selection versus structure_ explaining family type differences in contact with close kin

7/28/2019 Selection Versus Structure_ Explaining Family Type Differences in Contact With Close Kin

http://slidepdf.com/reader/full/selection-versus-structure-explaining-family-type-differences-in-contact-with 1/24

http://jfi.sagepub.com/ Journal of Family Issues

http://jfi.sagepub.com/content/29/11/1448The online version of this article can be found at:

DOI: 10.1177/0192513X08318154

2008 29: 1448 originally published online 27 May 2008Journal of Family Issues Trees De Bruycker

Contact With Close KinSelection Versus Structure : Explaining Family Type Differences in

Published by:

http://www.sagepublications.com

can be found at:Journal of Family Issues Additional services and information for

http://jfi.sagepub.com/cgi/alertsEmail Alerts:

http://jfi.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://jfi.sagepub.com/content/29/11/1448.refs.htmlCitations:

What is This?

- May 27, 2008OnlineFirst Version of Record

- Oct 15, 2008Version of Record>>

at The British Sociological Association on February 5, 2013 jfi.sagepub.comDownloaded from

http://jfi.sagepub.com/


http://jfi.sagepub.com/content/29/11/1448


http://www.sagepublications.com/

http://jfi.sagepub.com/cgi/alerts


http://jfi.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav


http://www.sagepub.com/journalsPermissions.nav

http://jfi.sagepub.com/content/29/11/1448.refs.html



http://online.sagepub.com/site/sphelp/vorhelp.xhtml



http://jfi.sagepub.com/content/early/2008/05/27/0192513X08318154.full.pdf


http://jfi.sagepub.com/content/29/11/1448.full.pdf









http://www.sagepub.com/journalsPermissions.nav


http://jfi.sagepub.com/subscriptions


http://www.sagepublications.com/





Selection Versus Structure

Explaining Family Type Differences

in Contact With Close Kin

Trees De BruyckerGhent University, Belgium

This article focuses on one aspect of family networks, namely, the frequency of

contact with close kin for adults living in different traditional and new family

types. Two mechanisms are hypothesized to account for the differences. The

first focuses on structural factors such as the number and type of persons in the

primary family network, availability of a second family network, and geograph-

ical proximity. The second is selection: Individuals with more postmodern

(family) attitudes and relatively strong orientation to friends rather than to

family may be selected into certain family types. Data from the Netherlands

Kinship Panel Study ( N = 8,155) give little support for the selection hypothesis

in explaining the differences in contact frequency found by family type. The

structural hypothesis, however, yields significant results, with network size and

geographical proximity being of key importance.

Keywords: family relations; intergenerational relations; family structure;

selection; contact

The 20th century, more especially its second half, saw profound changes

in family and living arrangements in Western societies. Characterized

by greater voluntarism in family formation, marriage, and childbearing, the

changes and their implications for the family have been reflected on at

length by many researchers in both sociology and demography (Bengston,

2001). The family decline hypothesis, as postulated by Burgess (1916) and

Journal of Family Issues

Volume 29 Number 11

November 2008 1448-1470

© 2008 Sage Publications

10.1177/0192513X08318154

http://jfi.sagepub.comhosted at

http://online.sagepub.com

1448

Author’s Note: I thank Hilary Page (Ghent University) for general advice and guidance and

Ronan Van Rossem (Ghent University) for his constructive comments on methodology and

results. The Netherlands Kinship Panel Study is funded by Grant No. 480-10-009 from the MajorInvestment Fund of the Netherlands Organization for Scientific Research and by the Netherlands

Interdisciplinary Demographic Institute, Utrecht University, the University of Amsterdam, and

Tilburg University. This article forms part of a larger research project on diversity in familial net-

works. Please address correspondence to Trees De Bruycker, Ghent University, Department of

Sociology, Korte Meer 3, 9000 Gent, Belgium; e-mail: [email protected].








later elaborated by Parsons and Bales (1955) and Popenoe (1993), captured

these trends by pointing to an increasing erosion of family life, family rela-

tions, and family values and formed the start of discussion pro and contrathe survival of the family. This resulted in, on one hand, studies on the

diverse new family types, often focusing on the weakness and fragility of

various contemporary family forms and living arrangements (e.g., Coleman,

Ganong, & Fine, 2000; Lansford, Ceballo, Abbey, & Stewart, 2001). On the

other hand, opponents countered the erosion hypothesis by highlighting the

importance of the broader family network (Bengston, 2001; Busschots &

Lauwers, 1994; Hogan, Eggebeen, & Clogg, 1993; Knijn, 2004). Recent work

on the second demographic transition both in Europe (Lesthaeghe & Surkyn,2007) and, particularly, in the United States (Lesthaeghe & Neidert, 2006) has

stimulated further interest in this discussion.

Theories concerning postmaterialism and postmodernism made clear that

the processes of change in the family in the last decennia were accompanied by

broader changes in values and attitudes, with increasing emphasis on expres-

sive individualism and voluntarism, and the weakening of social prescriptions

in general (Thornton, 1989; Thornton & Young-DeMarco, 2001). Studies of the

second demographic transition focus on the consequences of these for familyformation and dissolution, in particular on the decreasing importance of formal

marriage, the increasing acceptance of divorce, and the questioning of the

importance of having children (see, in particular, Lesthaeghe & Meekers, 1986;

Lesthaeghe & Neidert, 2006, pp. 669-672). Clearly, family construction can

increasingly be a matter of personal choice, what results in the now observed

destandardization of the life course and diversification of family types. Both the

family types in which individuals live and their relationships with kin can be

expected, therefore, to be at least in part the result of a selective process based

on personal attitudes, values, and ideas concerning the family. However,

although selection can be important in the way relationships with family

members are filled in, the size and structure of the family network can itself also

determine the relations between family members.

This article focuses on the impact of selection versus structure on the

differences in contact with close kin. More specifically, I examine the dif-

ferences in contact frequency with close kin between individuals living in

different family types and estimate the relative importance of selection and

structure. I focus on contact because contact can be considered effectivelya sine qua non for a personal relationship in general, and personal contact

is particularly important for social support (Attias-Donfut, 2003; Hogan

et al., 1993). With this focus, I try to get more insight into the differences

between the newer family types and more classic family forms.

De Bruycker / Selection Versus Structure 1449








Despite the continuing importance of the family network in an individ-

ual’s social capital and social support networks, as found in recent studies

(Agneessens, De Lange, & Waege, 2003; Attias-Donfut, 2003; Bengston,2001), little research has been carried out on the dynamics and diversity of the

family network of individuals living in the various contemporary family types.

Research has tended to concentrate on particular relationships, especially the

parent–child relationship (e.g., Lye, 1996; Schwarz, Trommsdorff, Albert, &

Mayer, 2005), the stepparent–stepchild relationship (e.g., Henderson &

Taylor, 1999; MacDonald & DeMaris, 2002; White, 1994), and the sibling

relationship. Less has been done at the level of the family network as such.

Moreover, where studies of the differences in the family networks of individ-uals with different living arrangements do exist, they have been largely

restricted to the elderly (i.e., Knipscheer, De Jong Gierveld, Van Tilburg, &

Dykstra, 1995; Pinquart, 2003). Furthermore, given their descriptive charac-

ter or their focus on outcomes such as loneliness rather than on the network

itself, the dynamics of the family network have received little attention.

In this article, I try to extend existing knowledge of family networks in

three ways. First, this article is broader in approach than many studies. It

addresses the characteristics of relationships with close kin in general ratherthan those of just one sort of close kin relationship such as the parent–child

relationship or the sibling relationship; it also studies the close kin networks

of all adults rather than those of just one or more particular subgroup such

as the elderly. Second, it focuses on network characteristics and dynamics as

outcome variables rather than as explanatory variables. Third, it examines

the impact of family type on family networks for all the main contemporary

family types, which makes it possible to identify possible differences

between the more traditional and the various new family forms.

Hypotheses

In the first place, this article aims at elaborating and refining our under-

standing of contemporary family life, in particular of differentials associ-

ated with the second demographic transition. According to the family

decline hypothesis, new or contemporary family types are expected to be

associated with weaker family life, which in this article means that individ-

uals living in the new family forms are expected to have less contact withtheir close kin. Our first hypothesis is, therefore, that cohabitants, divorced

and never-married single parents, divorced singles, and members of step-

families formed after divorce will have a lower frequency of contact with

their close kin than will individuals living in classic family types such as

1450 Journal of Family Issues








married couples, widowed singles and single parents, stepfamily members

formed following widowhood, and long-term singles (Hypothesis 1).

However, differences in contact with close kin can develop through twoquite distinct broad causal pathways—the selection and the structural

hypothesis—each containing several subhypotheses. The selection hypoth-

esis originated primarily from family studies and family demography (see,

in particular, the theory on the second demographic transition) and is the

most directly related to the original family decline hypothesis. Individuals

develop attitudes and values concerning the society in general and the

family in particular. These will have an impact on the importance individu-

als attend to formal marriage and to having children, which will have aneffect on their choices concerning the transitions to certain family types

(cohabitation, marriage, divorce, parenthood, etc.; De Jong Gierveld &

Liefbroer, 1998; Lesthaeghe, 2002; Lesthaeghe & Moors, 1994; Moors,

1996). These values will also have an impact on the relations that are main-

tained with family members living outside the household. In particular, atti-

tudes concerning family matters, like attitudes on family support and family

norms, as well as more general attitudes like postmaterialism, are found to

operate within the selection mechanism. Individuals with more postmod-ern, more postmaterialistic attitudes will be expected to be more likely both

to live in the new family types and also to have less contact with their close

kin (Hypothesis 2). Next, for those who are more strongly oriented toward the

family rather than toward friends, contact with family members is likely to be

more intense than for those who are oriented more to friends (Hypothesis 3).

Previous research has indicated that the connection between the family type

in which people live and their family attitudes, as well as their orientation

toward family or friends, is a result of selection (Moors, 1996). Married and

widowed persons tend to have more traditional family attitudes and to be

more family oriented than persons in other family types (Fischer, Sollie,

Soreel, & Green, 1989; Moors, 1996). Divorce may also have an impact on

ties with one’s own family (Johnson, 1992; Terhell, Broese van Groenou, &

Van Tilburg, 2004), and if people who are less oriented to family are more

likely to divorce, divorcees will also be a select group.

Network studies and social capital theory tend to focus more on the

effects of network size and structure. A central idea here is that people have

a certain need for contact. On one hand, the larger the number of biologi-cal kin, the more contact an individual is likely to have with kin. On the

other hand, individuals have limited time and resources. When their poten-

tial family networks are large, individuals will have to maintain contact

with more persons; because of limited time and/or resources, contacts have










to be spread more thinly (Hypothesis 4). Following the same reasoning, the

existence of a second network, for example, of in-laws, could result in less

contact with one’s own (biological) family among married or cohabitatingpersons (Hypothesis 5). A second pronounced structural determinant of

contact frequency can also be found in network studies: distance. In social

support research, we find confirmation that geographical proximity facili-

tates contact (Attias-Donfut, 2003; Rossi & Rossi, 1990). Hogan et al.

(1993), for example, found that contact that is motivated by personal

exchange is more frequent when the distance between the two persons is

relatively small (Hypothesis 6). Finally, having young children in the

household can also have important effects on contact with kin, althoughwhether the effects are positive or rather negative is still a subject of discus-

sion (Hypothesis 7). Hogan et al. (1993) found results that support the idea

that children, especially young children, bring the family together: A

preschool-age child, for example, leads to intensified contact as a result of

a greater need for support with child care. On the other hand, having

children limits the time and energy that can be invested in other family con-

tacts: Moore (1990) found that children led to a smaller effective network,

with fewer kin ties. Having other relatives living in the household (parentsor siblings) is also expected to have an impact on family relations. Because

parents are among the most important close kin (Agneessens et al., 2003;

Attias-Donfut, 2003), it is plausible that other family members will have

more contact when a parent is living in the household (Hypothesis 8). The

same reasoning can be followed for siblings living in the household of the

respondent (Hypothesis 9).

This brief review makes clear that both selection and structure can

account for differences, but the two mechanisms are not always in line with

each other. To test the relative importance of each, I estimate a path model,

as outlined in the following section.

Data, Measures, and Estimation Methods

Data

The Netherlands Kinship Panel Study (NKPS; Dykstra et al., 2004) con-

tains rich information permitting analysis of family networks, includingtesting of both the selection and the structure hypotheses. The NKPS col-

lected information between 2002 and 2004 on the family history, family

structure, and family relations of a random sample of adults living in pri-

vate households in the Netherlands. Information was obtained for a main









sample of 8,155 persons ages 18 to 79 via a computer-assisted personal

interview survey and a supplementary written self-completion survey.

Compared to the total population of the Netherlands, there is a slight over-representation of women, because of differences in response rates, and a

slight underrepresentation of the youngest age groups (for both men and

women) and of the oldest (only for women); people with children at home

are slightly overrepresented, whereas single women living alone and young

adults living with their parents are underrepresented (Dykstra et al., 2004).

Measures

Frequency of contact with close kin. In this analysis, I use a rather

restricted definition of close kin: the respondent’s own (biological) parents,

adult children, and siblings. This definition was chosen first for method-

ological reasons, particularly the need to compare specific kin for adults in

all family types, and second because biological family, especially parents,

children, and siblings, rather than in-laws or more distant kin, tend to be

seen as the key relatives with whom to have contact and to exchange sup-

port (Agneessens et al., 2003; Terhell et al., 2004; Wellman & Wortley,

1989). These last considerations are also reflected in the design of theNKPS, which collected quite detailed data on parents, children, and sib-

lings, and only more limited data on other relatives.

Contact itself can take many forms. Here, I focus primarily on face-to-

face contact, as this is the most important in the context of social support

(Rossi & Rossi, 1990). It is possible of course that telephone or electronic

contact may these days be a partial substitute for face-to-face contact, but

generally similar patterns were found when the analysis was repeated using

these measures rather than face-to-face contact.For each close kin member (father, mother, children, and siblings), the

respondent had to give information about the frequency of contact in the

past year, using the following seven categories (1 = not at all, 2 = once,

3 = a few times, 4 = at least once a month, 5 = at least once a weak, 6 = a few

times a week, and 7 = daily). For adult family members living in the house-

hold of the respondent, an assumption of daily contact was made. The sum of

these sums is taken here as an indicator of overall contact with close kin.

Family type. Twelve mutually exclusive family types are used, hereby

making a distinction between individuals living in more classic family types,

those in newer family forms, and those who have not (yet) started a long-

standing relationship. The distinction concentrates on the de facto situation

and is based on choices concerning marriage and/or partner forming and








dissolution and having children present in the household. I distinguish five

traditional family types—married individuals without children in the house-

hold, married individuals with children in the household, widowed singles,classic single parents (widows and widowers with children in the household),

and individuals living in classic stepfamilies (stepfamilies formed following

widowhood of one of both partners—no distinction is made here between a

respondent’s stepchild or the partner’s stepchild). I also distinguish five new

family types—divorced singles, new single parents (divorced or never mar-

ried persons with a child in the household), cohabitants with children in the

household, cohabitants without children in the household, and members of

new stepfamilies (stepfamilies formed after a divorce of one or both partners).For completeness, I include two categories for primary singles (never married

nor in a relationship that lasted longer than 3 years, and not cohabitating at

the moment of the interview): young primary (younger than 30) singles and

older primary (older than 30) singles. It is not easy to classify individuals in

these two groups as classic or as new family types. The first group is very

diverse, including both those who want to remain single and those who do

not. The second group is more uniform; its members are more likely, delib-

erately or not, to remain single. On one hand, there have always been primarysingles; on the other hand, their number has been increasing recently, sug-

gesting that they may be more new than classic.

Selection Variables

Based on the hypothesis of the selection effect, four variables were iden-

tified as valuable for this analysis: attitudes on family support, attitudes on

family norms, preference for postmaterialism or materialism, and orienta-

tion to family as opposed to friends.

1. A scale of 12 items, with five response categories, was used to estimate the

attitudes on family support. Items such as “You must be able to count on

your family” and “Children should look after their sick parents” (with the

possible scores on each item being 5 = totally agree, 4 = agree, 3 = don’t

know, 2 = don’t agree, and 1 = totally don’t agree) give an indication of the

scale. The scale has strong internal reliability (Cronbach’s α = .875).

2. The scale for attitudes on traditional family norms was constructed in a sim-

ilar way, with the same five response categories. It includes items like “Aman and a woman may cohabitate without marrying” and “A woman should

stop working when she gets a child,” which focus on new family trends and

gender equalities in contemporary family life. These items also result in a

scale with strong internal reliable (Cronbach’s α = .844).










3. The postmaterialism or materialism measure is based on the classic items

that were used by Inglehart (Abramson & Inglehart, 1998). Two dummy

variables were created, one for respondents who rank the two postmaterial-istic items “political participation” and “freedom of speech” highest, and

one for respondents who preferred the two materialistic items “law and

order” and “fighting rising prices.” Respondents who indicated a mixed

preference form the reference category. These three variables are expected

to have a direct selection effect as well as an indirect effect via the variable

on family orientation.

4. The strength of the individual’s family orientation is measured using a scale

based on five items contrasting family members with friends (“If I have

problems I can discuss them with members of my family rather than withfriends,” “I have more confidence in the members of my family than in my

friends,” “If I would need help, I would call on my friends rather than on

members of my family,” “I can rely more on my friends than on members of

my family,” and “I prefer to have friends come to visit than members of the

family”). Where necessary, the items were recoded so that they all scored

in the same direction (1 = strongly oriented toward friends to 5 = strongly

oriented toward family). This scale too has strong internal reliability

(Cronbach’s α = .805).

Structural Variables

In total, six structural variables are included in the model.

1. The size of the family network is simply the number of living parents,

children, and siblings of the respondent.

2. The availability of a second network is a dummy variable based simply on

whether the respondent knows his or her partner’s parents (0 = not known or

not alive or 1 = one or both parents-in-law known) because parents-in-laware the most important individuals in the family-in-law network (Rossi &

Rossi, 1990).

3. The geographical proximity of close kin is operationalized as the number of

close kin living within 10 kilometers of the respondent.

4. The three variables for coresidence of close kin are all dummies based on

the presence (coded 1) or absence (coded 0) of one or more kin of the type

concerned (children, parents, or siblings) living in the respondent’s house-

hold. In this analysis, young children are children under the age of 10,

because we can assume that up to this age children cannot operate indepen-dently from adult supervision, which results in a need for childcare.

Control Variables

Age, gender, and educational attainment are also included in the analyses

as control variables. Age is measured in single years; educational attainment









is entered as a 10-point scale based on the highest education completed,with scores ranging from 0 for those who have not finished elementary edu-

cation to 9 for those with postacademic education. The importance of age

(Ganong, Coleman, McDaniel, & Killian, 1998; Gauthier, 2002), sex

(Attias-Donfut, 2003; Kohli & Künemund, 2003; Marks & McLanahan,

1993), and education (Kalmijn, 2005) in the study of family relations and

family networks has been emphasized in other studies. Given the broad

support for the importance of these personal features, I included them in

this model, and their effects are included in the results. Because they are notcentral to the questions addressed in this particular article, they are not dis-

cussed in the results.

Estimation Method

The model underlying the hypotheses is shown in Figure 1. Ideally, the

various effects in the underlying model would be estimated via structural

equation modeling. However, the model includes endogenous variables that

are measured as nominal and categorical data. It is, therefore, not possibleto estimate the entire model using standard programs for structural equation

modeling. I have, therefore, had to limit the analysis to an examination of

each of the various relationships separately, using either regression or

(multinomial) logistic regression, depending on the dependent variable.

Figure 1

Underlying Model

Classic - Married without children- Married with children- Widowed single- Classic single parent- Classic stepfamilyNew

- Divorced single- New single parent- Cohabitant without children- Cohabitant with children- New stepfamilyPrimary singles - Young primary single- Older primary single

Family type

- Size of primary network- Presence of a second network- Geographical proximity of kin- Co-residence of children,

siblings, or parents

Structural aspects of thefamily network

- Orientation family versus friends- Family norms- Family attitudes- Postmaterialism

Selection

Frequency ofcontact

with close kin

Control variables: age, gender, and education








Results

I discuss first the differences in contact frequency observed betweenindividuals living in different family types and then the estimated direct

effects on contact frequency of family type, the structural variables, and the

selection variables. To evaluate the role of the structural and selection

mechanisms in producing the relationship between family type and contact

frequency, I then combine the findings for the direct effects of the structural

and selection variables on contact frequency with the estimated relationship

of these variables with family type.

Direct Effects of Family Type, Selection Variables, and

Structural Variables on Differences in Contact Frequency

A first regression model (Table 1, Model 1), confirms the existence of

differences in contact frequency for individuals living in the 12 different

family types. Controlling for age, gender, and education, living in almost

every one of the new family types or being an older primary single are asso-

ciated with having statistically significant less contact than the reference

group, namely, married individuals with children in the household. Older

primary singles score the lowest ( B = –5.181, p < .001), followed by cohab-

itants without children ( B = –2.072, p < .001) and with children ( B =–2.058, p < .01). Divorced singles ( B = –1.885, p < .001) and new single

parents ( B = –1.419, p < .05) score higher than these three family types but

still lower than married individuals with children, although the differences

are smaller. Married individuals without children, widowed singles, classic

single parents, and stepfamily members do not differ significantly from

married individuals with children in their households. The significant dif-ferences in contact frequency are mostly in line with the expectations of

Hypothesis 1, namely, that individuals living in the newer family forms will

have less frequent contact with close kin. However, the hypothesis does not

hold for new stepfamily members.

Table 1, Model 2, presents the results of a multiple regression analysis

of contact frequency that includes not only family type but also the struc-

tural and selection variables as independent variables. Note that most of the

effects of family type are now insignificant or markedly smaller, except foryoung primary singles, married individuals without children, and new step-

families, where a significant difference appears, and also that in general the








T a b l e 1

E f f e c t o f S t r u c t u r e a n d S e l e c t i o n o n C o n t a c t F r e q u e n c y : S t e p w i s e - B u i l t R e g r e s s i o n

M o d e l W i t h F a m i l y T y p

e s ( M o d e l 1 ) a n d S t r u c

t u r e a n d S e l e c t i o n V a r i a b l e s ( M o d e l 2 ) ,

U n s t a n d a r d i z e d a n d S t a n d a r d i z e

d R e g r e s s i o n C o e f f i c i e n t s

F a c e - t o - F a c e C o n t a c t

M o d e l 1

M o d e l 2

R e g r e s s i o n

B

β

B

β

C o n s t a n t

1 8 . 1 1 4 * * *

– 1 . 1 0 9

F a m i l y t y p e s a

C l a s s i c

M a r r i e d w i t h o u t c h i l d r e n

0 . 6 2 9

. 0 2 6

1 . 0 3 4 * * *

. 0 4 2

W i d o w e d s i n g l e

– 0 . 4 6 6

– 0 . 0 1 0

0 . 0 8 0

. 0 2 0

C l a s s i c s i n g l e p a r e n t

1 . 2 7 6

. 0 1 1

0 . 9 4 7

. 0 0 8

C l a s s i c s t e p f a m i l y

– 2 . 5 2 2

– . 0 1 2

– 1 . 3 6 9

– . 0 0 7

N e w D i v o r c e d s i n g l e

– 1 . 8 8 5 * * *

– . 0 5 5

– 0 . 4 8 2

– . 0 1 4

N e w s i n g l e p a r e n t

– 1 . 4 1 9 *

– . 0 2 7

– 1 . 0 9 6 * *

– . 0 2 1

C o h a b i t a n t w i t h o u t c h i l d r e n

– 2 . 0 7 2 * * *

– . 0 5 3

0 . 1 1 7

. 0 0 3

C o h a b i t a n t w i t h c h i l d r e n

– 2 . 0 5 8 * *

– . 0 3 3

0 . 3 3 3

. 0 0 5

N e w s t e p f a m

i l y

– 0 . 7 3 4

– . 0 1 1

– 1 . 4 8 4 * *

– . 0 2 2

P r i m a r y s i n g l e s

Y o u n g p r i m a r y s i n g l e

– 0 . 4 3 5

– . 0 1 1

0 . 9 6 2 *

. 0 2 4

O l d e r p r i m a r y s i n g l e

– 5 . 1 8 1 * * *

– . 1 2 8

– 0 . 7 2 0 *

– . 0 1 8

S t r u c t u r e v a r i a b l e s

N e t w o r k s i z e

2 . 3 6 9 * * *

. 6 8 4

A v a i l a b i l i t y s e c o n d n e t w o r k

– 0 . 6 4 0 * * *

– . 0 3 1

G e o g r a p h i c a l p r o x i m i t y

1 . 5 1 4 * * *

. 3 0 4

( c o n

t i n u e d )

1458








T a b l e 1 ( c o n t i n u e d )

F a c e - t o - F a c e C o n t a c t

M o d e l 1

M o d e l 2

R e g r e s s i o n

B

β

B

β

C o r e s i d e n t f a m

i l y b

P a r e n t s

4 . 9 0 8 * * *

. 0 7 5

Y o u n g c h i l d r e n

– 4 . 6 3 5 * * *

– . 1 9 1

S i b l i n g s

0 . 1 7 2

. 0 0 2

S e l e c t i o n v a r i a b l e s

F a m i l y v e r s u s

f r i e n d s o r i e n t a t i o n

0 . 3 1 9 * * *

. 1 1 3

A t t i t u d e s f a m i l y s u p p o r t

– 0 . 0 0 1

. 0 0 0

A t t i t u d e s f a m i l y n o r m s

– 0 . 0 0 9

– . 0 0 6

P o s t m a t e r i a l i s m c

– 0 . 4 9 8 *

– . 0 1 4

M a t e r i a l i s m

– 0 . 2 5 7

– . 0 1 0

C o n t r o l v a r i a b l e s

A g e

0 . 0 9 0 * * *

. 1 2 9

– 0 . 0 6 3 * * *

– . 0 9 1

S e x d

0 . 3 1 4

. 0 1 5

0 . 6 1 7 * * *

. 0 2 9

E d u c a t i o n

– 0 . 6 1 2 * * *

– . 1 3 6

0 . 2 0 8 * * *

. 0 4 6

R ²

. 0 7 6

. 7 3 0

A d j u s t e d R ²

. 0 7 4

. 7 2 9

C h a n g e d R ²

. 0 7 6 * * *

. 6 5 3 * * *

a . R e f e r e n c e c a t e g o r y f o r f a m i l y t y p e s i s m a r r i e d i n d i v i d u a l s w i t h c h i l d r e n .

b . R e f e r e n c e c a t e

g o r y f o r c o r e s i d e n t k i n i s n o c o r e s i d e n t k i n .

c . R e f e r e n c e c a t e g o r y f o r p o s t m a t e r i a l i s m o r m a

t e r i a l i s m i s t h e i n t e r m e d i a t e p o s i t i o n .

d . R e f e r e n c e c a t e

g o r y f o r s e x i s m a l e .

* p <

. 0 5 . * * p < . 0 1 . * * * p <

. 0 0 1 .

1459









effects of the structural variables are stronger than those of the selection

variables. However, to evaluate and elaborate the possible structure and

selection mechanisms, it needs to be examined whether those structural andselection variables that are significantly associated with contact frequency

are also significantly associated with particular family types.

The Structural Mechanism

The estimated effects of the structural variables on contact frequency are

presented in Table 1, Model 2. The estimated effects of family type on the

structural variables are shown in Table 2 (regression) and Table 3 (logisticregression). I examine the role of each of our structural variables in turn.

First of all, the size or range of the network (Hypothesis 4) is, as

expected, of primary importance in the frequency of contact with close

kin (see Table 1). Having more family members results in a higher score

on the overall frequency of contact with kin: One additional person in the

network produces an extra score of 2.369 ( p < .001) on the overall fre-

quency scale. This does not mean, however, that a larger network is asso-

ciated with more frequent contacts per kin member: The opposite is true.We can see this easily if we compare, for example, the 10th and the 90th

percentiles. These correspond with scores of 6 and 32, respectively, on

overall contact frequency; that is, they differ by 26. The same percentiles

differ by only 7 on network size. The coefficient of 2.369 applied to a dif-

ference of 7 in network size is only 16.583, which is considerably less

than the observed 26. In other words, the coefficient of 2.369 implies that

for each additional person in the network there is more frequent contact

in total, but less contact per person. Table 2 shows that all family types,

except single parents and classic stepfamilies, have significant smaller

networks than married persons with children. Primary singles and, in

decreasing order, cohabitants without children and divorced singles show

quite smaller networks. Married individuals without children, widowed

singles, and cohabitants with children have slightly smaller family net-

works, compared to the reference category. The distinctive smaller net-

works for divorced singles and cohabitants without children result in a

first structural explanation for the differences between classic and new

family types in contact frequency.The second structural variable, geographical proximity (Hypothesis 6),

also exhibits a strong effect on contact frequency (see Table 1) as could be

expected given the large support for the effect of geographical proximity in

previous research (Attias-Donfyt, 2003; Rossi & Rossi, 1990).









Primary singles, especially young ones, have the least amount of kin liv-

ing within 10 kilometers (see Table 2). Individuals living in new familytypes such as divorced singles, cohabitants without children, and new step-

families are also less likely to be living close to relatives. This combined

with the effect of geographical proximity on contact frequency provides

another partial explanation for their low contact frequency.

Table 2

Effect of Family Types on Structural Features: Regression on Size of

Family Network and Geographic Proximity, Unstandardized andStandardized Regression Coefficients

Structural Variables

Geographical

Size Network Proximity

Regression B β B β

Constant 5.906*** 3.757***Family typesa

Classic

Married without children –1.294*** –.182 0.124 .025

Widowed single –1.236*** –.098 0.049 .005

Classic single parent –0.600 –.017 –0.044 –.002

Classic stepfamily –0.220 –.040 –0.640 –.015

New

Divorced single –1.494*** –.150 –0.360*** –.052

New single parent –0.091 –.006 –0.209 –.020

Cohabitant without children –2.065*** –.180 –0.356*** –.044Cohabitant with children –0.689*** –.039 –0.193 –.016

New stepfamily 0.416* .022 –0.309* –.023

Primary singles

Young primary single –2.175*** –.193 –1.219*** –.154

Older primary single –2.857*** –.241 –0.500*** –.060

Control variables

Age 0.044*** .222 –0.004 –.028

Sexb –0.091 –.015 –0.090 –.021

Education –0.161*** –.125 –0.243*** –.270

R² .172 .101Adjusted R² .171 .100

a. Reference category for family types is married individuals with children.

b. Reference category for sex is male.

* p < .05. *** p < .001








T a b l e 3

E f f e c t s o f F a m i l y T y p e s

o n S t r u c t u r a l F e a t u r e s

: L o g i s t i c R e g r e s s i o n o n S e c o n d F a m i l y

N e t w o r k ,

C o r e s i d e n t P a r e n t , S i b l i n g ,

a n d Y o u n g C h i l d ,

O d d s R a t i o s

S t r u c t u r a l V a r i a b l e s

L o g i s t i c R e g r e s s i o n

S e c o n

d N e t w o r k

C o r e s i d e n t P a r e n t

C o r e s i d e n t S i b l i n g

C o r e s i d e n t Y o u n

g C h i l d

C o n s t a n t

4 8

5 . 0 4 9 * * *

1 3 0 . 5 5 0

3 . 6 2 2

1 4 9 9 4 0 0 * * *

F a m i l y t y p e s a

C l a s s i c

M a r r i e d w i t h o u t c h i l d r e n

0 . 4 5 6 * * *

0 . 0 0 0

0 . 0 0 0

0 . 0

0 0

W i d o w e d s i n g l e

0 . 0 2 9 * * *

0 . 0 0 0

6 . 5 8 5

0 . 0

0 0

C l a s s i c s i n g l e p a r e n t

0 . 0 2 3 * * *

0 . 0 0 0

0 . 0 0 0

0 . 2

9 5 *

C l a s s i c s t e p f a m i l y

4 . 6 4 3 *

0 . 0 0 0

0 . 0 0 0

1 . 1

4 7

N e w D i v o r c e d s i n g l e

0 . 0 3 6 * * *

2 . 4 1 8

2 . 7 3 5 *

0 . 0

0 0

N e w s i n g l e p a r e n t

0 . 0 3 9 * * *

0 . 0 0 0

0 . 0 0 0

0 . 6

9 8 *

C o h a b i t a n t w i t h o u t c h i l d r e n

1 . 0 1 6

0 . 2 5 6

0 . 2 0 5 *

0 . 0

0 0

C o h a b i t a n t w i t h c h i l d r e n

1 . 1 2 4

0 . 3 1 6

0 . 9 5 2

1 . 8

6 3 *

N e w s t e p f a m

i l y

1 . 1 8 5

0 . 0 0 0

0 . 0 0 0

1 . 2

1 0

P r i m a r y s i n g l e

Y o u n g p r i m a r y s i n g l e

0 . 0 0 9 * * *

7 . 3 4 6 * * *

7 . 5 4 2 * * *

0 . 0

0 0

O l d e r p r i m a

r y s i n g l e

0 . 0 1 5 * * *

9 . 7 6 3 * * *

3 . 6 6 5 * * *

0 . 0

0 0


A g e

0 . 8 9 2 * * *

0 . 7 5 2 * * *

0 . 8 7 7 * * *

0 . 7

1 5 * * *

S e x b

0 . 6 4 8 * * *

0 . 8 6 7

0 . 9 3 0

0 . 4

3 3 * * *

E d u c a t i o n

1 . 1 1 6 * * *

0 . 8 7 7 * *

0 . 8 0 7 * * *

1 . 1

8 5 * * *

N a g e l k e r k e R ²

. 6 1 3

. 6 3 7

. 4 7 3

. 8

4 1

a . R e f e r e n c e c a t e

g o r y f o r f a m i l y t y p e s i s m a r r i e

d i n d i v i d u a l s w i t h c h i l d r e n .

b . R e f e r e n c e c a t e

g o r y f o r s e x i s m a l e .

* p <

. 0 5 . * * p < . 0 1 . * * * p <

. 0 0 1 .

1462








The third structural variable, the availability of a second family network

(Hypothesis 5) also has a significant effect on contact frequency (see Table

1). It is also related to family type (see Table 3), exhibiting a very logical neg-ative relationship with all types of single individuals. On the other hand, a few

not-so-expected results become clear with a closer look at the results. Being

married without children has a significant negative effect on the availability

of a second network in comparison to being married with children. The same

is true for widowed singles. Although I controlled for age in general, these

results may be an effect of a concentration of older persons in these two

family types, especially in the category of widowed singles, resulting in a

higher chance that both parents-in-law are deceased. The combination of asignificant effect of a second network on contact frequency with several sig-

nificant relationships between family type and the availability of a second

network makes it possible to conclude that this can also account in part for

the differences in contact by family type, especially for singles.

The estimated effects of having family members living in the household

(see Table 1) suggest that these effects can be brought back to constraints

imposed by limited time and resources. Having a young child present in the

household has a large negative effect on contact with close kin ( B = –4.635,β = –0.191, p < .001; Hypothesis 7). This is in line with Moore (1990), who

said that having children intensifies the interaction within the household

and therefore leaves less time for maintaining relations with close kin out-

side the household. Combining this effect with the relationship between

family type and having a young child in the household, we see that this

mechanism largely accounts for the lower contact frequency of cohabitants

with children, given the fact that these persons have a greater chance of

having young children in the household. For single parents, it works in the

other direction, however, given the fact that children in these households

are mostly of older age. Table 1 shows that parents living in the household

also are of great importance for the frequency of contact with close kin

(Hypothesis 8). Given the fact that the score on contact frequency was aug-

mented with 7 (assumed daily contact) for adults living in the household,

the observed positive effect is logical, but its value shows that contact fre-

quency with other family members is lower. Having a parent living in the

household is most common for primary singles, who can more easily take

in a parent than can individuals with partners or children (see Table 3). Thisresults in overall positive effects only for singles though. From Table 1, we

see that having a sibling living in the household has no effect on contact fre-

quency and cannot, therefore, account for the differences in contact by

family type. Therefore we have to reject the Hypothesis 9.










The Selection Mechanism

The selection variables show fewer significant direct effects on contactfrequency than the structural variables do (Table 1, Model 2). In general,

therefore, we might expect the selection mechanism to be weaker unless the

selection variables have particularly strong effects on family type. The

results of multinomial logistic regressions for the latter are given in Table 4.

A relatively strong orientation toward family rather than friends does

have a direct positive effect on the frequency of contact with close kin

(Hypothesis 3): As expected, people who feel more strongly about their

families also have more contact with them (see Table 1). Family orientation

is also significantly related to certain family types: Individuals who are

more strongly oriented to family are less likely to be living as primary sin-

gles, new single parents, divorced singles, cohabitants without children, and

classic stepfamily members. The combination of these effects provides fur-

ther explanation for the low levels of family contact among individuals liv-

ing in some of the new family types. Family orientation does not have a

significant effect on the likelihood of being a cohabitant with children or

living in a new stepfamily, however.

Having more postmaterialistic ideas is related to less contact with closekin (see Table 1). Primary singles and cohabitants with children score

remarkably higher on postmaterialism (see Table 4), compared to the refer-

ence category, than do married or cohabitating individuals without children.

None of the other selection variables contributes directly to differences

in contact frequency by family type. Although primary singles, divorced

singles, and single parents have more positive ideas on support (see Table

4), and divorced singles, single parents, and cohabitants are more likely to

be found among those with less attention to classic family norms (see Table4), these variables cannot account for the differences in the overall fre-

quency of contact with their kin, because there is no direct effect of these

attitudes on contact frequency (see Table 1).

Therefore, if these variables play a role at all, it must be through their

effect on family orientation and, to a lesser extent, through having more

postmaterialistic values.

Conclusion and Discussion

With rising voluntarism in family life decisions, an increasing number of

individuals are likely to be living in new family types in the future; it is,

therefore, of interest to develop more insights in the family dynamics and








T a b l e 4

E f f e c t o f S e l e c t i o n V a r i a b l e s o n F a m i l y T y p e s : M u l t i n o m i a l R e g r e s s i o

n ,

O d d s R a t i o s

F a m i l y T y p e s a

C l a s s i c

N e w

P r i m a r y s i n g l e s

M a r r i e d

C l a s s i c

N e w

C o h a b i t a n t

C o h a b i t a n t

Y o u n g

O l d e r

M u l t i n o m i a l L o g i s t i c

W i t h o u t

W i d o w e d

S i n g l e

C l a s s i c

D i v o r c e d

S i n g l e

W i t h o u t

W i t h

N e w

P r i m a r y

P r i m a r y

R e g r e s s i o n

C h i l d r e n

S i n g l e

P a r e n t

S t e p f a m i l y

S i n g l e

P a r e n t

C h i l d r e n

C h i l d r e n

S t e p f a m i l y

S i n g l e

S i n g l e

S e l e c t i o n v a r i a b l e s

F a m i l y v e r s u s f r i

e n d s

o r i e n t a t i o n

0 . 9 9 3

0 . 9 7 0

1 . 0 0 3

0 . 8 6 1 *

0 . 9 0 8 * * *

0 . 9 0 5 * * *

0 . 9 6 7 *

1 . 0

0 8

0 . 9 7 8

0 . 8 9 8 * * * 0

. 9 3 1 * * *

A t t i t u d e s f a m i l y s u p p o r t

0 . 9 9 5

1 . 0 0 3

0 . 9 7 3

0 . 9 9 9

1 . 0 3 5 * * *

1 . 0 5 8 * * *

1 . 0 1 3

1 . 0

1 3

1 . 0 2 8 *

1 . 0 1 6

1

. 0 3 2 * * *

A t t i t u d e s f a m i l y n o r m s

1 . 0 2 0 * *

0 . 9 8 0

1 . 0 3 0

0 . 9 9 1

0 . 9 6 3 * * *

0 . 9 5 2 * * *

0 . 9 4 5 * * *

0 . 9

2 5 *

0 . 9 8 8

0 . 9 8 9

0

. 9 9 5

P o s t m a t e r i a l i s m b

1 . 3 4 6 *

1 . 0 7 6

1 . 2 6 5

0 . 8 8 3

1 . 7 4 6

2 . 0 3 4

1 . 4 9 3 *

1 . 8

2 1 *

1 . 4 9 4

2 . 0 1 8 *

1

. 5 0 9 *

M a t e r i a l i s m

0 . 9 1 0

0 . 9 7 3

0 . 9 1 9

0 . 6 9 8

0 . 9 7 4 * *

1 . 1 3 7 * *

0 . 9 8 6

1 . 2

2 4

0 . 7 9 3

0 . 6 2 0 *

0

. 8 5 3


A g e

0 . 8 9 0 * * *

1 . 1 2 8 * * *

0 . 9 5 9 * *

0 . 9 6 8

0 . 9 6 0 * * *

0 . 9 0 0 * * *

0 . 8 3 3 * * *

0 . 8

5 4 * * *

0 . 9 0 1 * * *

0 . 5 6 2 * * * 0

. 9 2 3 * * *

S e x c

1 . 2 3 2 * *

2 . 8 8 9 * * *

1 . 8 0 9 *

0 . 9 8 4

1 . 0 2 1 * * *

5 . 4 5 8 * * *

0 . 8 6 5

1 . 0

7 7

1 . 3 1 8

0 . 4 1 4 * * * 0

. 8 8 4

E d u c a t i o n

1 . 0 2 6

0 . 8 8 8 * * *

0 . 9 1 9

0 . 8 1 4

0 . 9 7 5

0 . 9 4 3

1 . 1 0 3 * *

1 . 0

8 6 *

0 . 9 0 8 *

1 . 1 2 9 * * 1

. 0 9 4 * * *

M o d e l

– 2 l o g l i k e l i h o o d =

2 2 , 6 3 7 ; χ 2 =

6 , 4 5 9 ; d f =

8 8 ; p s e u d o R 2 N a g

e l k e r k e =

. 6 1

a . R e f e r e n c e c a t e g o

r y f o r f a m i l y t y p e s i s m a r r i e d i n d i v

i d u a l s w i t h c h i l d r e n .

b . R e f e r e n c e c a t e g o

r y f o r p o s t m a t e r i a l i s m o r m a t e r i a l i s m i s t h e i n t e r m e d i a t e p o s i t i o n .

c . R e f e r e n c e c a t e g o

r y f o r s e x i s m a l e .

* p <

. 0 5 . * * p <

. 0 1 . * * * p <

. 0 0 1 .

1465








family relations of individuals living in the new family types. Based on the

family decline hypothesis and on studies of the second demographic tran-

sition, I hypothesized that individuals living in new family types wouldhave less contact with their close kin family, indicating the higher volun-

tarism in new family life. The analysis of the NKPS data reveals consider-

able differences in frequency of contact with close kin by family type, with

individuals living in the classic family types having a higher contact fre-

quency on average than those living in other family types. I have shown that

individuals living in nearly all the new family types associated with the sec-

ond demographic transition tend to have less contact with their close kin

than do those in the reference category (married individuals with childrenin the household). Cohabitating individuals and divorced singles in partic-

ular have lower frequency of contact. Unexpected results are found, how-

ever, for individuals living in a stepfamily formed after a divorce. Although

this family type can be clearly categorized as a new family form, it shows

no significant difference from the classic married-with-children type. Older

primary singles (individuals older than 30 who have not yet entered in a

longstanding or cohabitating relationship) have markedly lower contact fre-

quency, which can be traced both to weaker orientation to family and tohaving fewer close kin living relatively close by. This suggests that

although this family type has always existed, its increasing share in the total

population reflects a new family type rather than a classic family type.

Given the fact that contact frequency is linked with support (Hogan

et al., 1993; Marks & McLanahan, 1993) and solidarity (Silverstein &

Bengston, 1997), I can, based on the results, expect that most of the indi-

viduals in new family types will have less access to support from their rel-

atives. This means that I do not find confirmation for the idea that the

multigenerational bonds in new families have been taking over some of the

lost functions of the classic family, as Bengston (2001) argued.

Although the differences between individuals living in new or contem-

porary family types and those in the more classic family forms can be

explained by selection and by structure, structural characteristics of the

family network appear to be more important in determining contact fre-

quency than selection, based on family orientation and other attitudes. The

larger the number of close kin, the greater the overall contact frequency but

the thinner the contact tends to be spread over the various kin members, inline with the notion of constraints on the time and resources devoted to kin.

The effects of having young children in the household and the availability

of a second network and of proximity confirm the importance of time and

resource constrains. With individuals in new family types tending to live









further away from their relatives, proximity is one of the most important

paths in explaining the differences between new and traditional families.

The selection hypothesis is supported by the effect of family versus friendorientation, with most individuals in most of the new family types being

more likely to be more oriented toward their friends and by the higher post-

materialism of cohabitants. That selection tends to be less important in

explaining contact frequency differences can be due to the changed ratio of

new to classic family types. As more individuals start living in the new

family types, we could assume that living in those types will be more

accepted. Consequently, differences in attitudes and norms between indi-

viduals in new and classic family types can be assumed to become smallerover time and disappear eventually. Based on the longer tradition of infor-

mal partnerships, the postponement of having children, and rising divorce

rates, the Netherlands can be seen as rather ahead in the second demo-

graphic transition compared to the rest of the European countries (de Beer

& Deven, 2000). In spite of this, the majority of the Dutch are still members

of a classic family type and have a traditional life course (Dykstra &

Komter, 2006). It is possible that the selection effect and consequently the

differences in contact between individuals in classic and new family typeswill further slow down if more individuals enter new family types.

Finally, some limitations should be mentioned. First, the NKPS data set

has a few limitations. The response rate was only 45%, which is a normal

rate for surveys in the Netherlands (Dykstra et al., 2004), but still rather

low. Information on the nonrespondents is very limited, but it cannot be

ruled out that persons for whom family is less important are underrepre-

sented. Furthermore, although the NKPS is one of the most extensive sur-

veys on family life and family relations, practical constraints meant that

even here only part of the family network could be studied in detail and

choices as to which relations would be studied in detail had to be made.

Based on this data set, it was, therefore, not possible to make the link with

other personal and family networks.

Second, note that some of these theoretical concepts were not easy to

operationalize. This was particularly the case for the selection mechanism,

where I tried to combine several different types of potentially relevant

information. Family orientation was introduced as a key variable because of

its high relevance given that social capital and networks may be becomingmore oriented toward friends than toward family. However, the family ori-

entation variable does not capture underlying family and nonfamily values.

I therefore added scales based on attitudes to traditional family organization

and on the importance of family as a source of potential support and also a









basic indicator of postmaterialism. Despite the high internal strength of the

scales, neither they nor postmaterialism showed many direct effects. Any

effects they have are presumably channeled through family orientation.Last but not least, this particular article is limited to contact frequency

and to cross-sectional data. I have not addressed the content of kin contact,

nor have I explicitly addressed the possible impact of transitions to other

family types over the life course. The first of these is, however, currently

under study and will be the topic of another article.

References

Abramson, P. R., & Inglehart, R. (1998). Value change in global perspective. Ann Arbor:

University of Michigan Press.

Agneessens, F., De Lange, D., & Waege, H. (2003). Over het verband tussen sociale relaties

en attitudes, waarden en normen [About the relationship between social relations and atti-

tudes, values and norms]. In Administratie Planning en Statistiek (Ed.), Vlaanderen

Gepeild, 2003 (pp. 15-51). Brussels, Belgium: Ministerie van de Vlaamse Gemeenschap.

Attias-Donfut, C. (2003). Family transfers and cultural transmissions between three genera-

tions in France. In B. L. Bengston & M. Silverstein (Eds.), Global aging and challenges

to the families (pp. 214-250). New York: Aldine De Gruyter.

Bengston, V. L. (2001). Beyond the nuclear family: The increasing importance of multigener-

ational bonds. Journal of Marriage and Family, 63, 1-16.

Burgess, E. (1916). The function of socialization in social evolution. Chicago: University of

Chicago Press.

Busschots, M., & Lauwers, J. (1994). Familiale en sociaal culturele werken [Family and social

cultural work]. Leuven, Belgium: Acco.

Coleman, M., Ganong, L., & Fine, M. (2000). Reinvestigating remarriage: Another decade of

progress. Journal of Marriage and Family, 62, 1288-1367.

de Beer, J., & Deven, F. (2000). Diversity in family formation: The second demographic tran-

sition in Belgium and the Netherlands. Dordrecht, Netherlands: Kluwer Academic.De Jong Gierveld, J., & Liefbroer,A. (1998). Sociale herkomst, opvattingen over relaties en gezin

en leefvormkeuzes van jongvolwassenen [Social background, family attitudes and living

arrangement choices of young adults: Results of a panel study]. Sociologische gids, 45, 96-115.

Dykstra, P., Kalmijn, M., Knijn, T., Komter, A., Liefbroer, A., & Mulder, C. (2004). Codebook

of the Netherlands Kinship Panel Study: A multi-actor, multi-method panel study on soli-

darity in family relationships. Wave 1 [NKPS working paper] (Vol. 1). The Hague:

Netherlands Interdisciplinary Demographic Institute.

Dykstra, P., & Komter, A. (2006). Structural characteristics of Dutch kin networks. In P.

Dykstra, M. Kalmijn, T. Knijn, A. Komter, A. Liefbroer, & C. Mulder (Eds.), Family soli-

darity in the Netherlands (pp. 21-42). Amsterdam: Dutch University Press.Fischer, J. L., Sollie, D. L.,Sorell, G. T., & Green, S. K. (1989). Marital status and career stage influ-

ences on social networks of young adults. Journal of Marriage and Family, 51, 512-534.

Ganong, L., Coleman, M., McDaniel, A. K., & Killian, T. (1998). Attitudes regarding obliga-

tions to assist on older parent and stepparent following later life remarriage. Journal of

Marriage and Family, 60, 595-610.









Gauthier, A. (2002). The role of grandparents. Current Sociology, 50, 295-307.

Henderson, S. H., & Taylor, L. C. (1999). Parent–adolescent relationships in nonstep-, simple

step-, and complex stepfamilies. In E. M. Hetherington, S. H. Henderson, & D. Reiss(Eds.), Adolescent siblings in stepfamilies: Family functioning and adolescent adjustment

(pp. 79-101). Cambridge, MA: Blackwell.

Hogan, D. P., Eggebeen, D. J., & Clogg, C. C. (1993). The structure of intergenerational

exchanges in American families. American Journal of Sociology, 98, 1428-1458.

Johnson, C. L. (1992). Divorced and reconstituted families: Effects on the older generation.

Generations, 16, 17-21.

Kalmijn, M. (2005). Educational inequality and family relationships: Influences on contact

and proximity. European Sociological Review, 22, 1-16.

Knijn, T. (2004). Family solidarity and social solidarity: Substitutes or complements. In

T. Knijn & A. Komter (Eds.), Solidarity between the sexes and generations (pp. 18-33).Cheltenham, UK: Edward Elgar.

Knipscheer, C. P. M., De Jong Gierveld, J., Van Tilburg, T. G., & Dykstra, P. (1995). Living

arrangements and social networks of older adults. Amsterdam: VU University Press.

Kohli, M., & Künemund, H. (2003). Intergenerational transfers in the family: What motivates

giving? In V. L. Bengston & M. Silverstein (Eds.), Global aging and challenges to families

(pp. 123-142). New York: Aldine de Gruyter.

Lansford, J. E., Ceballo, R., Abbey, A., & Stewart, A. J. (2001). Does family structure matter?

A comparison of adoptive, two-parent biological, single-mother, stepfather, and step-

mother households. Journal of Marriage and Family, 63, 840-851.

Lesthaeghe, R. (Ed.). (2002). Meaning and choice: Value orientations and life course deci-sions. Brussels, Belgium: NIDI-CBGS Monographs.

Lesthaeghe, R., & Meekers, D. (1986). Value changes and the dimensions of familism in the

European Community. European Journal of Population, 2, 225-268.

Lesthaeghe, R., & Moors, G. (1994). Living arrangements, socio-economic position, and val-

ues among young adults: A pattern description for Belgium, France, the Netherlands and

West-Germany, 1990. In H. Van den Brekel & H. Deven (Eds.), Population and family in

the Low Countries 1994: Selected current issues (pp. 1-56). Dordrecht, Netherlands:

Kluwer Academic.

Lesthaeghe, R., & Neidert, L. (2006). The second demographic transition in the United States:

Exception or textbook example? Population and Development Review, 34, 669-698.Lesthaeghe, R., & Surkyn, J. (2007). When history moves on: The foundations and diffusion

of a second demographic transition. In R. Jayakody, A. Thornton, & W. Axinn (Eds.),

International family change: Ideational perspectives. Mahwah, NJ: Lawrence Erlbaum.

Lye, D. N. (1996). Adult-child parent relationships. Annual Review of Sociology, 22, 79-122.

MacDonald, W. L., & DeMaris,A. (2002). Stepfather–stepchild relationship quality: The step-

father’s demand for conformity and the biological father’s involvement. Journal of Family

Issues, 23, 121-137.

Marks, N. F., & McLanahan, S. (1993). Gender, family structure, and social support among

parents. Journal of Marriage and Family, 55, 481-493.

Moore, G. (1990). Structural determinants of men’s and women’s personal networks. American Sociological Review, 55, 726-735.

Moors, G. (1996). Gezinsvorming en processen van waardenselectie en–aanpassing [Union for-

mation and processes of selection and adaptation of values]. Mens en maatschappij, 71, 4-24.

Parsons, T., & Bales, R. F. (1955). Family: Socialization and interaction process. New York:

Free Press.









Pinquart, M. (2003). Loneliness in married, widowed, divorced, and never-married older

adults. Journal of Social and Personal Relationships, 20, 31-53.

Popenoe, D. (1993). American family decline, 1960-1990: A review and appraisal. Journal of Marriage and Family, 55, 527-555.

Rossi, A. S., & Rossi, P. H. (1990). Of human bonding. New York: Aldine de Gruyter.

Schwarz, B., Trommsdorff, G., Albert, I., & Mayer, B. (2005). Adult parent–child relation-

ships: Relationship quality, support, and reciprocity. Applied Psychology: An International

Review, 54, 396-417.

Silverstein, M., & Bengston, V. L. (1997). Intergenerational solidarity and the structure of

adult child–parent relationships in American families. American Journal of Sociology, 103,

429-460.

Terhell, E. L., Broese van Groenou, M. I., & Van Tilburg, T. (2004). Network dynamics in the

long-term period after divorce. Journal of Social and Personal Relationships, 21, 719-738.Thornton, A. (1989). Changing attitudes toward family issues in the United States. Journal of

Marriage and Family, 51, 873-893.

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, 1009-1037.

Wellman, B., & Wortley, S. (1989). Brothers’ keepers: Situating kinship relations in broader

networks of social support. Sociological Perspectives, 32, 273-306.

White, L. (1994). Growing up with single parents and stepparents: Long-term effects on

family solidarity. Journal of Marriage and Family, 56, 935-948.


selection versus structure_ explaining family type differences in contact with close kin

Documents