan, K. (2002). Cohabitation in Western Europe: Trends, issues and implications. In A.
Booth & A. C. Crouter (Eds.), Just living together: Implications of cohabitation on
families, children, and social policy. Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Kreiger, R. (2010). Increase in opposite-sex cohabiting couples from 2009 to 2010 in the
annual social and economic supplement (ASEC) to the current population survey (CPS).
Housing and Household Statistics Division, US Census Bureau.
Kroeger, R. A., & Smock, P. J. (2014). Cohabitation: Recent research and implications. In
J. K. Treas, J. Scott, & M. Richards (Eds.), The Wiley Blackwell companion to the
sociology of families (pp. 217–235). Chichester, UK: John Wiley & Sons, Ltd.
Kuperberg, A. (2009). Motherhood and graduate education: 1970–2000. Population
Research and Policy Review, 28 (4), 473–504.
Kuperberg, A. (2012). Reassessing differences in work and income in cohabitation and mar-
riage. Journal of Marriage and Family, 74 (4), 688–707.
Kuperberg, A. (2014). Age at coresidence, premarital cohabitation and marriage dissolution:
1985–2009. Journal of Marriage and Family, 76 (2), 352–369.
Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organiza-
tion of sexuality. Chicago: University of Chicago Press.
Lehrer, E. L. (2008). Age at marriage and marital instability: Revisiting the
Becker–Landes–Michael hypothesis. Journal of Population Economics, 21 (2), 463–484.
Lichter, D. T., & Qian, Z. (2008). Serial cohabitation and the marital life course. Journal of
Marriage and Family, 70 (4), 861–878.
Lundberg, S., & Pollak, R. A. (2007). The American family and family economics. Journal
of Economic Perspectives, 21 (2), 3–26.
Lundberg, S., Pollak, R. A., & Stearns, J. (2016). Family inequality: Diverging patterns in
marriage, cohabitation, and childbearing. Journal of Economic Perspectives, 30(2), 79–102.
Madrick, J., & Papanikolaou, N. (2007). The stagnation of male wages. Schwartz Center
for Economic Policy Analysis, New School for Social Research Policy Note. Retrieved
28 A. KUPERBERG

from http://www.economicpolicyresearch.org/images/docs/research/employment/Madrick
_Nicolas.pdf
Manning, W. (2010). Trends in cohabitation: Twenty years of change, 1987–2008 (FP-10-07).
National Center for Family & Marriage Research. Retrieved from https://www.bgsu.edu/
content/dam/BGSU/college-of-arts-and-sciences/NCFMR/documents/FP/FP-10-07.pdf
Mernitz, S. (2018). A cohort comparison of trends in first cohabitation duration in the
United States. Demographic Research, 38, 2073–2086.
Mooyaart, J. E., & Liefbroer, A. C. (2016). The influence of parental education on timing
and type of union formation: Changes over the life course and over time in the
Netherlands. Demography, 53 (4), 885–919.
Mosisa, A., & Hipple, S. (2006). Trends in labor force participation in the United States.
Bureau of Labor Statistics Monthly Labor Review, 129 (10), 35–57.
Nakonezny, P. A., Shull, R. D., & Rodgers, J. L. (1995). The effect of no-fault divorce law
on the divorce rate across the 50 states and its relation to income, education and religi-
osity. Journal of Marriage and the Family, 57 (2), 477–488.
Oppenheimer, V. K. (1988). A theory of marriage timing. The American Journal of
Sociology, 94 (3), 563–591.
Phillips, J. A., & Sweeney, M. M. (2005). Premarital cohabitation and marital disruption
among White, Black, and Mexican American women. Journal of Marriage and Family,
67 (2), 296–314.
Pleck, E. (2012). Not just roommates: Cohabitation after the sexual revolution. Chicago:
University of Chicago Press.
Raley, R. K. (2000). Recent trends and differentials in marriage and cohabitation in the
United States. In L. J. Waite, C. Bachrach, M. Hindin, E. Thomson, & A. Thornton
(Eds.), The ties that bind (pp. 19–39). New York: Aldine de Gruyter.
Rosenfeld, M. J. (2007). The age of independence: Interracial unions, same sex unions and
the changing American family. Cambridge: Harvard University Press.
Sassler, S., & Miller, A. (2017). Cohabitation nation: Gender, class, and the remaking of
relationships. Oakland, CA: University of California Press.
Schneider, D. (2017). The effects of the Great Recession on American families. Sociology
Compass, 11 (4), e12463.
Schneider, D., & Hastings, O. P. (2015). Socio-economic variation in the demographic
response to economic shocks: Evidence from the Great Recession. Demography, 52 (6),
1893–1915.
Schor, J. B. (1998). The overspent American: Upscaling, downshifting, and the new consumer.
New York: Basic Books.
Seltzer, J. A. (2000). Families formed outside of marriage. Journal of Marriage and Family,
62 (4), 1247–1268.
Smock, P. J., & Gupta, S. (2002). Cohabitation in contemporary North America. In
A. Booth & A. C. Crouter (Eds.), Just living together: Implications of cohabitation on fam-
ilies, children, and social policy. Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Smock, P. J., Manning, W. D., & Porter, M. (2005). “Everything’s there except money”:
How money shapes decisions to marry among cohabitors. Journal of Marriage and
Family, 67 (3), 680–696.
Snyder, T. D., & Dillow, S. (2012). Table 213: Enrollment rates of 18 to 24 year olds in
degree-granting institutions, by level of institution and sex and race/ethnicity of students:
1967 through 2010. In Institute of Education Sciences (U.S.) National Center for
Education Statistics Staff (Ed.), Digest of Education Statistics 2011. Government Printing
Office.
 MARRIAGE & FAMILY REVIEW 29

Stolzenberg, R. M., Blair-Loy, M., & Waite, L. J. (1995). Religious participation in early
adulthood: Age and family life cycle effects on church membership. American
Sociological Review, 60 (1), 84–103.
Suzuki, T., & Best, J. (2003). The emergence of trendsetters for fashions and fads: Kogaru
in the 1990s Japan. The Sociological Quarterly, 44 (1), 61–79.
Thornton, A., & Young-Demarco, L. (2001). Four decades of trends in attitudes toward
family issues in the United States: The 1960s through the 1990s. Journal of Marriage and
Family, 63 (4), 1009–1037.
Warren, E., & Tyagi, A. W. (2011). Why middle-class mothers and fathers are going broke.
In A. S. Skolnick & J. H. Skolnick (Eds.), Families in transition. Boston: Pearson.
Wiik, K. A. (2008). ‘You’d Better Wait!’—socio-economic background and timing of first
marriage versus first cohabitation. European Sociological Review, 25 (2), 139–153.
Wink, P., & Dillon, M. (2002). Spiritual development across the adult life course: Findings
from a longitudinal study. Journal of Adult Development, 9 (1), 79–94.
Wydick, B. (2007). Grandma was right: Why cohabitation undermines relational satisfac-
tion, but is increasing anyway. Kyklos, 60 (4), 617–645.