Parental Education and Family Dissolution: A Cross‐National and Cohort Comparison

• DOI: http://doi.org/10.1111/jomf.12461
• Date: 01 April 2018
• Published date: 01 April 2018

M.D. (A) B Netherlands Interdisciplinary Demographic Institute and

VU University Amsterdam

J H Stockholm University∗

Parental Education and Family Dissolution:

A Cross-National and Cohort Comparison

This is the rst study to systematically analyze

whether the association between parental

education and family dissolution varies

cross-nationally and over time. The authors use

meta-analytic tools to study cross-national vari-

ation between 17 countries with data from the

Generations and Gender Study and Harmonized

Histories. The association shows considerable

cross-national variation, but is positive in most

countries. The association between parental

education and family dissolution has become

less positive or even negative in six countries.

The ndings show that the association between

parental education and family dissolution is

generally positive or nil, even if the association

between own education and family dissolution

is in many countries increasingly negative.

Netherlands Interdisciplinary Demographic Institute, The

Hague (NIDI/KNAW),University of Groningen, P.O. Box

11650, 2502 AR The Hague, Netherlands (brons@nidi.nl).

Department of Sociology, VU UniversityAmsterdam, De

Boelelaan 1105, 1081 HV Amsterdam, The Netherlands.

∗Department of Sociology, Stockholm University,

Sociologiska institutionen, Demograska avdelningen 106

91 Stockholm, Sweden.

© 2018 The Authors. Journal of Marriage and Familypub-

lished by WileyPeriodicals, Inc. on behalf of National Coun-

cil on Family Relations.

This is an open access article under the terms of the Creative

Commons Attribution License, which permits use, distribu-

tion and reproduction in any medium, provided the original

work is properly cited.

Key Words: cross-national research, divorce, intergenera-

tional, parental education.

The authors nd suggestive evidence that the

association is related to the crude divorce rate,

but not to the generosity of the welfare state

in these countries. The implications of these

ndings for understanding the stratication in

family dissolution are discussed.

Advantaged family backgrounds pave the way

to higher education, higher incomes, and better

health (Breen & Jonsson, 2005; Elo, 2009).

Higher socioeconomic backgrounds are also

related to many favorable family demographic

outcomes, such as postponement of childbear-

ing beyond adolescence (Dahlberg, 2015) and

marriage with highly educated partners (cf.

Schwartz, 2013). Do favorable family back-

grounds also beget family stability and the

benets associated with it? Recent research has

paid much attention to the growing educational

disparities in family dissolution (e.g., Amato,

2010; Härkönen & Dronkers, 2006; McLahanan,

2004), but this interest has not been matched by

a similar focus on family dissolution patterns

by parental educational background. Because of

the importance of family background on indi-

viduals’ future life chances, this omission limits

our understanding of the social stratication in

family demography.

Previous studies on the association between

parental education and family dissolution have

produced intriguing ndings. In contrast to the

increasingly negative association between own

education and family dissolution in many soci-

eties (Härkönen & Dronkers, 2006), many stud-

ies have found a positive association between

426 Journal of Marriage and Family 80 (April 2018): 426–443

DOI:10.1111/jomf.12461

Parental Education and Family Dissolution 427

parental education and family dissolution (Swe-

den: Hoem & Hoem, 1992; the Netherlands:

Klijzing, 1992; Janssen, 2001; Finnish women:

Mäenpää & Jalovaara, 2014; Italy: Todesco,

2013), even when the relationship between

own education and family dissolution is neg-

ative (Norway: Lyngstad, 2004, 2006). This

suggests a nuance to perspectives of the lower

status character of family dissolution that have

come to dominate the literature on stratication

of family instability. However, other studies

reported zero relationships (Australia: Bracher,

Santow, Morgan,& Trussel, 1993; United King-

dom: Berrington & Diamond, 1999; Finnish

men: Mäenpää & Jalovaara, 2014) or a negative

association (United States: Bumpass, Martin,

& Sweet, 1991), suggesting that the associ-

ation may vary cross-nationally akin to the

relationship between own education and family

dissolution (Härkönen & Dronkers, 2006; Mar-

tin, 2006; Matysiak, Styrc, & Vignoli, 2014).

Many of the above studies are also rather dated,

raising the possibility that the association has

changed over time, potentially from a positive

to a negative one as has been reported for

the educational gradient of divorce in many

countries (Chan & Halpin, 2008; De Graaf &

Kalmijn, 2006; Härkönen & Dronkers, 2006;

Hoem, 1997; Raymo & Iwasawa, 2017).

This study presents the rst comparative

analysis of the association between parental edu-

cation and family dissolution. In light of the pre-

vious discussion, we rst ask whether parental

education is related to family dissolution in 17

European societies and whether this association

varies cross-nationally.Second, has this associa-

tion changed over time? Third, to understand the

causes of the variation across societal contexts,

we analyze whether cross-national and cohort

differences in the association can be linked to

two contextual-level variables that reect the

sociocultural and economic contexts of family

life, namely, the average crude divorce rate and

the generosity of the welfare state. Our analysis

contributes to the understanding of (variation

in) stratication of family dissolution and of

intergenerational effects on family dissolution,

other family demographic behaviors (Dahlberg,

2015; Dronkers & Härkönen, 2008; South,

2001; Wiik, 2009; Wolnger, 2003), and life

chances more generally. We use family history

data from the Generations and Gender Study

(GGS) and Harmonized Histories data sets. Our

outcome is the dissolution of rst childbearing

unions, which is more suitable than divorce as

a measure of family instability given the high

cohabitation rates in the countries we analyze.

B

Why Do Divorce Risks Vary by Parental

Education?

Theorizing about why parental education would

matter for their children’s union dissolution

has been sparse. The existing explanations for

the association between parental education and

union dissolution can be grouped into those

underlining socioeconomic and family demo-

graphic pathways and into those theorizing the

remaining net association between parental

education and union dissolution (e.g., Lyngstad,

2006; Todesco, 2013).

First, parents’ education can affect their off-

spring’s family dissolution risks because of the

intergenerational transmission of educational

attainment. The persistent positive association

between parental and offspring’s education is

among the most consistent ndings in the social

sciences (Breen & Jonsson, 2005), but whether

higher parental education promotes family

stability or not through this pathway depends

on the relationship between own educational

attainment and family dissolution. Higher levels

of education were in many countries related to

elevated family dissolution risks just a couple

of decades ago, but this relationship has today

largely disappeared or reversed to a negative

one (Goode, 1962; Härkönen & Dronkers, 2006;

Matysiak et al., 2014): As family dissolution

became more common, it is the lower educated,

rather than the higher educated, who experience

the highest family dissolution risks. Thus, the

role of intergenerational educational transmis-

sion in shaping the association between parental

education and family dissolution is contingent

on the educational gradient of family dissolution

that prevails in each society and time period.

Second, parental education can affect the

risk of family dissolution through family

demographic pathways. Parental separation is

a well-known predictor of individuals’ own

union dissolution, and this relationship is found

in a range of countries (Dronkers & Härkö-

nen, 2008; Wolnger, 2003). If education was

associated with separation risk in the parental

generation, then parental separation can be

one of the pathways linking parental educa-

tion to family dissolution. Again, because the