Spillover PDF
Spillover PDF
Spillover PDF
Year
Spillover
Patricia V. Roehling
Rosemary Batt
Hope
College
University, pem3@cornell.edu
Cornell University, rb41@cornell.edu
This paper is posted at DigitalCommons@ILR.
Cornell
http://digitalcommons.ilr.cornell.edu/hrpubs/24
Phyllis Moen
7
Spillover
Patricia V Roehling, Phyllis Moen, and Rosemary Batt
102
Historical Overview
Spillover Literature
of the
I.
I
workers
positive
12 pem
correlati
Segmen
characte
encountl
Ama
from th
of spillc
from or
you are
relying
moved
designs
studies).
Wor:
The'
has dev
and inte
role ov
(or mor
more (
work-fa:
an indi'
research
behavi01
and inte
into the
and the,
negative
contrast
the delt
spilloveJ
BecR
spillovel
simply::
used by
family r
Spillover
ng of the
example,
to higher
ls of pro~and conIIconflict,
tress, and
e concept
r between
lage their
and idenlWon our
incidence
lcteristics
low does
affect the
~draw on
tterviews)
nificantly
e correlaand satisltionship:
pensation
19 to find
lction are
compart .e domain
tisfaction
es III one
nonwork
ajority of
from the
lercent of
103
workers met the criteria for spillover (work satisfaction and nonwork satisfaction
positively correlated), 20 percent for segmentation (minimally correlated), and
12 percent for compensation (negatively correlated).9 It is estimated that the
correlation between work and nonwork satisfaction is between 0.40 and 0.48.10
Segmentation and compensation, when they do occur, are more likely to
characterize employees in nonprofessional occupations and workers who have
104
One explanation for the greater level of work-to-family spillover is the relative inflexibility of work life compared to family life. In most cases, employees
are required to work a set number of hours, with the scheduling and location of
work relatively fixed. Family roles, however, have no external guidelines for the
amount of time that must be devoted to family members or the location in which
family interactions must take place. When work-family conflicts do arise, workers
can hire others to perform many domestic duties (e.g., child care, cleaning, and
cooking), which is rarely the case for workplace responsibilities. As a result of
this relative inflexibility, demands atwork tend to invade and dictate the pace and
timing of family life.
A second explanation is that employees typically have less control over decisions in their work life than they do in their family life. For example, employers
dictate and may alter the hours, location, or conditions of work without notice,
whereas workers exert more control over such decisions on the home front. A
1995 study of health-care professionals (nonphysicians) showed that when
employees have greater influence over decisions at work, they are more able to
balance work and family demands.14Similar findings emerge in The Cornell
Couples and Careers Study.
Factors Influencing
Spillover
Role Involvement
According to the scarcity hypothesis, the more committed a person is to a particular role, the greater the chance of conflict or interference with other roles. 15
Consistent with this hypothesis, several studies have found that spillover is related
to level of role involvement. Scholars have found that the more involved or the
more central an individual's work role, the greater the reported level of work-
role also
actual h(
predict g
Invol'
of work
positive:
alumni,
with wo
Note, he
in the pi
spent in
positive
on the I
children,
the man
importar
Genl
The
be und(
spillover
role thm
expect 1
men. EJ
longitud
expeneIJ
studies,
greater j
children
on degn
that the1
spilloveJ
.
Then
to-famil
tive WOJ
Three st
expener
women
more tir
that, in r
and inte
Spillover
lfe Impor1SUSfinds
'amily life
it random
:, for both
feres with
s the rela~mployees
ocation of
1es for the
1 in which
e, workers
ming, and
a result of
~pace and
over deciemployers
)lit notice,
e front. A
that when
)re able to
re Cornell
is to a parler roles.
15
r is related
ved or the
1 of work'e measure
lterference
the family
105
role also predicts work-family conflict. Ego involvement in parenting (but not the
actual hours spent in the parental role) and high levels of family involvement
predict greater negative work-to-family spillover.18
Involvement in either work or family roles increases not only the likelihood
of work-family conflict and negative spillover, but also the likelihood of
positive spillover. In a study of Canadian managers and Canadian business school
alumni, scholars found greater parental role involvement to be associated
with workers' perceptions of greater benefits, both at home and at work. 19
Note, however, that such benefits are not linked to the amount of time spent
in the parenting role. In fact, for this sample of business school alumni, time
spent in active parenting is negatively correlated with positive spillover.2oThus,
positive spillover appears to be a function of the value that workers place
on the parenting role and the quality of the interaction between parents and
children, not the amount of time that workers spend in parenting. This echoes
the mantra that it is the quality not the quantity of the family time that is
important.
Gender
The relationship between gender and spillover is complex and can best
be understood by examining family-to-work spillover and work-to-family
spillover separately. Given that women identify more strongly with the family
role than do men21and spend more time with their children than do men,22we
expect that women also experience more family-to-work spillover than do
men. Empirical evidence regarding this hypothesis is mixed, however. A
longitudinal daily diary study23and two cross-sectional surveys24found that men
experience greater family-to-work spillover than women. However, two other
studies, one of which was also a daily diary study, found that women experience
greater family-to-work spillover than do men, particularly women with young
children.25Three other studies show no difference between men and women
on degree of farnily-to-work interference.26These studies lead to the conclusion
that there is no definitive story linking gender with negative family-to-work
spillover.
,
There is a clearer picture regarding the relationship between gender and workto-family spillover. Four studies find women experience greater levels of negative work-to-family spillover and work interference with family than do men.27
Three studies report no gender differences, and one provides evidence that men
experience more time-based work-to-family conflict than do women.28Given that
women report levels of work commitment similar to men29and that women spend
more time than men engaged in housework and child care, it is not surprising
that, in many studies, women report the highest levels of work- to-family spillover
and interference.
106
levels of
househ01
likely to
less spil
individu.
The contexts in which employees work and live influence their experience of
spillover. Several studies confirm that the more flexible and supportive the
workplace, the less interference and negative spillover there is from work to home.
Specifically, having a supportive supervisof'l and having supportive coworkers31
are related to lower levels of negative work-to- family spillover and conflict. Degree
of autonomy at work,32work variability,33and employee control over decisions at
work34are also related to less work-to-family spillover. Finally, substantial
evidence suggests that the use of flextime is related to lower levels of work-family
report eJi
LinkE
Pare:
To fu
mcorpor
Rather, t
impact'
example
associat{
tively ai
report he
family Ii
Strest
interacti<
days ch3
iorally a
from we
are ofter
Marit
crossov{
expenen
nal stud~
loaded a
Wives' f
feelings
couples
from th{
Ther
and ass(
and wiv
alleviate some of the stresses and conflicts associated with integrating work and
family roles.
On the other hand, some workplace conditions lead to higher levels of negative work-to-family spillover. Not surprisingly, jobs that are high in stress and
conflict are related to higher levels of negative spillover from work to home.36A
heavy workload and time pressures are also related to more work-family conflict
and spillover.37Finally, among men, schedule inflexibility is linked to greater
work-family conflict.38Thus, a demanding, stressful, and inflexible job often
results in the spillover of stresses and frustrations into the home.
Context: Family-Related
Variables
among (
This sUI
toward'
their an
career h
t
I
I
Spillover
rience of
rtive the
to home.
Norkers3!
t. Degree
::isions at
Ibstantial
~k-family
loyees to
IVorkand
of negatress and
lome.36 A
y conflict
0 greater
job often
Ie family
spillover.
e greater
vith child
I-to-work
Ited with
e number
reporting
y, studies
le parents
'port than
ner). This
~r among
also be a
lssociated
cially for
y responort lower
107
from the husbands' job to their wives, but not vice versa.52
The mutual work arrangements of a couple matter as well. Jeffrey Greenhaus
and associates have found that work-family conflict is greatest when husbands
and wives have very different levels of job involvement. And conflict is lowest
among couples in which both members have a high level of job involvement.53
This suggests that if both members of a couple share a common orientation
toward work, they may be less likely to put pressure on one another to change
their arrangements. Problems also arise when one spouse reports that his or her
career has higher priority than his or her spouse's career.54 Finally,Phyllis Moen
108
Meth
and Yan Yu found less conflict and stress among dual-earner couples if both
members work approximately the same full-time hours but neither spouse puts
in long (more than forty-five) hours.55
Couples
and
In our study, we extend the spillover research in several ways. First, we focus
on workers who share a particular context: dual-earner middle-class couples. The
demands of dual careers and professional work in our sample provide the opportunity to examine spillover when the range and intensity of spillover are likely
to be high. We also examine gender differences between men and women who
face quite similar professional demands at work.
Second, most of the conflict and interference and the spillover research has
focused on the negative aspects, the strains, of integrating work and family roles,
whereas we consider positive as well as negative spillover. We also identify the
strategies that dual-earner couples employ to maximize the benefits and minimize
the costs of meshing work and family life.
Third, by taking a couple-level perspective in our spillover analyses, we
examine the linked lives of working men and women, a dimension most scholars have ignored. Couple-level research suggests that spousal behaviors, emotions, and characteristics are likely to playa significant role in the other spouse's
experience of work-family spillover. We therefore use information gathered from
both members of our dual-earner couples to understand the relative levels of
spillover experienced by husbands and wives and how the work experiences of
one member of a couple affect their partners' sense of spillover.
Fourth, we take a life course perspective, which gives us a snapshot of the
work and family career dynamics that occur as men and women move through
different work and family roles. Previous studies that have taken a life stage
approach to understanding spillover have, with few exceptions, only defined two
life stages: parenthood and nonparenthood.56Our seven life-course stages permit
a better understanding of how the benefits and stresses associated with work and
family vary across the life course for both men and women.
Finally, we examine the effect of a variety of work and family characteristics
on spillover and the strategies that people use for managing work and family. We
consider family constraints (such as the time spent in household chores and
dependent care), as well as work-related variables (such as hours of work, workload, and control over scheduling). Of particular interest are family and work
strategies (the use of computer technology, telecommuting, and flextime) for
increasing the efficiency of running the home and/or to increase the flexibility of
work. Each of these characteristics is measured for both self and spouse.
II
i'
i
!
I
II
I
I
Our n
and Care
niques as
least sqm
sis of var
scores ar,
the coupl
are treate
not comp
pared the
specific e
Meas
Spillo
the John
Midlife j
with two
5 equals
spillover.
Life S
two n
(agt
four SI
0-5
est"
hon
one st
em]
t
!
!
[
jL
Varia.
spillover,
include
workers
lady use
and a po
work at
arrange ::
Spillover
es if both
)ouse puts
:, we focus
,uples. The
the oppor. are likely
omen who
search has
mily roles,
ientify the
1minimize
alyses, we
lost scholiors, emo~r spouse's
lered from
~ levels of
~riences of
109
Methods
Our respondents are dual-earner couples (N = 811) in The Cornell Couples
and Careers Study (see app.). We analyze the data using basic descriptive techniques as well as repeated-measures analysis of variance (ANOVA) and ordinary
least squares (OLS) hierarchical regressions. We use a repeated-measures analysis of variance rather than one-way ANOVAs because husband and wife spillover
scores are not independent of one another. With a repeated-measures ANOVA,
the couple is treated as the unit of analysis and husband and wife spillover scores
are treated as separate observations within the same unit. Previous studies have
not compared the spillover scores of husbands and wives; rather, they have compared the scores of employed men and women, without taking into account the
specific employment circumstances of their spouse.
Measures
Spillover We use a shortened version of the spillover scale developed by
the John D. and Catherine T. MacAruthur Foundation Network on Successful
Midlife Aging. The scale assesses four types of spillover, measuring each type
with two questions (scored on a five-point scale where 1 equals all the time and
5 equals never). Items were recoded so that a higher score represented greater
spillover. The items were then averaged to form an index.
Life Stage
two nonparent stages: young nonparents (ages 25-39) and older nonparents
(ages 40+)
;hot of the
ve through
life stage
efined two
.ges permit
1work and
four stages with children in the home: launching, preschool-age children (ages
0-5); early establishment, young school-age children (ages 6-12); later
establishment, adolescents (ages 13-18); and adult children (over 18 in the
home)
racteristics
family. We
;hores and
ork, workand work
:xtime) for
~xibility of
Ise.
one stage that includes parents of children who are grown and have left home:
empty nest5?
110
more ffei
spillover.
mance at
work life
Spillov~
Previo
conflictin
report gre
itive fam
small, the
our coup]
that was t
reporting
sample, v
workday~
days) thai
likely to 1
We de
tive work
wives di~
bands. It
female w
place bet
At the sa]
and spou
We de
positi ve \
pared to
work out
in the ho
between
a general
the chan!
Work-to-Family
Fam.ily-to-Work
versus
Spillover
Our findings on the relationship between work-to-family spillover and familyto-work spillover are consistent with previous studies. Specifically, even in this
middle-class sample, we find that work has more of a negative impact on family
than family has on work (see figure 7.1) for both women and men. Most employees in our study bring more worries and stresses home with them from work
than they take from home to work. To function effectively at the workplace, most
people compartmentalize their family concerns while at work. By contrast, family
members are less effective at shielding their concerns at work from home. Note,
however, that on average respondents report negative work-to-family spillover
somewhere between rarely and sometimes, meaning that for most people in our
sample, negative spillover from work to home (or vice versa) is not a major issue.
Moreover, respondents in our study also report that, for the most part, the benefits of combining work and family outweigh any drawbacks (see figure 7.1). On
average, respondents report that positive spillover from family to work occurs
women
1]
Patte:
Previe
as separa
I
I.
I
Spillover
., schedule
)b requires
111
more frequently (between sometimes and most of the time) than negative
spillover. Thus, family life enhances more than it hinders an individual's performance at work. Further, the couples in our sample report that home life enhances
work life to a greater degree than work life enhances home life.
to perform
Joint scale
Spillover
neasure of
ler responrkda y, the
the timing
1 home on
~r they are
lork. Items
very much
he responc, combin-
ciJy-related
\ the time
members.
nd familyven in this
on family
Most emfrom work
)lace, most
ast, family
Jme. Note,
y spillover
Jple in our
lajOr Issue.
rt, the benre 7.1). On
ark occurs
between
Husbands
and Wives
of Spillover
112
.
L2I
Negative Spillover
Positive Spillover
C)
c
~[t:
...
G)
>
'Q.
en
sometimes (3)
G)
C)
f!G)
rarely (2)
<>
never (1)
men
Family-to-Work
women
Spillover
men
women
Work-ta-Family Spillover
Figure 7.1 Work-to-family and family-to-work spillover among middle-class men and women
in dual-earner couples. Source: The Cornell Couples and Careers Study, 1998 (N ==1642). Significant differences occur between negative work-to-family and negative fami1y-to-work spillover
Figure 7.2
1998 (N == II
(t == 19.0, p < .01); between negative family-to-work spillover and positive family-to-work spillover
(t ==53.6, p < .01); and between men and women on positive work-to-familyspillover(t == 9.3,
p < .01), positive family-to-work spillover (t ==2.5, p < .05),and negativefamily-to-workspillover
(t ==3.4, p < .01).
responden
rdate to one another. To identify the frequency of patterns that could occur
between positive and negative spillover, we place respondents into categories
based on whether they report a high level (average between 4 and 5, meaning
they report experiencing spillover all or most of the time) or a low level (average
between 1 and 2, meaning they report experiencing spillover rarely or never) of
each type of spillover. Workers who report spillover sometimes (an average of 3)
are not included in either of the high or low categories and are therefore excluded
from these analyses. We then place respondents into categories based on their
level of both negative and positive spillover.60
Figure 7.2 shows how our middle-class families are distributed in these categories. Happily, the Family Optimal category is by far the most common pattern
of spillover, with the majority of workers reporting high positive and low negative
family-to-work spillover. Work Segmented is the second most common category.
These workers report very little spillover from work to home, suggesting that they
are able to compartmentalize experiences at work and not let them influence
emotions and behaviors in the home. Approximately 10 percent reported Work
Strain, with high negative and low positive work-to-family spillover. Family
Strain, representing high negative and low positive family-to-work spillover, is a
relatively rare category, representing only I percent of our sample. Most of our
that famil)
Spillove:
!
I
I
I
I
t
l
We ne)
under the
ent compe
family spi
health car,
use respor
a significa
Work Optj
whether ~
terns of Sf
Family
work in h
fall into tJ
members,
faculty ha
bility ma)
workers.
Spillover
113
[] Women
.Men
60
50
-5i
RI
2.a
(.)
...
~'_._~"""""--"---''''""''''"~-'~~~
~40
''-
~~~~'-=~'''~-'''''''''''--'''''''''~''~"_'~M~
30
&. 20
.~,.
T4-'-'-'~~
10
0
nen
,ill over
and women
542). Signif)rk spillover
ork spillover
ver (t = 9.3,
ark spillover
mId occur
categories
, meamng
~l(average
. never) of
~rageof 3)
~excluded
d on their
:hese cateIonpattern
-Nnegative
1category.
g that they
influence
,rted Work
~r. Family
llover, is a
lost of our
"__'M"'_~""M'~"M'M'MM_"
family
optimal
work
optimal
10
family
strain
work
strain
Spillover Pattern
Figure 7.2 Distribution of types of spillover. Source: The Cornell Couples and Careers Study,
1998 (N = 1604, 803 couples).
respondents find that their work lives are enhanced by their family lives. Few feel
that family is a detriment to their work.
114
We also find that a large number of men (67%), but fewer of the women (44%),
who work in the health-care sector report a Family Optimal spillover pattern. This
gender difference may be due to the different types of positions that men and
women in our health-care sector hold. Most of the women are in support or technical health-care positions (e.g., nursing and lab technicians), whereas most of
the men are in management positions. Women in these nonmanagement positions
tend to have fixed schedules, often including shift work, which may be less flexible and more difficult on the family. Employees in the utility and manufacturing
sectors were the least likely to report a Family Optimal pattern of spillover
(between 35% and 54%).
We find no significant differences among sectors on the Work Optimal pattern
of spillover. However, we do find a significant difference among sectors on the
Work Strain category. A greater percentage of men and women in the manufacturing sector report high levels of negative work-to-family spillover and low
levels of positive work-to-family spillover (15% and 14%, respectively, in the
manufacturing sector compared to 5%-9% of men and women in the other
sectors). This may be because both of the organizations in our manufacturing
sector were downsizing their workforces during our survey (see app.). Our focusgroup interviews reveal that professional employees at these organizations are
concerned about the security of their jobs and are also taking on the work and
responsibilities of those who have been laid off. Thus, the context of downsizing, rather than the sector itself, is a plausible explanation for the higher workto-family strain employees in the participating manufacturing firms experience.
However, these firms are part of the competitive and hard-driving global
economy, with (according to the focus-group interviews) demands and uncertainties clearly spilling over into the family lives of their employees.
Linked Lives between Husbands
Wives: Patterns of Symmetry in
Farnily-to-Work Spillover
Figure 7.3
and Careers
first and hi
notice and
Similar
expenenct
whenever
hire some
think is h
and
mommIes
To get;
family-toand two a
below the
Figure 7.3
metry for
groups w,
work spill
reports of
above or
spillover,
Curious about whether there are common patterns of spillover between husbands and wives, we categorized couples' spillover relationships as being either
symmetric, asymmetric, or independent. A symmetric relationship is one in which
both members of a couple experience similar levels of spillover from family to
work. An asymmetric relationship exists when one spouse reports high levels of
negative spillover and his or her spouse reports low negative spillover. A 46-yearold computer technician and mother of two children explains how, even among
couples who share family respo~sibilities, the experience of spillover can be
asymmetric: "... he [her husband] told me last night that he was headed to
Buffalo today and wouldn't be back until tomorrow night and could I meet the
sitter, stay late, all that kind of stuff. The routine has been that I get out the door
,
I
Spillover
1en (44%),
tttern. This
t men and
)rt or techas most of
Itpositions
e less ftexlUfacturing
f spillover
nal pattern
tors on the
~manufac:r and low
'ely, in the
the other
mfacturing
Our focuswtions are
:
work and
f downsiz;her work~xperience.
ing global
and uncer-
..
[]
II
Symmetric:
115
Asymmetric:
40
37
0
C)
Q)
33
IV
.s::.
(.)
IV
W
.5:
III
~
Co
30
26
20
:::I
0
U
-~
IV
10
cQ):
(.)
...
Q)
D..
Negative Spillover
Positive Spillover
Figure 7.3 Distribution of couple-level family-to-work spillover scores. Source: Cornell Couples
and Careers Study, 1998 (N = 803 couples).
first and he waits until the sitter arrives and then he leaves so it's like no advance
notice and all of a sudden this morning I had to stay."
Similarly, a 40-year-old accountant with two school-age children describes her
experiences regarding day-care arrangements: ". . . had someone very good, but
whenever there was a problem, I'd mention it to 'Charlie' and he'd say, 'Oh, just
hire somebody else.' Like you could just go out on the street. Another thing I
think is hilarious is that they have both phone numbers at school, you know,
mommies and daddies. And they never call the daddy."
To get a fuller understanding of the relationship between husbands' and wives'
family-to-work spillover, we divide couples into four categories (two symmetric
and two asymmetric) based on whether each member of the couple is above or
below the median score for men and women on negative family-to-work spillover.
Figure 7.3 illustrates the percentage of couples falling into each category of symmetry for negative and positive family-to-work spillover (the differences between
groups was significant at the .01 level for both positive and negative family-towork spillover). Overall, we find that a majority of couples are symmetric in their
reports of family-to-work spillover. Over one-half of the couples are either both
above or both below the median on positive and negative family-to-work
spillover, meaning that they have a similar level of spillover experiences. When
116
3.2
,,,,,-,,,,.,,
"""~-'o"""'~m""'-'
...
Q)
Work-to-Family Spillover
2.8
~
'0.
2.6
;:
.
.2:.~ 3.8
.s 3.9
"-~-~"""'
"""->~-=""""'X<,<""",-,*,.,"""A---=""""""~_="
Q)
4.1
...
0
z 2.4
2.2
4.2
.:0::
,~
en
~
;::
ca
C)
:u
>
0
'0.
en
LL
~ 3.7
;::
.;n
3.6
3.5
early establishment
(grade school children)
non parents
age < 40
launching
(preschool children)
nonparents
age 40+
late establishment
(teenage children)
nonp:
age <
adult children
at home
life Stage
Figure
7.4 Negative spillover by life stage in middle-class dual-earner couples. Source: 1998
Cornell Couples and Careers Study (men N = 802, women N = 802). Family-to-work spillover difference between life stages: F = 14.9, p < .01. Work-to-family spillover difference between life
stages: F = 12.5, P < .01.
Life
For each form of spillover, negative and positive, family-to-work and workto-family, we perform an analysis of variance to assess whether spillover scores
differ across life stages for both men and women. We are particularly interested
in the impact of parenthood on spillover.
Figure 7.5
Source: 1998
F=14.9,p<
is consisten
of employn
One unexp
married mt
young peol
occupation
Regardi
difference'
tive family
a scale fro
the home,
years. Takt
care for thl
on women
Work-1
Family-to-Work
Spillover
Our analyses reveal that the level of negative family-to-work spillover does vary
significantly by life stage, for both men and women (both significant at the .01
level). Negative family-to-work spillover is relatively high for workers who have
young children in the home and decreases as children get older (see figure 7.4). This
Negativ
the .01 Ie,
similar for
to family i
at which n
Spillover
117
4.3
~
>
0
~
en
4.2 -~~---~--~~~-'--'~'~~'--'-'~--4.1
,~>=<=
-''''-''''''~~''''''''''''''''''
.x:
...
.~,,
~~--<='-"'.~-~~""""
0
3:
~=
.2
.
3.9
.~
LL
3.8
3.7
3.6
--.--
:;::
'iij
~
nest (children
ger at home)
...
3.5
nonparents
age < 40
Husbands'
Positive
Spillover
early establishment
(grade school children)
launching
(preschool children)
late establishment
(teenage children)
nonparents
age 40+
Life Stage
)ource: 1998
spillover difbetween life
ports high
fhese findand wives
e bears the
interested
Figure 7.5 Positive family-to-work spillover across the life course in dual-earner couples.
Source: 1998 Cornell Couples and Careers Study (men N = 802, women N = 802). Life stage
F = 14.9, P < .01. Life stage x gender interaction: F = 3.2, p < .01.
is consistent with the scarcity hypothesis,61which predicts that combining the roles
of employment and parenting is associated with higher levels of negative spillover.
One unexpected finding is the relatively high negative spillover among young
married men and women without children. This may reflect the struggles these
young people face as they simultaneously adjust to both their marriages and their
occupational careers.
Regarding positive family-to-work spillover, we find a statistically significant
difference by life stage for women, but not men (see figure 7.5). Although positive family-to-work spillover scores are generally high (averaging close to 4 on
a scale from 1 to 5), positive spillover is generally lower when children are in
the home, but increases by the time the youngest child reaches his or her teen
years. Taken together, we see that, as children become older and better able to
care for themselves, they have less of a negative and more of a positive impact
on women's and men's work life.
Work-to-Family
r does vary
: at the .01
; who have
e 7.4). This
Spillover
118
into launching families. In our sample of young married couples without children, work evidently intrudes into the home at a relatively high rate. These men
and women have few family commitments and tend not to separate their jobs
from their home lives.
Interestingly, among dual-earner couples, parenthood is associated with less
negative work-to-family spillover compare9 to men and women of comparable
ages without children (young nonparent and older nonparent stages). We speculate that the presence of children may act to buffer parents from the stresses at
work. Parents may be less willing to let the hassles and pressures from work
invade their life at home, making a conscious effort to separate their work life
from their family life. Children may also act to distract parents from the issues
that they face at work. The bottom line, however, is that parenthood appears to
help men and women achieve a greater separation of work from home (but not
vice versa). Overall, negative work-to-family spillover tends to gradually diminish across the life course. As employees settle into their careers, they may learn
how to juggle work and family so that they do not interfere as much with one
another. Life stage is statistically unrelated to positive work-to-family spillover
for both men and women.
Predicting
the same VE
the lower t}
negative m
Roles ir
whose wive
work spillc
likely that 1
hold tasks
bands to pe
at home. J
impact of c
Issue.
Work st
wife work~
tent with p1
high levels
conflict,63I
portive of
in the hon
seems to b
report grea
hours (reci
Spillover
these
WOIT.
greater ne~
Flexiblt
spillover.
,
negativesl
trast, men
only. Amo
without thl
responsibll
when they
stress in tt
Pre die
Several
First, for I
works is a
workload j
husbands'
tive work-
The more tasks and roles that individuals perform in the home, the greater the
negative family-to-work spillover. Specifically, men who spend more time on
household chores, and men and women who spend more hours at work (leaving
fewer available for the home) report higher negative family-to-work spillover. In
,!
Spillover
pillover, we
)ositive and
nily-to-work
steps. First,
e stage. For
~espondents '
~r.Third, we
rk strategies
orientations
of spillover.
: that in the
,les and flex-
the same vein, the more free time that husbands and wives report having at home,
the lower their reports of negative family-to-work spillover (recall we control for
negative mood and life stage).62
Roles in the home also have implications for positive spillover. Husbands
whose wives care for an infirm relative report lower levels of positive family-towork spillover than husbands whose wives do not have this additional role. It is
likely that women who care for an ill relative are less available to perform household tasks and to meet the needs of their immediate family, leaving their husbands to perform some of these tasks, which then detracts from their experiences
at home. As the population ages and elder care becomes more prevalent, the
impact of caregiving on worker and family well-being will become a more acute
Issue.
Work strategies also impact family-to-work spillover. The more hours that a
wife works, the more positive spillover reported by her husband. This is consistent with previous research that shows that when husbands and wives have similar
high levels of commitment to work, husbands report lower levels of work-family
conflict.63It may be that wives who work relatively long hours are more supportive of their husbands' work and more willing to discuss work-related issues
in the home, resulting in more positive home-to-work spillover. The opposite
seems to be true for women. Wives whose husbands work relatively long hours
report greater negative family-to-work spillover. Because of their husband's long
hours (recall that the men in our sample work longer hours than the women),
these women may shoulder a larger burden of home-related tasks, resulting in
greater negative family-to-work spillover.
Flexible work strategies have a paradoxical relationship with family-to-work
spillover. Women who bring work home report higher levels of both positive and
negative spillover. They feel both enhanced and burdened by this strategy. In contrast, men who bring work home report greater positive family-to-work spillover
only. Among men, working at home is a strategy that enhances their work life,
without the corresponding downside that we find with women. Women, who are
responsible for the majority of household tasks, may be overburdened at home
when they also have to perform work-related tasks, leading to greater fatigue and
stress in the workplace, which is not found among men.
Predicting
e greater the
ore time on
ork (leaving
spillover. In
119
Work-to-Family
Spillover
120
number of
report low
relatively b
ity of cases
bear the br
Third, Ii
expenence
careers ane
are more Ii
is particuh
tend to re
among hm
related to I
dren may 1
Fourth,
significant
enhance p<
Flexible w
both hush
related to
telecomml
at home te
Thus, not
but their h
these strat
bilities, th
of spillov(
Note tt
ulation. 0
is, individ
parents in
class coul
have mon
should in<
eralized t<
Our an
different]
family re]
family he
well as th
experiences at home.
A more complicated picture emerges when we examine the relationship
between spillover and workplace strategies that are designed to facilitate the
meshing of work and family. For husbands and wives, bringing work home has
the paradoxical effect of increasing both negative and positive work-to-family
spillover. Further, wives whose husbands bring work home tend to report lower
positive work-to-family spillover. The use of telecommuting, another strategy
touted as an aid for working families, is related to lower levels of positive
spillover among wives and also among wives whose husbands' telecommute. The
use of flextime also has some negative consequences. When husbands use flextime, wives report lower levels of positive work-to-family spillover. Interestingly,
the use of these family-friendly policies appears to have more of a detrimental
effect on the wive's spillover than on the husbands'. In addition, wives appear to
be negatively impacted by their husbands' use of these family-friendly policies,
but husbands are not impacted by their wives' use of the policies.
Although flexible work strategies seem to help workers cope with the multiple demands of work and family, they are not a panacea for the working couple
(especially for the wife) but seem to be a double-edged sword. Strategies such as
bringing work home are simultaneously related to enhancing and detracting from
husbands' and wives' experiences at work and at home. Other flexible work strategies, such as telecommuting and flextime, are related to decreases in positive
spillover for women and for wives whose husbands use these strategies.
Summing
Up
Spillover
eet workers'
relationship
'acilitate the
:k home has.
rk-to-family
report lower
her strategy
of positive
mmute. The
tds use flexnterestingly,
detrimental
es appear to
jly policies,
h the multi'king couple
gies such as
:acting from
work stratein positive
51es.
~r in several
0 show how
5 of positive
mon pattern
dents report
ltive familyKist, it is not
l sample.
)w negative
expenences
ings suggest
:;es III expe-
ructural lag
ion of work
l significant
121
number of couples who have asymmetric spillover. For example, some husbands
report low negative spillover from home to work, whereas their wives experience
relatively high negative spillover. Our interpretation is that in a substantial minority of cases the traditional relationship of asymmetry still exists; that is, the wives
bear the brunt of negative spillover for both partners.
Third, life stage plays a significant role in how much spillover working spouses
experience. Life stage captures the dynamic interplay between individuals' work
careers and family careers. Husbands and wives with the greatest family demands
are more likely to report higher levels of negative family-to-work spillover. This
is particularly true for the wives in our study. Both spouses in later life stages
tend to report less negative work-to-family spillover. Surprisingly, however,
among husbands and wives under the age of fifty, having children in the home is
related to less negative work-to-family spillover. These findings suggest that children may buffer their parents from the stresses of work.
Fourth, home conditions, work conditions, and flexible work strategies are
significant determinants of spillover. Control over work schedules appears to
enhance positive spillover, whereas heavier workloads predicts negative spillover.
Flexible work strategies have complex effects that vary for men and women. For
both husbands and wives, bringing work home acts as a double-edged sword,
related to higher levels of both positive and negative spillover. Wives who
telecommute and wives whose husbands telecommute, use flextime, and/or work
at home tend to experience less positive spillover from their work and home lives.
Thus, not only do women's own use of flexible work policies influence spillover,
but their husband's use of those policies also affects spillover. Therefore, although
these strategies make it easier for couples to combine work and family responsibilities, they may come at a cost to wives' (and a lesser extent to husbands') sense
of spillover.
Note that our findings should be interpreted in the context of the sample population. Our sample does not include single individuals or single parents-that
is, individuals who have opted not to marry or who have more challenges as single
parents in balancing work and family. In addition, although a sample of middleclass couples is expected to have high work demands, these professionals also
have more resources than do lower-income families. Future research, therefore,
should include a wider range of workers to test whether our results can be generalized to individuals in other household and economic circumstances.
Our analysis of positive and negative spillover among dual-career couples at
different life stages provides important insights into the complexities of workfamily relationships. The effects of spillover from family to work and work to
family heavily depend on the characteristics and strategies of the household as
well as the circumstances at work, for individuals and for their spouses.