Parental Divorce and Offspring Marriage: Early or Late?

Abstract

At least 25 separate studies have examined the impact of family
structure on offspring marriage timing. Some find that parental divorce
makes marriage more likely, while others show that it delays or deters
marriage. This research analyzes data from the 1973-94 NORC General
Social Survey with the intention of shedding light on the debate. The
extraordinarily varying results of prior studies can probably be
attributed to change across two dimensions of time, individual
life course and historical period. In 1973 parental divorce greatly
increased the chances of marriage but by 1994 people from divorced
families were slightly less likely to marry than were people from intact
families. Furthermore, parental divorce raises the likelihood of teenage
marriage, but if the children of divorce remain single past age 20 they
are disproportionately likely to avoid wedlock.

Full text

Parental Divorce and Offspring Marriage: Early or Late?
Author(s): Nicholas H. Wolfinger
Source:
Social Forces,
Vol. 82, No. 1 (Sep., 2003), pp. 337-353
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/3598148 .
Accessed: 23/01/2014 18:14
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Social Forces.
http://www.jstor.org
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental
Divorce and Offspring Marriage:
Early
or Late?*
NICHOLAS
H. WOLFINGER,
University of Utah
Abstact
At least 25 separate
studies have examined the
impact
offamily
structure on offspring
marriage
timing.
Somefind
thatparental
divorce makes
marriage
more
likely,
while
others
show that it delays
or deters
marriage.
This research
analyzes
data
from the
1973-94 NORC General Social
Survey
with the intention
of shedding light
on the
debate. The
extraordinarily
varying
results
of prior
studies can
probably
be attributed
to change
across two dimensions
of time,
individual
life
course and historical
period.
In 1973
parental
divorce
greatly
increased the chances
of marriage
but
by 1994
people
from
divorced
families
were
slightly
less
likely
to marry
than were
people from
intact
families.
Furthermore,
parental
divorce raises the likelihood
of teenage
marriage,
but
if the
children
of divorce remain
single past age
20 they
are
disproportionately likely
to avoid wedlock.
At least 25 separate
studies have examined
the relationship
between parental
divorce and the age at which adult offspring marry.
The findings have been
anything
but consistent. Most studies show that nonintact parenting
leads to
early marriage
(Aquilino 1994;
Axinn & Thornton 1992, 1993;
Carlson 1979;
Glenn & Kramer 1987; Goldscheider & Goldscheider 1993, 1998; Keith &
Finlay 1988; Kiernan 1992; McLanahan & Bumpass 1988; McLeod 1991;
Michael & Tuma
1985;
Mueller & Pope 1977;
Ross & Mirowsky
1999;
Thornton
1991; Waite & Spitze 1981), but a sizable minority find it delays or deters
* Eric
Kostello,
Lori
Kowaleski-Jones,
Matt McKeever,
Kelly Raley,
Sonia Salari,
Ken Smith, the
Utah Population
Studies Seminar,
and three anonymous
reviewers
provided
helpful comments
on previous
drafts. This article
has also benefited
from the research assistance
of Angela Cassidy,
Al Hernandez,
Ann House,
Andrea
McGinn,
and Kim
Shaff,
as well as generous supportfrom
the
Bireley
Foundation. An earlier version was
presented
at the
2001 annual meeting
of the American
Sociological
Association,
Anaheim.
Direct
correspondence
to Nicholas
H. Wolfinger,
Department
of Family
and Consumer
Studies,
225 South 1400 East,
AEB
228, University
of Utah, Salt Lake
City, UT 84112-0080. E-mail:
Nick.Wolfinger@fcs.utah.edu.
Social
Forces,
September
2003,
82(1):337-353
? The University
of North Carolina
Press
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

338
/ Social Forces
82:1,
September
2003
marriage (Avery,
Goldscheider & Speare 1992; Goldscheider
& Waite 1986,
1991; Kobrin & Waite 1984; Li & Wojtkiewicz 1994; South 2001). Three
additional
studies
show no relationship
between
family
of origin and marriage
timing (Amato & Booth 1997; Cherlin, Kiernan & Chase-Lansdale 1995;
Musick & Bumpass 1998).
This debate
has important
implications.
Early
wedlock
greatly
increases the
chances
of divorce
(Bumpass,
Martin & Sweet 1991;
South 1995),
while people
who never
marry report
lower
levels
of overall
well-being
than
their
married
peers
(Waite 1995; Waite & Gallagher 2000). Given that approximately 50% of
marriages
now end in divorce (Bramlett & Mosher 2001; Kreider & Fields
2001), tens of millions of Americans can be affected.
Various factors make it difficult to reconcile prior research
in this area.
Some studies (Aquilino 1994; Goldscheider & Goldscheider 1998; Li &
Wojtkiewicz
1994;
Michael & Tuma 1985;
Thornton 1991) find that the effects
of family structure
on marriage
rates vary by type of nonintact family.
The
offspring
of stepfamilies
appear
to have
high marriage
rates,
while
people raised
by single parents have marital rates comparable to their peers from intact
families.
Goldscheider
and Goldscheider
(1993), however,
find high rates for
stepchildren
and lower-than-intact-family
rates
for people from single-mother
families. Other studies
(e.g., Michael & Tuma
1985;
South
2001;
Waite
& Spitze
1981) do not consider
whether
nonintact families are the product of divorce,
death, or out-of-wedlock
birth. Parental
death, for instance,
generally
has far
fewer negative effects on the marital behavior of offspring than divorce
(Bumpass, Martin & Sweet 1991; McLanahan & Bumpass 1988; Wolfinger
2000).
Other limitations
of prior
research concern data and methods.
Many
studies
(e.g., Avery,
Goldscheider & Speare 1992; Axinn & Thornton 1992, 1993;
Goldscheider
& Goldscheider
1993; Michael & Tuma 1985;
Thornton 1991)
analyze only respondents in their early twenties or younger, and so
comparatively
little is known about how parental divorce might affect the
marital
behavior of anyone older.
A similar issue applies to two studies that
examine historical variation
in the relationship
between
parental
divorce and
the marital behavior of offspring (Li & Wojtkiewicz 1994; McLanahan &
Bumpass
1988).
Both use a single
cross
section,
a research
design
that truncates
the age range
of more recent birth
cohorts.
Finally, many
studies have
employed
statistical methods that are not sensitive to temporal variation in the
relationship between parental divorce and the timing of marriage. The
frequently
used Cox proportional
hazards model (e.g.,
Axinn &
Thornton
1993;
Goldscheider & Goldscheider
1998;
McLanahan
& Bumpass 1988;
Thornton
1991)
forces the effects of family
structure to remain constant across the hazard
function and therefore does not permit insight into whether
parental
divorce
affects rates of marriage differently
at different
ages. South (2001) allows the
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce and
Offspring Marriage
/ 339
effects
of parental
divorce to vary by age but constrains variation to linearity.
This is an untenable
assumption,
as my results
will show.
The current
study attempts
to reconcile the contradictory
results
produced
by prior
research.
I analyze
data
from the 1973-94 General Social
Survey
(GSS)
that include respondents
of all ages. Survey
items permit the differentiation
of parental
divorce from other family structures and remain consistent over
the entire time series,
providing
a better
test of historical
change
than analyses
based on single cross sections. Finally,
a discrete-time
event-history analysis
allows me to ascertain how the effect of parental
divorce on marriage timing
varies across
the life course.
My results show that the relationship
between family structure of origin
and marriage
varies across two dimensions of time, the individual life course
and the historical period. Teenagers from divorced families have
disproportionately high marriage
rates,
but if they remain single past the age
of 20 their chances of matrimony dip below those of their peers from intact
families. Furthermore,
this relationship
has changed over time. In 1973, the
beginning of the time series analyzed, parental
divorce greatly
increased the
likelihood of marriage.
By 1994, the children of divorce were somewhat less
likely to marry than were people from intact families. Taken
together,
these
findings
can probably
account for why prior
research has produced
conflicting
results.
Why Parental Divorce Affects Marital Behavior
Does parental divorce raise or lower rates of marriage? Strong evidence
supports
both positions. The arguments
for a decrease in marriage
rates are
simpler,
mostly centering
on the effects that parental
divorce has on children's
feelings about romantic relationships. Divorce often imbues children with
comparably
unfavorable
attitudes toward marriage (Amato & Booth 1991;
Axinn & Thornton 1996).
This in itself
may
be enough to discourage
marriage.
When they do marry, people from divorced families often exhibit behaviors
not conducive to maintaining a lasting union (Amato 1996; Silvestri 1992;
Webster,
Orbuch
& House 1995).
Assuming they are present
before marriage,
these behaviors may interfere
with the formation of intimate relationships.
Another explanation for low marriage rates references the alternative of
cohabitation.
It is well established that children of divorce often live with their
partners out of wedlock (Amato & Booth 1997; Axinn & Thornton 1993;
Cherlin,
Kiernan & Chase-Lansdale
1995;
Thornton 1991). Fearful
of repeating
their parents' experiences, they may opt for cohabitation as a "safe" alternative
to marriage.
Three
arguments
can be offered
in favor of the hypothesis
that the children
of divorce
will have
high rates of marriage,
or be inclined to marry
young. First,
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

340 / Social Forces 82:1,
September
2003
parental
divorce and remarriage
often create
an unpleasant
home environment,
and so teenagers may marry
as a way of getting out. The second explanation
cites the grievous
consequences
that divorce
has on women's
incomes
(see, inter
alia, Hoffman & Duncan 1988; Smock 1993). People from impoverished
families often marry
an early age because they lack alternatives,
such as the
opportunity
for a higher education (Axinn & Thornton 1992;
Waite
& Spitze
1981).
The third argument
to account for high marital rates for the children of
divorce is more speculative. Perhaps
growing
up in a divorced
family imbues
offspring
with an inner neediness that leads them to seek out romantic involve-
ment. It is evidence for this proposition that the children
of divorce are dis-
proportionately likely to become sexually
active at an early age (Hogan, Sun
& Cornwell 1998; Moore & Chase-Lansdale
2001; Wu, Cherlin & Bumpass
1997)
and to become pregnant
out of wedlock
(Moore
& Chase-Lansdale
2001;
Wu 1996;
Wu & Martinson
1993). Parental divorce
even accelerates
menarche,
the onset of menstruation
in young women (Hetherington
1993). Even
if pa-
rental
divorce
engenders
negative
attitudes toward
marriage,
it may be an in-
evitable consequence
of premature
romantic activity.
EXPLAINING CHANGE OVER TIME
The children of divorce should be less likely
to get married
in recent
years
for
two reasons:
(1) An overall decline in the negative
effects of growing up in a
divorced
family,
which may have pushed teenagers
into marriage
prematurely;
and (2) the increasing
availability
of cohabitation
as an alternative.
Many
researchers contend that the normalization
of divorce
in contemporary
America has weakened its negative
impact
on children. Goldscheider and Waite
(1991) foresaw
this development
but did not attempt to verify it. Kulka and
Weingarten
(1979)
found that
parental
divorce had fewer
negative
effects on survey
respondents interviewed
in 1976 than it did for a comparable sample from
1957. Through
an exhaustive
meta-analysis,
Amato and Keith (1991) provide
additional evidence that the negative consequences of parental
divorce had
weakened over time. More recently, Wolfinger
shows that the intergenerational
transmission of divorce,
the propensity
for the children
of divorce to end their
own marriages,
declined
approximately
50% over the last 30 years (Wolfinger
1999). Finally,
and most relevant for the current
research,
family
structure
had
a stronger
effect on the marriage
rates for people born in the 1940s than for
more recent cohorts (Li & Wojtkiewicz
1994).
Wolfinger
(1999) contends that the declining
rate of divorce transmission
can be explained
in part
by the diminishing
social and legal
barriers to divorce.
As a result, couples no longer wait as long before ending a marriage.
Hence,
their children
do not suffer as much acrimony
as they did in years
gone by; in
contrast,
when divorce was rare,
quarreling
spouses stayed together
far longer
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce
and
Offspring
Marriage
/ 341
before
calling
it quits.
When couples
finally
ended their
marriages,
the situation
may have deteriorated
far more than is typical
in most modern divorces.'
This argument predicts a decline in marriage rates. If the children of
divorce marry young to escape unpleasant
home environments,
they should
have less incentive to leave if the typical divorcing family
is now less conflict-
ridden
than it used to be. Leaving
home before
marriage
has also become much
more common (Goldscheider
& Goldscheider
1993), and so now the children
of divorce are under less pressure
to marry
before moving out on their own.
In addition,
teenagers
may be less likely
to act out sexually
if parental
divorce
no longer subjects them to as much conflict and upheaval as it once did.
Parental conflict can account for many
of the negative consequences
of growing
up in a divorced
family (e.g., Amato, Loomis & Booth 1995;
Emery
1982).
The second argument for declining marriage rates for the children of
divorce concerns the alternative of living with a partner out of wedlock.
Cohabitation has become increasingly
common in recent years (Bumpass
&
Lu 2000; Bumpass
& Sweet 1989;
Casper
& Cohen 2000). In 1970, only one
American
was cohabiting
for every
hundred
married;
by 1996,
seven cohabited
for every hundred
marriages
(Saluter
& Lugaila
1998). Moreover,
increases
in
cohabitation have
largely
offset declines
in marriage (Bumpass,
Sweet & Cherlin
1991). This may hold particularly
true for the children of divorce,
who cohabit
at much higher rates
than do people from intact families
(Axinn & Thornton
1993;
Cherlin,
Kiernan
& Chase-Lansdale
1995;
Kiernan
1992;
Thornton
1991).
As cohabitation became more acceptable, perhaps an increasing
number of
people from divorced families were inclined to opt for it in lieu of marriage.
Methods
DATA
This research uses the 1973-94 General Social Survey
(GSS) (Davis & Smith
1994),
an ideal
data set for studying
trends because of its consistency
over time.
A national probability sample of English-speaking households within the
continental
U.S.,
the GSS has been conducted
annually
or biennially
since 1972.
The 1972 and post-1994 surveys
omit key variables
and are therefore not used.
Analyses
also exclude
the black
oversamples
in 1982
and 1987. The final
sample
size for the years
studied is 23,195.
Weighted
and unweighted analyses produce
almost identical
results,
and so the unweighted
estimates
are presented.
Although the sampling
design of the GSS
has undergone
various changes
over the years,
one constant has been cluster
sampling.
This presents
a problem
for most statistical
packages,
which assume simple random sampling in the
calculation of standard errors.
Artificially
inflated
significance
levels
may result
from standard errors
derived from cluster-sampled
data, and so I report the
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

342 / Social
Forces 82:1,
September
2003
TABLE 1: Percentage or Means for Variables in Analyses
Respondent experienced parental
divorce
Yes 9%
No 91
Education of head of
respondent
parental family
Less
than
high
school 42%
High
school
graduate 40
Junior
college graduate 2
College graduate 8
Postgraduate 5
Data
missing 4
Respondent
family
size
Respondent
was
only
child 5%
Respondent
had
siblings 95
Respondent urbanicity
at
age
16
Lived
in rural area
or small town 62%
Lived
in
city
or suburbs 38
Respondent gender
Male 44%
Female 56
Respondent
race
White 88%
Black 10
Other 2
Respondent religion
Catholic 26%
Not Catholic 74
Respondent
education
Less than
high
school 25%
High
school
graduate 53
Junior
college
graduate 4
College
graduate 13
Post
graduate 6
Age
wed 22
Survey
year 83
Respondent
has ever
married
Yes 82%
No 18
Note:
Percentages
may
not sum to 100 as a result
of rounding
error.
results of significance tests performed on Huber-White standard errors (Greene
1993).
Missing data are trivial in number and therefore deleted listwise for all
variables except parental education (N = 801). For this item an additional
dummy variable is coded for cases missing data. More sophisticated missing-
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce and Offspring
Marriage
/ 343
data techniques, such as multiple imputation, do not perform appreciably
better (Paul et al. 2002).
VARIABLES
The GSS includes two items that measure the structure of respondents' families
of origin: (1) Respondents were first queried about household composition at
the age of 16; (2) if respondents were not living with both biological parents,
a second item ascertained the reason. Analyses are based on the 83% of GSS
respondents who reported three varieties of parental family structure: intact
two-parent families, mother-only families resulting from divorce or separation,
and mother/stepfather families resulting from divorce or separation.
Respondents reporting other living arrangements are omitted from the analysis,
as are those whose situations at age 16 were the product of parental military
service, parental incarceration, or parental death. Unfortunately the GSS does
not permit identification of respondents born out of wedlock. The family
structure items are recoded as a dummy variable measuring whether a
respondent hails from a divorced family. Percentages or means for this and all
other variables appear in Table 1; the coding for all variables appears in
Appendix A.
Preliminary analyses
show that respondents from stepfamilies
have significantly
higher marriage rates than do people from mother-only families. Nevertheless,
I opt to combine them for two reasons. First, statistical significance is easily
achieved with over 20,000 respondents but the actual difference between
coefficients is not large. Both groups have much higher zero-order marriage
rates than people from intact families. More important, the coefficients
measuring change over time are almost identical for respondents from single-
mother families and stepfamilies. Second, the analysis contains complicated
interaction structures. Measuring family structure with only one variable greatly
simplifies the interpretation of results. A detailed analysis of differences in
marriage rates by family type appears in Li and Wojtkiewicz (1994).
Additional variables are used to ascertain whether the marital behavior of
people from divorced families can be explained by sociodemographic
differences between respondents. These include respondent education, parental
education, sex, race, Catholicism, presence of siblings, and rural versus urban
origins. These are essentially the same control variables used by other studies
of parental divorce based on the GSS (e.g., Glenn & Kramer 1985, 1987;
Wolfinger 1998, 1999) and there are no other variables available that might
conceivably affect the relationship between family of origin and the marital
behavior of offspring.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

344 / Social Forces 82:1, September 2003
TABLE 2: Discrete-Time Event-History Analyses of Entry into First
Marriage on Family Structure of Origin, Historical Period, and
Other Variables
Model
(1) (2) (3)
Respondent
from divorced
family
Survey year
Survey year
x divorced
family
Divorced
family
x age14
Divorced
family
x age
15
Divorced
family
x age
16
Divorced
family
x age
17
Divorced
family
x age
18
Divorced
family
x age 19
Divorced
family
x age
20
Parent education
Less than
high school
High school graduate
Junior
college graduate
College graduate
Postgraduate
Data
missing
Nonurban at age 16
Only
child
Catholic
Male
Race
White
Black
Other
.003
-.094
-.232***
-.232***
-.096*
.129***
.027
-.232***
- - _~~-.515***
- -.333***
-_ -.189***
Respondent
education
Less
than
high school
Junior
college graduate
College graduate
Postgraduate
Intercept -3.310***
Log
likelihood
.161***
-.128***
-.445***
-.457***
-3.288*** -3.445***
-60596.07 -60528.17 -59207.49
Notes:
Analyses
control for duration dependence.
Parameter estimates and standard errors for
zero-order
duration
dependence
terms available from the author. N for all models is 23,195;
293,694
person-years.
* p < .05 ** p < .01 p < .001 (two-tailed tests)
1.930***
-.013***
-.022***
1.351***
-.013***
-.018***
1.140**
.671*
.886***
.778***
.582***
.408***
.299**
1.443***
-.008***
-.020***
1.146**
.678**
.896***
.793***
.602***
.427***
.311***
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental
Divorce and Offspring
Marriage
/ 345
ANALYSIS
I conduct an event-history
analysis
of entry
into marriage
using a discrete-time
model, estimated
via complementary
log-log regression.
The complementary
log-log is a better estimator than logit or probit when discrete data approximate
a continuous time process (Allison 1995). The discrete-time model permits
testing of hypotheses about interactions between parental divorce and duration
dependence. As the GSS measures age at marriage in years, little would be
gained by using a continuous time model. The hazard function is captured by
a dummy variable for each year of age, up to 39 (40 is the omitted category).
Life tables revealed little meaningful variation in the risk of marriage beyond
age 40, and so for purposes of defining the hazard function age is top-coded at
40.
Survey year is interacted with parental divorce to measure trends in
marriage timing for people from divorced families. The choice of year as the
temporal "index" was not consequential: A model based on birth cohort
produced similar results, as did including variables for both survey year and
birth cohort.
Results
Table 2 shows the effects of parental divorce on the entry into marriage. The
coefficients for family background and the survey year x family background
interaction are both statistically significant (p < .001), indicating that parental
divorce has an effect on marriage rates (model 1). The negative coefficient for
the interaction term shows that the effect of family background has weakened
between 1973 and 1994.
The magnitude of the trend in the consequences of parental divorce can
be learned by substituting different values of the year variable (measured as
the last two digits of the survey year) into the following equation, derived from
model 1:
Hazard ratio = exp(1.930 - .022 x year) (1)
The results appear in the first row of Table 3. For 1973, the equation yields a
figure of 1.38, indicating that respondents from divorced families interviewed
then were 38% more likely to marry than were people from intact families. By
1994, the hazard ratio had declined to .87, suggesting that people from divorced
families had become somewhat less likely to marry than were people from intact
families. This result is consistent with Li and Wojtkiewicz (1994), who found
that family structure had a stronger impact on marital rates for people born
in the 1940s than for those born in later decades.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

346
/ Social
Forces
82:1,
September
2003
TABLE 3: Odds Ratios Comparing
Likelihood
of Marriage
for Respondents
from Divorced
and Intact Families,
by Age and Survey
Year
Respondent Age 1973 1994
Without interactions between age
and
duration
dependence 1.38 .87
18 1.86 1.27
20 1.40 .96
Older than 20 1.04 .71
These estimates constrain trends in the relationship between parental
divorce
and respondent
marriage
to linearity.
I test this assumption by pooling
the data
into nine groups
based on survey year,
then for each group
examining
the relationship
between parental
divorce and the entry into marriage.
This
analysis
confirmed
linearity
in the changing
effects of parental
divorce (result
not shown).
The results presented so far have been based on the assumption that
parental
divorce affects the probability
of marriage equally
for respondents
of
all ages.
This turns out not to be the case.
Interactions between
parental
divorce
and each of the dummy
variables
measuring
duration
dependence
were added
to model 1, with statistically significant terms retained in model 2. The
interactions are all positive, indicating that the children of divorce are
especially likely
to marry
at ages 14-20. Exploratory
analyses
revealed no three-
way interactions between survey year, parental divorce, and duration
dependence;
in other words, the relationship
between parental
divorce and
early marriage
holds throughout
the 1973-94 time series.
These interactions have a substantial effect on the changing
marital
rates
of the children of divorce. This can be shown by adjusting
equation 1 for the
interactions
included in model 2:
Hazard ratio = exp(1.351 -.018 x year + age inti) (2)
In this equation
age
inti
is the interaction
between
marrying
at age i and coming
from a divorced
family.
The results of equation 2 for various values on year
and age
int are shown in the bottom three rows of Table 3. They represent
the
odds of marriage compared
to someone of the same age from an intact
family.
Although
the chances of marriage
have declined for all respondents
from
divorced families,
there are considerable
age differences.
Children of divorce
younger than 20 have high odds of marriage,
both in 1973 and 1994. By 20,
things are different. In 1973, the children of divorce were about 40% more
likely to marry
than were people from intact families. By 1994, 20 year-olds
from divorced
families had almost the same marriage
rate as their peers from
intact families. The rates
go down even more for people over 20, who by 1994
comprised the great majority
of those marrying
for the first time (Saluter
&
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce and
Offspring
Marriage
/ 347
Lugaila
1998). In 1973, children of divorce older than 20 had approximately
the same chances of marriage
as did people from intact families.
By 1994 they
were 29%
less likely
to marry.2
Of the 82%
of GSS
respondents
who have ever
been married,
59% (N = 11,348)
wed after
the age of 20. In contrast,
4% (N =
759) were 16 or younger when they first married.
Since the GSS measures
parental
family
structure at age 16, there is a small
chance of measurement error
for respondents marrying
at 14 or 15. However,
most parents probably ended their marriages
before respondents' sixteenth
birthdays,
so their divorces would be captured
by the family
structure variables.
Note also that the pattern of youthful marriage holds past age 16, when
measurement
error is far less likely.
None of the control variables affect the changing relationship between
parental
divorce and getting married. Model 3 (Table
2) includes
measures of
parental
education,
respondent
education, race, Catholicism,
sex, presence
of
siblings, and urbanicity.3
The coefficients measuring the effect of parental
divorce hardly change at all from model 2, indicating that the weakening
relationship
between parental
divorce and marriage
cannot be explained by
fundamental
demographic
differences
between respondents
or their parents.
Furthermore,
interacting family background
with the demographic
variables
does not substantially
affect trends in marriage
for the children of divorce
(result not shown).
These findings
indicate that marriage
rates for the children
of divorce
have
declined
dramatically
over the study
period,
with the exact
rate
in any given
year
dependent
upon one's
stage
in the life course.
Furthermore,
changes
in marriage
timing
cannot
be explained by key
demographic
differences
between
respondents.
Conclusion
Parental divorce has a significant effect on marriage timing, although the
relationship
varies
greatly
by respondent
age and survey year.
The children of
divorce
have
high marriage
rates
through
age 20, but if unmarried
by this point
they are disproportionately
likely to remain so. Marriage
rates also declined
considerably
over the course of the study,
for both the over-20 and under-20
age groups. Together,
these results can explain why prior studies produced
conflicting results.
Since
people from divorced families
experience
both high and low marriage
rates
at different
ages,
all the arguments
considered
earlier
may hold true. The
children of divorce may marry as teenagers to escape unhappy home
environments,
as the inadvertent
result of premature
sexual
activity,
or simply
to assuage psychic wounds. After the age of 20, interpersonal
problems may
get in the way of marriage, or people may opt for cohabitation. The only
hypothesis
that can be rejected
is the notion that the socioeconomic
well-being
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

348
/ Social Forces
82:1,
September
2003
of single mothers leads to high levels of youthful marriage.
Controlling for
parental and respondent education did not affect marital rates, and so the
children
of divorce
probably
do not marry
for lack of other
opportunities.
This
result should be qualified on the grounds that education taps only certain
elements of socioeconomic well-being. Be that as it may, other research has
shown that parental
divorce
affects
marriage
rates
irrespective
of both parental
income and education (Axinn & Thornton 1992;
South 2001).
Two explanations
were proposed to explain historical
trends in the mar-
riage rate. First, since the weakening
consequences of parental
divorce have
produced a decline in the rate of divorce transmission,
they may also have
affected
children's entrance into marriage.
Thus, the children
of divorce
may
now be achieving
more "normal"
patterns
of marriage
formation.
Also,
divorce
and stepparenting
can push teenagers
into marriage
as a means of escaping
an
unpleasant
home environment.
Nowadays teenagers
will have less reason to
leave if parental
divorce is less unpleasant
than it used to be. Second,
the chil-
dren of divorce
may now be more inclined than ever to cohabit
as an alterna-
tive to getting married. The evidence for this proposition is straightforward:
the children of divorce are disproportionately likely to cohabit (Axinn &
Thornton 1993; Cherlin, Kiernan & Chase-Lansdale 1995; Kiernan 1992;
Thornton 1991), and declines
in the marriage
rate have been largely
offset by
rising levels of cohabitation
(Bumpass,
Sweet & Cherlin 1991). Moreover,
pa-
rental divorce
substantially
decreases the chances that a cohabitant
will marry
his or her partner
(Wolfinger
2001). Finally,
it should be noted that cohabita-
tion with one partner
does not rule out marriage
to another,
and so for some
people cohabitation
may delay matrimony
without replacing
it.
The GSS lacks data on cohabitation and therefore does not permit a test
of these theories.
Nevertheless,
the cohabitation
explanation
is favored in one
respect:
it can account for the finding that the children of divorce are now
less likely to marry
than are people from intact families.
Having experienced
the upheaval
of parental
divorce,
perhaps
today's
young adults seek to avoid
marriage altogether by resorting
to an increasingly
acceptable
alternative.
If
the weakening consequences
of parental
divorce were solely responsible,
there
is no reason that the marriage
rate for people from divorced
families should
have slipped below that of respondents
from intact families.
Future research should attempt to adjudicate
between these explanations
by exploring
the changing
role of cohabitation
in the relationship
formation
behavior of people from divorced families.
No matter which one is right, the
results reported in this article reflect a dramatic transformation in the
demography
of marriage timing.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce and Offspring Marriage / 349
Notes
1. No adequate historical data exist, and so it is impossible to know for certain whether
modern divorces are indeed less acrimonious. However, the indirect evidence is
compelling. Amato and Booth (1997) recently concluded that less than one-third of
modern divorces are preceded by serious conflict, a finding sharply at odds with historical
depictions of marital breakdown (e.g., May 1980; Phillips 1991).
2. A note on this calculation: For respondents over 20, age int. was not statistically
significant and is therefore omitted from equation 3. Thus Table 3 does not differentiate
between over-20 respondents.
3. Including education in model 3 introduces the risk of measurement error, since
respondents marrying as teenagers may have not yet completed their formal educations.
Repeating this analysis without the education variable did not affect the results.
Unfortunately time-varying measures of education are not available in the GSS.
References
Allison, Paul
D. 1995.
Survival
Analysis
using
the SAS
System:
A Practical Guide.
SAS Institute.
Amato, Paul R. 1996. "Explaining
the Intergenerational
Transmission
of Divorce."
Journal
of
Marriage
and the
Family
58:628-40.
Amato, Paul R., and Alan Booth. 1991. "The Consequences of Divorce for Attitudes toward
Divorce and Gender Roles."
Journal
of Family
Issues. 12:306-22.
. 1997. A Generation
at Risk: Growing Up in an Era of Family Upheaval. Harvard
University
Press.
Amato, Paul R., and Bruce Keith. 1991. "Parental
Divorce and the Well-Being of Children:
A Meta-Analysis."
Psychological
Bulletin
110:26-46.
Amato, Paul R., Laura Spencer Loomis, and Alan Booth. 1995. "Parental
Divorce, Marital
Conflict, and Offspring Well-being
during Early
Adulthood." Social Forces 73:895-915.
Aquilino, William S. 1994. "Family
Structure and Home-Leaving:
A Further
Specification of
the Relationship."
Journal
of Marriage
and the
Family
53:999-1010.
Avery, Roger,
Frances
Goldscheider,
and Alden Speare
Jr.
1992. "Feathered
Nest/Gilded Cage:
Parental
Income and Leaving
Home in the Transition
to Adulthood."
Demography
29:375-
88.
Axinn, William G., and Arland Thornton. 1992. "The Influence of Parental
Resources
on the
Timing of the Transition to Marriage."
Social Science Research 21:261-85.
.1993. "Mothers,
Children,
and Cohabitation:
The Intergenerational
Effects
of Attitudes
and Behavior." American
Sociological
Review 58:233-46.
.1996. "The Influence
of Parents'
Marital Dissolutions on Children's Attitudes toward
Family
Formation."
Demography
33:66-81.
Bramlett,
Matthew D., and William D. Mosher.
2001. First
Marriage
Dissolution,
Divorce,
and
Remarriage:
United States. Advance Data from Vital Health Statistics;
no. 323. National
Center for Health Statistics.
Bumpass, Larry
L., and Hsien-Hen Lu. 2000. "Trends
in Cohabitation and Implications for
Children's
Family
Contexts
in the United States."
Population
Studies 54:29-41.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

350 / Social Forces 82:1,
September
2003
Bumpass, Larry
L., Teresa Castro Martin, and James
A. Sweet. 1991. "The Impact of Family
Background
and Early
Marital Factors on Marital
Disruption."
Journal
of Family
Issues
12:22-42.
Bumpass, Larry
L.,
and James
A. Sweet.
1989. "National Estimates of Cohabitation."
Demography
26:615-625.
Bumpass, Larry
L., James
Sweet, and Andrew J. Cherlin. 1991. "The Role of Cohabitation in
Declining Rates of Marriage."
Journal
of Marriage
and the
Family
53:913-27.
Carlson,
Elwood. 1979.
"Family Background,
School, and Early Marriage."
Journal
of Marriage
and the
Family
41:341-53.
Casper,
Lynne
M., and Phillip N. Cohen. 2000. "How Does POSSLQ
Measure
Up? Historical
Estimates of Cohabitation."
Demography
37:237-45.
Cherlin,
Andrew
J.,
Kathleen
E. Kiernan,
and P.
Lindsey
Chase-Lansdale. 1995. "Parental
Divorce
in Childhood
and Demographic
Outcomes
in Young
Adulthood."
Demography
32:299-318.
Davis, James
A., and Tom W. Smith. 1994. The General Social
Surveys:
Cumulative
Codebook,
1972-1994. National Opinion Research Center.
Emery,
Robert E. 1982. "Interparental
Conflict and the Children of Discord and Divorce."
Psychological
Bulletin 92:310-330.
Glenn, Norval D., and Kathryn B. Kramer. 1985. "The Psychological Well-being of Adult
Children of Divorce."
Journal
of Marriage
and the
Family
47:905-12
.1987. "The
Marriages
and Divorces of the Children of Divorce."
Journal
of Marriage
and
the
Family
49:811-25.
Goldscheider, Francis K., and Calvin Goldscheider. 1993. Leaving Home before
Marriage:
Ethnicity,
Familism,
and Generational
Relationships. University
of Wisconsin Press.
.1998. "The Effects
of Childhood Family
Structure on Leaving
and Returning
Home."
Journal
of Marriage
and the
Family
60:745-56.
Goldscheider,
Frances
K.,
and Linda
J.
Waite. 1986. "Sex Differences in the Entry
into Marriage."
American
Journal
of Sociology
92:91-109.
Greene,
William H. 1993. Econometric
Analysis,
2d ed. Prentice Hall.
Hetherington,
E. Mavis 1993.
"An
Overview of the Virginia Longitudinal
Study
of Divorce and
Remarriage
with a Focus on Early
Adolescence."
Journal
of Family
Psychology
7:39-56.
Hoffman,
Saul
D., and Greg
J.
Duncan.
1988. "What Are the Economic
Consequences
of Divorce?"
Demography
25:641-45.
Hogan, Dennis P.,
Rongjun
Sun, and Gretchen T. Cornwell. 1998.
"Cohort
Differences,
Family
Structure,
and Adolescent Sexual
Activity."
Paper presented
at the annual meeting of the
American
Sociological
Association,
San Francisco.
Keith,
Verna M., and Barbara
Finlay. 1988. "The Impact of Parental
Divorce on Children's
Educational
Attainment,
Marriage Timing,
and Likelihood of Divorce."
Journal
of Marriage
and the
Family
50:797-809.
Kiernan,
Kathleen E. 1992.
"The
Impact
of Family Disruption
in Childhood on Transitions
Made
in Young-Adult
Life."
Population
Studies 46:213-34.
Kobrin,
Frances
E., and Linda J.
Waite. 1984. "Effects
of Childhood Family Structure on the
Transition to Marriage."
Journal
of Marriage
and the Family
46:807-16.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental Divorce and Offspring Marriage / 351
Kreider,
Rose M., and Jason
M. Fields.
2001. Number,
Timing,
and Duration of Marriages
and
Divorces:
Fall 1996. U.S. Bureau of the Census, Current Population Reports, P70-80.
Government
Printing
Office.
Kulka,
Richard
A., and Helen Weingarten.
1979. "The
Long-Term
Effects of Parental
Divorce in
Childhood in Adult
Adjustment."
Journal
of Social Issues
35:50-78.
Li,
Jiang
H., and Roger
A. Wojtkiewicz.
1994. "Childhood
Family
Structure and Entry
into First
Marriage." Sociological Quarterly
35:247-68.
May,
Elaine Tyler.
1980. Great
Expectations: Marriage
and Divorce in Post-Victorian
America.
University
of Chicago Press.
McLanahan,
Sara
S., and Larry
L. Bumpass. 1988.
"Intergenerational
Consequences
of Family
Disruption."
American
Journal
of Sociology
94:130-52.
McLeod,
Jane
D. 1991.
"Childhood
Parental
Loss and Adult Depression."
Journal
of Health and
Social
Behavior
32:205-20.
Michael, Robert, and Nancy Brandon Tuma. 1985. "Entry
into Marriage
and Parenthood by
Young
Men and Women: The Influence of Family Background." Demography
22:515-44.
Moore, Mignon R., and P. Lindsay
Chase-Lansdale. 2001. "Sexual Intercourse
and Pregnancy
among African
American Girls
in High-Poverty
Neighborhoods:
The Role of Family
and
Perceived
Community Environment."
Journal
of Marriage
and Family
63:1146-57.
Mueller,
Charles
W., and Hallowell
Pope. 1977.
"Marital
Instability:
A Study
of Its Transmission
between Generations."
Journal
of Marriage
and the
Family
39:83-93.
Musick, Kelly,
and Larry
L. Bumpass. 1998. "How Do Prior Experiences
in the Family
Affect
Transitions to Adulthood?"
NSFH Working Paper No. 81, Center for Demography and
Ecology, University
of Wisconsin Madison.
Paul,
Christopher,
Daniel McCaffrey,
William M. Mason,
and Sarah
A. Fox. 2002. "What
Should
We Do about Missing
Data?
A Case
Study
Using Logistic
Regression
with Data Missing
on
a Single Covariate."
Unpublished
paper,
RAND.
Phillips,
Roderick.
1991. Untying
the Knot:
A Short
History of
Divorce.
Cambridge University
Press.
Ross, Catherine
E., and John Mirowsky.
1999. "Parental
Divorce,
Life-Course
Disruption, and
Adult Depression."
Journal
of Marriage
and the
Family
61:1034-45.
Saluter,
Arlene F.,
and Terry
A. Lugaila.
1998. Marital Status and Living
Arrangements:
March
1996. U.S. Bureau of the Census, Current Population Reports, P20-496. Government
Printing
Office.
Silvestri, Silvio. 1992. "Marital Instability in Men from Intact and Divorced Families:
Interpersonal
Behavior,
Cognitions,
and Intimacy."
Journal
of
Divorce and
Remarriage
18:79-
108.
Smock,
Pamela
J. 1993. "The Economic Costs of Marital
Disruption
for Young
Women over the
Past
Two Decades."
Demography
30:353-71.
South, Scott J. 1995. "Do You Need to Shop Around?
Age at Marriage,
Spousal Alternatives,
and Marital Dissolution."
Journal
of Family
Issues 16:432-49.
. 2001. "The
Variable Effects of Family
Background
on the Timing of First Marriage:
United States,
1969-1993."
Social Science
Research 30:606-26.
Thornton, Arland. 1991. "Influence of the Marital History of Parents on the Marital and
Cohabitational
Experiences
of Children."
American
Journal
of Sociology
96:868-94.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

352 / Social Forces 82:1,
September
2003
Waite,
Linda
J. 1995. "Does Marriage
Matter?"
Demography
32:483-507.
Waite,
Linda
J., and Maggie Gallagher.
2000. The Case
for Marriage:
Why
Married
People
Are
Happier,
Healthier,
and Better
Off
Financially. Doubleday.
Waite, Linda J., and Glenna D. Spitze. 1981. "Young Women's Transition to Marriage."
Demography
18:681-94.
Webster,
Pamela S., Terri L. Orbuch, and James
S. House. 1995. "Effects
of Childhood Family
Background
on Adult Marital
Quality
and Perceived
Stability."
American
Journal
of Sociology
101:404-32.
Wolfinger,
Nicholas H. 1998. "The Effects of Parental
Divorce on Adult Tobacco
and Alcohol
Consumption."
Journal
of
Health
and Social
Behavior 39:254-70.
. 1999. "Trends
in the
Intergenerational
Transmission
of Divorce."
Demography
36:415-20.
.2000. "Beyond
the Intergenerational
Transmission
of Divorce: Do People Replicate
the
Patterns
of Marital
Instability They Grew
Up With?"
Journal
of Family
Issues 21:1061-86.
. 2001.
"The Effects
of Family
Structure
of Origin
on Offspring
Cohabitation
Duration."
Sociological Inquiry
71:293-313.
Wu, Lawrence
L. 1996. "Effects of Family Instability,
Income,
and Income Instability
on the Risk
of a Premarital
Birth."
American
Sociological
Review 61:386-406.
Wu, Lawrence
L., Andrew J. Cherlin, and Larry
L. Bumpass. 1997. "Family
Structure,
Early
Sexual Behavior,
and Premarital Births."
Unpublished paper, Department of Sociology,
University
of Wisconsin
Madison.
Wu, Lawrence
L.,
and Brian C. Martinson. 1993.
"Family
Structure
and the Risk
of a Premarital
Birth."
American
Sociological
Review 58:210-32.
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions

Parental
Divorce and Offspring Marriage
/ 353
APPENDIX A: Coding of Variables
Variable Coding
Age at marriage
Education
of head
of respondent
parental family
Race
Respondent
education
Respondent
family
structure
Religion
Respondent
urbanicity
at age 16
Continuous variable measured
in years,
recoded
as
person years
Set
of five dichotomous indicators,
each coded
1 if family
head is not a high school graduate,
is a junior college graduate,
is a four-year college
graduate,
has a post-graduate degree,
or if
data are missing; high school graduate is the
reference
category
Set
of two dichotomous indicators,
each
coded 1 if respondent
is black or other
(nonblack,
nonwhite);
white is the
reference
category
Set of four dichotomous indicators,
each coded
1 if respondent
is not a high school graduate,
is a junior college graduate,
is a four-year college
graduate,
or has a postgraduate degree; high
school graduate
is the reference
category
Coded 1 if respondent
lived with a divorced
single
mother or remarried divorce mother at age 16,
0 if respondent
lived with biological parents
Coded 1 if Catholic,
0 if not Catholic
Coded 0 if respondent
lived in a city of 50,000 or
more people or in a suburb of a large
city,
coded 1 if
respondent
lived in a town of under
50,000
persons
or in a rural
area
Coded 0 if female, 1 if male
Siblings
present
Survey year
Coded 1 if only child, 0 otherwise
Last two digits
of calendar
year;
continuous variable
Sex
This content downloaded from 155.97.178.73 on Thu, 23 Jan 2014 18:14:09 PM
All use subject to JSTOR Terms and Conditions