w-fl.

V «I l 0
3:31? y!!! f?

 

 
    
           

 
     

Ni" '3'

L1: 5 .
=2‘ 4 g i‘
= afiéfi'“! 5!!

‘ , . i‘iW7;!!:'\ I,

'9'
‘ b
I

V
.0.
' :

  
 
  
    
 
 
  
 
 

"E3"

.- “A.

WE.

 

"ma. =

cu

    
  
     
 
 
     

Mu~¢

     
  

  
 
 
 
  

H. 5:
We)?“ ’7' f ‘

      
 
  

, .Es
, .
"IS" a"
,:”- '
P

     
    
  
  
 

       

       
    
 
 

         

     
 
      

1
I .
a

n t
. 'l ,
" z'lj‘ ‘ A . ny'lzl‘
:‘x‘1i‘:"> : 1‘ ‘ .h! 'g. " ”0, ‘3: 1'. A;
WWWWWMWWV Mm wwwflflt
:fifimmgmwgmwmmmmwyfiga (VaWF£&q *3 fig H1géie
if!“ 'gfl32~«;\'9=»1v9;i..'-gi'wa‘ Hi 29’; --‘ WW -1$ _. - '=$ ”WW
{a}. ..é§@‘%-f”3*tist" :zz!:§*.%.:gh.:. g~- ‘ ” ‘I‘S’f’t’fi' '
‘tfili3'1'1; - L!!'d""»;ji;xi .1134". jfi,‘ .‘|;'1_V.‘ : ~ ‘_ "‘1‘ ’ 5. :
fa’ 15133.!“ L‘ . li'ii‘g‘t' “4 ! Uh"! ! nii‘ll . x . - » - - 21”" " Ki :3
qfifiwmr fi§$fiwm%§mwfi m fi'“* wuzé- “!§fi
[:1 .1 ‘t [1‘.. J: '6!»

.ga, '1 w 3*
,E‘gggfi: 'fig‘lfi‘Yrfli .
. g. 3!”. ‘.‘= m“ " “
' 33w: ram?“-
Jihtifia' {W4 g V

V! ‘ i
g ‘ i!!!
25*?" l
.‘ ”.VL I, 1-, f‘ I
1” ‘WWVMEW 4’ !
rm
“2h

    
   
   
    

  
   
  

" l . .
""'r'v’<
mflni

VIM

 
  

         

        
    
    
    
  
  
 
      
    
    
     
       

:' I: t
I; t?,

131$;ng

£3

o.

1!
”r
1

       

‘Ir'
?5
i 1,?

 
 
   
 

 
  
  

.4- .
an

'25. .
a: ..u

   
   

      

  
 
   

 

51‘
:M‘ H
3537?}? Mn 5' '

  
    
   

       

     

 

.‘cmwi

__,.....:.
133?“?-

  
   
    

. .«U "
J.”

c .
u... rm

em
a.

A- 1...:-

    
 

nu- EA...” .
_ . - ‘ , g .
. ‘ _ ‘ . .2. .
. . 0 ‘ ' ' ‘ ' ~ ‘
'2‘» ”gun- ‘ ; ...
M. 0.”. , . .
”V 3:: - ,. - *‘ -
g ' ‘ O r. . _ V ‘
....

 
      

v‘ '
.. .m-
«nu-v

  
 
  

    
     

  
    
 
 

fifgh‘i'
.gfij‘gi1‘2‘5" 3 2%;
#233!!!“ 4, 3:35-
: 3g? T - i1 -
, : _ - . i ' -
«is:

v.
.-

 

'pn
"gr--

 
 

     
    

!
.3}

2V.

5! I}. .g',

- . . [..'

‘ ; “‘0 ‘
:l

.L,

    
 
  
    
 
   
   
 
     
   
   
     

   
  
  

 
 
 
 
      
   
 
   
 
  
   
 

  

 
  
 
 

 
   
   
  
  

   
   
 

    
     

 
   
 
   
 
 
  
       
     
  
 
  

  
  
 
 
 

 

  
    
    
  
    
 
  

  
 

     
    

      
   
 
 
  
    
  
 
  

    
     
 
    
     

       
     
   
 
    
     
          
   
 
     

     

      

      
 
 

   
    

 

  
 
  

     

 

  

w
u

k. LEI-
“ . “‘33P?! \ z"
. _--~._ m‘: 1:1” . _E. - r '
. I'.N': “I“! u .. 1'“! ’EE:' ‘.: _
1. . -."..'-.'. --'.~ 11""..1 .' . '~
“1'21. , fig 3"."gl‘m'fifis .
I g m . . ”a -.t':‘:%:£”4';s:u" . , i.
. f.) ,th .‘ .7 , . 'L . . .ng 1;:Jr‘Ji‘t! 35!! ~€§I Li . . f
,'~ with " " i" :1;;:+‘;-- gnu
- va#‘*vewm gw“' .&%wmg.
; ‘ "NJJJ'HLW .1. .; '1. ' ‘q‘w‘z ‘z‘ N _, ‘ mi.» » '2'“ : {finufilzfiééfikg 2 ff 9“
v W%””>mWH-Www~QWWng WWfiWfiw
n11. 1.5m '=:'.-' g 1 var-WE. ' z.‘wJM§" r' fig!-‘“ “ - 'v i" "~ ‘2.“
. g, «4.. ._ ‘ p75,]! 3 ,-.r- .. .g 5‘ I i W IV _ U. f _ _ 2"
”Mggw§_mWM " $WWxMWM ' ‘mw
V w l» :52u4's..;m;":.may Mm Mi; w 2:: * = -— " ' "a
x‘ A.“ .1“ 1!!! ,.,_.| 13!, .u ’1» \H K. '5 Ln]. . i
'1 - '1: W'hmewg‘imex , gumuw ark!!!“ ’I"I"-::;:’w um .
‘J‘I'H‘hw. um I.» ‘w'gdni .iu»! (I: I: ’50: Jul”: n ‘
WW .r-JW “0‘ g l! “a" ‘5‘!!! W! ‘1 "Wl' ‘ ”*1!!ng '1 ivixg i H t .
‘Z'N‘Jiiw :‘;It‘-' ( 4:.u n . .V
4hhfiwrmk%3fifimflfihfiHflVfiiWW!!!@ 4% "5
""iibflh‘1!]!"utinggflfiuflg. My! 1"4 ’!§35}'€n , , '
v. u: gg-Egg'ufimgg} .l!€}<~i|{1!13}.3>?$ 11.! i 1!!!! nsyx'p . . ,1 ;
W Niv!‘ “' ~!!’H"s!'-‘l-~;r!!~ 3M! E I 'W“! 'I.‘ K "
"‘4' 'H‘".!h”' 4 WW {M- ‘. "iii: W M’ 1 M
MU"! 1‘ " Ii"? !““’!‘§‘!t ? 1 '1' in ! ‘ "if! -
a .1 ‘1 . 'a " 2
.g i ! !t ”“3!!! _ i S

w...—
.. ‘
or.
"M
..-
‘
a—c

un-

.4.-
__..
v.

.’
:33:

Illlllillllllllllllllllllllllllilllll\Illlllllilllilllilli

3 1293 02058 202

This is to certify that the

dissertation entitled

The Development of the Parenting Partnership

Over the Transition to Parenthood

presented by

Laurie A. Van Egeren

has been accepted towards fulfillment
of the requirements for

Ph.D. Psychology

degree in

 

 

; Major priesfi ;

 

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDDUE . DATE DUE DATE DUE

lll ----

031303

MAY 1 1 2004
* 312953

 

 

 

 

 

‘ FEB E 32025.!

M-

 

 

 

 

 

 

 

 

 

 

11/00 animus-m4

 

THE DEVELOPMENT OF THE PARENTING PARTNERSHIP
OVER THE TRANSITION TO PARENTHOOD

By

Laurie A. Van Egeren

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Psychology

1999

ABSTRACT

THE DEVELOPMENT OF THE PARENTING PARTNERSHIP
OVER THE TRANSITION TO PARENTHOOD

By

Laurie A. Van Egeren

The current study is an exploratory and descriptive account of the
development of the parenting partnership from its initiation, the birth of the first
child, through the ensuing 6 months. Married couples (N_ = 101 couples) were
assessed during the third trimester of their first pregnancy and 1, 3, and 6 months
after the child’s birth. Using multiple measures of coparenting and marital quality,
structural equation modeling analyses provided evidence for the specificity of the
parenting partnership construct and the nature of the association between the
coparentingand marital relationships. Additionally, hierarchical linear modeling
was used to identify developmental trajectories for perceptions of the parenting
alliance over the first 6 months. The average parenting alliance trajectories for the
sample as a whole was shown to be quite high and stable; however, fathers were
significantly more satisfied with the parenting partnership than were mothers.
Despite the overall high quality and stability of the perceived coparenting
relationship, factors measured both prior to and after the child’s birth accounted
for individual differences in parenting alliance trajectories. Pre-birth predictors of

the parenting alliance for one or both parents included the marital relationship,

differences in childrearing philosophies, assimilation of the parenthood role, ego
development, reactance, and socioeconomic status. Moreover, the parenting
alliance continued to be predicted by the post-birth marital relationship, violated
expectations of childcare and housework, and parenting efficacy.

In general, the results point to differences in the ways that mothers and
fathers experience and affect each other’s experiences of the parenting
partnership, and particularly emphasize the critical role of the marital relationship
in organizing coparenting. It is suggested that mothers regard coparenting as a
distinctly different part of the relationship than do fathers, and that fathers are
more likely to experience a general relationship quality that encompasses marital,
family, and parent-child interactions. In addition, mothers appear to drive the
systemic development of the parenting partnership, directly affecting fathers’
perceptions, and indirectly affecting their own through interactions with fathers.
The need for a common language and expanded definitions of the parenting

partnership is discussed.

For Tess, the pilot study
and

Theo, the follow-up.

iv

ACKNOWLEDGEMENTS

I would like to express my appreciation to my advisor, Hiram E. Fitzgerald,
who imparted the message, whether intentionally or not, that obstacles are only in
one’s mind; go for it! Without Hi’s casual conviction, I would have attempted far
less. I would also like to thank the members of my committee, Ellen Strommen,
Alexander von Eye, and Susan Frank, for their valuable comments and support
during this long process.

A special thanks goes to the families who participated in this time-
consuming project out of the goodness of their hearts, as well as to Expectant
Parents Organization, Ingham Regional Medical Center, and St. Lawrence
Hospital. In addition, I must recognize the dedication of Adrienne Auten, Sarah
Davis, Alicia Dawson, Emalee Fields, Marcy Hoopingarner, Kristine Kolarovic,
Junna Mita, Kristen Sturman, Rene Teper, Renee Veryserr, Beth White, and
especially Joan Cook, Dyane Hawkins, Rick Lower, Marina Nelson, and Beth
Robinson.

Finally, I am especially grateful to my mother, Nancy Reese, my sister,
Johanna Tousey, and my grandmother, Arline Hertenstein for being there; Mollie
Poorman, fellow traveler, venting partner, and study participant; Renai Vodick,
real-world mom; and of course, my husband, Michael, who always gives me

something to think about.

This research was supported by grant MHS 8461-01 from the National
Institute of Mental Health, an American Association of University Women
American Award Dissertation Fellowship, grants from the Michigan
Psychological Association and Abigail Associates, and the Michigan State

University Department of Psychology.

vi

TABLE OF CONTENTS

LIST OF TABLES ................................................. ix
LIST OF FIGURES ................................................ xii
INTRODUCTION .................................................. 1
REVIEW OF THE LITERATURE ..................................... 9
The Parenting Partnership Construct: Definitions,
Typologies, and Correlates ...................................... 9
Perceptions of the parenting alliance ........................ 9
Self-reported coparenting behaviors ........................ 13
Observed coparenting behaviors ........................... 15
Convergence in parenting ................................ 22
Men, women, and coparenting ............................ 22
Comparison of subjective and objective parenting partnership
evaluations ............................................ 23
Change in the Parenting Alliance ............................... 24
Potential Predictors of the Parenting Alliance ...................... 27
Differences between mothers and fathers .................... 29
Individual parent characteristics ........................... 30
Contextual characteristics ................................ 42
Child characteristics .................................... 49
HYPOTHESES ................................................... 52
METHODS ...................................................... 5 6
Participants ................................................. S6
Characteristics of the sample ............................. 57
Non-participants ....................................... 5 7
Missing data .......................................... 62
Incentives ............................................ 63
Procedure .................................................. 63
Third trimester ......................................... 63
One and three months post-birth ........................... 66
Six months post-birth ................................... 66
Measures ................................................... 68
The parenting partnership ................................ 68
Individual psychological resources ......................... 75

vii

Contextual characteristics ................................ 81

Child characteristics .................................... 89
RESULTS ....................................................... 90
Missing Data Estimation ...................................... 9O
Construct Validity of the Parenting Partnership .................... 91
Analytic strategy ....................................... 91
Results for the hypothesized model ........................ 93

Post hoc analyses ....................................... 95
Parenting Alliance Trajectories ................................ 104
Analytic strategy ...................................... 104

The Baseline Trajectory of the Parenting Alliance ................. 110
Assumptions and reliability of parameter estimates ........... 11 1
Model specification .................................... 11 1
Baseline Couple Parenting Alliance Trajectories .................. 117

Explanatory Model A: Stable Predictors of the Parenting Alliance . . . . 125
Explanatory Models B: Process Variables and Parenting Alliance

Trajectories ................................................ 136
Post hoc analyses ...................................... 145
DISCUSSION ................................................... 151
The Construct Validity of the Parenting Partnership ............... 153
Parenting Alliance Trajectories ................................ 157
Pre-Birth Predictors of the Parenting Alliance .................... 161
The Parenting Alliance in a Changing Context .................... 174
Summary and Implications .................................... 181
Limitations and Future Directions .............................. 186
APPENDIX A: Newsletter ......................................... 193
APPENDIX B: Informed Consent .................................... 196
APPENDIX C: Measures .......................................... 199
APPENDIX D: Preliminary Analyses for Explanatory Model A:
Stable Predictors ................................................. 227
REFERENCES .................................................. 242

viii

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

Table 10:

Table 1 1:

Table 12:

Table 13:

Table 14:

LIST OF TABLES

Characteristics of the Sample ............................ 58
Characteristics of Non-Participants ........................ 61
Schedule for Data Collection ............................ 69
Loadings for Coparenting Behavior PrincipalComponents

Analysis ............................................. 74
Distribution of Ego Levels for Mothers and Fathers .......... 77

Interrater Reliability (Intraclass Correlations) for IDCS Marital
Interaction Ratings ..................................... 83

Loadings for IDCS Marital Interaction Principal Components
Analysis ............................................. 85

Loadings for Ideas About Parenting Measure Components
Analysis ............................................. 87

Intercorrelations, Means, and Standard Deviations for 6-Month
Parenting Partnership and Marital Relationship Variables ...... 94

Fit Indices for Competing Two- and One-Factor Models of the
Parenting Partnership and the Marital Relationship Parameter
Estimates for the Baseline Model ......................... 97

Parameter Estimates for Baseline Model .................. 1 13

Percentage of Parents Whose Parenting Alliance Changed From
1 to 6 Months Post—Birth ............................. 117

Mean Parenting Alliance Scores Over Three Timepoints
for Mothers and Fathers ............................... 118

Reliabilities for Baseline Couple Model Parameter Estimates . . 119

ix

Table 15:

Table 16:

Table 17:

Table 18:

Table 19:
Table 20:
Table 21:

Table 22:

Table 23:
Table 24:
Table 25:
Table 26:
Table 27:

Table 28:

Parameter Estimates for the Baseline Couple Model ......... 120

Percentage of Mothers and Fathers Whose Parenting Alliance

Changed From 1 to 6 Months Post-Birth ............... 121
Intercorrelations and Means Among Pre-Birth Individual
Characteristic Predictor Variables for Mothers (upper halt) and
Fathers (lower half) ................................... 126
Intercorrelations and Means Among Pre-Birth Contextual
Characteristic Predictor Variables for Mothers (upper half) and
Fathers (lower halt) ................................... 127
Correlations Among Mothers’ Pre-Birth Individual and

Contextual Characteristic Predictor Variables .............. 128
Correlations Among Fathers’ Pre-Birth Individual and

Contextual Characteristic Predictor Variables .............. 129
Variables Selected for Further Exploration Through

Preliminary Analyses .................................. 131

Final Explanatory Model A for Stable Pre-Birth Predictors . . . . 133

Means and Standard Deviations for Post-Birth Predictor
Variables by Assessment ............................... 137

Explanatory Model Bl: Relationships with Post-Birth Marital
Adj ustrnent ...... '. .................................. l 3 8

Explanatory Model B2: Relationships with Post-Birth Violated _
Expectations for Division of Childcare .................... 139

Explanatory Model B3: Relationships with Post-Birth Violated
Expectations for Divison of Housework ................... 140

Explanatory Model B4: Relationships with Post-Birth Parenting
Efficacy ............................................ 141

Explanatory Model B5: Relationships with Post-Birth
Perceptions of Infant Fussy Temperament ................. 142

Table Fl:

Table F2:

Table F3:

Table F4:

Table F5:

Table F6:

Table F7:

Table F8:

Table F9:

Table F10:

Table F11:

Table F12:

Table F13:

Preliminary Explanatory Model A: Ego Development ......... 228

Preliminary Explanatory Model A: Positive Childbearing
Motivations .......................................... 229

Preliminary Explanatory Model A: Negative Childbearing
Concerns ............................................ 230

Preliminary Explanatory Model A: Triangulation in

Family of Origin ...................................... 231
Preliminary Explanatory Model A: Positive Parenting in

Family of Origin ...................................... 232
Preliminary Explanatory Model A: Marital Conflict in

Family of Origin ...................................... 233
Preliminary Explanatory Model A: Reactance .............. 234

Preliminary Explanatory Model A: Demographic Contextual
Variables ............................................ 235

Preliminary Explanatory Model A: Self-Reported Marital
Adjustment .......................................... 236

Preliminary Explanatory Model A: Mothers’ and Fathers’ Own
Marital Interactions .................................... 237

Preliminary Explanatory Model A: Partner’s Marital Interactions 238

Preliminary Explanatory Model A: Differences in Childrearing
Philosophy ........................................... 239

Preliminary Explanatory Model A: Infant Sex ............... 240

xi

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Figure 5:
Figure 6:
Figure 7:

Figure 8:

Figure 9:

Figure 10:

Figure 1 1:

Figure 12:

Figure 13:

Figure 14:

LIST OF FIGURES

Hypothesized Model for Multitrait-Multimethod Model Testing
Construct Validity of Parenting Partnership Confirmatory Factor
Analysis of Coparenting and Marital Variables .............. 92

Parameter Estimates for Two-Factor Model of Self-Reported
Parenting Partnership and Marital Relationship .............. 98

Parameter Estimates for Two-F actor Model of Observed
Parenting Partnership and Marital Relationship ............. 100

Parameter Estimates for F our-F actor Model of Dyadic

Relationship ......................................... l 02
Individual Parenting Alliance Trajectories ................. 114
Distribution of Slope Estimates for Baseline Model ......... 115
Distribution of Quadratic Estimates for Baseline Model ...... 116

Individual Parenting Alliance Trajectories of Mothers and
Fathers ............................................. 122

Distribution of Slope Estimates for Mothers and Fathers ...... 123
Distribution of Quadratic Estimates for Mothers and Fathers . . 124

Cross-Lagged Stability Model for Parenting Alliance and
Marital Adjustment for Mothers (above) and Fathers (below) . . 146

Cross-Lagged Stability Model for Parenting Alliance and
Violated Expectations for Childcare for Mothers ............ 147

Cross-Lagged Stability Model for Parenting Alliance and
Partner’s Violated Expectations for Childcare for Mothers . . . . 148

Cross-Lagged Stability Model for Parenting Alliance and
Violated Expectations for Housework for Mothers .......... 149

xii

INTRODUCTION

For the most part, investigations of associations between farme
relationships and child development have concentrated on mother-child or, to a
lesser extent, father-child dyads. Additionally, explorations into the contributions
of the marital relationship to child outcomes are abundant. Recently, however,
inquiries into family interactions have expanded from the study of dyadic
relationships to consideration of family-level variables (McHale & Cowan, 1996).
According to Belsky, Putnam, and Cmic (1996), “Whole-family-dmamics include
events and processes that involve all family members together or a family
subsystem (parent-child or husband-wife) that affects and is affected by the other
subsystems in the family” (p. 46). Following this framework, one subsystem that
comprises a family-level variable is the copaental system, or the relationship
between two parents directed specifically at the children and childrearing. This
coparenting alliance has been described by family systems theorists (Bowen,
1978; Haley, 1976; Minuchin & F ishman, 1981), but has only recently become a
focus for developmental researchers.

Coparenting has proven difficult to operationalize. Assessment of three or
more individuals is likely to necessitate complicated, and probably nonlinear,
models to represent their systemic interrelationships. In addition, coparenting can

be manifested at multiple subjective and behavioral levels, and requires reliable

differentiation from related relationships such as the marriage. In response to this
lack of coherence in definitions of coparenting, McHale (Cowan & McHale,
1996; McHale, 1999) has identified a critical next step among coparenting
researchers to be the development of a common language. At a general level, the
coparenting partnership encompasses the ways that parents work together as a
team, including both facilitative (vs. indifferent/obstructive) behaviors, as well as
perceptions of support and feelings of respect for partners as they implement
parenting functions (Abidin & Brunner, 1995; Cohen & Weissman, 1984; Gable,
Belsky, & Cmic, 1992; Minuchin & F ishman, 1981; Weissman & Cohen, 1985).

Currently, one predominant method examines the subjective feelings of
satisfaction and support in coparenting that Cohen and Weissman (1984) have
termed the parenting alliance (Abidin & Brunner, 1995; Frank et al., 1991). From
their perspective, derived from family systems and psychoanalytic theory and
applied predominantly within the context of a traditional marriage, the parenting
alliance is a relationship between the parenting partners, different from but a
component of the marital relationship. The multifaceted parenting alliance
construct includes investment in the parenting role, esteem for the other parent’s
involvement and respect for his/her parenting decisions, and the desire to
communicate together about parenting issues. These phenomenological aspects of
the coparenting relationship represent the subjective feelings of mutuality in

parenting, a common commitment to the child’s well-being, and a sense of

comprising a united front as the couple fulfills the organizational functions of the
coparenting subsystem. In the absence of an integrated parenting alliance, “parents
proceed on the basis of their own belief systems, whether or not these clash with
those of their partner” (McHale, Kuersten, & Lauretti, 1996, p. 8), thereby
opening the door to resentment and discord.

An alternative conceptualization of the parenting partnership relies on
observations (Camara & Resnick, 1989; Gable, Belsky, & Cmic, 1995; Katz &
Gottman, 1996; Lindahl, Clements, & Marlonan, 1997; McHale, 1995; Schoppe,
Mangelsdorf, Charlton, Dorris, & Frosch, 1999) and self-reports (Ahrons, 1981;
Brody & Flor, 1996; McHale, 1997) of the type and frequency of particular
coparenting behaviors to assess the process and quality of the coparenting
partnership. Backing up the other parent’s directives, competing for the child’s
attention, facilitating the other parent’s efforts to teach the child a task, and
invoking the other parent’s name (warmly or derogatorily) in his/her absence are
examples of the behavioral approach to coparenting measurement.

Only within the past five years has attention been explicitly directed to the
influence of the parenting partnership, apart from the marital relationship, in
family development. Prior studies have revealed that marital satisfaction,
adjustment, and interactions have significant effects on child-centered concerns
such as parenting confidence and parent-child relationships (Belsky, Youngblade,

Rovine, & Volling, 1991; Cox, Tresch-Owen, Lewis, & Henderson, 1989; Erel &

Burman, 1995; Floyd & Zmich, 1991; Goldberg & Easterbrooks, 1984; Kerig,
Cowan, & Cowan, 1993; Stoneman, Brody, & Burke, 1989). Much discussion has
also centered around the association between marital and/or interparental conflict
and child adjustment and behavior problems (Brody, Flor, & Neubaum, 1998;
Brody, Stoneman, & Flor, 1995; Camara & Resnick, 1989; Cummings, 1994;
Davies & Cummings, 1994; Goldberg & Easterbrooks, 1984; Grych & Fincham,
1990; Howes & Markman, 1989; Jouriles, Farris, & McDonald, 1991a; Jouriles,
Pfiffner, & O'Leary, 1988; Lindahl, Clements, & Markman, 1998; Owen & Cox,
1997; Reid & Crisafulli, 1990). However, the emerging literature on the
coparenting partnership substantiates the differentiation of the parenting
relationship from the marital relationship. Further support is found in studies of
divorce, which indicate that a cooperative parenting partnership is related to better
conflict resolution and more successful child outcomes and parent-child
relationships despite the dissolution of the marital relationship (Ahrons, 1981;
Buchanan, Maccoby, & Dombusch, 1991; Camara & Resnick, 1989; Dozier,
Sollie, Stack, & Smith, 1993; Maccoby, Depner, & Mnookin, 1990).

Given the consistent correlations between marital variables and child
adjustment, as well as our incipient understanding of the role of the parenting
partnership, it seems likely that the well-documented influence of the marriage on
children and parenting is at least partially, if not primarily, an indirect relationship

mediated by the parenting alliance. Indeed, Floyd, Gilliom, and Costigan (1998)

provided recent evidence of the mediational role of the parenting alliance in the
relationship between marital quality and parenting perceptions and behaviors, and
coparenting has been shown to be associated with child behavior problems even
after accounting for the influence of marital adjustment (Bearss & Eyberg, 1998;
Brody etal., 1994; McHale & Rasmussen, 1998) and individual mother-child
relations (McHale et al., 1996).Our knowledge of the parenting partnership must
be greatly expanded in light of the far-reaching implications for both intact and
divorced families and for child outcomes.

In explicating developmental processes, systems theorists have encouraged
researchers to focus on particular transition points that are most likely to be
reflected in perturberations in the system of interest (Parke, Omstein, Rieser, &
Zahn-Waxler,- 1994; Smith & Thelen, 1993; Thelen, 1992; Thelen, 1995), and P.
Minuchin (1988) has specifically emphasized the importance of transitional
periods for investigating family subsystem development. One milestone
experienced by the majority of wedded couples, and indeed by most individuals, is
the transition to parenthood. Over this period, a number of systemic changes occur
as spousal relationships are renegotiated, familiar roles are revised, and new roles
are taken on. Theorists who initially addressed the transition to parenthood
postulated that the first child’s birth precipitates a crisis in the existing marital
dyad that forces “a drastic reorganization of statuses, roles, and relationships,”

(Dyer, 1963, p. 196) and from which “the large majority of couples appear to have

made a quite satisfactory recovery” (p. 201; see also LeMasters, 1957). However,
subsequent investigations dismissed the “crisis” model of family change and
reconceptualized the transition as a normative developmental stage, encompassing
a multiplicity of changes, both stressful and rewarding, resulting from the entrance
of a third individual into an established dyad (Hobbs, 1965; Hobbs, 1968; Hobbs
& Cole, 1976; Hobbs & Wimbish, 1977; Jacoby, 1969; Nock, 1981; Rossi, 1968;
Russell, 1974). The pre-birth husband-wife dyad is represented by the marital
relationship; yet the new triad is characterized not only by the marital relationship,
but also by each parent’s individual relationship with the child an_d the new
parenting partnership.

To date, research on the couple’s transition to parenthood has
overwhelmingly concentrated on its effects on marital relationships (e.g., Belsky,
1985; Cowan & Cowan, 1988b; Heinicke & Guthrie, 1996; Isabella & Belsky,
1985; Ruble, Fleming, Hackel, & Stangor, 1988; Wallace & Gotlib, 1990), and, to
some extent, on the development of parenting abilities (F erketich & Mercer, 1995;
Isabella & Belsky, 1985; Mercer & F erketich, 1995; Palkovitz, 1985; Wilkie &
Ames, 1986) and change in adult personality (Antonucci & Mikus, 1988; Deutsch,
Brooks—Gunn, Fleming, Ruble, & Stangor, 1988; Leifer, 1977; Sirignano &
Lachman, 1985). Despite burgeoning interest in the parenting partnership, no
published studies have examined the development of coparenting at the time at

which it initiated, upon the birth of the first child. However, work is progressing;

a symposium at the 1999 meeting of the Society for Research in Child
Development presented preliminary findings from three programs (including that
of the current study) examining pre-birth predictors of coparenting over the
transition to parenthood. For the most part, these studies related parental
representations of relationships measured prior to the child’s birth with post-birth
coparenting behaviors at one point in time. No studies have integrated both
subjective and behavioral aspects of the parenting partnership, nor has there been
any multi-wave longitudinal investigation of the course and correlates of the
parenting partnership over the transition to parenthood. Following Parke and
colleagues’ (1994) recommendation to explore deveIOpmental processes by
implementing short-term longitudinal studies over transitional periods presumably
characterized by rapid change, the aim of the current research was to examine the
development of the parenting alliance, as a newly emergent relationship during the
transition to parenthood, in the initial months after the birth of the first child. Four
questions were addressed in this exploratory and descriptive study:
1. Does evidence exist for the construct validity of the parenting partnership
as a separate dimension from the marital relationship?
2. How does the experience of the parenting partnership change over the
initial months of parenthood?
3. What pre-birth characteristics predict the quality of the parenting

partnership and individual patterns of change over time?

What is the interrelationship between the parenting partnership and other
variables that may change or develop as a result of the transition to

parenthood?

 

REVIEW OF THE LITERATURE

This review of the literature first presents studies that have looked
specifically at the parenting partnership. Subsequently, variables postulated as
likely to predict the development of a new parenting partnership and factors that
may affect change in the parenting partnership are discussed.

The Parenting Partnership Construct: Definitions, Typologies, and Correlates

Investigations of the parenting partnership have typically operationalized
c0parenting in one of four ways: perceptions of the parenting alliance, self-
reported coparenting behaviors, direct observation of coparenting behaviors
within the family context, or measurement of differences between parents on some
parenting variable, with the degree of convergence between parents connoting the
coparenting dimension.

Perceptions of the parenting alliance. Studies support the distinction
between the parenting alliance and the marital relationship. Whereas correlations
between the two constructs have generally reached statistical significance, the
magnitude of the relationship has been low to moderate, with stronger
associations for fathers. Specifically, correlations between parenting alliance and
marital adjustment self-report measures have ranged from .20 to .38 for mothers
and from .44 to .67 for fathers; and correlations between mothers’ and fathers’

reports of the parenting alliance have ranged from .33 to .50 (Abidin & Brunner,

1995; Floyd & Zurich, 1991; Frank, Jacobson, Hole, Justkowski, & Huyck, 1986).
Evidence thus supports the posited linkage of perceptions of the spousal and
coparental relationships, but also confirms that a large percentage of unique
variance in predicting the parenting alliance beyond that attributable to marital
satisfaction remains to be explained.

Allocation of power within the couple may moderate the relationship
between the experience of the parenting alliance and couple interactions. Laub
(1990) found that father-dominant couples reported the most positive perceptions
of the parenting alliance, but manifested the most negative behaviors in a
problem-solving task directed toward a parenting issue. Conversely, mother-
dominant couples reported the most negative perceptions of the parenting alliance,
but their problem-solving outcomes were not significantly different from either
father-dominant or egalitarian couples. It appears plausible that traditional father-
dominant couples may have the strongest need for social desirability, but lack the
communication abilities to effectively work through real problems.

Perceptions of the parenting alliance have been shown to exhibit
theoretically predicted associations with other parenting variables. For example, in
a sample of parents with 4- to 6-year-old children, mothers and fathers who
reported a positive parenting alliance were less likely to report high levels of
parenting stress (Abidin & Brunner, 1995). They were also more likely to endorse

a warm, authoritative parenting style; in contrast, marital satisfaction and

10

parenting style were unrelated. In addition, Floyd and Zmich (1991) reported that
a positive parenting alliance was related to greater parenting confidence and more
positive parent-child interactions among parents of school-aged mentally retarded
and typically developing children. Although these cross-sectional correlational
studies preclude firm conclusions regarding the causal effects of the parenting
alliance, theoretically, the supportive function of the parenting alliance would be
expected to directly decrease stress related to childrearing and indirectly promote
more efficacious feelings and constructive parenting behaviors.

The parenting alliance appears to be particularly salient for fathers’
relationships with their children. Abidin and Brunner (1995) found that for
fathers, the parenting alliance was related to self-reported attachment to the child,
whereas maternal attachment to the child was independent of the parenting
relationship. Furthermore, fathers’, but not mothers’, experiences of the parenting
alliance were related to their child’s adjustment across multiple raters (mothers,
fathers, and teachers). Frank et al. (1991) demonstrated that fathers reporting a
stronger parenting alliance experienced greater strain when their child had
frequently been ill. In contrast, fathers reporting a weaker alliance reported no
relationship between stress and child illness. Perceptions of the parenting alliance
did not relate to mothers’ experiences of stress and child illness. Taken as a whole,
the above results suggest that a father who considers himself to be part of a

successful parenting team is likely to place a higher priority on parenting and to be

11

more involved in child-related functions, potentially with beneficial effects on
child adjustment. Ironically, the Frank et a1. study portends that highly involved
fathers may be particularly vulnerable to stress when faced with child illness,
whether emotionally as they empathize with their child’s discomfort, or due to
logistical complications as they attempt to alter work commitments in the service
of doctor visits and childcare. Conversely, fathers who do not perceive themselves
as part of a functional coparenting dyad are likely to be not only uninvolved
coparents, but less involved with parenting in general, and thereby buffered
against stressful events related to the child.

Recently, McBride and Rane (1998) conducted an exploratory factor
analysis of the Parenting Alliance Inventory (Abidin & Brunner, 1995), producing
three factors: appraisal of the other spouse’s parenting, perceptions of the other
spouse’s confidence in one’s own parenting, and shared philosophy and
perceptions of parenting. Both fathers’ and mothers’ perceptions of the parenting
alliance predicted father involvement; however, for mothers, all three factors were
associated with father involvement, but for fathers, only wives’ confidence in their
(fathers’) parenting skills was related. After accounting for hours of maternal
employment, marital satisfaction accounted for little variance beyond the effects
of parenting alliance perceptions. This study highlights the complex processes that
occur within the family system; in the prediction of father involvement, it is

mothefi’ perceptions and opinions of their husbands’ parenting abilities, as well

12

as of their own parenting confidence, that appear to be critical.

Overall, it appears that for men, more so than for women, a well-
functioning parenting partnership is likely to be accompanied by more positive
parent-child relationships. Nonetheless, one study indicated that mothers’, but not
fathers’, reports of a weaker parenting alliance were related to reports of fairly
severe behavior problems in typically developing children, although not in
mentally retarded children (Floyd & Zmich, 1991). Therefore, while the parenting
alliance appears especially important for fathers, by no means is it incidental for
mothers.

Self-reported coparenting behaviors. Whereas measures of the parenting
alliance tap individuals’ personal feelings about the quality of the parenting
partnership, an alternative method has focused on self-reported behaviors that
serve coparenting functions. Noting that coparenting can occur even in the
absence of the parenting partner, McHale (1997) asked parents to individually rate
how often they manifested particular coparenting behaviors, either in the presence
of the partner or when alone with the child. Cluster analysis was then used to
identify five coparenting dyad types: Disconnected coparents (19% of the
families) were characterized by a lack of warmth and little engagement in
discipline activities; Supportive coparents (31%) promoted family integrity, even
in the partner’s absence (e.g., invoked the partner’s name when alone with the

child) and evidence little conflict or criticism; Average coparents (21%) had

13

average ratings on all behaviors, the exception being higher paternal involvement
in discipline issues; Distressed-Conflicted coparents (9%) did not promote family
integrity and manifested high levels of criticism and conflict; and Passionate
coparents (20%) both promoted family integrity gag reported criticism and
conflict.

In the McHale (1997) study, Supportive couples had the highest levels of
marital satisfaction, Disconnected and Distressed—Conflicted couples the lowest
(with the Distressed-Conflicted group falling into the clinical range), and both
Average and Passionate families reported average marital satisfaction. Thus, a
commitment to representing the family and the coparenting subsystem as an
integrated Whole, rather than a lack of conflict, appeared to distinguish couples
with better relationship functioning overall. Similarly, Kerig (1995) found that
mothers, fathers, and children who rated the spousal subsystem as having a
tendency to triangulate the child were‘more likely to be rated as having a
conflictual marriage by all involved.

Self-reported coparenting behaviors are also associated with child
adjustment. In an ongoing study of family relationships and child outcomes in
rural African-American families, perceptions of more effective coparenting
relationships were linked to whole-family interaction quality and greater self-
regulation in 9- to 12-year-olds, which in turn predicted fewer behavior problems

and better academic achievement (Brody & Flor, 1996). Interestingly, fathers’

14

perceptions of coparenting support were only indirectly related to child self-
regulation through the quality of family interactions; however, mothers’
perceptions of coparenting support, in conjunction with an interparental conflict
variable combined from the reports of both parents, directly predicted self-
regulation. Moreover, in another study, parents and teachers rated 4-year—old
children as lower in internalizing and externalizing problems when the fathers
were committed to promoting family integrity or mothers refrained from
criticizing their spouse in the child’s presence (McHale, 1997). These studies
again underscore the differentiated but interconnected pathways that mothers and
fathers traverse in maintaining the coherence of the family system.

Observed coparenting beh_aviors. Coparenting interactions rated by outside
observers show independent associations from individual parent-child interactions
(Belsky et al., 1996; Lindahl et al., 1997; McHale & Cowan, 1996), indicating that
“whole-family processes are not simply extensions of mother-child and father-
child relationships” (McHale & Cowan, 1996, p. 13). A caveat must be noted: To
date, coding systems have focused on three farmly members participating in a
laboratory-based play interaction; when more than three members have been
present, behaviors have been coded in relation to a target child. Research on
family-level interactions is, therefore, currently limited to the family mad, either
actual or created through the exclusion of other family members.

Naturalistic study of coparenting behaviors has demonstrated that parents

15

were more likely to support one another in coparenting exchanges than to criticize
or undermine each other’s parenting efforts (Gable et al., 1995). However, fathers
were significantly more likely to support mothers in coparenting than mothers
supported fathers, although there were no sex differences in the likelihood of
undermining each others’ efforts. This is not to say that mothers were
unsupportive; in this home setting, mothers were probably also more likely to
initiate opportunities for cooperative parenting, thereby providing more occasions
for fathers to exhibit coparenting behaviors directed toward support functions.
Relatively speaking, fathers were the recipients of a greater proportion of
unsupportive coparenting experiences, and may potentially be more vulnerable to
feeling denigrated by their wives.

Poor coparenting appears to occur primarily in the context of stressors,
with the most frequently investigated stressor being marital discord (e. g., Belsky
& Volling, 1987; McHale, 1997). For example, one set of findings indicates that
marital difficulties and triangulation of children in a family interaction task go
hand in hand for both parents (Lindahl et al., 1997; Lindahl et al., 1998).
However, other studies suggest that fathers’ reactions within marital interactions
may trigger, or at least mark, the formation of different types of alliances. On the
one hand, marital withdrawal by husbands is associated with the formation of
alliances at the expense of either the mother or father (Katz & Gottrnan, 1996;

Paley, Cox, Kanoy, Harter, & Margand, 1999). On the other hand, greater marital

16

1113111151 is associated with more frequent coparenting disagreements and more
intrusive but less positive parenting by fathers, but with more positive parenting
by mothers (Katz & Gottrnan, 1996). Similarly, Belsky et a1. (1991) found that
when feelings of love had decreased for one or both spouses, fathers were more
likely to be intrusive and mothers to be positive when interacting with their child.
Both of these studies suggest the formation of a mother-child alliance, presumably
as mothers attempt to buffer children from the effects of interparental conflict.
Another possibility is that mothers who experience marital dissatisfaction may
invest disproportionately in their role as parents (Brody, Pellegrini, & Sigel, 1986;
McHale, 1995), and be likely to feel that the parenting alliance is weak as well
(Floyd et al., 1998).

Marital problems, and specifically marital hostility (Katz & Gottrnan, 1996)
appears to be related to coparenting difficulties, with hostile marital interactions
mirrored by unsupportive, competitive, uncooperative coparenting interactions. To
complicate matters, McHale (1995) found that poor coparenting, evident only
among maritally distressed couples, was contingent upon the child’s sex. Parents
of boys were likely to evidence hostile and competitive behaviors as they
attempted to engage the infant together, while parents of girls showed a pattern of
maternal involvement and father withdrawal. This may reflect a greater desire on
the part of fathers to continue involvement with their sons in the face of an

unhappy marriage, concurrently exposing the parenting partnership to “spillover”

l7

of resentrnents from the marital relationship. Regardless, most of the work on
coparenting has examined mother-father differences, but has not considered
interactions with the child’s sex; although further complicating an already
labyrinthine endeavor, it is clearly imperative that we make it a habit to investigate
parent-by-child sex interactions in the study of family processes.

Apart from the negative influence of marital problems, chronic day-to-day
stress appears to exacerbate difficulties in the parenting partnership. Belsky,
Cmic, and Gable (1995) observed coparenting interactions in homes of parents
with lS-month-old sons. More negative, angry, and critical coparenting behaviors
were predicted by greater differences between mothers and fathers in personality,
such as extraversion, sensitivity to others, and comfort with intimate relationships,
but not by demographic differences or differences in attitudes toward using
discipline and control with the child. Personality differences were especially
predictive of unsupportive coparenting when parents were experiencing many
hassles (although it should be noted that the researchers did not specifically
investigate whether marital conflict played a role in these difficulties, nor did the
sample include female children).

Additional support for the relation between coparenting and child behavior
is apparent in a follow-up study with the Belsky et al. (1995) sample.
Unsupportive coparenting interactions were associated with changes in inhibition

between ages 1 and 3; specifically, those boys who would be predicted on the

18

basis of early temperament to be more inhibited as preschoolers were actually
substantially less inhibited if their parents demonstrated more unsupportive
coparenting interactions (Belsky et al., 1996). The influence of poor coparenting
was evident even beyond the effect of greater parental insensitivity and
intrusiveness toward the child, which was independently related to declining
inhibition. One might conjecture that unsupportive coparenting is a factor in
impelling these children toward greater externalizing problems. However,
Schoppe et a1. (1999) found that more supportive coparenting and less
unsupportive coparenting measured at age 3 was related to fewer child
externalizing problems as rated by parents and teachers at age 4. It may be that
although these inhibited boys grow less inhibited over time, their greater overall
levels of inhibition may protect them from manifesting significant externalizing
problems as preschoolers in the face of unsupportive individual parenting and
coparenting.

Mechanisms through which the parenting partnership affects parenting
processes and parent-child relations are just beginning to be identified. In a rare
study examining differential coparenting between younger and older siblings,
coparental cooperation, rather than competitiveness, was a critical factor in the
types of control strategies parents used and the likelihood of child compliance
(Gorvine, 1999). When parents were more actively cooperative with one another,

older siblings (M age = 46.9 months), but not younger siblings(M_ age = 16

19

 

months), were more likely to comply immediately with mothers’ directives and
were unlikely to ignore fathers’ directives. Additionally, in more cooperative
couples, fathers relied on gentle guidance to a greater extent than negative control
strategies. Whether these compliance/control relations then mediate child
adjustment remains to be seen, but the Gorvine study provides a first step toward
the specification of elaborated coparenting processes.

Taking a different tack from that of the previous studies, F ivaz-
Depeursinge and colleagues (Fivaz-Depeursinge & Corboz-Wamery, 1999; Fivaz-
Depeursinge, Frascarolo, & Corboz-Wamery, 1996; Frascarolo, Fivaz-
Depeursinge, & Corboz-Wamery, 1999) have developed a program of research
specifically designed to explore farme dynamics at the level of the family rather
than the dyad. They have proposed a hierarchical model of triadic alliances in
which the coparents are the “framing unit” and the infant is the “developing unit.”
Family interactions with young infants (beginning at 3 months) are coded
microanalytically for participation, task attention, role organization, and affective
contact. Four types of progressively unhealthier family alliances have been

hypothesized: in Cooperative alliances, all family members participate and attend

 

to the task, with defined roles and consistent affective communication; in Stressed
alliances, all members participate and have clear roles, but affective contact and
task attention are inconsistent; in Collusive alliances, all family members

participate, but roles are poorly defined, and there is poor task attention and

20

emotional contact; and in Disordered alliances, participation, role organization,
attention, and affective contact are all poor. In a preliminary longitudinal study
with 12 families, the hierarchical triadic model was confirmed when the children
were 3, 6, and 9 months old, although it was most successfully applied at younger
ages. Parents in families with more organized, integrated alliances reported pg
clinical symptoms in their children in an interview conducted when the children
were 4 years old, whereas parents in families with less organized alliances
indicated that their children evidenced behavior problems (Fivaz-Depeursinge et
al., 1996); however, the groups did not differ on behavior problems measured by
the CBCL.

Complicated interrelationships among family members have been revealed
using the above method that would not be evident or necessarily predicted by
studies of dyadic parent-child interaction. During triadic play, even 3-month-old
infants appear particularly engaged by positive affective states of fathers and
subsequently behave more positively toward mothers (Shapiro, 1999; Von
Klitzing, Simoni, & Buergin, 1999a; Von Klitzing, Simoni, & Buergin, 1999b).
On a general level, mothers may arbitrate the degree of contact that fathers have
with their children. However, once fathers are part of the interaction, evidence
increasingly points to the pivotal role that fathers occupy as a catalyst for

behavioral and emotional exchanges for family processes as a whole.

21

 

Convergence in parenting. Finally, a number of studies have centered on an
indirect measure of the parenting partnership, congruence-between mothers’ and
fathers’ parenting beliefs, values and attitudes. Parental disagreements about
discipline have been shown to be related to lower marital satisfaction for parents
of girls, but not boys (Stoneman et al., 1989), and congruent childrearing
philosophies when children were 3 V2 significantly discriminated between couples
who were still married versus those who were divorced 10 years later (Block,
Block, & Morrison, 1981). In addition, when mothers reported more frequent
disagreements related to childrearing issues, they also reported more behavior
problems in their three-year-old boys; in fact, the frequency of childrearing
disagreements was a better predictor of behavior problems than marital adjustment
(J ouriles et al., 1991b). Similarly, child behavior problems were predicted by
mothers’ reports of spousal disagreements over childrearing issues (Snyder, Klein,
Gdowski, Faulstich, & LaCombe, 1988). Congruence as opposed to disagreement
over childrearing values and discipline tactics may serve as a marker of a more
successful parenting partnership; however, congruence should more appropriately
be conceived of as a predictor of coparenting quality, while disagreements over

childrearing issues are likely to be an outcome of weak coparenting alliances.

 

Men, women, and coparentrp' g. The findings described above across
multiple methods of assessing the parenting partnership provide support for other

research suggesting that marriage and parenting are inextricably intertwined for

22

fathers (Belsky, Rovine, & Fish, 1989; Belsky et al., 1991; Brody et al., 1986;
Easterbrooks & Emde, 1988; Feldman, Nash, & Aschenbrenner, 1983; Goldberg
& Easterbrooks, 1984; Howes & Markman, 1989; Lindahl et al., 1997; Owen &
Cox, 1997). The social construction of women as caregivers and nurturers may
encourage women to perceive the parenting role as a given, regardless of the state
of the marriage (Lindahl et al., 1997). Men’s participation in parenting, however,
has traditionally been more discretionary, contingent on successful dyadic
relationships with their wives. Therefore, men who are happier in their
relationships with their wives are more interested in parenting and more satisfied
coparents, whereas men who express dissatisfaction with their marital
relationships tend to be distanced parents and to feel burdened in the coparenting
role. Furthermore, mothers typically spend far more time with children apart from
the father than fathers spend alone with the child; in other words, for fathers, time
with the child is more likely to be whole-family time, whereas mothers participate
in distinct dyadic mother-child and whole-family contexts (Lirn & Clements,
1999). It may be more difficult, therefore, for fathers to make a distinction
between individual parenting, coparenting, and marital relationships, since all tend
to occur in the context of the whole family.

Comparison of subjective and objective parenting partnership evaluations.
The above review highlights a gap in the existing parenting partnership literature,

in that no published research has addressed the relationship between the subjective

23

experience of the parenting relationship and coparenting interactions. However,
Laub (1990), in an unpublished master’s thesis, found an association for mothers,
but not for fathers, between self-reports of the parenting partnership and
observations of the couple’s problem-solving behaviors. Although the problem-
solving discussion. centered around a specific child discipline problem, problem-
solving and conflict resolution are also an integral component of marital
interactions. Marital quality was not accounted for, so that the parenting
relationship and marital relationship may have been confounded. One goal of the
current study was to assess coparenting and marital quality by multiple methods,
and to then attempt to differentiate between the two in order to extend the
evidence supporting the construct validity of the parenting partnership.
Change in the Parenting Alliance

Another aim of the study was to delineate the average trajectory of
perceptions of the parenting alliance in the initial months of parenthood. Within
the constraints of this study (i.e., observations of the parenting alliance at three
timepoints), which reconrrnended the use of a linear rather than curvilinear model,
three possibilities exist, all of which seem equally plausible. On the one hand,
perceptions of the parenting alliance may increase over time, as men and women
negotiate and solidify their respective parenting roles and become better skilled at
cooperative parenting. Two studies of observed family behavior provide some

indirect support for this hypothesis. Fivaz-Depeursinge et al. (1996) determined

24

that the type of triadic alliance families displayed when the child was 3 months of

 

age remained fairly consistent when the child was 6 and 9 months old. However,
of the 12 families, none deteriorated and four, those with more problematic
alliances, ultimately improved. Gable et al. (1995) found that among parents of
15-months-olds assessed again at 21 months, supportive coparenting exchanges
remained stable, but unsupportive exchanges declined by a third. Given the
differences in parenting demands relative to this study (i.e., toddlerhood versus
early infancy), one cannot directly extrapolate from the Gable et al. results.
However, they do hint at flexibility and reorganization of the coparenting
partnership over time. How the behavioral changes in the above studies might be
reflected in parenting alliance perceptions is unclear, but they suggest that greater
satisfaction in coparenting over time is a viable possibility.

Conversely, new parents might be subject to what Kamey and Bradbury
(1997) referred to as “disillusiomnent” in their investigation of marital change in
newlyweds; in short, nascent feelings of warmth and mutuality in the parenting
partnership may decline as the realities of childcare set in, infants become more
active and require more monitoring, and mothers return to work. These factors
may well promote greater stress, which might be manifested by a decrease in the
parenting alliance over time. Indirect support comes from the marital literature in a
comparison of marital satisfaction assessed during pregnancy and at l and 6

months post-birth (Wallace & Gotlib, 1990). Results revealed that marital

25

adjustment peaked at 1 month, followed by a steep decline by 6 months. This
study suggests an initial “honeymoon period” in the marriage may occur
immediately following the birth of the child, which may be reflected in the
parenting alliance.

Finally, it is possible that perceptions of the parenting alliance remain
unchanged over the first six months. From a developmental contextual perspective
(e.g., Lerner, 1993), the parenting alliance is likely to be fairly stable as children
remain within developmental stages (e.g., infancy, toddlerhood, school—age,
adolescence). However, parenting demands, emphases, and challenges depend to
some extent on the child’s developmental stage, which may precipitate
perturberations to the system. According to this model, the parenting partnership
may need renegotiation, therefore, as the child moves between developmental
stages, but should appear consistent during infancy, the period measured during
this study. Both the F ivaz-Depeursinge et a1. (1996) and Gable et al. (1995)
studies discussed in the context of an improving parenting alliance can also be
construed as evidence for the stability of the parenting partnership. In the F ivaz-
Depeursinge investigation, two-thirds of the couples did not change in type of
triadic alliance over the child’s first year; notably, all of the stable couples
displayed the most favorable category of alliance according to the researchers’
classification scheme. In the Gable et al. study, supportive coparenting exchanges

remained stable and unsupportive exchanges decreased. If perceptions of the

26

parenting alliance are more strongly representative of supportive coParenting
behaviors, we would expect parenting alliance ratings to remain stable; however,
if experiences of the parenting partnership are a corollary of unsupportive
coparenting behaviors, increasing satisfaction with the parenting alliance would
be more likely.

Although one purpose of this study is to identify an mp parenting
alliance trajectory, this average can be misleading (Belsky & Rovine, 1990). It is
probable that individual differences will contribute to substantial variation in the
initial level and trajectory of change in the parenting alliance over the first 6
months, as certain factors may systematically contribute to perceptions of the
parenting alliance as more or less successful. The following section reviews
variables that are hypothesized to be meaningful predictors of the development of
the parenting alliance.

Potential Predictors of the Parenting Alliance

Characteristics of individual spouses and of their dyadic relationship pr_io_r
to the child’s birth can be presumed to “set the stage” for the development of a
successfirl or problematic parenting partnership. Adaptive personality
characteristics, attitudes, and expectations, and constructive, organized
relationships exemplify resources that can nurture the development of successful
coparenting relationships, even, perhaps, in the face of marital dissolution.

In addition to pre-birth variables that predict the subsequent development

27

of the parenting alliance, factors that change over time may concurrently affect or
be affected by the changing parenting alliance. Belsky and Hsieh (1998) neatly

illustrated the distinction between what might be termed stable predictors and

 

more m'_oc_e§-oriented variables in a study of marital change in families with a 10-
month-old son who were reassessed when the boys were 27, 36, and 60 months of
age. Presumably stable personality characteristics such as neuroticism,
agreeableness, and extraversion distinguished between marriages that remained
consistently good or poor; however, deteriorating marriages were marked by a
relatively high percentage of unsupportive coparenting exchanges, which the
authors interpreted as dynamic relationship processes. Thus, stable characteristics
were associated with consistent relationship quality, and process characteristics
were associated with relationship change.

For the current study, stable (predominantly pre-birth) and process (post-
birth) factors hypothesized to be associated with individual differences in
parenting alliance experiences were identified. To date, few longitudinal
predictors of the coparenting partnership have been specified. Since the parenting
partnership is presumed to share variance with the marital relationship and
individual parenting abilities, variables that relate to marital adjustment and
individual parenting may also predict the parenting alliance; some may even be
shown to have no relationship with marital and individual adult variables once the

parenting alliance is accounted for. The marital and parenting literature was used,

28

therefore, to identify potential predictors of the parenting alliance.

Differences between mothers and fathers. Men’s and women’s differential
marital and parenting experiences are well documented (e. g., Levy-Shiff, 1994;
Thompson & Walker, 1989), as are their separate experiences of the transition to
parenthood (Cowan et al., 1985; Woollett & Parr, 1997). The first task, therefore,
was to determine whether the typical parenting alliance trajectories were similar or
distinct for mothers and fathers. F irst-time mothers experience more pervasive life
changes and shoulder greater burdens for housework and childcare (Belsky &
Pensky, 1988; Belsky, Spanier, & Rovine, 1983; Cowan, Cowan, Coie, & Coie,
1978; Cowan et al., 1985; Hoffman, 1978), particularly in the first 6 months
(Cowan & Cowan, 1988a), than do fathers. I hypothesized that mothers would
consistently experience less satisfaction with the parenting alliance, but would
also report increasing satisfaction over time in response to the more active
involvement that fathers evidence as infants grow older (Rustia & Abbott, 1993);
conversely, I hypothesized that fathers’ perceptions of the parenting alliance
would be consistently higher than mothers’ and would either remain stable over
time or decrease as more involvement in caretaking was expected. As an analytic
strategy, if, as expected, different trajectories for each sex wept}, specified,
subsequent predictors would be examined separately for men and women.

Beyond sex differences, which might act as a marker for a multitude of

social, cognitive, and/or biological variables, several other predictors were

29

assessed. Belsky (1984) has presented a model of the detenrrinants of parenting
which, although directed toward dyadic parent-child relations, should also be
applicable to the coparenting relationship. Using this model, coparenting quality is
determined by three sources: individual psychological or personality
characteristics, contextual characteristics in which the coparenting partnership is
embedded, and characteristics of the individual child. Belsky also hypothesizes
differential degrees of influence for each parenting determinant, with individual
parent characteristics comprising the most important factor; however, this will not
be explicitly assessed in the current study.

Individual parent characteristics. One determinant of coparenting suggested
through this model is the psychological resources of the individual parent. Stable
pre-birth psychological factors have been shown to relate to more successful
parenting and marital relationships. For example, mothers’ adaptation,
competence, and capacity for positive relationships reported before the birth of
their child related to their perception of general family adjustment and to their
responsiveness to the infant (Heinicke, Diskin, Ramsey-Klee, & Given, 1983;
Heinicke & Guthrie, 1992). Additionally, mothers’ capacity for impulse control
related to less decline in marital adjustment as they became parents (Levy-Shiff,
1994)

These studies suggest that ego development, a construct discussed by

Loevinger (1976; Loevinger & Wessler, 1970), and considered one measure of

30

socioemotional maturity, may contribute to the development of a successful
parenting alliance. Ego development is conceptualized as encompassing one’s
perspective of the self and world and includes cognitive complexity, impulse
control, and interpersonal differentiation. According to the revised edition of the
Washington Sentence Completion Test manual (Hy & Loevinger, 1996), there are
seven codable stages: a) Impulsive, characterized by black and white thinking,
little psychological insight, lack of understanding of cause-and—effect relations,
and an orientation toward immediate satisfaction of desires; b) Self-Protective,
characterized by a desire to avoid punishment, a preoccupation with issues around
control, and simplistic perceptions of self and others; c) Conformist, marked by an
increasing concern with social norms, perceptions of the world in terms of cliches,
and decreasing egocentricity; d) Self-Aware, with a greater ability to conceive of
an inner life, a distinction between the self and the group, and comprehension of
alternative possibilities; e) Conscientious, distinguished by self-evaluated
standards rather than blind adherence to societal rules, an awareness of people as
complex and differentiated, and an appreciation for others’ perspectives; f)
Individualistic, characterized by a sense of individuality, tolerance for differences
in others, and an understanding of psychological causation; and g) M,
demonstrating recognition of autonomy needs in others, respect for the rights of
others to make their own mistakes, and a quest for self-fulfillment. A final stage,

h) Integrated, is extremely rare and thus is not well-studied, but is hypothesized to

31

represent the self-actualized person. Higher levels of ego development have been
linked to a greater sense of responsibility and propensity to nurture others (Hauser,
1978; Helson & Wink, 1987; Westenburg & Block, 1993) , characteristics that
would be important for the development of a sound coparenting partnership. In
addition, ego development assessed prior to the transition to parenthood has been
related to more sensitive parenting for mothers and to less marital conflict and
more positive attitudes toward the infant and the parenting role for fathers (Cox et
al., 1989; Owen & Cox, 1997).

Theoretically, individuals at higher levels of ego development should have
a greater capacity for sustaining sensitive, fulfilling, and organized relationships.
A few investigations have examined the relation between ego development and
marital quality at a single point in time, with rrrixed results. For example, one
study found that among women, but not men, ego development is related to the
ability to develop more open, intimate relationships (White, Houlihan, Costos, &
Speisman, 1990). On the other hand, Nettles and Loevinger (1983) found no
differences in ego level between couples in distressed marriages in comparison to
those in nondistressed marriages. Furthermore, in a cross-sectional study of ego
development in couples over different stages of the life cycle, only among older
couples did higher levels of ego development result in significantly greater
expressions of love (Swensen, Eskew, & Kohlhepp, 1981). To complicate matters,

a group in India found that higher levels of ego development were associated with

32

le_ss marital harmony (Agarwal & Srivastava, 1989), as did the Mills study of
middle-aged American women (Helson & Wink, 1987). Barring possible cultural
variations, presumably, high-level individuals would be more likely to
acknowledge contradictions and problems in the couple relationship than those at
more conforrrrist levels, while still evidencing positive coparenting behaviors.
This opens the possibility that ego development and the subjective experience of
the parenting alliance may relate in a curvilinear manner, with conformist
individuals reporting the highest satisfaction with the parenting alliance.

Little research has addressed the contribution of ego development to
relationship change; one study, however, examined change in marital quality
during a stressful transitional period from relative health to patient status (cancer
diagnosis). Based on retrospective reports, patients at higher levels of ego
development indicated that their marriages had become stronger since their illness
(Fuller & Swensen, 1992). The difference between this transition and the
transition to parenthood, as well as the use of retrospective reports of marital
functioning, make extrapolation of the results from the Fuller study to predictions
of coparenting change extremely tentative. Nonetheless, couples functioning at
higher levels of ego development presumably have a greater likelihood of
tolerating and working through differences in each spouse’s individual parenting
decisions, eventuating in increasing satisfaction with the coparenting experience

over time.

33

A second individual characteristic that may predict the development of the
parenting alliance is the degree to which the parent has assimilated, or internally
incogporated, the idea of being a parent. Mebert (1991) views the transition to
parenthood as a process that, for some couples, begins not upon the delivery of the
child, but much earlier, in some cases even prior to the child’s conception. Parents
who have planned their pregnancy, have strong cultural codes about childrearing
roles, or strongly anticipate parenting may have an “internal working model,” or a
set of expectations that provides a framework for parenting. Parents in this mode
(which Mebert terms “assimilation”) are likely to have discussed and problem-
solved about many of the potentially conflictual issues that parents who are less
cognitively prepared (the “accommodation” mode) may find surprising and
stressful. Operationalizing the assimilation mode as a strong motivation for
parenthood, Mebert (1991) found support for her hypothesis in that more highly
assimilated women had fairly stable perceptions of their marital quality over the
period before and after the baby’s birth, and that their pre-birth expectations of
their infant’s temperament were highly correlated with their post-birth
perceptions. In contrast, women who were less assimilated had much less stable
perceptions of the marriage and the child’s difficulty. It appears that women who
have assimilated the parental role may have a more organized internal working
model or set of preconceived notions that contribute to the experience of

continuity over the transition to parenthood. Interestingly, the pattern of

34

correlations was not particularly different for more assimilated versus less
assimilated husbands and were generally lower than for mothers. This supports the
idea that women have a more central and considered idea of themselves in the
parenting role than do fathers (Bielby & Bielby, 1989; Duxbury & Higgins, 1991;
Hoffman, 1983) and suggests that the process of becoming a parent may have less
of an immediate psychological impact on fathers than mothers.

A third individual variable hypothesized to relate to the couple’s success in
developing an effective parenting partnership is each person’s experiences in the
family of origin. Retrospective reports of positive relationships with parents have
been associated with self-reports of more satisfied marriages and, to a lesser
extent, with marital interactions, with stronger associations for women than for
men (Frosch, Mangelsdorf, & McHale, 1998). Belsky and Isabella (1985)
explored the association between retrospective reports of perceptions of parenting
styles and marital quality in the family of origin and change in the marital
adjustment in couples by the time the first child was 9 months old. The researchers
found that women who experienced more accepting, nurturing relationships with
their parents, and men who evaluated their parents’ marriages as more successful,
reported less decline in their marital quality over the transition to parenthood.
Furthermore, when the family of origin was rated as cold and rejecting and the
parents’ marital quality was judged poor, couples reported the greatest degree of

negative change in their own marriages. Similarly, men and women who described

35

the family of origin as more healthy reported better marital adjustment than those
who rated the family of origin in a negative manner (Lane, Wilcoxon, & Cecil,
1988)

These findings about the effect of the perceived family of origin on the
marital relationship can be extended to the development of the parenting
partnership. Individuals who have had the opportunity to observe a competent
parenting partnership may have a more functional framework with which to foster
their own parenting alliance. Some support was provided by Paley et al. (1999),
who demonstrated that husbands rated as having an insecure attachment status as
adults were more susceptible to problematic coalitions among family members,
but only when they and their spouses tended to engage in escalating patterns of
marital conflict.

One construct derived from the treatment process and outcome literature
but of interest as a predictor of relationship development is the personality trait
reactance. Reactance is described as an internal force which motivates the
individual to maintain and restore personal freedoms (Brehm, 1966); high levels
of reactance, therefore, denote a lack of susceptibility to the influence of others.
Research on personality correlates has delineated a portrait of reactant individuals
as having high levels of anger and impulsivity, and, especially among males,
being judgmental and intolerant of others’ beliefs (Beutler, Sandowicz, Fisher, &

Albanese, 1996; Dowd & Wallbrown, 1993; Dowd, Wallbrown, Sanders, &

36

Yesenosky, 1994; Frank et al., 1998). Findings have been inconsistent with regard
to self-image, with some studies indicating that reactant individuals are more
susceptible to depression (Hong & Faedda, 1996; Joubert, 1990), but others
finding higher levels of self-esteem (Hellman & McMillin, 1997; Hong & Faedda,
1996). People high in trait reactance tend to be more avoidant and distanced in
relationships and less likely to affiliate with others (Beutler et al., 1996; Dowd &
Wallbrown, 1993; Frank et al., 1998), but also more lonely (Joubert, 1990). They
have strong needs for control and independence and are less concerned than most
with social approval (Beutler et al., 1996; Dowd et al., 1994). Despite this
constellation of potentially maladaptive traits, Dowd and Wallbrown (1993)
postulated that highly reactant people might also make effective, persistent
leaders.

Still, given the qualities described above, reactant individuals may have
difficulty compromising or taking spousal requests into consideration. For
example, individuals in violent relationships have been shown to be highly
reactant (Hockenberry & Billingham, 1993). Since most studies in non-clinical
samples find that men are more highly reactant than women (e.g., Hellman &
McMillin, 1997; Hong & Faedda, 1996; see Hong, 1990, and Hong & Page, 1989,
for exceptions), this may play out in the development of the parenting partnership
through greater dissatisfaction on the part of wives as a result of husband’s lack of

c00peration, or through husbands’ irritation at wives’ demands for participation in

37

infant care. The relations are likely to be complex, as illustrated by a study
evaluating the association between reactance and outcome after a marital
enrichment program (Oliver, Mattson, & Moore, 1993). For men, reactance was
not related to outcome status. However, men evidenced both increasing reactance
over time an_d a lack of response to treatment. Women, on the other hand, did
significantly improve on average as a result of the intervention; but highly reactant
women showed less response to treatment. In general, greater reactance is
hypothesized to be predictive of a more problematic parenting alliance.

Ego development level, assimilation of the parenting role, and reactance
are all proposed as stable pre-birth predictors of the parenting alliance. Other
individual characteristics emerge as a result of the child’s birth and, along with the
parenting alliance, may be subject to change over time. One of these process
variables, the individual parent’s self-perception of competence as a parent, might
be expected to reciprocally determine his/her perceptions of the parenting
partnership. Parental efficacy can be conceptualized as both an outcome of
previous coparenting experiences and a predictor of the subsequent parenting
alliance, which would thus suggest a separate but related developmental
trajectory. Mercer and Ferketich (1995) provide support for this contention in a
comparison of inexperienced mothers and experienced mothers over the first 8
months after a child’s birth. Whereas experienced mothers showed no differences

in their parental efficacy over time, first-time mothers moved from feeling less

38

competent at 1 month to more competent by 4 and 8 months. Another study of the
transition to parenthood corroborated these results over the first 6 months
(Woollett & Parr, 1997). In a corollary study with fathers, F erketich and Mercer
(1995) found that the parental competence trajectory of inexperienced fathers did
not differ from that of experienced fathers; both groups of fathers had similar
trajectories to inexperienced mothers (in the Woollett & Parr, 1997, study, first-
time fathers’ parenting confidence remained high and stable). For first-time
mothers and fathers, parenting efficacy was predicted by a sense of internal locus
of control, as well as by better family functioning for fathers (Ferketich & Mercer,
1995; Mercer & Ferketich, 1995).

A well-functioning parenting alliance might be expected to enhance
individual parenting efficacy as each parent receives support for childcare
decisions and is validated in his/her parenting role. Furthermore, the importance
of family harmony for fathers in predicting more successful parenting is once
again highlighted, underscoring the prospect that the parenting alliance is a
particularly critical feature of the parenting process for men. Teti and Gelfand
(1991), in a study of mothers only, determined that maternal efficacy around
infant care mediated the relationship between social-marital support and observed
parenting competence. However, the association between social-marital support
and parenting behavior did approach significance (p < .055) in this sample of 89

mothers. It remains to be investigated whether both a direct and an indirect effect

39

might emerge in a larger sample (as well as if the marital relationship were
assessed as a separate indicator rather than in conjunction with other forms of
social support) and the manner in which these effects might evolve over time.

The nature of the relationship between the parenting partnership and
parenting efficacy over time is an interesting question: Does a successful alliance
elicit greater parenting confidence, does a more efficacious parent promote a more
effective partnership, or is the interrelationship even more complicated? Floyd et
al. (1998) found a nonreciprocal relationship: Among parents of school-age
mentally retarded children, declines in ratings of the parenting alliance predicted
feelings of low parenting competence 18 to 24 months later; however, perceptions
of one’s self as an ineffective parent were unrelated to subsequent satisfaction
with the parenting alliance (but did predict declining marital satisfaction among
mothers). Ironically, fathers who were more satisfied with their marital
relationships also reported feeling less confident as parents in the face of spousal
criticism. Thus, more distanced fathers appeared to be buffered, while involved
fathers were more vulnerable.

A final individual process variable examined in the current study relates to
the division of labor, which has received a fair amount of attention in
investigations of change in the marital relationship over the transition to
parenthood. Regardless of the pre-birth division of labor, roles become

increasingly traditional after the transition to parenthood, with mothers assuming

40

the majority of childcare and household responsibilities (Belsky & Pensky, 1988;
Belsky etal., 1983; Cowan et al., 1978; Cowan et al., 1985; Hoffrrran, 1978;
Kluwer, Heesink, & Van De Vliert, 1997; Moss, Bolland, F oxman, & Owen,
1986; Sanchez & Thomson, 1997). This is especially apparent in the first 6
months (Cowan & Cowan, 198 8a), and occurs even among couples consciously
committed to an egalitarian approach (Cowan et al., 1985). In fact, for mothers,
“fathers’ daily involvement in housekeeping responsibilities and childcare is
considered the central source of ecological support during the post-birth period”
(F eldman, Greenbaum, Mayes, & Erlich, 1997, p. 153). Problematically, both
mothers and fathers perceive their share of household tasks to be greater than their
partners give them credit for (Cowan et al., 1985), thereby setting the stage for
resentment and conflict.

LCM involvement in tasks appears to be a less salient predictor of
declining marital quality than either satisfaction with or expectations about
involvement. For example, pregnant mothers typically overestimate the amount of
childcare assistance they will receive from their spouse, and the more help they
expect, regardless of the amount they actually receive, the more difficult their
adjustment to motherhood (Kalmuss, Davidson, & Cushman, 1992). Additionally,
individuals (especially wives) whose post-birth experiences in a variety of
domains were more negative than expected reported a greater decline in marital

quality and poorer adjustment to parenthood (Belsky, 1985; Kach & McGhee,

41

1982; Kalmuss et al., 1992). Violated expectations of labor, measured even before

the couple marries, are related to higher levels of depression for both mothers and

 

fathers of young children (Strazdins, Galligan, & Scannell, 1997)

Different relations are revealed depending on whether expectations of
childcare or household chores are violated. Cowan and Cowan (1988a) found that
men who were satisfied with the household division of labor reported higher
levels of marital satisfaction, while men who were more satisfied with the
childcare division of labor experienced less parenting stress. Similarly, among
women, violated expectations around the division of housework were associated
with feeling less close to their husbands, but violated expectations around the
division of childcare did not affect their perceptions of their spousal relationship
(Ruble et al., 1988). These authors suggest that childcare may be more
intrinsically rewarding than household tasks; thus, despite the fact that more time
was spent in childcare than expected, it did not elicit the resentment that assuming
an unexpected proportion of housework might. Since the division of labor
changes over at least the first 2 years after the child’s birth (Cowan & Cowan,
1988a), assessment of violated expectations at a single timepoint would be
misleading. Violated expectations of childcare and housework are, therefore,
assessed multiple times after the child’s birth.

Contextual characteristics. A second category of coparenting determinants

proposed is contextual characteristics, which includes demogr_aphic factors (e.g.,

42

age, socioeconomic status) and relational factors (i.e., variables related to the
couple’s existing dyadic relationship into which the child enters). Demogr_aphic
contextual variables that have been shown to relate to marital satisfaction and/or
more positive parenting are older parental age, longer duration of marriage, and
higher socioeconomic status (Maccoby, 1984; Moss et al., 1986; Wright,
Henggeler, & Craig, 1986). However, findings are particularly mixed regarding
the influence of age. Whereas several studies have found a positive relationship
between age and successfirl marital and parenting outcomes, others show no
relation (Frosch et al., 1998) , and Gable et al. (1995) found that older parents
evidenced feyv_e_r supportive coparenting behaviors than younger parents. Why the
results for parental age are inconsistent is unclear, but may be related to the
specific way coparenting is measured, to the developmental stage of the child, or
because of an indirect link with the marriage, which is likely to be of a longer
duration for older parents.

Among wedded couples, the marital relationship constitutes the relational
context most germane to the development of the parenting partnership. Ample
research, described throughout this chapter, attests to the overall link between
marriage and coparenting. It seems plausible that the marital relationship,
measured prior to the child’s birth, would provide a barometer of the couple’s
ability to constructively interact and emotionally sustain one another. Few

longitudinal studies exist to test the prediction that pre-birth marital adjustment

43

relates to the post-birth coparenting partnership. However, those that have been
conducted uphold the association, at least in part. For example, families in which
fathers display marital withdrawal prior to the child’s birth have poorer
interactions and are more prone to forming coalitions during whole-family play
tasks when the children are age two (Paley et al., 1999). Furthermore, Lindahl and
colleagues (Lindahl et al., 1997; Lindahl et al., 1998) found that families were
more likely to present themselves as a warm family unit and to show more
adaptive maternal coparenting behavior five years after the child’s birth when
mothers evidenced better pre-birth subjective and observed marital adjustment.
The same associations did not hold, however, for fathers’ pre-birth marital quality.
Instead, only concurrent paternal marital satisfaction was related to family
outcomes, and only to dyadic father-child interactions. These inconsistent results
necessitate further investigation into the validity of pre-birth marital quality as a
predictor of parenting partnership.

The discussion has heretofore conceived of the marriage as a stable context
from which the parenting alliance unfolds. The marital relationship itself is,
however, a mutable entity, and the transition to parenthood appears to be a
landmark in its development. Both self-reports and observations of marital
adj ustrnent consistently reveal small but significant declines over the transition to
parenthood, particularly for wives (e.g., Belsky, Lang, & Rovine, 1985; Belsky &

Rovine, 1990; Belsky et al., 1991; Cowan & Cowan, 1992; Cowan et al., 1985;

Crohan, 1996; Engfer, 1988; Levy-Shiff, 1994; Miller & Sollie, 1980; Moss etal.,
1986; Ruble et al., 1988; Tomlinson, 1987; Waldron & Routh, 1981; Woollett &
Parr, 1997)‘. However, investigations of patterns of marital relationships over time
reveal that although some marriages do indeed grow worse after the child’s birth,
many maintain the status quo and others actually improve in systematic ways
(Belsky & Rovine, 1990; Lewis, 1988).

These changes in marital quality may depend on a number of factors,
including parental sex and age. Cowan and Cowan (1988b) found that over the
transition to parenthood, women’s marital dissatisfaction peaked at 6 months after
the child’s birth, but husbands reported the greatest declines between 6 to 18
months post-birth. In addition, older couples were more likely to experience
change than younger couples (Cowan & Cowan, 1992), possibly because longer-
standing, fixed patterns of interaction were more vulnerable to disturbance in
response to a new infant (Frosch et al., 1998); younger parents are often
simultaneously in the process of establishing stable patterns of marital interaction

and may in some ways tolerate the addition of a new family member better.

 

1It must be noted, however, that most studies of the transition to parenthood assess
couples in the third trimester and then after the child’s birth, and may only be
tapping change that naturally occurs in all married couples over time. Evidence
from prospective studies of newlyweds has been inconclusive, with some
researchers finding that marital adjustment declines regardless of whether children
are born (Kamey & Bradbury, 1997; Lindahl et al., 1998), and others finding
significantly less decline in childless couples (Belsky & Pensky, 1988; Crohan,
1996)

45

Most likely, the interrelationship between change in the parenting alliance
and change in the marital relationship as a process variable will vary among
families. The parenting alliance may function as a buffer to minimize or even
reverse the post-birth decline in marital adjustment. However, Belsky and Hsieh
( 1 99 8) suggest that marital decline is m by the coparenting relationship, and
specifically by a disproportionate frequency of unsupportive coparenting
behaviors. As they describe, a lack of spousal support in parenting decisions and
disagreement over parenting goals is likely to be reflected in decreasing
experiences of marital satisfaction, although they also note that these influences
are probably reciprocal and perhaps self-perpetuating. Another possibility,
therefore, is that change in the parenting alliance may exacerbate difficulties in the
marital relationship over time. Belsky and Hsieh (1998) measured coparenting
processes at only one point in time. This study aims to expand upon previous
research by assessing both the parenting alliance and the marital relationship as
time-varying variables, and examining associations between the two trajectories.

Another stable relational factor that sets the context for parenting
partnership development is differences in childrearing philosophy, or the degree to
which parents agree or disagree regarding the appropriate way to raise children.
Naturalistic study in the home revealed that in families of toddlers, disagreements
over childrearing attitudes and styles of parenting were the most common catalyst

for critical, unsupportive coparenting exchanges (Gable et al., 1995). Nonetheless,

46

there was no relation in this sample between differences in childrearing
philosophy and frequency of supportive or unsupportive c0parenting (Belsky et
al., 1995). Other investigations have found that greater similarity between parents
regarding childrearing values was not related to perceived marital quality (Kolak,
1999; Russell & Russell, 1994). Many studies, however, do find significant
associations between attitudes about childrearing and family outcomes. For
example, an index of parental agreement around childrearing orientations assessed
in parents of 3-year-olds predicted both child behavior problems at age 5 and
marital dissolution by the time the children were age 13 (Block et al., 1981).
Some evidence suggests that wives’ marital satisfaction may be more
vulnerable to disagreements resulting fi'om divergence in childrearing
philosophies. In one study, disagreement regarding childrearing orientations
significantly contributed to the prediction of women’s marital satisfaction beyond
the influence of parental effectiveness (Deal, Halverson, & Wampler, 1989). In
addition, parental disagreement on childrearing has been linked not only to level,
but to change in wives’ marital adjustment over the transition to parenthood.
Greater disagreement regarding traditional childrearing beliefs was related to
decreases in wives’ marital intimacy ratings between 1 month and 3 years after the
child’s birth; these effects were primarily attributable to the period fiom 15
months on (O'Brien, Peyton, & Roy, 1999). Most likely, as children become more

active and demand greater autonomy and thereby require more limit setting,

47

interparental conflicts would be increasingly likely to occur. The process by which
these conflicts translate into decreasing satisfaction among wives with their
spousal relationships is not clear. It may be that as husbands become increasingly
involved with children as they grow older (Rustia & Abbott, 1993), their opinions
regarding appropriate methods of childrearing impinge on mothers’ heretofore
primary judgments.

Deal et al. (1989) examined the issue of parental agreement in greater
detail: They concluded that parental agreement was fundamentally an additional
index of functional parenting, along with such characteristics as authoritative
attitudes, warmth, and sensitivity. In fact, the mothers and fathers who agreed
appeared to reference a general standard of good parenting, in that they were
effective parents who agreed not only with one another, but also with the other
“good” parents in the sample. According to this line of thinking, these would also
most likely be parents with high ratings of the parenting alliance. In contrast,
mothers and fathers who did not agree were ineffective parents who disagreed
both with their spouses and the other parents; in short, as a group, they appeared
to lack a well—organized reference point from which to base their philosophy of
childrearing. Moreover, and as might be expected, attitudes toward desirable child
behaviors are less controversial than toward undesirable behaviors. Kolak (1999)
found when parents had disparate perceptions of the parenting partnership, they

were also significantly more likely to disagree about what constitutes undesirable

48

child behaviors. Without common criteria for acceptable parenting practices when
faced especially with discipline issues, even if those criteria are less child-centered
than the current social ideal, mothers and fathers who disagree seem exceptionally
susceptible to conflict, criticism, and .lack of support or even attempts thwart
parenting efforts.

Most researchers have investigated the influence of agreement on
childrearing philosophy in parents of toddlers, focusing on issues of authoritative
versus authoritarian control. Among parents of young infants, however, control is
a less consequential concern than are care and protectiveness. Belsky et a1. (1996)
postulate evolutionary differences in male-female attitudes toward protectiveness
of young children. They describe such examples as a father who stimulates a baby
by tossing it into the air while the mother anxiously cautions him to be carefirl, or
a mother who, when faced with a child who has fallen, says “Oh, that really hurts,
doesn’t it,” while the father makes light of the pain or tries to divert the child’s
attention. For the current study, which examines families of infants up to the age
of 6 months, parental differences in protectiveness, as opposed to authoritative or
authoritarian philosophies (Baumrind, 1967) are postulated to be most predictive
of unsuccessful parenting partnerships.

Child characteristics. A third determinant of parenting, according to Belsky
(1984), is comprised of attributes of the child. The current wave of whole-family

research points to the unique influence of the child on dyadic and triadic

49

interaction (Fivaz—Depeursinge et al., 1994; Von Klitzing et al., 1999b;
Vuchinich, Emery, & Cassidy, 1988). The literature is inconclusive regarding the
effect of the child’s s_e_x_ on the parenting partnership. In the few studies of
coparenting, most have either not addressed child gender or have examined only
boys. However, in a study of the subjective experience of the parenting alliance,
Floyd and Zmich (1991) found no differences related to the sex of the child.
Alternatively, McHale (1995) discovered differential patterns of coparenting
relationships among maritally distressed couples according to child gender, with
parents of boys evidencing high levels of competition and parents of girls showing
a pattern of high maternal involvement and low father involvement. One
possibility is that behavioral variations associated with child sex are not reflected
in global self-reports of the parenting alliance. However, particular items on
parenting alliance instruments may be more predictive of these differences, as may
differences between mothers and fathers in parenting alliance ratings.

Finally, child difficult temperament has demonstrated a consistent negative
association with marital and parenting quality, particularly for women (however,
see Tomlinson [1987] for an exception). For example, wives reporting less marital
satisfaction have children with more difficult temperaments as rated by both
mothers and observers (Sheeber & Johnson, 1992; Wright et al., 1986). More
difficult observed infant behaviors have also been related to a decrease in

women’s marital satisfaction over the transition to parenthood (Levy-Shiff, 1994),

50

 

although they were no longer predictive once parental interactions with the infant
were taken into account. Infant temperament has been shown to improve the
ability to discriminate marriages that declined and improved in quality across the
transition to parenthood (Belsky & Rovine, 1990). Lastly, the child’s soothability
at one month of age predicted the mothers’ responsiveness at one year (Heinicke
et al., 1983) and related to higher maternal efficacy (Ahuja et al., 1999). A fussy
baby is likely to heighten stress levels, as well as require parents to make more
judgments about how to care for and soothe the infant. Coparenting may suffer as
stress takes its toll and relatively few rewarding experiences enable the
development of a supportive relationship. Since perceptions of the infant’s
temperament are likely to vary over the first few months (Rothbart, 1981),

probably improving, temperament is conceptualized as a process variable.

51

 

HYPOTHESES

This study addresses four issues: a) The construct validity of the parenting
partnership as a relationship associated with but distinct from the marital
relationship; b) Change in the parenting alliance over the first 6 months after the
first child’s birth; 0) Stable, pre-birth predictors of the developing parenting
alliance; and d) The relationship between the parenting alliance and other time-
varying process variables that may affect and be affected by the developing
parenting alliance. It must be noted that given the variable nature of previous
research on the parenting partnership, particularly regarding issues of change and
development, thesehypotheses must be considered conj ectural and exploratory.

The following hypotheses will be tested.

1. Construct validity will be demonstrated as mother and father self-reports
and observed coparenting behaviors will form one latent variable, and self-
reports of marital satisfaction and observed marital interactions will form a
distinct but related latent variable while simultaneously accounting for the
method effects of maternal self-report, paternal self-report, and
observational ratings.

2. A specific hypothesis regarding the trajectory (increasing, decreasing, or
stable) of the average parenting alliance was not made. It is hypothesized,

however, that there will be a significant amount of variability in the level

52

(e. g., the degree to which the parenting alliance is perceived as stronger or
weaker overall) and change in parenting alliance, denoting substantial
individual differences. Therefore, the average growth curve will provide a
summary of the early parenting alliance trajectory (and will enable
comparison to other studies examining group means). Parent sex is
predicted to contribute to individual differences in parenting alliance level
and change, with mothers expected to rate the parenting alliance as poorer,
but as increasing over time, and fathers expected to rate the parenting
alliance as better, but as decreasing or remaining stable over time. The
following hypotheses will examine additional predictors of individual level
and change in the parenting partnership for mothers and fathers separately,
unless no parental sex differences emerge.

Stable individual, contextual, and child characteristics will predict
individual differences in the development of the parenting alliance.
Specifically, the development of a stronger parenting alliance (higher mean

level and/or increasing rate of change) will be predicted by:

a. Individual characteristics of the mother and father:
i. Higher levels of ego development
ii. Greater assirrrilation of the parental role

iii. Retrospective reports of a stronger parenting partnership in

one’s family of origin

53

iv. Lower levels of reactance

Contextual characteristics:

i. Older age and higher socioeconomic status

ii. Greater pre-birth marital satisfaction

iii. More positive and fewer negative marital interactions

iv. More agreement on childrearing philosophies, particularly
toward child protectiveness

Child characteristics:

i. Child sex will be examined in relation to the parenting

alliance, but no specific relation is hypothesized

Level and change in the parenting alliance will relate in meaningful ways

to level and change in individual, contextual, and child process variables

that vary over time:

a.

Level and change in the parenting alliance and marital adjustment
will be positively related; thus, a positive relationship will indicate
that they are both strong or both wead and/or change in the same
direction.

Level and change in the parenting alliance and violated expectations
for the division of labor will be negatively related; that is, a better
parenting alliance will be marked by less violated expectations, and

an increase in parenting alliance will be accompanied by a decrease

54

in violated expectations. This is expected to hold particularly for
violated expectations around childcare, but not necessarily for
violated expectations around housework.

Level and change in the parenting alliance and parenting efficacy
will be positively related.

Level and change in the parenting alliance and perceptions of

difficult infant temperament will be negatively related.

55

 

METHODS

Participants

Participants were 101 couples recruited from childbirth preparation classes2
in central Michigan over the period from September 1997 to October 1998, and
their first-born infants (49 boys, 52 girls). Approximately 51% of couples
recruited agreed to participate’. Previous research indicates that parenting
variables can be affected differently for delayed and premature children as well as
for children of multiple births (Floyd & Zurich, 1991; Harrison, 1990; Lytton,
Watts, & Dunn, 1987) relative to singletons and typically developing children.

Couples were eligible, therefore, if: a) pregnant with their first child and neither

 

partner had other children; b) married (M = 3.53 years, SD = 2.69, range = 2
weeks to 13 years); c) it was not a multiple birth; (1) the pregnancy was normal and
healthy. All infants were healthy and nearly all were firll-term; two infants were
born approximately one month early, but weighed over 2500 grams at birth and

did not require special care.

 

2Recruitment fliers were also placed in most obstetricians’ offices in the
Lansing/East Lansing area, as well as at the W1C office and the Department of
Public Health, without success. Two couples entered the study after reading a
newspaper editorial about the project research.

3Demographic information was also requested from couples who were eligible to
participate but declined to do so. Some couples may not have completed this
information, inflating acceptance rates. However, most or all eligible, non-
participating couples appeared to provide demographic data.

56

Characteristics of the 8%. Table 1 details the demographic
characteristics of the sample. Both men and women ranged in age from 19 to 42
years (M = 30.55 years for men, __S_12 = 4.90; M = 28.97, S1; = 4.58 for women). A
wide range of educational levels were represented, with some parents having not
finished high school and others holding advanced degrees; for both mothers and
fathers, the median educational level was a 4-year college degree; however, more
than a third of the sample had at least some graduate training. The average
occupational level, measured using the Duncan index (Stevens & F eatherrnan,
1981), denoted the professional/technical occupations. About 54% of the sample
reported a family income (average yearly gross) of between $30,000 and $69,999,
but income ranged from less than $10,000 to over $100,000. Ninety percent of the
sample was Caucasian and 10% were ethnic minorities. In short, this can generally
be described as a predominantly white, middle-class, well-educated sample. Most
couples indicated that they had planned their pregnancies, but about 8% reported
that the pregnancy was unplanned and 20% described it as not specifically
planned but not avoided.

Non-participants. Information on age, education, and income was collected
from individuals who declined to participate and who completed a brief
demographic questionnaire (168 wives, 131 husbands, sometimes from the same
couple but more often from different couples—in retrospect it became apparent that

often one spouse was designated to complete the form) to determine whether

57

Table 1

Characteristics of the Sample

 

 

 

 

 

 

Characteristic Mother
M SD Range Range
Age 28.97 4.58 19-42 19-44
Occupation’ 52.61 24.56 0-88.4 56.59 22.53 0-88.4
Years married 3.53 2.69 2 weeks-
(mother report) 13 years
Frequency % Frequency %

Education

Less than high school 2 2.0

High school graduate 7 6.9 9 8.9

Some college 22 21.8 28 27.7

Bachelor’s degree 34 33.7 27 26.7

Some graduate 12 11.9 13 12.9

school

Advanced degree 26 25.7 22 21.8
Ethnicity

White, non-Latino 89 88.1 92 91.1

African American 5 5.0 4 4.0

Latino 2 2.0 2 2.0

Asian 5 5.0 1 1.0

Otherb 2 1.0

(Cont)

58

Table 1 (cont)

 

 

 

 

 

 

Characteristic Frequency %
Income (father report)
Less than $10,000 2 2.0
$10,000-$29,999 23 22.8
$30,000-$49,999 34 33.7
$50,000-$69,999 21 20.8
$70,000-$99,999 13 12.8
$100,000 and up 8 7.9
Mother Father
Plannedness of pregnancy
Planned 73 72.3 72 71 .3
Unplanned 8 7.9 9 8.9
Unplanned, but not 20 19.8 20 19.8
avoided
Males Females
Child sex 49 52

 

aOccupation measured using Duncan scores.

bOther includes Native American and Palestinian.

59

differences existed between those who volunteered and those who did not.
Characteristics of non-participants are presented in Table 2. Men ranged in age
from 20 to 44 years (M = 29.40, _S_D_ = 4.63), and women ranged in age from 17 to
43 years (M = 28.01, SI; = 4.52). Median educational level for women was some
college attendance and for men was split about equally between some college
attendance and a bachelor’s degree; only about 12% of non-participants reported
graduate training. Family income was in the $30,000 to $59,999 range for about
54% of the sample. Approximately 7% of women and 12% of men were ethnic
minorities.

Independent groups t-tests were used to determine whether participants
differed from non-participants in age, education and income. Age and income
were similar between the two groups; however, both women and men who
participated were significantly more educated than those who did not (for sample
as a whole, t = 6.89, p < .0001). In addition, among women, the distribution of
ethnic groups differed, with Asian females significantly more likely to participate
(X2 (4, IE = 254) = 14.32, p < .001); in fact, all five Asian females recruited joined
the study. Recruitment within a university community appears to have resulted in
a sample that is more educated than the population at large, and although income
did not differ between the groups, this is most likely due to the student status of
several participants. This is the first study in the coparenting and transition to

parenthood literatures to explicitly investigate the representativeness of

60

Table 2
Characteristics of Non-Participa_n_ts

 

 

 

 

 

 

 

Characteristic Mother Father
M. S_D Range M SD Range
Age 28.01 4.52 17-43 29.40 4.63 20-44
Frequency % Frequency %

Education

Less than high school 1 .6 2 1.4

High school graduate 18 10.7 20 14.5

Some college 70 41.7 48 34.8

Bachelor’s degree 59 35.1 51 37.0

Some graduate 11 6.5 10 7.2

school

Advanced degree 9 5.4 7 5.1
Ethnicity

Anglo American 141 93.4 106 88.3

African American 4 2.4 7 5.8

Latino 2 1.3 2 1.7

Asian

Other 4 2.6 5 4.1
Income Frequency %

Less than $10,000 5 3.7

$10,000-$29,999 33 24.3

$30,000—$49,999 34 25.0

$50,000-$69,999 39 28.7

$70,000-$99,999 17 12.5

$100,000 and up 8 5.9

 

61

participants, and, as suspected, reliance on volunteers increased the risk of bias
toward more affluent and educated, and possibly more motivated and better-
functioning, couples. Despite such limitations, similar samples (e.g., Belsky &
Rovine, 1990; Cowan et al., 1985) have been sufficiently varied so as to allow for
the investigation of individual differences. Nonetheless, the potential biases
inherent in this sample must be noted in considering the generalizability of the
study results.

Missing data. Of the couples recruited, nine agreed to participate but either
had their babies prior to the first assessment (3 = 8) or lost the infant to
intrauterine death (p = 1). Two additional couples completed the initial assessment
and subsequently dropped out of the study, citing time constraints or discomfort
with the questions. The sample size of 101 is based on all remaining couples.
Camera malfirnction precluded coding four of the pre-birth interviews and one of
the post-birth interviews. Six families relocated during the study; all completed
the questionnaire measures and returned them by mail, but were unavailable for
the 6-month videotaped interview and coparenting play session. In addition, one
couple separated and reunited during the course of the study; whereas the mother
completed the 6-month questionnaire assessment, the father did not, nor did the
couple complete the 6-month observations. The sample size is slightly smaller,
therefore, for analyses that include observational ratings. Missing data is

otherwise minimal. Two couples and one father (described above) did not

62

complete one assessment, all at different timepoints.

Incentives. In order to encourage participation, a congratulations card,
including the newspaper birth announcement if available, was sent to each family.
When the infants were 5 months old, a newsletter that described the progress of
the study, introduced the research team members, and provided information on
infant development and parenting was mailed to each family (see Appendix A). At
the end of the study, couples were presented with an infant t-shirt and a copy of
their videotaped interviews and play session. Twelve families, drawn at random,
received either a $50 or $100 savings bond in the child’s name.

Procedure

Assessments were conducted during the third trimester of the pregnancy,
and 1, 3, and 6 months after the child’s birth. Measures included questionnaires
and coded videotaped observations of marital and coparenting interactions.

Third trimester. An interviewer (either the author or one of three trained
senior undergraduate research assistants) visited each home during the third
trimester, typically 4-6 weeks before the baby’s due date. The interviewer
reviewed the informed consent (see Appendix B), answered any questions, and
obtained each spouse’s signature. Subsequently, three questionnaires most likely
to be affected by the interview tOpics (marital adjustment, division of labor, and
motivations around parenthood) were administered so that responses would not be

contaminated by couple discussion of these issues. Couples were asked to

63

complete the questionnaires (randomly ordered) separately and not to discuss their

answers with each other. The interviewer was available to answer questions and

read items if necessary.

Next, a semi-structured interview with the couple was conducted, using a
procedure described in McHale (1995). This interview, typically 30-40 minutes
long, was videotaped and served as the sample of behavior from which marital
interaction ratings were coded; ratings were based on the interactions between the
couple during the interview, not the content of the couple’s response to the
interview questions. To begin, couples were asked a standard question (I ’d like
you to describe how you anticipate the baby will change your life) and instructed
to discuss the answer between themselves. The interview was subsequently guided
by the following questions based on an interview developed by Chavez et al.
(1987) which was used in McHale (1995) for couples with children and modified
for this study to be relevant to pre-birth couples:

1. Mother’s return to work (scripted questions: How do you plan to deal with
that? At what point did you make the decision to return or not return to
work? What issues came up for each of you during the decision-making
process?)

2. Child care (scripted questions: Do you plan to put your child in child care?
What are the arrangements? "420 made them? What issues and concerns

have each of you had in deciding about child care? How happy are you

64

with this decision?)

3. Role arrangements (scripted questions: How will you decide who attends to
the baby when you ’re home together? What problems might come up
around the division of responsibilities ?)

4. Time for the self (scripted questions: How do you plan to get enough time
for yourselves after the baby is born? Do you anticipate any difliculty?
What issues and concerns have each of you had about how things might
change around this?)

5. Time for the couple relationship (scripted questions: How do you anticipate
the baby changing your relationship as a couple? How will you try to
spend time together? What do each of you see as getting in the way of your
spending time together?)

Two types of probes were used. The first explored conflicts in responses between

partners (e.g., Do you see this the same way he/she does ?), and the second

examined decision-making processes (e. g., Are the two of you able to discuss this?

What usually happens when this topic comes up ?). The goal of the interview was

to obtain a representative sample of behavior within the context of the marital

dyad; probes were, therefore, designed to encourage couples to discuss issues
together rather than simply answer the interviewer. Although all topics were
addressed, the order and amount of time spent on a specific topic varied with each

couple depending on their responses.

65

After the interview, a second set of questionnaires, measuring ego
development, perceptions of coparenting in the family of origin, reactance, and
childrearing philosophy, was given to each partner and completed during the visit.
The entire visit typically took 1% to 2 hours.

One and three months post-birth. For the Time 2 and Time 3 assessments,
undergraduate research assistants visited participants with questionnaire packets.
Data gathered at this time included information on the perceptions of the
parenting alliance, the marital relationship, the division of household labor and
childcare, parenting efficacy, and perceptions of infant temperament. The research
assistant cared for the child while the parents completed the questionnaires,
usually within 20-45 rrrinutes.

Six months post-birth. An interviewer, accompanied by a research assistant,
made the final home visit at a time when all three family members were present.
Questionnaires assessing perceptions of the parenting alliance, the marital
relationship, the division of household labor and childcare, parenting efficacy,
perceptions of infant temperament, and reactance were mailed to the couple one
week prior to the visit so that they could be completed before and returned during
the assessment.

In most cases, unless the child was asleep when the research team arrived,
the first activity during the home visit was a 10-minute videotaped play

interaction, modeled on McHale’s (1995) protocol for 9-month-olds and modified

66

for use with 6-month-olds. The infant was placed in between the two parents
(infants who were not yet able to independently sit up were propped by pillows),
and a toy, changed every few rrrinutes, was placed on the ground equidistant from
both parents. Toys were chosen to be interesting but slightly above the
developmental level of an average 6-month-old infant. First, a soft book was
presented, and parents were instructed to “try to teach __ how to turn the pages”
(3 min). Second, a crayon and paper was given, and parents were instructed to “try
to teach __ how to scribble on the paper” (3 min). Finally, two toys, a set of
keys with different activities on each key, and a “clacker” toy, were presented, and
parents were instructed to “play with these toys any way you want” (4 min).
Parents received no further direction during the interaction.

Next, couples completed their questionnaire packets if they had not already
done so. The post-birth counterpart of the Time 1 marital interaction was then
conducted. While the research assistant cared for the infant, couples were
videotaped participating in the original interview from McHale (1995) with a
modified procedure. Couples were asked a standard question (I’d like you to
describe a typical week at home since the birth of the baby) and asked to discuss
the question between themselves before answering the interview. The interview
was subsequently guided by the following questions (the probes remain the same):

1. Mother’s return to work (scripted questions: How is that working for the

two of you? At what point did you make the decision to return or not return

67

to work? What issues came up for each of you during the decision-making
process?)

Child care (scripted questions: Is your child currently, or will he or she
soon be, in child care? What are the arrangements? Who made them?
What issues and concerns have each of you had in deciding about child
care? How happy are you with how things are going?)

Role arrangements (scripted questions: How do you decide who attends to
the baby when you ’re home together? What problems have come up
around the division of responsibilities ?)

Time for the self (scripted questions: Are each of you getting enough time
for yourselves these days? VWzat things, realistically, would need to change
in order for you to get the time you need?)

Time for the couple relationship (scripted questions: Are you getting any
regular time together as a couple these days? What do each of you see as

getting in the way of your spending more time together?)

At the end of the visit, parents were presented with t-shirts for their infants. The

entire visit took approximately 1% hours on average.

The schedule for data collection is shown in Table 3.

Measures

All measures are presented in Appendix C.

The parenting partnership. The parenting partnership measures assess both

68

Table 3

Schedule for Data Collection

 

 

Variables Third trimester One month Three months Six months
Parenting 0 Parenting - Parenting - Parenting
relationship alliance alliance alliance

° Coparenting
interactions
Individual - Ego ' Parenting 0 Parenting 0 Parenting
characteristics development efficacy efficacy efiicacy
- Assimilation - Division of - Division of 0 Division of
- Family of labor labor labor
origin
- Reactance
° Division of
labor
Demographic - Age
contextual 0 Education
variables - Occupation
° Income
- Length of
marriage
Relational - Marital 0 Marital 0 Marital - Marital
contextual adjustment adjustment adjustment adj ustrnent
variables - Marital - Marital
interactions interactions
0 Differences
in
childrearing
philosophy
Child ° Perceptions 0 Perceptions - Perceptions
characteristics of infant of infant of infant
temperament temperament temperament
0 Infant sex

 

69

the individual subjective experience of the parenting alliance between the two
partners and the behavioral interactions between the partners as they engage in
parenting the child.

1. Parenting alliance. The General Alliance subscale of the Family
Experiences Questionnaire (Frank, Jacobson, & Avery, 1988) was given at 1, 3,
and 6 months post-birth to assess each parent’s perceptions of the parenting
partnership. Thirty-one items are rated from 1 (strongly disagree) to 4‘(strong1y
agree). The scale assesses the degree to which the parent respects the partner’s
parenting judgments, feels supported by the partner in parenting decisions, is
satisfied with the partner’s contribution to parenting, and feels that parenting has
increased the degree of cohesiveness in the relationship. Internal reliability for the
parenting alliance scale was high, with the average coefficient alpha over the three
administrations at .93 for fathers and .94 for mothers. Frank and Tuer (1988)
reported a three-week test-retest reliability coefficient of .86 for a sample of
parents of older adolescents. Construct validity for the scale has been
demonstrated through associations with parental reports of greater father
participation in child care (Pirsch, 1990), better interparental interactions and
parent-child interactions (Floyd et al., 1998; Laub, 1990), decreased parenting
stress (Frank et al., 1991) and more parenting competence (Floyd et al., 1998), and
less severe ratings of child behavior problems (Floyd & Zmich, 1991).

2. Coparenting behaviors. The coding system developed by McHale

70

(1995) was used to assess coparenting interactions videotaped at 6 months post-
birth. Research assistants received 25 hours of training in the form of reviewing
videotapes and discussing the coding system with the primary investigator, then
practicing coding with protocol videotapes of family interactions. Raters were
blind to all other information about the participants.

Videotapes were coded for five dyadic dimensions and two individual
parent-child dimensions. A manual described behaviors characteristic of each
value for all dimensions. For the dyadic dimensions, a five-point Likert-type scale
was anchored at 1 and 5, with higher numbers connoting more conflictual,
indifferent, or parent-centered interactions. The scales included: a) beneficial
Mon (actively interfering with or sabotaging the partner’s initiatives and
interventions with the child); b) W (working as a team,
actively supporting the partner’s interventions with the child); e) W
(exchanging critical, sarcastic, and/or hostility-tinged remarks during play with the
child); (1) W (level of warm, positive affect shown toward
the partner during play with the child); and e) 911W (pace
and flow of session dictated exclusively by baby as opposed to pace and flow of
session dictated exclusively by parents.

The two individual parent-child dimensions were rated on 7-point scales: f)
Wt, assessing the parent’s participation in play and reflecting withdrawal

versus overinvolvement, ranged from 1 (extremely involved, possibly enmeshed,

71

child’s performance appears to be reflection of parent’s self) to 7 (completely
disengaged); and g) warmth toward child.which had a scale of 1 (extremely
expressive; use of speech, touch, active eye contact to convey warmth may appear
gushy) to 7 (stiff, wooden parent who fails to smile or speak in warm tones). In
order to avoid bias generated by comparing one partner to the other, three raters
coded each tape; one rated the five dyadic dimensions, one rated the two mother-
child dimensions, and one rated the two father-child dimensions (the two latter
coders alternated rating mothers and fathers). Twenty videotapes were coded by a
second rater in order to assess interrater reliability. Intraclass correlations were .81
for behavioral competition (M = 2.93, SD = .99), .56 for cooperation between
partners (M = 2.86, S1; = .97), .64 for verbal sparring (M = 1.99, S_D = 1.27), .80
for warmth between parents (M = 3.09, SD = .94), .62 for child-centeredness (M =
3.42, _S_I_)_ = 1.00), .82 for mother investment (M = 3.11, S_D = .91) .74 for father
investment (M = 3.49, S_D = .95), .80 for mother warmth (M = 2.90, S]; = 1.05),
and .74 for father warmth (M = 3.26, S_D = 1.03). All scale points were used for
all scales except the lowest extreme (7) for mother and father warmth and father
investment, and the two lowest extremes (5 and 6) for mother investment.

Two discrepancy scores, investment discrepancy and warmth discrepancy,
were calculated using the absolute difference between the mother’s and father’s
ratings. Greater discrepancy scores thus tap withdrawal or less positive affect

toward the child on the part of one parent relative to the other parent. Both

72

discrepancy scores ranged from 0 to 3 (maximum possible = 6). The mean warmth
discrepancy was 1.16 (S_D = .82) and the mean investment discrepancy was .98
(S_D = .92).

In a factor analysis of the coparenting variables, McHale (1995) derived
three factors which were replicated in the current study. A principal components
analysis with varimax rotation of behavioral competition, cooperation between
partners, verbal sparring, warmth between parents“, child—centeredness, and the
two discrepancy scores resulted in a three-component solution that accounted for
65% of the variance (loadings are presented in Table 4). The coparenting harmony
factor was comprised of the cooperation between partners and warmth between
parents scores, and accounted for 25% of the variance; the hostilig-
competitiveness factor was comprised of the behavioral competition, verbal
sparring, and child centeredness (reversed) scales, and accounted for 21% of the
variance; and the parenting discrepancy factor was comprised of the warmth and
investment discrepancy scores, and accounted for 19% of the variance. Scale
scores were calculated by summing the standardized scores for each coded
variable’s primary loading on their respective factors. Alphas for the three scales

were .75, .50, and .43 for coparenting harmony, cooperation-competitiveness, and

 

4McHale (1995) included a composite variable called family warmth comprised of
dyadic warmth, mother warmth, and father warmth in the factor analysis. Since the
purpose of the current study was to examine coparenting rather than whole-family
interactions, and since the family warmth variable overlapped with the warmth
discrepancy variable (both included individual warmth ratings), only the dyadic warmth
variable was entered into the factor analysis.

73

Table 4

Loadings for Coparenting Behavior Principal Components Analysis (E = 94)

 

 

Scale 1 2 3
Cooperation between parents .89 .03 .10
Warmth between parents .84 .12 .1 1
Verbal sparring .04 .83 -. 1 7
Competition between parents .46 .62 .06
Child-centeredness (R) .06 .59 .20
Investment discrepancy .01 .18 .82
Warmth discrepancy .22 -.11 .74

 

Note. Boldface denotes the primary loading for each scale; R = reverse scored.

parenting discrepancy, respectively.

McHale (1995) reports that construct validity was demonstrated by
comparing the coparenting factors to independent staff ratings of similar
behaviors assessed during waiting room assessments, with self-reports of play
time, home coparenting, and satisfaction with own and spouse coparenting
involvement. Hostile-competitive coparenting was also related to marital distress
in families of boys, while large discrepancies in parenting involvement were
related to marital distress in families of girls (McHale, 1995). The cooperation
scale has been related to parental control strategies and child compliance

(Gorvine, 1999).

74

Individual psychological resources. Measures have been chosen to reflect
the individual partner’s contribution to the development of parenting (Belsky,
1984), and more specifically, to the parenting partnership.

1. Ego development. The Washington University Sentence Completion
Test (SCT; Hy & Loevinger, 1996; Loevinger, 1985; Loevinger, Wessler, &
Redmore, 1970) is a semiprojective measure of personality given at the first
assessment. The original SCT was comprised of 36 items with alternate male and
female forms. However, the two halves of the 36-item SCT were designed to meet
criteria for parallel forms in order to provide a shorter version if necessary
(Loevinger, 1985); Novy and Francis (1992) confirmed that the two forms are
equivalent. The revised 18-item version of the SCT used in the current study
(Form 81-2) was created to make the items more comparable for men and women
(although the forms continue to be sex-specific). Participants were instructed to
complete each sentence stem but were given no further instruction. A scoring
manual (Hy & Loevinger, 1996) was used to code each response and denote a
developmental level based on the cognitive complexity, perspective-taking
abilities, and psychological awareness of the individual. Response frequencies
may then be used to derive a total protocol rating (TPR) based on a frequency
ogive, or each level may be given a numerical rating and the ratings then summed
to form an item sum score. Because the item sum score is usually more reliable

than TPR ratings and designated cutoffs for the item sum score are usually

75

equivalent to TPR ratings (Hy & Loevinger, 1996), the item sum score was used
in the current study.

All protocols were transcribed and coded by the primary investigator. To
assess interrater reliability, 40 protocols were coded by a second rater. Interrater
reliability as assessed by an intraclass correlation was .94. The short SCT has
demonstrated an internal reliability of .84 and .81 for the first half and second half
of Form 81, respectively, with a correlation (corrected for unreliability) between
the two halves of .96 (N ovy & Francis, 1992). For the current sample, alphas were
.90 for mothers and .83 for fathers. No test-retest information is available for the
revised SCT, but the original SCT has demonstrated test-retest reliability
coefficients of .64 to .91 (Redmore & Waldman, 1975). The construct validity of
the SCT is supported by theoretically consistent relations with sex differences,
interpersonal behavior, behavior problems, and family environment (e.g., Cohn,
1991; Hauser et al., 1984; Noam et al., 1984).

For purposes of description, the distribution of ego levels using cutoffs
suggested by Hy and Loevinger (1996) is presented in Table 5. No participants
were categorized below the conforrnist level, and most were classified as
conscientious, although a large proportion of fathers were also classified one level
below in the self-aware category. Three participants were classified at the highest
and rarest level, integrated. Mothers (M = 97.36, E = 8.20) had significantly

higher item sum scores than fathers (M = 93.31, SD = 8.55), t = 3.99, p < .000.

76

Table 5

Distribution of Ego Levels for Mothers and Fathers

 

 

Ego level Mothers Fathers
Conformist 1 % 8%
Self-Aware 17% 32%
Conscientious 53% 43%
Individualistic 17% 14%
Autonomous 12% 2%
Integrated 1% 2%

 

2. Deggee of assimilation of the parental role. Parents who are highly

motivated to become parents or intensely concerned about parenthood are likely to

have considered, discussed, and researched issues related to parenthood, thus

forming a set of expectations of the demands involved in the parental role (e. g.,

have an “internal working model” of parenthood). These parents might be

considered more “assimilated” into the notion of parenthood. Thus, pre-birth

motivations for the positive aspects of parenthood and pre-birth concerns about

the problems of parenthood were used to measure the degree to which parents are

assimilated into the parental role. The Childbearing Questionnaire (CBQ; Miller,

1981; Miller, 1995) has two scales that assess the degree to which the parent is

focused on positive and negative aspects of childbirth and the parenting process.

77

The Positive Childbearing Motivation (PCM) scale contains 28 items and includes
subscales labeled: (a) joys of pregnancy, birth, and infancy; (b) traditional
parenthood; (c) satisfactions of child rearing; (d) feeling needed and connected;
and (e) instrumental values of children. The Negative Childbearing Concerns
(NCC; originally labeled Negative Childbearing Motivations) scale consists of 21
items that tap: (a) discomforts of pregnancy and childbirth; (b) fears and worries
of parenthood; (c) negatives of childcare; and (d) parental stress. Items are rated
on a 4-point scale anchored at very desirable consequences and not desirable
consequences. Miller (1995) investigated the psychometric properties of the CBQ
in two groups of married couples; couples in one group were experiencing
parenthood for the first time, while the couples in the other group were childless.
Both scales related as expected to childbearing desires and intentions, and the
PCM scale was associated with attempts to conceive (Miller, 1995). Mean test-
retest correlations for 2 weeks, 1 year, and 3 years were .92, .72, and .66 for the
PCM scale and .82, .59, and .54 for the NCC scale, respectively (Miller, 1995). In
the current study, internal consistency coefficients for mothers and fathers were
.92 and .89 for the PCM scale and .83 and .78 for the NCC scale, respectively.
Mean scale scores between the two scales were essentially uncorrelated (for
mothers, r = -.13; for fathers, ; = .06).

3. Family of origin experiences. To assess each spouse’s recollection

of his/her own parents’ coparenting partnership, participants completed the

78

Perceptions of Parent Conflict Scale (PPCS; Frank, 198 8) at the first assessment.
Scales include Positive Parenting Partnership (6 items), Triangulation (vs.
boundary maintenance; 4 items), and Marital Harmony (6 items). Items in the
original measure assess current perceptions of the parental relationship; items

were reworded to tap memories of parental behaviors in the family of origin. The
scale ranges from 1 (never true) to 4 (very true). The PPCS was found to mediate
the relationship between parent marital status and college students’ experiences of
connection and autonomy (Frank & Burke, 1992), and moderates the relationship
between separation-individuation in college students and their insecurities in
relation to their parents (Frank, Poorman, & Charles, 1994). Coefficient alphas for
women and men, respectively, were .79 and .85 for Positive Parenting Partnership,
.81 and .80 for Triangulation, and .86 and .87 for Marital Harmony. Higher mean
scores indicated more positive relationships.

4. Reactance. Individual reactance, or lack of susceptibility to
influence by others, was measured in the third trimester and at 6 months by the
Therapeutic Reactance Scale (TRS; Dowd, Milne, & Wise, 1991). Twenty-eight
items are scaled from 1 (strongly disagree) to 4 (strongly agree). Internal
consistency was .76 for mothers and .69 for fathers. High mean scale scores
represented a stronger tendency toward reactance. High reactance as measured by
the TRS is associated with anger, impulsivity, and avoidant relationships.

5. Parenting efficacy. The Maternal Efficacy Questionnaire (MEQ;

79

Teti & Gelfand, 1991) is a 10-item self-report measure of efficacy around
childcare tasks such as soothing the infant, understanding the infant’s needs, and
entertaining the infant. Items are rated on a 4-point scale with anchors that vary,
but range from least competent to most competent; high mean scores indicate a
greater sense of parenting efficacy. Designed for use with mothers, the instrument
was also administered to fathers. Average coefficient alphas over the three
timepoints were .81 for mothers and .74 for fathers. In a previous study, maternal
efficacy scores correlated -.75 with the (reverse-scored) Parenting Stress Index
(Abidin, 1986) parenting competence scores (Teti & Gelfand, 1991).

6. Violated expectations of the division of labor. Actual and expected
division of labor was measured using a scale developed and modified by Ruble
and colleagues (Hackel & Ruble, 1992; Ruble et al., 1988). Participants rated their
perceptions of the division of household labor and childcare according to the
percentage done by themselves and their spouse (e.g., wife = 80%, husband =
20%), so that each item is rated on a scale from 0-100%. The Household Tasks
scale assesses labor around laundry, cleaning, cooking, shopping, and dishes. The
Childcare Iasks scale taps labor around feeding the baby, changing the baby,
playing with the baby, and soothing the baby when fussy. In the third trimester,
husbands and wives independently completed the measure based on their
expectations for division of labor after the birth of the child. At 1, 3, and 6 months

following the child’s birth, husbands and wives rated their perceptions of the

80

actual division of labor for housework and childcare. Alphas for expected division
of housework were .63 for mothers and .73 for fathers and for expected division
of childcare were .70 for mothers and .67 for fathers; average alphas, which were
stable over time, over 1, 3, and 6 months for actual division of housework were
.67 for mothers and .64 for fathers, and for actual division of childcare were .80
for mothers and .79 for fathers. Scores for the Household Tasks and Childcare
Tasks scales were computed by subtracting actual division of labor at each of the
post-birth assessments from the expected division of labor. Negative scores,
therefore, represent me responsibility for labor than was expected prior to the
child’s birth; although positive scores also indicate violated expectations in that
parents do less than they had expected, violated expectations in the direction of
more responsibility (negative violations) are expected to‘be more pertinent to the
development of the parenting alliance.

Contextual characteristics. Measures of the context are designed to tap the
environmental system in which the parenting alliance develops.

1. Demographic context. At the third trimester, a brief questionnaire
collected information from each partner on age, education, occupation (measured
by the revised Duncan Socioeconomic Index ; Stevens & Featherrnan, 1981),
family income, ethnicity, and years married. A composite SES variable was made
by standardizing and summing scores for each partner for educational level,

occupation, and income. Age and years married were analyzed separately.

81

2. klatieng mtext.

A. W. The Dyadic Adjustment Scale (DAS;
Spanier, 1976), one of the most widely used self-report measures of marital
adjustment, was administered at all four data points. The DAS has 32 items
designed to tap each partner’s perception of general satisfaction, consensus,
cohesion, and affection in the marital relationship, resulting in a total dyadic
adjustment score. Average internal consistency coefficients for the omnibus scale
over the four timepoints were .89 for both wives and husbands. Test-retest
reliabilities are reported to be above .70 (Belsky et al., 1983).

B. W The videotaped couple interviews were
coded using the Interactional Dimensions Coding System (IDCS; Julien, Markman,
& Lindahl [1989]) by senior undergraduate research assistants who completed 90
hours of training. Videotape samples were divided into three equal segments.
Coders rated each partner’s behavior individually for each consecutive segment
(e.g., mother for segment 1, father for segment 1, mother for segment 2, father for
segment 2, etc.) on nine dimensions (Conflict, Dominance, Denial, Withdrawal,
Support/Validation, Communication Skills, Problem Solving, Positive Afi‘ect, and
Negative Affect) on a 9-point scale. The three segment scores for each dimension
were then averaged to form nine individual dimension ratings for each partner’s
ratings. Interrater reliability was assessed for each of the separate dimensions using

the intraclass correlation coefficient, with reliabilities presented in Table 6. The

82

Table 6

Interrater Reliabilig (Intraclass Coefficients) for IDCS Marital Interaction Ratings

 

 

 

Dimension Mother Father

T1 T2 T1 T2
Conflict .59 .50 .58 .71
Dominance .76 .74 .52 .82
Denial .47 .60 .52 .70
Withdrawal .29 .33 .65 .41
Support/validation .66 .56 .62 .72
Communication skills .68 .69 .69 .71
Problem solving .62 .69 .60 .15
Positive affect .72 .53 .59 .66
Negative affect .54 .19 .72 .45

 

Note. Boldface denotes interrater reliabilities less than .50 and subsequently
removed from further analysis. ‘

 

83

intraclass correlations ranged from .15 to .82 (M = .60). Th e coding system was
originally used to rate a problem-solving interaction, and we had some difficulty
getting adequate reliability on some scales due to low frequencies of particular
behaviors. Because of the variable interrater reliabilities, all tapes were coded by
two raters and the scores averaged. In addition, any scales in which interrater
reliability fell below .50 at any point were removed from further analysis (denial,
withdrawal, problem-solving, and negative affect).

A principal components analysis with varimax rotation was conducted for
wives and husbands combined on the remaining Time 1 scales: conflict,
dominance, support/validation, communication skills, and positive affect. Two
components emerged with eigenvalues greater than 1 that accounted for 79% of
the variance. Table 7 presents the final solution. The first component, labeled
positive marital interactions (PMI), accounted for 49% of the variance and
consisted of positive loadings for communication skills, support/validation, and
positive affect. The second component, labeled negative marital interactions
(NMI), accounted for 30% of the variance and is comprised of positive loadings
for conflict and dominance. The solution was replicated when conducted
separately for wife and husband ratings and for the 6—month ratings. Scale scores
were calculated by averaging the individual scales for each component. Alphas
were .86 for mother PMI, .85 for father PMI, .64 for mother NMI, and .53 for

father NMI.

84

ff‘n

Table 7

Loadings for IDCS Marital Interaction Principal Components Analjgis

 

 

Variable l 2
Communication skills .87 .27
Support/validation .89 -.16 '
Positive affect .79 -.05
Conflict -.34 .84
Dominance .34 .84

 

Note. Boldface indicates designated component for each variable.

3. Differences in childrearing philosophy. The Ideas About Parenting
Questionnaire (Heming, Cowan, & Cowan, 1990) was used to assess each
partner’s childrearing attitudes and values prior to the child’s birth. The measure
has 46 items with a 9-point scale ranging from “very much agree” to “very much
disagree” and is designed for parents of children from 0 to 18 months of age.
Reliable subscales were not available for this version (although scales had been
developed for a version designed for parents of children over 18 months;
Hinshaw, Zupan, Simmel, Nigg, & Melnick, 1997). The items were therefore
subjected to a principal components analysis with varimax rotation. Nineteen
items that focused on gender differentiation, parental efficacy, or other domains
not of interest to this study were eliminated prior to analysis. After three items

with loadings less than .35 and one that loaded equally on multiple components

85

were removed, the most interpretable solution had three components that
accounted for 28.6% of the variance, cohered with Baumrind’s (1967) typology,
and supported the findings of Hinshaw et a1. (1997). The final solution is
presented in Table 8. The first component, labeled Authoritative Beliefs, had 9
items and appeared to tap attitudes around the encouragement of emotional
expression and curiosity, as well as a desire to provide security while avoiding
overprotectiveness (12.2%; alphas = .65 for mothers, .58 for fathers). Component
2, labeled Authoritarian Beliefs, consisted of 5 items emphasizing strict control,
suppression of emotion, and a punishment orientation (9.1%; alphas = .69 for
mothers, .59 for fathers). The final component, labeled Permissiveness/
Protectiveness, had 9 items addressing child—centeredness, lack of directiveness,
and provision of security (7.3%; alphas = .61 for mothers, .56 for fathers). The
authoritative and authoritarian scales were negatively correlated (g = -.29, p < .01
for fathers, 1 = -.28, p < .01 for mothers), but neither was associated with the
permissiveness/protectiveness scale (1’s ranged from .05 to .11). Correlations
between mothers and fathers were .23 for authoritarian beliefs and for
permissive/protective beliefs and .20 for authoritative beliefs (all p < .05).
Absolute differences between the spouses’ mean scale scores were calculated,

with high scores indicating greater differences in childrearing philosophies.

86

Table 8

Loadings for Ideas About Parenting Measure Components Analysis

 

 

 

# Item 1 2 3
31 Children should be encouraged to express their .66 -.10 .10
anger as well as their more pleasant feelings.
34 I like to see a child have opinions and express .59 -.29 .05
them, even to adults.
29 There are times in the lives of small children .59 .10 -.01
when they need to be with people other than
their mothers and fathers.
9 Babies should be held when they are upset so .50 -. 12 .19
they will be secure in their parents’ love for
them.
42 One of the joys of parenthood is encouraging a .46 -.19 .09
child’s natural curiosity.
46 A certain amount of fi'ustration and upset is .44 .34 -.15
necessary for a child’s emotional growth.
Because of this, parents should not be too
protective.
35 It’s all right for parents to go on a vacation .41 .05 -.23
together and leave the baby with a good sitter.
1 Too much cuddling spoils a child. -.38 .21 -.02
13 Babies wish their parents would stop fussing -.38 .10 -.08
over them so much.
15 A child should not talk back to a parent. -.21 .72 -.01
19 When a child is called, he or she should come .00 .62 .02
immediately.
2 A parent should not allow his or her mind to be -.33 .56 .14
changed by a child.
44 A child should be punished for breaking his or -.07 .53 .02

her own toys in a fit of anger.

 

87

Table 8, cont.

 

 

# Item 1 2 3

26 Good parents should keep a firm hold on their -.21 .48 .20
child’s expression of angry feelings.

17 Parents should try to avoid open conflict with a -.21 .26 .57
child.

5 Children, one year old or younger, don’t feel -.23 -.04 .50
their parents are really “with them” if the parents
are doing something else at the same time.
8 If parents are ready to go out for the evening and —.23 -.24 .57

their baby reacts by crying and screaming, it is
best for them to cancel their plans.

25 Since new babies are fragile and delicate, they .15 .17 .49
must be handled with extreme care.

32 I want to try to tell my child what to do as little .13 .00 .48
as possible.

43 Other people will probably think I am permissive .06 -.30 .44
with my child.

39 An infant’s needs must always come before the .17 .23 .44
parent’s needs.

41 I will try to anticipate situations which could .03 -.04 .43
cause my child pain or discomfort and help my
child avoid them if at all possible.

23 Babies are fiequently so demanding that their .11 -.05 .39

Note. Boldface indicates designated component for each variable.

parents have no time for anything else.

88

Child Characteristics:

1. Infant sex. Infant sex was recorded at the 1-month assessment.

2. Perceptions of infant temperament. The Fussy/Difficult subscale of
the Infant Characteristics Questionnaire, 6-month version (Bates, Freeland, &
Lounsbury, 1979), was completed by mothers and fathers at 1, 3, and 6 months.
Six items on a scale from 1 to 7 (descriptors vary, but high scores denote more
difficult temperament) assess the degree to which the infant is irritable and
difficult to soothe. Parent reports on the Fussy/Difficult scale correlate with
independent behavioral observations (Bates et al., 1979). Over the three
timepoints, the average internal consistency was .87 for mothers and .82 for

fathers, with little variability across time.

89

RESULTS

Two methods were used to test the hypotheses. To assess the construct
validity of the parenting partnership as distinct from the marital relationship
(Hypothesis 1), structural equation modeling with LISREL 8.20 (Joreskog &
Sorbom 1998) using a confirmatory factor analytic model was conducted. The
development of parenting alliance trajectories (Hypotheses 2 through 4) was
exarrrined by implementing hierarchical linear modeling with HLM/2L, Version
4.03, software (Bryk, Raudenbush, & Congdon, 1996).

Missing Data Estimation

Structural equation modeling with the LISREL8.20 program does not allow
missing data. Therefore, only couples with complete parenting alliance and marital
data at 6 months were included in the LISREL analyses (N = 94). On the other
hand, HLM allows data to be missing over time (at Level 1). If many observations
were missing, estimates would be at risk of systematic bias. However, minimal
data was missing over time and parameters should not be affected. HLM does
require complete data at Level 2. Examination of the pre-birth data revealed 20
scale scores missing across 12 individuals. Most of the missing data was from the
marital interaction data of families where the camera malfunctioned during the

interview and the data could not be coded. In addition, six participants did not

complete the SCT. To avoid discarding these couples fi'om analysis, data

90

imputation was completed for all missing scale scores using regression equations
based on variables predictive of those missing from couples with complete data.
The amount of missing data was considered insufficient to require tests of
differences between groups with and without missing data (HLM N = 101).
Construct Validig of the Parenting Partnership

The first research question referred to the related but distinct nature of the
parenting and marital relationships (Hypothesis 1).

Analytic strategy. Assessed by multiple methods and across multiple raters
using cross-sectional data gathered 6 months after the birth of the child, structural
equation modeling with LISREL 8.20 (Joreskog & Sorbom, 1998) was used to
test the appropriateness of a multitrait-multimethod model (Campbell & Fiske,
1969) with the parenting partnership and the marital relationship as traits and
mother report, father report, and observer ratings as methods (Figure 1). Adequate
model fit would confirm that the parenting partnership is a separate construct from
the marital relationship while accounting for method effects. Alternatively,
inadequate model fit would indicate that the model does not sufficiently
encompass the interrelationships in the observed data; specific method effects may
not be as specified, or the parenting partnership and the marital relationship may
not be discriminable dimensions.

For the test of the hypothesized model, mother and father general alliance

scores, DAS dyadic adjustment scores, positive and negative marital interaction

91

.mcosofioufi .858
gamma: u :2 Z 3228825 33.88 035.3 .I. 9E Son—smocomme messy—ma n .9285 Sconce: meteocaoo u 308.8:
28300 mmmo=o>£uomEo9b==mos n .QEoOB—zmo: “Baum Eon—amigos 2ng n m<Q ”cogs—E masseuse u Adand 13.02

a53— toaoa :83—
szano A use:
a 325-:

.22 S.— Ea . . .38 35 3o .35. .55.
as: see: as; use: sec-5 €288 2:8: as... .33: Beam use:

Eases;
.352

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\

      

35985.8“—
mama—2am

 

92

ratings based on the marital interview, and hostility-competitiveness, coparenting
harmony, and parenting discrepancy scores rated from the family interaction
comprised the observed variables. Intercorrelations, means and standard
deviations for the observed variables are displayed in Table 9. Four indices were
used to evaluate model fit, with consistency across indicators heralding adequate
fit: a) A nonsignificant X2 would indicate that the observed covariance matrix did
not differ significantly from the hypothesized model; b) the goodness of fit index
(GFI; Joreskog & Sorbom, 1989, cf., Tanaka & Huba, 1985) compares how well
the model fits relative to no model at all, with values that range from .00 to 1.00;
values above .90 are considered adequate fit; c) the non-normed fit index (NNFI;
(Beutler & Bonett, 1980) assesses the degree to which the model fits better than a
model in which all variables are uncorrelated, or an independence model;
although these values are designed to range from .00 to 1.00, they can exceed
1.00; and 4) the root mean square error of approximation (RMSEA; Steiger, 1990)
accounts for the error of approximation of the estimate in the population per
degree of freedom, with values .05 or less indicating good fit and .08 or less
representing a reasonable fit (Browne & Cudeck, 1993).

Results for the hypothesized model. Maximum likelihood estimates of the
hypothesized model via analysis of covariance matrices produced an inadrrrissable
solution (i.e., standardized parameter estimates >1), with the problem appearing to

be the result of misspecification of the relationships between the trait and method

93

.‘1

5o. V9.21. .5. Van... .mo. vm... .ofi. V“.
.323 Be 28:22.6 23:3,.” d8 :0 8m memo:— .mo>u&.8mov 0525; wcccoaaoo com .Eeowflv 05 :0 2m 8052 can
82:08 20033 mecca—2.80 xmcowfie of 323 2a 8052 new mecca—oboe income? 05 gone 98 2052: no.2 28:22.80 1'0qu

 

 

 

E. 8. 2.2 mm. mm
82 a: «3 I
8.“ 3% 8.82 mom 2 see-.2
2. 8. 8.2 8. ma
co. co. co. .-
88 New 2!: :8 2 see:
1.8.- 2. 2. 8.- 2 .- ram.- caaaeeae 8828 .e
*8; t8.- 8.- *8. *8. .2. cases: mesa-.380 e
2. :8..- *8. .. z...- 2: 8.- anaeseecaeceeaaom .m
2830285
.. a. .8.- 8.- .2.st ..8- :8.- .2 .- Ease seamaz .e
mcomuowuouam
8.- .8. 8.- *8; 32lo- 2. :8. BEE nausea .m
2.- Ln... mot tarmm. SINM. tibia“ ***wm. “coaumzmcm 33.82 .N
8.- .2. . 8.- 2:8.- .........8. tram. in... 85:3 8:55 .2
e e m a m N _ aeSeE>

 

 

 

 

facets—om Rum-82 was aim-852$ 538mm 5:026 com 253.1259 Eng—Sm use 2802 88. 28.8835

a 035-

.--I11

94

latent variables. That is, the problem did not converge as hypothesized. The model
was assessed for mother and father variables separately, but the same problems
emerged for each. Thus, using the most conservative test, Hypothesis 1 was not
supported.

Post hoc analyses. Additional analyses were conducted in hopes of defining
a more simplified model that would differentiate between coparenting and marital
self-reports and observational measures. Mother and father data were analyzed
separately to decrease error likely to exist due to sex differences. To test the
hypothesis that coparenting and the marital relationship are related but distinct
constructs, a two-factor model was specified in which the parenting partnership
and the marital relationship were correlated latent variables. This model was tested
against a one-factor model in which the coparenting and marital variables were
hypothesized to form one “dyadic relationship” latent variable. The two
competing models were examined separately for self-report data and for
observations. For the self-report measures, the parenting partnership latent
variable was comprised of two parallel subscales derived from the general alliance
scales, and the marital relationship latent variable was comprised of the four DAS
subscales, dyadic satisfaction, dyadic cohesion, dyadic consensus, and dyadic

affectional expression. For the observational measures, the parenting partnership

 

5The general alliance scale has 31 items. One item that did not have a matching
item was dropped to equalize the number of items in each scale, resulting in two
parallel scales of 15 items.

95

latent variable consisted of hostility-competitiveness, coparenting harmony, and
parenting discrepancy, and the marital relationship latent variable was made up of
positive and negative marital interactions. In all, four sets of analyses were
conducted, with separate tests examined for mother and father self-reports and
observations.

Fit indices for each of the models are presented in Table 10. Although
support was generally provided for the hypothesis, results for the self-report data
were the most conclusive. For both mothers and fathers, the two-factor model
differentiating the parenting partnership and the marital relationship provided an
excellent fit to the observed data. All factor loadings (shown in Figure 2) were
Significant, ranging from .48 to 1.00, ts > 3.30, p < .001, as were the correlations
between the factors, mothers’ 5 = .58, t = 7.31, p < .001; fathers’ ; = .60, t = 7.16,
12 < .001. No additional paths were suggested by modification indices.

Furthermore, the two-factor model was a considerable improvement over the one-
fa<>tor model, mothers’ A x2 (1, _r_r = 94) = 95.05, p < .001, fathers’ A {(1, p = 94) =
67 - 03, p < .001. A simultaneous group analysis of the two-factor model for both
p arents with factor loadings allowed to vary determined that the model was
comparable for both parents, )6 (23, p = 94) = 29.91 , p = .15 , OF I = .94, NNFI =
98, RMSEA = .06.

The results for the observational variables were more equivocal. As with

the self-report data, fit indices and a lack of modification indices indicated that the

96

Table 10

Fit Indices for Competing Two- and One-Factor Models of the Parenting
Partnership and the Marital Relationship

 

Two-factor One-factor

x2“ GFI NNFI RMSEA x2” GFI NNFI RMSEA

Self-report
Mother 12.59 .96 .97 .08 107.64*** .72 .57 .34
Father 12.26 .95 .96 .08 79.29*** .76 .61 .30
Observations
Mother 4.07 .98 .98 .01 7.69 .97 .60 .08
Father 3.68 .98 1.01 .00 4.71 .98 .98 .00

 

Note. GFI = goodness of fit index; NNFI = non-normed fit index; RMSEA = root mean
square error of approximation.

3F or self-report measures, df = 8; for observational measures, df = 4; p = 94.

bFor self-report measures, df = 9; for observational measures, df = 5, p = 94.

***p<.001.

97

hr

.25. v m E 38$:me 83:58 282523 =< .3202 0.8 805mm new .32? 2a .9559: new $56508

uofiwcmvcfim £28898 Quezon-me 0:53 u mm<m<Q 8:38:00 069$ n ZOUm<D ”commence 069? u IOUm<Q
“catamaran 33% u H<mm<Q MN 2:808 028:? mesa—Ema u Nag/E x 03me8 028:? messaged u 1.5;; .Illoaoz

 

 

 

 

 

 

 

 

 

 

 

 

583 :83 3.2..
a. 2. 8.
3.. 8. 8.
8.—
82.83:— 528.5.
.552 «855

cc.
an.

 

BEES—om 33.82 98 £805.23 £323 to o ova—om me 382 .5 cam-03H 8m 8383mm 280 9:5

135—

~ 253m

98

two-factor model fit well for both parents. However, two of the six path
coefficients in the mothers’ model did not reach significance (parameter estimates
shown in Figure 3). In addition, the one-factor model was less distinct from the
two-factor model than had been the case for the self-report data. For mothers,
although the NNFI of .59 indicated a poor fit, the other fit indices suggested that
the one-factor model fit the data well. Moreover, the fit of the two models was not
significantly different, although there was a trend in favor of the two-factor
model, A x2 (1, p = 94) = 3.62, p < .10. For fathers, there were virtually no
differences in fit between the one— and two-factor models, )6 (1, p = 94) = 1.03, E.
To determine which model could be applied better across both parents,
simultaneous group analyses were used to evaluate each model for both parents,
with factor loadings allowed to vary. Both models fit well across mothers (all
factor loadings became significant) and fathers, two-factor model 12 (14, p = 188) =
9.69, _r_r__s_, GFI = .98, NNFI = 1.16, RMSEA = .00; one-factor model {(15, n = 188)
= 16.42, ne, GFI = .97, NNFI = .94, RMSEA = .03. A comparison of the models
favored two factors over one, A x’ (1, p = 188) = 6.73, p < .01. On the basis of these
results, it was concluded that the parenting partnership and the marital relationship
formed separate but related constructs using both self-report and observational
measures.

A final set of analyses examined whether self-reported perceptions and

observed interactions were interrelated or separate aspects of the couple

99

.8. v at... .5. v as... .8. v a...
.30—DD 0.8 mac—tam Mom .D>Onm ohm wave—HOE ho.“ mucosa-“000 UONm—ubw—uflaum .mmo=o>3wwon~500rb=flm0£ H .n—EOUE—fimom 6qu

 

 

 

 

 

 

8.2838:— mcozoflo.£ 3:30.85 taco—Ea: .anU
.252 .252 we. :03. o. a» _ no
02332 03.76.. .. a— .c Q B: I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

mindsets—om
3:32

Esau—Ea;
wczconam

   
    

X...

 

 

mimeouflom .252 one @380:th 525.8; 323.20 no 632 288803-2- 89 8383mm 230833
m PEN—m

100

relationship. A model was specified in which the four latent relationship variables,
self-reported parenting alliance and marital adjustment and observed coparenting
and marital interactions, were hypothesized to form different but related
dimensions of the dyadic relationship. The solution (shown in Figure 4) fit well
for both mothers, )6 (38, g = 94) = 42.76, 11$, CF I = .92, NNFI = .98, RMSEA =
- O4, and fathers, x2 (38, g = 94) = 42.83, n_s, GFI = .92, NNFI = .95, RMSEA = .04;
all factor loadings in the measurement model were significant and no changes
were suggested by modification indices. A simultaneous model for both mothers
and fathers with factor loadings allowed to vary resulted in x2 (93, r; = 188) =
1 O3 .80, ng, GFI = .89, NNFI = .97, RMSEA = .04. Although the GFI was just
below the conventional cutoff of .90, in conjunction with the other measures of
fit, which were excellent, the model was judged to confirm the adequacy of the
factor structure across parents.

A series of models attempting once again to derive an overall method (self—
r313011 or observation) and/ or trait (parenting partnership or marital relationship)
factor all resulted in inadmissable solutions. Inspection of the intercorrelations
among the latent variables provided a probable explanation. Although some
aSSOCiations were as expected, others were not; in addition, the pattern of
Correlations differed for mothers and fathers. For both parents, the two self-report
measures were moderately related (mothers’ r = .58, fathers’ r = .60, p < .001), but

the Observational measures were strongly associated only for fathers (mothers’ r =

101

no. v m EEEEE a «a “:85:me 83:58 .650 =m ”no. A m. .anmmcwmm on o. vofimofiona: macaw—oboe

@0508 8:865 one: combo—on agoficwfi on 8 35853»: macaw—oboe tab 380:2: .888 venue—om 6.830885
3:85 033w: u :22 ”ma—28885 3E2: 03:89 u E: $053.8va mas—~28 u .38me ”mmo=o>:zoafioo.b_Ema:
n .QEoUB—smom 623830 Enough? omega u mm<m<m “macaw—So 0:53 u ZOUm<Q Eofioaoo 23% n IOUm<Q
Ectoflmsmm 33.8 n H<mm<n ”N 23.88 8.3:? massaged u mafia : 2338 3.3:? wagging n 1513 94%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. .QEO
:8 in. 5293 48mg 83.. .233 3.23 SE .9585 :85: 3.3%.
A, ,
a... . 8. S. . 8. .
N? a we 8. NW. . a. 8
we. we 2 3..
8.

 

 

 

     
   

    
    

, 9.53:!—

e a o a o 3:22.23. 2585.8 .
Janey)" a .552 «£225; meta—2am
32.630 Bang—om 3383.. 82030

 

mm. 3..
”Q. QM. . emu.
NV. 5?“
Mb. oN
6w.

 

102

.37, n_s_, fathers’ r = .66, p < .001). Whereas the marital variables were more highly
correlated with each other for mothers (mothers’ r = .84, fathers’ r = .53, p <
.001), the coparenting variables were not strongly related to each other for either
parent (mothers’ r = .25, n_s, fathers’ r = .34, p < .001). Thus, there was only
partial evidence for the hypothesized trait and method effects. Furthermore,
correlations that should have been small according to the hypothesized model
were quite high (e. g., between self—reported parenting alliance and marital
observations and between self-reported marital adjustment and coparenting
observations, rs ranged from .29 to .73).

Overall, the results suggest that each of the latent variables comprised a
particular dimension of the dyadic relationship. More satisfied perceptions of
those aspects of the relationship related to coparenting were likely to be
accompanied by greater happiness with the marriage, and, especially for mothers,
Were reflected in more positive marital interactions. However, perceptions of the
parenting alliance were not particularly strongly related to how the couple actually
behaved when playing together with their child. Furthermore, warm, cooperative
couple behaviors in the context of the child were accompanied by more positive
marital interactions for fathers, but appeared to be a separate aspect of the

relationship altogether for mothers.

103

Parenting Alliance Trajectories

The remaining hypotheses explored the longitudinal development of the
parenting alliance. The overall aim was to specify trajectories using data gathered
at 1, 3 and 6 months after the baby’s birth, and to explain individual differences in
these trajectories using stable pre-birth factors and time-varying process factors.

Analyg'c strategy. Growth curve modeling (Bryk & Raudenbush, 1992;
Duncan, Duncan, & Hops, 1996; Duncan, Duncan, Strycker, Li, & Alpert, 1999;
Kamey & Bradbury, 1995; McArdle & Epstein, 1987; Raudenbush, Brennan, &
Barnett, 1995; Willett & Sayer, 1994) provides a conceptual model for examining

Specific aspects of longitudinal trajectories. Unlike traditional methods of
examining change, which utilize information from only two points in time via
analysis of residual change or difference scores, growth curve modeling
incorporates information from all data points. For each person, a within—person, or

Level 1, model was specified that characterized his/her individual trajectory and

 

Was potentially comprised of three parameters: an intercept, representing the le_ve_1
of parenting alliance at the midpoint of the study; a sloPe, indicating the linear rate
of change in the parenting alliance over time; and a quadratic filnction
representing a curvilinear component estimating the rate of acceleration or
deceleration in change. The parameters for the Level 1 model were estimated for
each person and used to specify an overall average parenting alliance trajectory

(Hypothesis 2), or baseline model, by the following equation:

104

Y, = no, + 7r”(rate),-, + 7:2,.(curve),, + 6,, ,
where Y,, was the parenting alliance trajectory for Participant 1' and Time t.
Orthogonal polynomial contrasts were used to define the linear and

quadratic functions, with 1, 3, and 6 months weighted as -l, O, and 1 for the linear
and -.333, .667, and -.333 for the quadratic (Kurdek, 1998)). Equal weighting
over time was considered defensible in light of the relatively equal spacing of the
timepoints (2 and 3 months apart). In addition, measurement error was estimated
by using the two previously derived parallel measures of the parenting alliance, so
that each person had two scores for each timepoint that represented a one latent
parenting alliance score at each assessment plus measurement error (Raudenbush
et al., 1995). Typically, in a study with three timepoints, the three observations per
person would allow only an intercept and a slope parameter to be estimated, plus
the estimation of the within-person error variance. The construction of parallel
measures (previously utilized to in post-hoc analyses for Hypothesis 1), however,
resulted in six observations per person and enabled the quadratic to be estimated.
Raudenbush and colleagues (1995) note that when data are limited to three
timepoints, the best interpretation of the quadratic may be as an indicator of the
temporal instability of the trajectory (i.e., a marker of the degree to which the
parenting alliance fluctuates versus develops as a consistent linear function). The
estimation of these three parameters saturated the model, so that the within-person

error was an index of the internal consistency of the general alliance scale. The

105

elements were then defined as:

7:0,. = average initial level of parenting alliance;

7r“. = average rate of change in Y for Participant i over all timepoints;
75,: average temporal instability in Y for Participant 1'; and

e,, = random within-person measurement error.

These three parameters (7:0,, 7r“, 7:2,) defined each individual’s parenting
alliance trajectory across time and became the outcome variables in subsequent
analyses exploring between-person differences. That is, the specification of
growth parameters at the level of the individual resulted in distributions of
intercepts, slopes, and quadratics which were then regressed onto stable between-

person predictors (termed Level 2 variables) to attempt to account for individual

 

differences in specific aspects of the growth curve trajectory (Hypothesis 3) in an

explanatory model:

7:0,. = 300+ fl0,(predictor 1),.+. . . [301(predictorj), + um
,610+ ,6”(predictor 1),.+ . . . fidpredictorj), + u”

7:2,. = 520+ 521(predictor 1),.+ . . .fl21(predictorj),._ + “2:

Because trajectories from mothers and fathers were likely to be dependent
Upon one another within a particular couple, an extended model that accounted for
the nesting of persons within couples was also defined. This combination of a
longitudinal model for individual change with a cross-sectional hierarchical model

for matched pairs resulted in a longitudinal model of change in matched pairs

106

(Barnett, Raudenbush, Brennan, Pleck, & Marshall, 1995; Kamey & Bradbury,
1 997; Kurdek, 1998; Raudenbush et al., 1995). The Level 1 baseline couple
model then became:

1”,, = (female), [7%,- + nth-(rate), + nth-(curve), ]+

(male),, [7rm0i + 7rm1i(rate),., + 7rm2,(curve),., ] + e,, ,
in which two intercepts, two slopes, and two quadratics were specified, one for
each parent, that took into account the covariation of the other partner’s
parameters. The Level 2 model took the same form as above for six as rather than
three as in the baseline model for the entire sample without differentiation
between mothers and fathers.

Explanatory models were also developed that examined the
interrelationship between the parenting alliance trajectories and other time-varying
Variables (Hypothesis 4) (Raudenbush etal., 1995). Using self-reported marital
adjustment as an example, this method separated the relationship between
parenting alliance trajectories and post-birth marital adjustment into two
components. The time-varying, or Level 1, component, represented the degree to
which changes in marital adjustment were associated with changes in the
parenting alliance; scores used in analysis were the deviation of the marital rating
at each timepoint from the person’s mean marital adjustment across the three
assessments. The time-invariant, or Level 2, component provided the relationship

between the individual’s mean marital adjustment over the three timepoints and

107

his/her mean level of parenting alliance. This Level 1 model was then:
1’, = (female), [7%.- + Entrate). + 75m(CUTVC).-t

+ 7:9,.(female marital rate)“ + 7rf4,(male marital rate),,] +

(male)it [75mm + 7rmli(rate)it + 7rmZi(curve)ir

+ anal-(male marital rate)“ + ”m4,(female marital rate),,] + e,,
For example, 7:8,- indicated the expected increase in a mother’s parenting alliance
rating per unit increase in her marital rating. Because the six parameters estimated
in the baseline couple model saturated the model, leaving no degrees of freedom
available for estimation of additional parameters, the four coefficients

representing the temporal component of marital adjustment were constrained to

have fixed effects. The Level 2 model for the 10 parameters became:

7%,. = [3,00 + ,6m(female marital mean),+ ,6m(male marital mean),+ um,
ani = ,Bmoo + mOl(male marital mean),+ ,Bmoz(female marital mean) ,. + umO,
75m = .Bno + “m

75ml: = .ero + “mi

7591 — 1690 + “a:

”"12: — flmZO + “mz.

75:31 = floo

”m3: _ flmo

”r4: = IBM

7tm41 - ,Bm40

108

This model was particularly applicable to research questions related to systemic
family relationships because it assessed not only the effects of the spouse’s own
perceptions on family relationships, but considered the effects of the partner’s
perceptions as well.

Hierarchical linear modeling (HLM; Bryk & Raudenbush, 1992) has a
number of advantages over alternative methods of growth curve analysis, such as
structural equation modeling (Bryk & Raudenbush, 1992; Kamey & Bradbury,

1 995). First, HLM does not assume that the data are collected simultaneously from
all participants at equal intervals (i.e., perfectly balanced). Because HLM uses all
available data from each individual to compute the trajectory, missing data is
allowed (for this study, individuals were required to have multiple measurements).
Second, individual parameter estimates are weighted by their precision;

traj ectories that can be estimated more precisely are weighted more heavily than
those estimated less precisely. These Bayesian estimates result in more
Conservative estimates of variance than traditional optional least squares (OLS)
methods. Third, HLM simultaneously estimates effects for each timepoint, so that
each parameter is controlled for when estimating other parameters.

Growth curve analysis via hierarchical linear modeling was conducted

using HLM/2L, Version 4.03, software (Bryk et al., 1996). First, a baseline model

for the entire sample without predictors specified an average parenting alliance

traj ectory. Next, the longitudinal mode] for matched pairs tested a baseline couple

109

_r_n_od_el to examine whether men and women differed in their parenting alliance
trajectories. A number of explanatory models were then identified. In explanatory
model A, the influence of stable pre-birth factors on parenting alliance trajectories
was investigated. In explanatory models B1 through B5, interrelationships
between the parenting alliance and time-varying factors, including the marital

relationship, the division of labor, parenting efficacy, and perceptions of infant

temperament, were assessed.

The Baseline Trajectog of the Parenting Alliance

Average parenting alliance ratings were 3.50 (S_D =.30 ) at one month,

3.51 (SD = .28) at three months, and 3.48 (S_D = .38) at six months, indicating
virtually no change in level across time. On the one hand, the remarkable
consistency of these means might suggest that attempts to define a slope parameter
Would be fruitless. Nonetheless, one purpose of the current study was to describe
the trajectory of the initial parenting alliance more precisely according to specified
Parameters, regardless of whether they were significant. Furthermore, individual
di fferences in the change parameters might make the baseline trajectory
IniSIeading. To illustrate, one possibility is that the parenting alliance is a stable
relationship that remains unchanged over time for most couples; in this case, the
c115-urge parameters would have little variance. On the other hand, some parents
may report increasing satisfaction with the parenting alliance, while others

beeome more unhappy. This scenario could also result in a stable baseline

110

trajectory, but would also leave significant variance to be explained in the change
parameters. The HLM analyses were therefore considered a suitable next step.
Assumptions and reliabiligg of the parameter estimates. HLM procedures
assume that variables are normally distributed and that the Level 1 variances are
homogeneous. Descriptive statistics and plots indicated that all variables were
within the bounds of normality, and a test of the Level 1 variances confirmed
homogeneity, 38094, E = 1202) = .32, pp. In addition, the reliability of the
parameter estimates was assessed. In the context of growth curve analysis, the
reliability of a parameter estimate is defined as “the proportion of observed
variance in a parameter that can be treated as true variance” (Kamey & Bradbury,
1995, p. 1101). This definition of reliability differs from internal reliability as
assessed by coefficient alpha, for which higher values are generally preferred. As
a guideline, Bryk and Raudenbush (1992) recommend that when reliabilities are
less than .05, the variance of the parameter is essentially zero. For this study,
baseline model reliabilities were considerably higher, estimated at .98 for the
intercept, .88 for the slope, and .82 for the quadratic. These reliabilities indicate
that nearly all the variance in the intercept and the majority of variance in the
change parameters can be construed as true variance, and imply that residual
variance should be able to be explained if the correct covariates are identified.

Model specification. The overall parenting alliance trajectory was

determined by specifying an unconditional between-person model in which the

111

parameters of the within-person model were allowed to vary randomly, providing
unbiased estimates of the population mean and population variance of the
intercepts, slopes, and quadratics. t-tests assessed the hypothesis that each
parameter estimate was different from zero. This hypothesis is meaningless in the
case of the intercept, which must differ from zero because it is based on an
arbitrary scale of 1 to 4. However, if the slope and quadratic are significantly
different from zero, change is indicated in the parenting alliance over time.
Additionally, the significance of variation among the parameters was examined
using the chi square statistic; significance indicates that variation in the parameter
remains to be explained, so that the average slope may not be a good
representation of rate of change in the sample as a whole.

Restricted maximum likelihood estimates of the baseline parenting alliance
trajectory described a trajectory in which the parenting alliance is quite strong and
unchanging, with estimates, standard deviations, and _t_ and )6 statistics presented
in Table 11. Effect sizes are also provided to facilitate interpretation, with .10, .3 O,
and .50 indicating small, medium, and large effects, respectively (Cohen, 198 8).
Despite the apparent lack of change over time, significant variation was
eVident around all three parameters. Figure 5 illustrates the individual parenting
alliance trajectories for all individuals in the sample. This figure demonstrates that
Whereas on average, parenting alliance scores were high and remained stable over

the 6-month period, individuals differed substantially in their coparenting

112

Table 1 1

Parameter Estimates for Baseline Model

Parameter estimates

 

 

Parameter Unstand. S_D Effect size r 1(203)a x2 (196)b
coeff.

Level 3.51 .11 --- 147.47*** 10023.70***

Rate —.01 .15 .05 -.86 1573.02***

Curve .01 .20 .04 .70 1082.05***

 

Note. Unstand. coeff. = Unstandardized coefficient.
“ Refers to significance of the parameter estimate.

b Refers to significance of the variance component.

***p< .001.

experiences, with both increasing and decreasing trajectories resulting in the
W of no change over time at the group level. The variation in change is
highlighted by the distributions of the slope (Figure 6) and quadratic (Figure 7)
eStirnates for individual participants. Using two cutoffs for comparison, Table 12
Presents the percentages of parents whose perceptions of the alliance increased,
ren‘lained unchanged, or decreased between 1 and 6 months post-birth. Using a
change in parenting alliance score of 1/2 SD (.15) from month 1 to month 6 as the
criterion, about half of the sample did report a stable parenting alliance, but the
re St were equally divided between increasing and declines. When a cutoff of 1 SD

(‘3 0) was used, three-fourths were consistent over time, with 11% reporting

113

Figure 5

Individual Parenting Alliance Trajectories

 

Parenting Alliance Score
N
LII
%

 

I l

 

 

1 minim 3 months 6 months

114

Figure 6

Distribution of Slope Estimates for Baseline Model

 

 

 

Frequency

 

 

 

 

 

 

 

 

-.6O -.4O -.2O .00
Slope Estimates

115

.20

 

 

 

Figure 7

Distribution of Quadratic Estimates for Baseline Model

 

 

 

 

12 ;
10‘ i
8‘ l
E" I
3 6- E
3 I
H I
IL I 3
44 w 3
I’lil ; i
2~ i! ' ' I
Ii‘lz i I r
i‘;i‘ I”‘ 1' i
I I II
..J I-II E'i!

0 I I I I I

 

 

 

 

-.85 -.7O -.55 -.4O -.25 —.10 .05 .20 .35 .50 .65 .80
Quadratic Estimates

116

Table 12

Percentage of Parents Whose Parenting Alliance Ratings Changed From 1 to 6
Months Post-Birth

 

 

 

Percentage
Direction of change Changed by 1/2 SD Changed by 1 SD
Increased 25% 1 1%
Stayed the same 53% 75%
Decreased 23% 14%

 

increasing satisfaction and 14% decreasing satisfaction.

The three parameters were found to be unrelated. The correlation between
level and rate was .07, between level and the curvilinear component was .03, and
between rate and curve was .07. Overall, then, individuals who reported more
satisfaction in the parenting alliance were not prone to greater declines, nor were
they more likely to maintain their satisfaction, than those who rated the parenting

alliance as lower.

Bgeline Couple Parenting Alliance Trajectories

Means and standard deviations for mothers’ and fathers’ parenting alliance
scores at each of the three timepoints are presented in Table 13 and suggest that
fathers and mothers differ in the level of parenting alliance, with mothers less

satisfied than fathers. Mothers’ and fathers’ parenting alliance ratings were

correlated within each timepoint, r = .43, p < .01 at one month, ; = .25, p < .05 at

117

Table 13

Mean Parenting Alliance Scores Over Three Timppoints for Mothers and Fathers

 

 

Parent 1 month 3 months 6 months
Mothers
_M_ 3.38 3.39 3.39
_S_I_) .41 .39 .41
Fathers .
M 3.59 3.59 3.57
S_D .32 .31 .34

 

three months, and _r_ = .3, p < .01 at six months.

To define separate parenting alliance trajectories for mothers and fathers
while accounting for the nesting of parents within couples, an unconditional
between-couples model was specified in which within-couple parameters were
allowed to vary randomly. All parameter estimates, shown in Table 14, were
highly reliable, although the change parameters were slightly less so for fathers.
The parameter estimates for the baseline couple model are presented in Table 15.
The baseline couple model replicates the findings of the overall baseline model in
that the two trajectories did not change over time. However, a multivariate test
indicated that mother and father trajectories differed, x2(3, l_\l_ = 101) = 17192.07, p

< .000. Follow-up univariate tests confirmed that although the mother and father

rates and curves were equivalent (i.e., no change was apparent in either group),

118

Table 14

Reliabilities for Baseline Couple Model Parameter Estimates

 

 

 

 

Reliability
Parameter Mother Father
Level .98 .98
Rate .9 1 .8 1
Curve .87 .70

 

fathers rated the parenting alliance as significantly higher than did mothers, x2(l,
fl = 101) = 17133.85, p < .000.

Furthermore, although all six parameter estimates continued to have
significant residual variance, indicating that individual differences exist in the
degree of change and in the stability of the parenting alliance, each parameter
varied to a greater extent for mothers than for fathers. Inspection of Table 16
reveals that mothers were more prone to change in experiences of the parenting
alliance than were fathers. These differences in variance are shown in Figure 8,
which depicts individual trajectories for mothers and fathers separately, and
Figures 9 and 10, which display the distributions of the slope and quadratic
parameters for each group. The couple model was a significant improvement over

the general baseline model, Ax2(15, _l\_I = 101) = 68.70, p < .000, indicating that a

model accounting for the different experiences of mothers and fathers represents

119

Table 15

Parameter Estimates for the Baseline Couple Model

 

 

Parameter Unstand. SQ Effect size 1(100)’ x2(93,
coeff. _r_ _N = 101)b

Mothers

Level 3.41 .34 98.58*** 5389.25***

Rate -.00 .18 .01 -.29 1038.94***

Curve .00 .24 .01 .22 731.69***
Fathers

Level 3.60 .29 121.35*** 3681.88***

Rate -.01 .12 .10 -1.10 $21.16*"

Curve .02 .15 .09 1.03 322.61***

 

Note. Unstand. coeff. = unstandardized coefficient.

 

aTests the significance of the parameter estimate from zero.

bTests the significance of the residual variance from zero.

***p_ < .0001.

120

Table 16

Percentage of Mothers and Fathers Whose Parenting Alliance Ratings Changed
From 1 to 6 Months Post-Birth

 

 

 

 

Percentage
Changed by 1/2 SD Changed by 1 SD
Direction of change Mothers Fathers Mothers Fathers
Increased 3 1% 1 8% 16% 6%
Stayed the same 40% 63% 68% 78%
Decreased 29% 16% 17% 12%

 

the observed data better than one which does not.

Within couples, the level of parenting alliance was correlated between
mothers and fathers (r = .42, p < .01), as were the quadratic functions (r = .26, p <
.05), suggesting that parents tended to experience unstable trajectories together.
Surprisingly, between—spouse slopes were unrelated (r = .12, pg), indicating that
change in perceptions of the parenting alliance in one parent was not concurrently
dependent upon change in the other parent. Subsequent analyses control for the
dependency of the parameter estimates within couples. The level of the parenting
alliance was not associated with either of the change parameters (the correlation
between level and rate was .07 for women and .10 for men, and between level and

curve was -.02 for women and .12 for men).

Slopes and quadratics were correlated for both women and men, but in

different directions (I = .21 for women, p < .05, r = -.26 for men, p < .01).

121

Figure 8

Individual Parenting Alliance Trajectories of Mothers and Fathers

Mothers

 

 

9"
U1
1
I

(A)
1
I

 

 

N
jI

Parenting Alliance Score
N
or

_x
01
1
I

 

 

 

 

 

 

 

 

 

1 .L I I
1 month 3 months 6 months
Fathers
4 _ $1” 7 Q- ...-..- ’4 “7’5 ‘ _"_____ -—-—-
» -«-=s--'::'f L... ‘ ' ‘7,
93. '3’ §==4~§$m
8 3 5 .._ m.‘i;" “'7' War—Ag
3 3 -_ \
t:
.‘2 ”
2 2.5 -- .. g _g - -- -.._..--—-~- W" "'2
g .4.-
‘e 2 “
’ca‘ 1.
a. 1.5 ~-
1 month 3 months 6 months

122

Figure 9

Distribution of Slope Estimates for Mothers and Famers

Frequency

Frequency

 

 

 

 

 

 

 

 

 

 

 

 

Mothers
15
10 ‘ j
5 " I I :
0_I -. 11 .1...111:1. . 1311- "I I
-.60 -.4O -.20 .00 .20 .40 .60
Slope Estimates
Fathers
15
10 “
5 _
0 IT I
-.60 -.40 -.20 .00 .20 .40 .60
Slope Estimates

123

Figure 10

Distribution of Quadratic Estimates for Mothers and Fathers

Frequency

Frequency

Mothers

 

 

-.85

llll

-.65 .55

-.25 -.05 .15 .35
Quadratic Estimates

-.45

Fathers

.75

 

 

-.85

 

 

W

 

i

-.65

.35
Quadratic Estimates

-.45 -.25 -.05 .15

124

 

Because there are only three timepoints, these correlations cannot be interpreted as
an index of the relationship between rate of change and acceleration or
deceleration. Rather, they suggest that change is associated with some instability in
trajectories; individuals tend not to follow a straight linear path, but instead may,
for example, rate the parenting alliance as higher at 3 months than at l or 6
months. However, the differing directions for men and women do indicate that the
greatest discrepancy between men and women’s parenting alliance scores is at 3
months post-birth, which is supported by the finding that the lowest correlation
between mother and father ratings is at 3 months.

E M 1A: lePr i o oftheParen in 11'

Because the above results provided evidence for difl‘erent trajectories for
mothers and fathers, subsequent explanatory models built on the baseline couple
model. The first explanatory model evaluated Hypothesis 3 by using stable pre-
birth factors (as well as infant sex, which, although not a pre-birth variable, is also
a stable predictor) to attempt to account for individual differences in parenting
alliance trajectories. Means and standard deviations for and intercorrelations
among the predictor variables are presented in Tables 17 through 20. Because of
the large number of predictors, the variables were clustered into distinct subsets in

order to use the strongest predictors in the final model (Bryk et al., 1996).

Results for the separate analyses used to identify the most likely predictors

are presented in Appendix D. Based on these analyses, variables were selected for

125

.50. v9.3... .o. vat. .mo. vm... .Ecomfie 05 so 08 805$ was $0508 $0503 macaw—0&8
30:0me 05 3209 08 805$ 8m macaw—0:00 258$? 05 2690 0.8 30508 c8 05:22.80 .Ewto mo EBB.“ I On .0. Ho. Z

 

 

 

 

mm. 3. 3. 3. G. 3. m3 mm .553

3% 32 32 33 new Sm :3 .2- been

8. 3. E. ee. 3. 9.. 33 mares:

SN 30 w? 35 3a 3.”. 8.3 .2. 5:32

fl 3.2. .2. 2.2.- 3.- 3. *8. 8388M s

.8. 3 t... K. 1.3.- 3. 2. 3.- 385: 358 em .e

2. :33. _lo. tame.- 3. 3. 3.- 33.355 0a .n
9230:th

e _ .- 13m.- ..19... 3... 8.- :.- S. 3323 02:83 0e .3.
«Eoocoo

3.- 3. 3. S. 31. 3. .8.- meeaeeeie 03302 .m
:28388

1.2.. S. 3. 3. 2 .- q 3.- 35323”. gee-em .m

3. :. 3. S.- 2. 3.- .....3 50320.60 emm ._

e e m e m N _ 838;
fig

 

 

: £0522 c8 w03ulula> 58605 054-58320 3:232: 505-05 :oE< 0502 new 80:20 0825

: use-3

126

Aoo. V9.20... Ao- Vmi. .mo. Va...

000005 0000 0050000000 00500000 .00 0000000000 0 0m mmmg .0w0 0050.0 000 005000 .00 00000 05 00 0m< .

2000m05 05 00 000 00050.0 000

0005000 000500 0005200000 2000w05 05 3200 000 00050.“ 00.0 0005200000 2000m0=0 05 0>0n0 000 0005000 00.« 0005200000 mg

 

 

 

 

8. 8. 8.8 0.0. 020000
8. 0.0.. 8. 8. 00.0. I

8.0 3.0 8.80 02 02000

8. 8. 8.0 04008002
8. 8. 3.0 8.8 8.8 I.

00.0 3.0 8.80 2 0082

0000000005
8.- 8.- 8. 3.- 8. 8.- 8.- 808800020000050 .0
8.- 00. 8.- 8 0 .- m 0 .- N0.- 8.- 0880000000 030809002 .8
8.- 0 0. 8. 8. 8. 8.- 8 0. 080200000 50500802 .0
8.- 8. o 0. 1.3.. 8.- 0 0 .- 8.- 8. 08080200 030008 0300002 .0
8. o 0 .- 8. 8.- 18.0.. *8. *8. E. 8208025 0300200 2000080 .0.
8. 8. E.- .........8.- *8. 3:81. 8. 8.- 0588.08 30.08 8082-0000 .0
8.- .2.- 8.- La.- :. o0. 3:00. .000 .N
8.- 8.- 80. 2.. 8. 8.- 30.8. .000. .0
0 8 0 0 0. m N 0 0288>
8.4980

 

 

 

0 000502 05 003200> 0000:0000“ 0520000000000 5050000000 505-05 008< 00002 000 000520008025
00 0300

127

.5. v9... .3. Va...

.0885 23 £23988 £23033 mo 260928 a 2 mMmn~

.0? $52 98 .552: no :38 05 2 $5

.mooceo-a€ n Mac 638888362823

n Emma 6236052“ H >D< Eatfitofisa H HD< 3:28.285 n .8885 35852.23 H .3533 mfimto mo £55 n Om .mflfl

 

 

128

8. 3. *3. 2. 8. tom.- 3- 2. 8:983. s
No. 8. 3. 3.- 2.- :.- 3.- 2.- 285: .938 On .3
2. 3.- 8. S. 8.- 2.- 2.- :.- 83353 On .m
2585.832
_ _.- No. 8.- 3.- 8. no. 3.- 3.- 32.23 2,283 8 .e
mac-0:00
2.- 2.- 2.- 3. 8.- E .- 3. S. 332220 2:302 .3
:28388
2.- S. 2.- 2.- 8.- :8. 3.- 2.- mag-522° 03:83 .N
S. 3.- 3.- 3. S. 8.- oo. 3.- “5832033 omm ._
.23 .26 >3. .26 .22 .6225 .3825 .523 .25 .03 8§E>
32mm 355 356 .958
03332 03:25 8:82
.23

 

 

w0_£wg> howomfiokm ozwtouoaumfiv EHSXDHCOU flaw —mD—v_>:u£ fitmmuoum Mug-:02 €05 mcomfifiobo

 

2 29¢

.83. vat: .8. vm...

.0885 28 £289.08 £23260 mo 0:89:00 0 2 mum;

0? 0050-“ 98 “0508 .«o 508 05 mm 039

.m0oc0u0tmv u “:6 ”03000853202803

.1- San-E 633E053 u >D< ”5:02:05?“ u HD< ”8288005 n .8802: 35803.80 n 4253 mimic mo 3:5 u Om .MSIZ

 

 

2. mo; :.- mo. 5. 3.- no: 2.- 00:0800M <-
3- ...8. 8.- 3.- ...2- 2:3..- 2.- 8.- 8°53; .358 On .0
o". 3. 3; no. 3.- LVN.- 3: nor cows—$53 Om .m
2:30:th
w _ .- 8. 8.- 8.- w _. tan. 3. 8.- 885.3 2:88 Cu ..‘
mE0ocoo
o_.- 2. mo. wor vor mo: 2.. S. misfit—£0 0>30w02 .m
.8203on
S. n 2. 3. cor oo.- 3. 3.- C .- war—«09220 03225 .N
nor 8.. . m _ .- *ON. $1.5”. No; 3. Z. 6083—03—0 0mm .—
u-Eu u-Eu >D< “:6 HD< 4080:: .8302: 4253 ammm .0w< 3052:;
San-E 32.88 .02-SE But-«E
0>zmw0z 03:00am c0280“
-.=0m

 

 

83%; 88605 2858220 338:8 25 3:03:05 £25.05 .2059.— =oE< macaw—02o

om 03.0-H

129

or.

aSE
mt
traj
adc'
per-
per-
011g
mat-
33¢-
59ml”

all p.

further investigation if p< .10. Potential predictor variables are listed in Table 21.
As might be expected, these individual analyses indicated that the marital
variables were the most robust predictors of the parenting alliance trajectory, but
only for the level (e. g., whether the parenting alliance was perceived as stronger or
weaker in general). Similar findings emerged for SES, although there was a trend
for SES to predict the maternal quadratic parameter as well. Ego level was
associated with fathers’ parenting alliance level, but not mothers’. Childbearing
motivations and concerns were also consistently related to the parenting alliance
trajectory, but to slope and quadratic functions, depending on the parent, in
addition to the intercept. Differences in childrearing authoritative and
permissive/protective childrearing philosophy were associated only with mothers’
perceptions of the parenting alliance, but fathers’ perceptions of the family of
origin predicted both mothers’ and fathers’ parenting alliance curves. Finally,
maternal reactance predicted the maternal curve parameter. Child sex, parental
age, fathers’ childbearing concerns and negative marital interactions, mothers’
family of orgin variables, and differences in authoritarian beliefs were unrelated to
all parenting alliance parameters.

Selected variables were examined simultaneously, thereby controlling for
the influence of one another, and entered only for the parameter(s) with which the
preliminary exploratory analyses indicated they were associated. This procedure

resulted in most of the trends at the preliminary exploratory level becoming

130

Table 21

Variables Selected for Further Exploration Through Preliminm Analyses

 

 

Variable Mother Father
Individual
Ego development X
Assimilation
Positive childbearing motivations X x
Negative childbearing concerns . X
Family of origin
Triangulation X
Positive parenting X
Marital conflict X
Reactance X
Demographic context
Age
SES X
Relational context
Marital relationship
Self-reported marital adjustment X X
Positive marital interactions X X

Negative marital interactions X

Differences in childrearing philosophy

Authoritarian
Authoritative X
Permissiveness/protectiveness X

Infant sex

131

insignificant (e. g., all father family of origin variables, even when entered
individually, and differences in authoritative philosophies). In addition, decisions
were required regarding which marital variables to retain, as both mother and
father variables and self-report and observational variables tended to be associated
with each parent’s trajectory and thereby cancel one another out. The decision was
made that when both mother and father versions of the same variable were related,
variables based on report or observations of the same parent whose parenting
alliance rating was being predicted would be retained. Similarly, when both self-
report and observational variables were associated but suppressed one another, the
self-report measure was be retained since a method effect would be expected to
make self-reported parenting alliance and marital quality related. A number of
different explanatory models were possible, and despite the choices made for the
final model, it could be argued, that a more interesting approach would be to keep
in the observational measures or the partner’s perceptions, thereby decreasing the
chance of a method effect and examining cross-parent systemic influences.

The final explanatory model, based on the above criteria, for stable pre-
birth predictors for parenting alliance trajectories for each parent, controlling for
the associations with the other parent, is presented in Table 22 (one trend is
included for mothers and two for fathers). Overall, more pre-birth predictors were
associated with mothers’ rather than fathers’ parenting alliance trajectories.

Furthermore, prediction of parenting alliance level was more successful than of

 

132

 

\ -.i...-< {mm-u.

Table 22

Final Explanatog Model A for Stable Pre-Birth Predictors

 

 

Parameter Unstand. coeff. 1(100) Effect size _r_
Mothers

Level 3.41 126.55*** ----
SES .18 4.67*** .47
MMARNEG -.14 ~4.18*"‘* .42
MCONCERNS -.18 -3.13** .31
FDAS .01 ' . 2.85“ .29

Rate -.00 _ -.O9 .01
PERMDIFF -.08 -3.17** .32
MMOTIV .06 1.70t .17

Curve .00 .06 .Ol
PERMDIFF -. 12 -3.34*** .33
MREACT .23 2.23“ .22
AUTVDIFF .06 1.98* .20

Fathers

Level 3.60 147.55*** ----
FDAS .02 6.33*** .63
SES .10 2.7 8*" .28
FEGO .01 2.39“ .24

Rate -.01 -.95 .10
FMOTIV -.OS _ -l.8'8t .19

Curve .02 .94 .09
MMOTIV .06 1.91t .19

 

ote. Unstand. coeff. = Unstandardized coefficients. M_ and F_ denote mother and
father report, respectively; CONCERNS = negative childbearing concerns; DAS = Dyadic
Adjustment Scale; MARNEG = negative marital interactions; MOTIV = positive
childbearing motivations; EGO = ego development; AUTVDIFF = differences in
authoritativeness; PERMDIFF = differences in pennissiveness/protectiveness; MARPOS
= Positive marital interactions. tp_ < .10. *p<.05. **p<.01. ***p<.001.

 

133

par
pre
and
van
con!
para
their

dism

posit
and;

man

mat}
Part:
“3). l

«Jami

glmgg parameters. For the most part, effect size rs for parenting alliance level
were solidly in the medium range, but were small to medium for the change
parameters. After taking parental sex into account, among mothers, the sets of
predictor variables accounted for 40% of the variance in level, 11% in the SIOpe,
and 18% in the quadratic, and among fathers, predictors accounted for 33% of the
variance in level, 2% in the slope, and 6% in the quadratic. Individual and
contextual characteristics did not differentially predict parenting alliance
parameters (i.e., neither was specific to a particular type of parameter), confirming
their use as a heuristic convention rather than as a meaningful method of
distinguishing between levels of influence.

For the level of the parenting alliance, higher SES was related to more
positive perceptions of the parenting alliance for both mothers (effect size L = .47)
and fathers (effect size 1 = .28). In addition, both self-report and observational
marital variables predicted parenting alliance levels. Interestingly, better pre-birth

marital satisfaction for fathers was associated with a more satisfactory parenting

 

alliance after the birth for both parents, although the association was particularly
strong for fathers (effect size L = .29 for mothers and .63 for fathers). Even when
mothers’ DAS ratings were used instead of fathers’ ratings to predict maternal
parenting alliance levels, the estimate only reached the trend level (effect size L =
.18). For both parents, m marital interactions were unrelated to parenting

alliance trajectories after accounting for paternal marital satisfaction. However,

134

mothers, but not fathers, who exhibited more pre-birth m marital behaviors
rated the parenting alliance as significantly weaker (effect size I = .42). Among
mothers only, negative childbearing concerns were related to a less satisfactory
parenting alliance (effect size I = .31), and among fathers only, higher ego
development was associated with a more satisfactory parenting alliance (effect
size I = .24).

Significant predictors of change were found only for mothers, with
differences in childrearing philosophy and reactance emerging as the critical
variables. When one spouse endorsed a more indulgent, child-centered
(permissive/protective) stance toward parenting relative to the other spouse,
mothers were more likely to feel that the parenting alliance deteriorated over time
(effect size L = .32). Moreover, the greater the differences in permissiveness/
protectiveness g in authoritative philosophies, the more likely mothers were to
experience fluctuations in their perceptions of the parenting alliance (effect size L
= .33 for permissiveness/ protectiveness, .20 for authoritativeness). Greater
maternal reactance was also predictive of instability in maternal parenting alliance
ratings.

In addition to these significant findings, three trends emerged for positive

 

motivations for childrearing as a predictor of change for both parents. Among
mothers, the more positively pregnant women anticipated the birth of children and

focused on the rewards of childrearing, the more likely it was that their parenting

135

alliance would become increasingly satisfactory over time (effect size I = .17) and
that fathers would report fluctuations in their ratings of the parenting alliance
(effect size I = .19). A trend for fathers’ reports of childrearing motivation

suggested that more eager fathers may experience a decrease in the parenting

 

alliance by 6 months post-birth (effect size 1 = .19).

Explanatory Models B: Process Variables and Parenting Alliance Trajectories

 

The effects of post-birth time-varying variables on parenting alliance
trajectories were examined for marital adjustment, violated expectations for the
division of childcare and housework, parenting efficacy, and temperament. Means
and standard deviations for the predictors at 1, 3, and 6 months are presented in
Table 23. To the baseline couple model, between-couple time-invariant scores,
denoting the mean score of the predictor variable across the three timepoints, were
added at Level 2. At Level 1, the time-varying component, assessed by deviation
scores at each timepoint from the individual’s mean, was included as a fixed effect.
Results are shown in Tables 24 through 28. Significant effects emerged for all
variables except perceptions of fussy temperament, which evidenced a single
trend.

Post-birth marital adjustment showed the largest effects on the parenting
alliance. As predicted, each parent’s overall mean marital quality was positively
related to his/her own level of parenting alliance (effect size r = .44 for both

parents). There was also a trend (9 = .053) for fathers’ level of parenting alliance

136

Table 23

 

Means and Standard Deviations for Post-Birth Predictor Variables by Assessment

 

 

 

1 month 3 months 6 months
Vafiable _M S_D. M SD. M. .82
Marital adjustment
Mother 150.64 10.33 149.17 12.11 146.52 13.20
Father 150.96 11.09 150.62 11.06 149.43 11.70
Violated expectations
for childcare
Mother -32.33 47.59 -28.43 48.22 -15.26 52.56
Father 43.28 42.29 31.61 46.99 18.47 51.15
Violated expectations
for housework
Mother -9.69 88.09 -26.59 72.50 -28.81 76.99
Father 47.50 86.31 73.81 83.07 69.11 89.46
Parenting efficacy
Mother 3.23 .38 3.49 .29 3.53 .29
Father 3.06 .34 3.24 .32 3.34 .32
Fussy temperament
Mother 3.27 1.01 2.87 1.00 2.67 .94
Father 3.36 .88 3.02 .86 2.88 .87

 

a;

137

Irv-t

Table 24

Explanatory Model Bl: Relationships with Post-Birth Marital Adjustment

 

 

Parameter Unstandardized t(100) Effect size 3
coefficient
Mothers
Level 3.41 118.97*** ----
MDAS mean across time .02 4.43*** .44
FDAS mean across time .00 .95 .10
Rate .03 1.53 .15
Curve .00 .06 .01
Fathers
Level 3.60 154.83*** -----
MDAS mean across time .01 1.93‘ .19
FDAS mean across time .01 4.35*** .44
Rate -.00 -.33 .03
Curve .02 1.09 .11
Father’s change in DAS -.01 -5.26*** .53
Mother’s change in DAS -.01 -6.57*** .66
Father’s partner’s change in -.00 -.58 .06
DAS
Mother’s partner’s change in -.00 -.18 .02
DAS

 

fiote. DAS = Dyadic Adjustment Scale; MDAS = mother report; F DAS = father
report.

‘2 < .10. ***p<.001

138

Table 25

Explanatory Model B2: Relationships with Post-Birth Violated Expectations for
Division of Childcare

 

 

Parameter Unstandardized t(100) Effect size I
coefficient
Mothers
Level 3 .41 102.3 8*** ----
MVIOC mean across time .00 2.62* .26
FVIOC mean across time .00 .01 .00
Rate -.03 -1.82‘ .18
Curve .01 .40 .04
Fathers
Level 3.60 127.24*** -----
MVIOC mean across time -.00 -.S7 .06
FVIOC mean across time .00 2.56* .26
Rate -.01 -.75 .08
Curve .02 .98 .10
Father’s change in VIOC -.00 -.26 .03
Mother’s change in VIOC -.00 -3.02** .30
Father’s partner’s change in .00 .82 .08
VIOC
Mother’s partner’s change in .00 2.12* .21
VIOC

 

NEE. VIOC = violated expectations for division of childcare; MVIOC = mother
report; FVIOC = father report.

‘2 < .10. *p<.05l **p_<.01. ***p<.001

139

lablz‘

r-r-r
I; p.
a; . ‘3
(I) :.T‘

Mot

Path.

Falhl

11ml

”PM

:50

Table 26

Explanatogg Model B3: Relationships with Post-Birth Violatid Expectations for
Divison of Housework

 

 

Parameter Unstandardized jt_(100) Effect size I
coefficient
Mothers
Level 3.41 98.12*** ----
MVIOH mean across time .00 .07 .01
FVIOH mean across time -.00 -1.00 .10
Rate .00 .07 .01
Curve .01 .28 .03
Fathers
Level 3.60 120.16*** -----
MVIOH mean across time .00 .88 .09
FVIOH mean across time .00 .93 .09
Rate -.01 -.87 .09
Curve .02 1.13 .11
Father’s change in VIOH .00 .14 .01
Mother’s change in VIOH -.00 -2.56* .26
Father’s partner’s change in -.00 -.92 .09
VIOH
Mother’s partner’s change in -.00 -.83 .08
_VIOH

 

Note. VIOH = violated expectations for division of housework; MVIOH = mother
report; FVIOH = father report.

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

140

Table 27

Explanatory Model B4: Relationships with Post-Birth Parenting Efficacy

 

 

Parameter Unstandardized 1(100) Effect size 1
coefficient
Mothers
Level 3 .41 102.75*** --—-
MEFF mean across time .36 2.92** .29
F EFF mean across time .18 1.48 .15
Rate -.03 -1.23 . 12
Curve -.01 -.26 .03
Fathers
Level 3.60 130.71*** -----
MEFF mean across time .22 2.20* .22
FEFF mean across time .33 3.49*** .35
Rate -.01 -.45 .05
Curve .02 1.10 .11
Father’s change in EFF .05 .89 .09
Mother’s change in EFF -.10 -1.28 .13
Father’s partner’s change in -.00 -.03 .00
EFF
Mother’s partner’s change in -.07 -.84 .08
EFF

 

Note. EFF = parenting efficacy; MEFF = mother report; FEFF = father report.

*p<.05. **p<.01. ***p<.001.

141

Table 28

Explanatory Model BS: Relationships with Post-Birth Perceptions of Iriant Fussy
Temperament

 

 

Parameter Unstandardized 1(100) Effect size r
coefficient
Mothers
Level 3.41 97.75*** ----
MFUSS mean across time -.02 -.43 .04
FFUSS mean across time -.00 -.07 .01
Rate -.02 -.88 .09
Curve .00 .08 .01
Fathers
Level 3.60 124.87*** -----
MFUSS mean across time -.01 -.25 .03
FFUSS mean across time -.10 -1.88‘ .19
Rate . -.01 —.77 .08
Curve .02 1.05 .1 1
Father’s change in FUSS .02 .85 .09
Mother’s change in FUSS .04 1.26 .13
Father’s partner’s change in -.02 -1 .13 .11
FUSS
Mother’s partner’s change in .00 .10 .01
FUSS

 

Note. FUSS = perceptions of infant fussy temperament; MEFF = mother report;
FEFF = father report.

tp < .10. ***p<.001

142

to be predicted by mothers’ marital satisfaction. Unexpectedly, however, change

 

in each parent’s marital adjustment was negatively related to change in his/her
own perceptions of the parenting alliance, with stronger effects for mothers (effect
size r = .66) than fathers (effect size I = .53), x2 (1, _N_ = 101) = 75.93, p < .000.
That is, increases in the parenting alliance were accompanied by decreases in
marital satisfaction after controlling for change in the partner’s relationship
quality. The partner’s marital satisfaction was not related to either the level or
change in the parenting alliance.

As expected, violated expectations for the division of childcare showed
stronger associations than for the division of housework, with small- to medium-
sized effects. For both mothers and fathers, more violated expectations in a
negative direction (i.e., doing more than anticipated) for childcare overall were
related to more negative perceptions of the parenting alliance (effect size _rs = .26
for both parents). However, change in violated expectations for childcare was
associated only with change in mothers’ parenting alliance. Increases in mothers’

violated expectations predicted a decrease in maternal perceptions of the parenting

 

alliance (effect size r = .30), and increases in fathers’ violated expectations

predicted an increase in maternal perceptions of the parenting alliance (effect size

 

r = .21). Notably, after taking into account change in violated expectations for
childcare, the r_ajt§ of parenting alliance change begins to approach significance for

mothers, suggesting a possible decline in satisfaction with the parenting alliance

143

fi‘

161

.31“;
“5513

leffr.

over time that is masked by increasing satisfaction with the division of childcare
as the infant ages. Change in violated expectations for childcare did not predict
change in fathers’ parenting alliance. For violated expectations for the division of
housework, only one significant effect was evident, with mothers’ increases in
violated expectations predicting a decline in perceptions of the parenting alliance
(effect size r = .26). Change in fathers’ parenting alliance and level of parenting
alliance for both parents was not related to violated expectations for housework.
In partial support of the hypothesis, parenting efficacy was associated with
level of parenting alliance, but not with change, and resulted in small to medium
effects. For mothers, greater maternal efficacy predicted a better parenting alliance
(effect size I = .29). For fathers, the efficacy of m parents was positively related
to parenting alliance (effect size 1 = .22 for maternal efficacy, .35 for paternal
efficacy), with stronger effects for father’s self-efficacy, x2 (1 , _N_ = 101) = 20.12, p
< .000. The effect of mother efficacy on mother parenting alliance was stronger
than the effect of mother efficacy on father parenting alliance x2 (1, N = 101) =
9.54, p < .01, but less than the effect of father efficacy on father parenting
alliance, x2 (1 , N = 101) = 20.24, p < .000. Perceptions of infant difficult
temperament were not strongly related to parenting alliance trajectories, with the
exception of one trend in the expected direction: Father perception of infant
fussiness tended to be associated with lower paternal parenting alliance ratings

(effect size r = .19).

144

Post hoc analyses. The growth curve analyses demonstrated that in some
cases, change in the parenting alliance and change in other process variables were
related, but did not address causality within those change relationships. To
examine causal effects in the development of the parenting alliance over time,
structural equation modeling was used to test a cross-lagged stability model for
each of the significant change relationships identified in the HLM analyses (i.e.,
parenting alliance and marital adjustment for both parents, and parenting alliance
and violated expectations for childcare and housework for mothers only). The
model controls for the stability of each variable and assumes that the residuals of
the criteria at 6 months are uncorrelated, so that associations between the factors
at 6 months are completely accounted for by earlier predictors (Rogosa, 1979).

Initially, an attempt was made to use latent variables formed from the two
parallel parenting alliance scales and the four DAS scales, but problems with error
variances prevented an admissible solution. Instead, the single general alliance
and overall DAS dyadic adjustment scores for each timepoint were analyzed.
Furthermore, as might be expected, constraining the correlation between parenting
alliance and marital adjustment to zero resulted in admissible solutions but poor
overall model fit. Nonetheless, this analysis should still enable the evaluation of

causal paths.

Figures 11 through 14 display the coefficients associated with each path (N

= 95 mothers, 94 fathers). Of the five analyses, only the model testing the

145

 

K N ad.\-.l\..~..\

Ace. V 9.3.3.. ._o. V as...”
.maoofio
:2sz 886 Emu—bawfi mouocov ovate—om .323 0.8 3053 he,“ .268“ 2m $059: How $56508 wouzflmucflm .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

802
*eiwo. anion.
:39. .38.
35:3 028:? mecca—E
wasp—2mm wages,” 532$
0o mo. no. Nor rm.
8 8 :3. 8.- runs. 6
M
23533 A 6253.8.“ 285%:th E08633
3:32 .352 . 32.8.2 . 3:32
«name. sit—w ***_m
***Nh. iiivu. «aimw.
5:2: 0 5:2: m 5:2: ~ £59-05

 

 

 

 

 

 

 

 

 

 

 

 

 

 

: oSwE

Aoo. Vm***

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

85% :12. 855a :.8. 38%
3.- 3.- 8... 8.
***cv.
28230 Sm 28230 .89 0.30230 Sm
833832 9833898 2853035

 

 

833$ .53. 3305 :.Q. 3335

 

 

 

 

 

 

 

 

 

5:08 o 5.88 m 558 ~

 

S oSwE

147

 

w . \ rdLs- MNNRN

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

85% ran: 85% 2:8. 38::
mgfiwam 33:23 39:88
o _ .. no. 8. 3.

030220 :8 chow—Eu 8.: 28220 :8
8033891 3658393 333035
Bum—e? . 33¢? . 323.»
355.3.— ..i: «anon-Ham :amw ragga
5:08 9 5:8. m 5:9: _

agawmr

.So. vat...

 

148

 

 

 

 

35:?
Quantum

 

 

 

.8383: .8
3038898

3%;

     

 

8&2:

aaaNc.

 

 

35:98.”

 

 

 

r8383: 5
8635830

 

 

 

5:2: 9

{*aee.

3%;

 

h

 

 

£58 m

       

 

85:?
95:93.“

 

 

 

838.5: 5
goufiooqxo

32o;

  

 

 

5:2: ~

.88. vat... .8. vm...

 

E 2:me

149

relationship between the parenting alliance and marital adjustment showed
significant causal paths. For both mothers and fathers, better pre-birth marital
quality predicted more positive ipipial perceptions of the parenting alliance. In
addition, for mothers, marital quality at 1 month also predicted the parenting
alliance at 3 months. However, no instances emerged in which the parenting
alliance predicted marital adjustment. These results suggest that the quality of the
marriage prior to the child’s birth does indeed set the stage for the development of
the parenting alliance, and that for mothers in particular, marital quality in the first
month continues to affect their satisfaction with the coparenting relationship. By
three months, however, the parenting alliance appears to become less vulnerable
to change in the marital dyad. Notably, at no point does the parenting alliance
affect marital satisfaction, nor does a direct causal effect at any time appear

between the parenting alliance and the division of labor.

150

DISCUSSION

The current study is an exploratory and descriptive account of the
development of the parenting partnership from its initiation, the birth of the first
child, through the ensuing 6 months. Using multiple measures of coparenting and
marital quality, the specificity of the parenting partnership construct and the nature
of the association between the coparenting and marital relationships were
examined. In addition, paralleling the 20-year literature on marital change over the
transition to parenthood, developmental trajectories for the subjective parenting
alliance over the first 6 months were investigated. First, average parenting alliance
trajectories for the sample as a whole and then for mothers and fathers separately
were defined. Factors measured prior to the child’s birth were subsequently used
to account for individual differences in the level and change in perceptions of
coparenting. Finally, associations between change in perceptions of the parenting
partnership and change in post-birth time-varying factors were examined and an
attempt was made to determine causal relationships.

Because this research was conducted with a comparatively well-educated
and ethnically restricted sample whose attitudes and knowledge about parenting,
relationship qualities, and resource availability may differ from those of other
groups, it is best considered as a starting point for further investigation. When the

findings for the present study are considered as a whole, however, four related

151

points emerged that suggest a number of avenues for future research with both
similar and more diverse groups. First, although undeniably linked to issues
nonspecific to the child, the results of both the CFA and HLM analyses confirm
that the parenting partnership is a separate component of the couple system from
the marital relationship, with different trajectories and correlates. Second,
al though the parenting partnership and the marital relationship were confirmed as
specific aspects of the couple relationship, the marital context was shown to
provide a fundamental framework for the development of the initial coparenting
relationship, consistently evidencing the strongest effects among all variables
investigated. Third, different pathways and correlates of coparenting development,
as well as distinctive associations among relationship dimensions, attest to the
Specificity of relationship experiences for men and women. In particular, the
I“5381116 suggest that men tend to experience an overall relationship quality that
generalizes across multiple subsystems of the family, but that women’s
perceptions of familial relationships are more contingent upon the specific
c<>litext. Fourth, this study highlights the dynamic process of the couple
relationship as an interacting system of two individuals’ expectations, perceptions,
and personalities. Both mothers’ and fathers’ parenting alliance trajectories were
t‘Olllld to be related to certain characteristics of the other parent, and some
co171erence was evident in these cross-parent relations. Moreover, the results

SL‘ggest that the mother provides the impetus for the specific deve10pment of the

 

152

parenting partnership, at least in the early months, directly as well as indirectly
through couple interactions. Specific findings that support these conclusions, and
their implications, are discussed below.
The Construct Validity of the Parenting Partnership
The essential question in establishing the construct validity of the parenting
partnership is whether aspects of the couple relationship specific to mutual
parenting can be distinguished from those that exist regardless of the couple’s
parenthood status (e. g., finances, intimacy, dealings with fanrily and fiiends). Prior
research has evaluated whether self-reported parenting alliance ratings are distinct
from self-reported marital satisfaction (Abidin & Brunner, 1995; Floyd et al.,
1 998; Floyd & Zmich, 1991; Frank et al., 1986) or have reported correlations
between coparenting and marital observations (Katz & Gottrnan, 1996; McHale,
1 995), but none have simultaneously included subjective and objective
assessments of both relationship variables. To address this issue, Hypothesis 1 of
the current study assessed the validity of the parenting partnership as a distinct but
11313:1th construct from the marital relationship using both self reports and observer
ratings.
The results generally supported the hypothesis, although not by the means
Originally stipulated. In the context of a multitrait-multimethod confirmatory
factOr analysis of the four relationship measures at 6 months post-birth, the

parenting partnership and marital relationship trait factors were not separable from

153

the three method factors of father report, mother report, and observer ratings. On

the other hand, adoption of a simpler approach enabled perceptions and
observations of coparenting and marital quality to be discriminated. From this
perspective, a four-factor model, comprised of subjective and objective
coparenting and subjective and objective marital quality, provided a good fit to the
observed data. These results suggested that each measure assessed a differentiated
aspect of the couple relationship, with the four factors best conceptualized as
multiple dimensions within the overall dyadic relationship.

Although four separate dimensions were identified, significant correlations
among several of the factors uphold the linkage between the coparenting and
marital relationships. Indeed, inspection of the factor intercorrelations reveals the
difficulty in partitioning out the hypothesized trait and method variance, which
Was further compounded by different patterns of association for mothers and
fathers. With respect to the hypothesized traits, self-reports and observations of
the parenting partnership had been postulated to form a distinct latent variable.
HOWever, the coparenting variables were significantly correlated only for fathers,
and the degree of correlation was weak. This suggests that the two measures
taPped quite different aspects of coparenting, with satisfaction with the parenting

endeavor not necessarily reflected in warm, cooperative dyadic interactions during
play with the child. More consistency was manifested between the two marital

relationship dimensions, particularly for mothers; nonetheless, the CFA results

154

indicated that the marital self-reports and observations also were more
appropriately conceived as separate dimensions.

The method of assessment also showed parent-specific relations. Whereas
ratings of the parenting alliance and marital adjustment were positively associated
for both parents, pointing to a potential method bias in the use of self-reports, the
observational measures revealed striking differences. Coparenting and marital
observations were strongly related for fathers, but not significantly correlated for
mothers. Thus, when parents behaved harmoniously in the coparenting context,
fathers were strongly disposed to display warmth, support and communicativeness
during the marital interview; yet mothers’ behavior in one setting was unrelated to
their behavior in the other setting.

The consistency of the model structure across parents confirms that
mothers and fathers experienced and displayed similar markers of relationship
quality. On the other hand, the different patterns of factor intercorrelations suggest
that they comprehended the complexities of the post-birth couple relationship in
different ways. Mothers tended to perceive both aspects of the dyadic relationship
collgruently; happiness with the parenting alliance was accompanied by
Satisfaction with the state of the marriage. Furthermore, how mothers felt about
@- dimension of the couple relationship was mirrored in their maria;
interactions, with correlations among the three dimensions ranging from .58 to

'84- Thus, mothers were likely to feel similarly about both aspects of their couple

155

relationship and behave in a way that was compatible with those feelings when
alone with their Spouse. These findings replicate those of Floyd and Zmich
( 1 991), although the pattern of results held for mph parents in that study.
Coparenting behaviors (which were not assessed in the Floyd and Zmich study),
however, appeared to be separate and unrelated to other maternal feelings and
behaviors (correlations ranged from .25 to .37 and none reached significance);
whether mothers enjoyed either aspect of the couple relationship and
communicated this to their spouse within the marital context were not consistently
related to the degree of positivity expressed toward the spouse during family play.
The four dimensions were far less distinct among fathers, with correlations, all
significant, ranging from .34 to .66 and supporting previous research (Abidin &
Brunner, 1995; Floyd & Zmich, 1991; Frank et al., 1986). The strongest
associations for fathers were between the two self-report dimensions and between
the two observational dimensions, suggesting that fathers were more prone to a
general method effect than were mothers. Put otherwise, fathers tended to feel
about their dyadic relationship one way, and act in the context of that relationship
dufing marital and family interaction tasks in a way that was relatively consistent
With but not identical to their subjective experience.
Taken as a body, the results point to the importance of examining the
Parenting partnership and the marital relationship, and in multiple ways, to reveal

a fIlller picture of dyadic functioning in the family context. This study confirms

156

that aspects of the couple relationship defined by childrearing are both different
from and linked to those that are nonspecific to parenting (Abidin & Brunner,
1 995; Cohen & Weissman, 1984), and are experienced somewhat differently by

mothers as compared to fathers, but additional research is necessary to delineate

particular correlates of each domain.

Parenting Alliance Trajectories

Whereas Hypothesis 1 included observations of coparenting behaviors, the
remainder of the study, which examined trajectories of the parenting partnership
over the first 6 months, focused exclusively on self-reported perceptions of the
parenting alliance. The first step was to identify an overall developmental
traj ectory for the sample as a whole, based on assessments at 1, 3, and 6 months
post-birth. Lacking precedent, Hypothesis 2 did not presuppose a specific
direction for change, but postulated that significant residual variance would
remain for further exploration of individual differences. Not surprising in this
group of well-functioning adults, the parenting alliance was generally considered
to be very satisfactory and remained so across all three assessments. No evidence
emerged for an initial honeymoon period followed by disillusionment and
dissatisfaction with the coparenting relationship, nor did couples appear especially
prone to conflicts during the newborn period and subsequently negotiate more

effective parenting partnerships.

These results contrast with the body of research on marital adjustment over

157

the transition to parenthood, in which small but consistent declines in marital

functioning are the rule (Belsky etal., 1985; Belsky & Rovine, 1990; Belsky et al.,

1 983; Levy-Shiff, 1994; Miller & Sollie, 1980; Waldron & Routh, 1981).
Notably, marital satisfaction did indeed decline significantly in this group, albeit
only for mothers. Additional support is provided, therefore, for the distinctiveness
of the coparenting and marital components of the couple relationship, especially
for women. Why do perceptions of marital quality decline while the parenting
alliance remains consistent? “Transition” may consist of different events for
different aspects of family relationships. The marital relationship, as an existing
relationship with established patterns, may be vulnerable to destabilization in the
event of a child’s birth; thus, for the marital relationship, the first-time parenthood
actually is a transition from one set of organized processes to another. In contrast,
the parenting partnership gums upon the first child’s birth. Having no
Organization prior to the child, the transition to parenthood is not a transition per

Se for the coparenting aspect of the couple relationship. Instead, we can conjecture

 

that the parenting alliance is stable within developmental periods (e.g., infancy,
toClcllerhood, and so forth), but becomes less stable during u’ansitions from one
Period to another as parents encounter different challenges and demands. Thus,
We tnight expect to see a stable parenting alliance until a relevant developmental

t1”allsition that should require parents to renegotiate the coparenting system, such

as the move into toddlerhood, is encountered.

158

Despite the remarkable consistency of the average parenting alliance,
support for Hypothesis 2 was apparent as significant variance remained to be
predicted in the level, rate of change, and degree of variability over time. The
mean parenting alliance trajectory was therefore misleading; depending on the
cutoff used to delimit change, as many as half of the parents reported either an

increase or decline in their ratings of the parenting partnership over the course of
tfre study. In addition, many of the trajectories deviated from a straight linear
progression. In the context of this short-term study, the quadratic or curve
parameter indexes instability in the new relationship. The finding that fluctuating
perceptions were fairly common supports the expectation that experiences of the
parenting alliance might be especially marked in the first few months after the
infant’s birth as the inexperienced mother and father engage in the process of
negotiating their parenting partnership.

Furthermore, cross-time correlations indicated that parents showed the
greatest disparity in their ratings when children were 3 months old, suggesting that
Parents’ perceptions of the coparenting relationship may have been moving in
oPposite directions since the 1-month assessment. A couple of factors are likely to
c0IltJibute to this shift. For one, many mothers return to work within the period
betVVeen 1 and 3 months after birth. Employed mothers continue to assume the
maj ority of the childcare and housework, as husbands of working women are no

mote likely to do housework than husbands of nonworking women (Shelton &

159

John, 1993). The process through which this reorganization takes place is unclear,
but is likely to affect evaluations of coparenting support. In addition, changes that
take place in the infant around 3 months of age, including greater alertness and an
enhanced ability to interact socially, make this period a landmark for parents.
Fathers may desire more extensive contact with the increasingly interesting infant,
but renegotiation of coparenting roles may be required, since babies are often
virtually the sole responsibility of their mothers, who can be reluctant to relinquish
tlleir favored position (De Luccie, 1995).

Even among these couples, in which fathers were observed as a rule to be
quite enthusiastic and involved in parenting, mothers and fathers reported
different levels of parenting alliance. Although more pleased than not with the
parenting alliance, mothers were less likely than fathers to feel that parenting was
a mutual endeavor, confirming Floyd and Zmich’s (1991) findings with parents of
bOth typically developing and mentally retarded children. This is unsurprising, as
methers typically have the majority of responsibility for the child’s care, especially
during the period of early infancy (Cowan & Cowan, 1988). Fathers were
Considerably more confident that their spouse would provide assistance and
SuPport in meeting the child’s needs, as well as opportunities to take a break from
Parenting, than were mothers. Additionally, far more variability was evident in all
parameters for mothers than fathers, whose trajectories tended to take a more

conStrained path and imply a potential ceiling effect for their ratings. To illustrate,

160

as many as 60 percent of mothers’ parenting alliance ratings changed over the
course of the study, compared with only 34 percent of fathers’ ratings. Several
mothers could actually be considered somewhat dissatisfied with the parenting
partnership; in contrast, only one father reported real unhappiness with
coparenting. These results provide more evidence that mothers and fathers
experience the parenting relationship in different ways, and hint that maternal

evaluations of coparenting are more determined through a more complex process

than are paternal perceptions.

Pre—Birth Predictors of the Parenting Alliance

The finding that significant residual variance existed around all parenting
alliance parameters enabled the evaluation of Hypothesis 3, which addressed the
influence of factors measured prior to the child’s birth in predicting c0parenting
traj ectories. These variables were conceptualized as providing a relatively stable

context in which the parenting alliance develops. Although the predictor variables
Were classified into individual and contextual characteristics, the categories did
n0t differentially predict particular aspects of the trajectories. Instead, both
individual and contextual factors were related to parenting alliance level and
change, and all factors will be discussed from here on without regard to their
heIllistic classification.

Partial support was found for Hypothesis 3, with six domains measured

tlefore the child’s birth emerging as most critical in predicting subsequent

161

parenting alliance trajectories: the marital relationship, socioeconomic status,
differences between parents in childrearing philosophy, ego development,
reactance, and assimilation into the parenthood role. However, for the most part,
the 1§y_e_1_ of the parenting alliance was associated with different influences than
was c_hp_ng§. Specifically, for both parents, marital quality and SES predicted level
(as did assimilation for mothers only and ego deve10pment for fathers only),
differences in childrearing philosophy and reactance predicted change for mothers

and assimilation tended to predict change for both. The short-term nature of the

 

study and the composition of the sample may have restricted variability, and
systematic individual differences in overall satisfaction with the parenting
partnership were more easily discemable than in the rate of change or in
instability; approximately two to three times as much variance was accounted for
by the predictors in the level parameter as in either of the change parameters.
Predictors of change may be more identifiable in the context of a longer study
period.

Whether the parenting alliance was perceived as more or less successful

Was most contingent, as expected, on the pre-birth marital relationship. For
fathers, out of all the pre-birth variables, self-reported marital adjustment was the
\Srt‘ongest predictor of parenting alliance level. For mothers, however, both marital
Self-reports and interactions independently predicted parenting alliance level.

Furthermore, it was fathers’ reports of marital satisfaction that predicted mothers ’

 

162

perceptions of the parenting alliance; even when examined separately, mothers’
reports of marital adjustment reached only the trend level. Preliminary analyses
indicated that mothers’ perceptions of the marriage were strongly related to
parenting alliance reports; the nonsignificant results were most likely due to
covariance with maternal negative marital interactions. Mothers who manifested
conflictual, dominating behaviors prior to the child’s birth were more likely to rate
the parenting alliance as weaker. These results confirm a prior study by Laub
(1990), wherein couples in which the mother was dominant during a problem-
solving interaction reported the most unsatisfactory parenting alliances.

The strong relationship between the extant marital relationship and the
parenting alliance was anticipated; the question instead becomes, why do maternal
negative interactions independently predict mothers’ parenting alliance, and why
are paternal negative interactions completely unrelated? Perhaps the answer lies in
What is considered normative behavior for men and women. Women tend to
Communicate more clearly, display higher levels of positive affect, and be more
reSponsive during interactions with their spouse (Klinetob & Smith, 1996; Noller
& Fitzpatrick, 1991; Thompson & Walker, 1989). Thus, relatively frequent
Positive marital interactions would be more normative for women than for men.
Conversely, for men, confrontation, assertiveness, and domination are expected

relatively more than for women (Lips, 1997). Certainly, men exhibit positive

mteractions; nonetheless, “negative” behaviors are more normative for men than

163

for women.

This explanation may also account for the finding that only fathers’ ego
development was predictive of parenting alliance level. In the preliminary
analyses for fathers, both ego development and positive marital interactions
separately predicted parenting alliance level. However, the correlation between the
two resulted in only one being retained in the final solution. Ego development is
conceptualized as an underlying cognitive and personality characteristic that
guides behavior. Thus, more positive marital behaviors are likely to be evidenced
by fathers who have a greater capacity for perspective-taking and more investment
in mutuality; in short, who are at a higher level of ego development. These
behaviors are more expected of women; moreover, as a group, the women in this
study displayed higher levels of ego development than did the men, although a
meta-analytic investigation showed no sex differences in adults overall (Cohn,
1991). Thus, in some ways, those men with higher ego levels may have been
distinguished in a variety of ways from the general group.

When partners interact in nontraditional or unexpected ways, perceptions
of the parenting alliance may be affected. In particular, openly negative women
may be more critical of their partners and willing to voice this unhappiness when
addressing coparenting issues, although it is unclear whether this is due to a
general negativity in personality, the influence of depressive symptoms, or a

realistic view of problems that exist within the couple. Conceivably, negative

164

behaviors by women may mark a marriage that is more at risk than most, in that
these women may be less invested in the facilitative, organizational functions that
maintain the couple relationship and are the greater responsibility of women
(Gottrnan, 1994; Strazdins et al., 1997). Withdrawal is another marital behavior
that has been found to mark family problems, especially when displayed by men
(Paley et al., 1999). However, it was not included in the index of negative marital
behaviors in this study due to poor interrater reliability. Possibly, if withdrawal
had been included in the negative marital interaction score, father negative
behavior might also have predicted parenting alliance development.

It must also be reiterated that because the marital variables were related
across measure and across parent, the model described herein is only one of
several possible solutions. When compared to the results of the preliminary
analyses, the problems of using multiple domains to predict the parenting alliance
become clear. Had this study concentrated on the predictive effects of pre-birth
marital interactions, the results would have been quite different. In the exploratory
analyses, both parents’ perceptions of the parenting alliance were associated with
mothers’ positive and negative interactions, as well as with fathers’ positive
interactions. In short, only negative interactions from fathers were unrelated to
parenting alliance level. However, in conjunction with variables correlated with
marital interaction quality, only maternal negative interactions independently

predicted parenting alliance level.

165

In addition, the final solution reported herein was most conservative. The
decision criteria specified that when correlated variables effectively suppressed
one another’s influence, self-reports would supersede observational variables and,
following O'Brien et al. (1999), variables based on one’s own reports and
behaviors would supersede reports and behaviors obtained from the partner. A
case could be made for incorporating variables that are theoretically important
from a systems perspective; for example, marital observations could be used to
predict the parenting alliance rather than self-reported marital quality, and one
partner’s marital adjustment could be used to account for variance in the trajectory
of the other partner’s parenting alliance. Furthermore, a model nright be
appropriate in which the relationship between marital behaviors and the parenting
alliance is mediated by the evaluations of marital quality.

In addition to the marital relationship, for both parents, SES was predictive
of level, but not change. In other words, more educated, affluent parents rated the
parenting alliance as higher than did lower-SES parents, but were neither more
nor less likely to experience alterations in their coparenting evaluations over time.
More education and greater freedom from financial burden are likely to promote
an overall sense of satisfaction. Furthermore, these parents may have additional
resources, such as access to outside help and money for a babysitter, that decrease
the need to rely solely upon one another. Parental age, on the other hand, did not

independently predict parenting alliance ratings; younger parents were as likely to

166

be pleased with their coparenting as older parents. The previous literature has
been inconsistent with regard to the relationship of age with parenting variables.
In this sample, which included a number of university students, the higher
education level of some of the younger participants may have obscured relations
that may be apparent in a more varied sample.

Whereas the marital relationship and SES predicted the level of parenting

 

alliance, the magnitude of the differences between spouses in parenting
philosophies were related to parenting alliance flapgg. Specifically, differences
around perrnissiveness and protectiveness promoted an overall decrease in
maternal, but not paternal, parenting alliance ratings over time. Moreover,
differences in both permissiveness/protectiveness and authoritativeness were
related to instability in mothers’ perceptions of the parenting alliance. The
pernrissiveness/ protectiveness dimension focused on putting the child’s needs
before the parents, and included items such as “If parents are ready to go out for
the evening and their baby reacts by crying and screaming, it is best for them to
cancel their plans.” The authoritativeness dimension addresses similar issues, but
centers more on providing a child with security while encouraging autonomy.
Perhaps perrnissiveness and security is less an issue during the period of early
infancy, parents are likely to concur that the primary goal is to cater to the infant’s
needs and protect him/her at all costs. Over time, however, parents who differ in

their attitudes of child-centeredness versus parent-centeredness may find

167

themselves at odds if one spouse wishes to devote more time to the couple
relationship while the other continues to disregard mutual time in the service of
the child. As the infant grows older, therefore, feelings of mutuality and support in
parenting issues would be likely to lessen. Notably, mothers and fathers in this
study did not differ in their attitudes toward perrrrissiveness/protectiveness, despite
the contention of Belsky (Belsky et al., 1996) that evolutionary sex differences
related to protectiveness exist in coparenting. Why, then, were only mothers’
parenting partnership ratings affected by differences in childrearing philosophy?
On the one hand, mothers have been shown to carry greater responsibility for the
“emotional wor ” in the family (Strazdins et al., 1997). Conflict over differences
in childrearing might, therefore, predominantly affect mothers’ well-being and be
reflected in perceptions of the parenting alliance. On the other hand, the present
study only measured attitudes and perceptions, and provides no information about
ways in which conflicts were resolved. If, as seems likely, women were more
likely to acquiesce to their husband’s preferences, whether to be more or less
child-focused, they would certainly be prone to experience greater dissatisfaction
with coparenting as a whole.

The findings for differences in childrearing philosophy may be linked to
the findings for maternal reactance, the other predictor of maternal parenting
alliance instability. Trait reactance is conceptualized as the need to maintain

personal fieedom and can be manifested by resistance to influence by others

168

(Dowd et al., 1994). In the context of a reactant mother, the coparenting
relationship would seem particularly vulnerable to problems when fathers attempt
to make caregiving decisions, particularly if parents do not agree about their
childrearing philosophy. Whereas it is likely that many new mothers are somewhat
protective of their special role as the child’s primary caregiver, a reactant mother
may feel personally invaded or attacked if her judgments are not accepted and may
even engage in behavior designed to limit fathers’ contact with the child (De
Luccie, 1995). However, maternal reactance was not related to overall level or
linear change in perceptions of the parenting alliance, only to fluctuations, and not
to fathers’ parenting alliance ratings. Thus, it appears that more reactant mothers
do not consistently feel threatened by fathers’ involvement, but attempt to
negotiate ways to manage coparenting cooperatively with varying success over the
first 6 months. Reactance may become even more critical among parents of older
infants. During infancy, fathers are on the whole more likely to defer to mothers’
childcare judgments; but as fathers grow confident in their ability to parent, which
occurs as children age (Ferketich & Mercer, 1995) and wish to join in decision-
making, individuals with more reactant traits may find it more difficult to
compromise and cooperate in the service of functional coparenting.

Finally, the degree to which an individual had assimilated the notion of
him/herself as a parent prior to actually becoming a parent was associated with

both level and change. For the purposes of this study, assimilation was

169

oper
abet
alliat
abou
and 2
W35 5
had 2
parer
result
WOm
that 1';
that s
Itsea,
Mate
for in
Heini
appee
tell.
inn,
ill-«Kim

bi

“Fife.

Operationalized in two ways: as negative concerns and as positive motivations
about birth and childrearing. First, concerns were related only to parenting
alliance level, and only for mothers. Mothers who expressed strong concerns
about potential problems such as painful childbirth, temperamental difficulties,
and altered post-birth lifestyles reported a whey coparenting relationship. This
was a somewhat surprising finding, as it was hypothesized that individuals who
had anticipated problems would hold a more realistic view of the vicissitudes of
parenthood and would thereby experience greater assimilation. However, the
results of this study suggest otherwise, at least for women. One possibility is that
women who report high levels of concerns hold a generally pessimistic outlook
that is also manifested in their parenting alliance ratings. A related alternative is
that strong pre-birth maternal concerns are a mark of excessive anxiety. Several
researchers have reported that pre-birth maternal anxiety, measured globally, is
related to later insensitivity in mother-child interactions and places infants at risk
for insecure attachment (Bosquet, 1999; Cox et al., 1989; Feldman et al., 1997;
Heinicke et al., 1983; Leifer, 1977). Anxiety specific to the parenting endeavor
appears to contribute, or at least mark, dissatisfaction in subsequent coparenting as
well. The current research does not indicate whether maternal concerns also
influence actual coparenting behaviors, but one reasonable hypothesis is that
anxiety may be directly linked to poor parent—child interactions, as well as

indirectly linked through difficulties in the coparenting partnership.

170

Concerns about childrearing were related to parenting alliance 1_QV_el, but
positive motivations tended to be associated with parenting alliance m.
Although all three results for positive motivations were at the trend level, their
consistency in predicting change parameters in both parents warrants discussion,
and it is recommended that further research examine this issue in greater detail.
For mothers who eagerly anticipated the joys and rewards of parenthood,
perceptions of the parenting partnership tended to grow increasingly more positive
across the course of the study. However, the husbands of these motivated mothers
were prone to greater instability in their perceptions of the parenting alliance. In
many ways, the initial months of parenthood are relatively unrewarding compared
to later, when infants are more social and interactive; yet it is likely that when
pregnant women envision parenthood and family life, it is these older, more
responsive babies with whom they see themselves. Thus, although these
expectations may be unrealistic immediately after the baby’s birth, greater
fulfillment occurs as infants develop, with concomitant increases in the mutuality
of parenting. In addition, highly motivated mothers may be more committed to
and successful at involving their spouses, and thereby enjoy coparenting that
much more. These results, pending further investigation, suggest that childrearing
motivation may be a more appropriate measure of assimilation than childrearing
concerns, at least for women, with the caveat that mothers may be assimilated into

the notion of parenting a slightly older infant; perhaps it is impossible to prepare

171

adequately for the experience of the first newborn!

The trend for instability of the parenting alliance for the husbands of highly
motivated mothers is interesting, and may be linked to the fact that new fathers
look to mothers, who may be just as inexperienced, for direction regarding how to
behave with an infant. Husbands who have wives with unrealistically high
expectations for the rewards of parenting in the first months may experience
greater confirsion in parenting than those whose wives have more moderate
expectations, which may be then reflected in fluctuating assessments of the
coparenting partnership. Furthermore, fathers who were themselves more excited
about the prospect of parenthood tended to rate the parenting alliance
progressively l_ow_er over the study period, suggesting that childrearing motivation
may function differently for mothers and fathers. Two explanations come to mind.
One possibility is that the realities of parenthood begin to coincide with mothers’
expectations by the time infants reach 3 months of age, but that eager fathers may
have a longer time period in mind when they anticipate parenting rewards . For
example, they may look forward to fatherhood with a child who can play, run,
roughhouse, and converse. As infancy progresses and children, although certainly
more interesting, are far from achieving the abilities of a toddler, fathers may find
parenting less involving than they had anticipated and coparenting not particularly
satisfying. An alternative possibility that speaks more directly to parenting alliance

outcomes is that highly motivated fathers may be denied the parenting

172

involvement at the level they would prefer, as mothers implement “gatekeeping”
behavior (De Luccie, 1995). Feelings of exclusion would be likely to cause fathers
to report the parenting alliance as decreasing over time, especially as children
become more interesting and exciting to play with and as fathers become more
sure of their caregiving abilities (Ferketich & Mercer, 1995).

Several variables were not included in the final model of pre-birth
parenting alliance predictors. For fathers, perceptions of triangulation, positive
coparenting, and marital conflict in the family of origin predicted parenting
alliance level in both parents during the preliminary analyses, but were unrelated
when analyzed in conjunction with marital satisfaction. That is, fathers who
perceived their parents to be successful coparents were happier in both their
marital and coparenting relationships and had wives who perceived coparenting as
more supportive. Thus, it is incorrect to conclude that the family of origin is
unrelated to coparenting for fathers; rather, family life appears to contribute to the
general relationship quality that men experience when developing families of their
own and that expands into their wife’s perceptions. Furthermore, these
(non)findings provide more support for the contention that mothers differentiate
Various aspects of relationships, whether related to parenting, the marriage, or the
faIrrily of origin, more so than do men.

In addition, differences in authoritarian childrearing attitudes were

unrelated to the parenting alliance. Potentially, this variable may become more

173

salient as children encounter new developmental tasks. For example, as
hypothesized, differences in pennissiveness/protectiveness were associated with
parenting alliance development during infancy, when children are solely
dependent on the investment and abilities of their caregivers for survival. As
children become toddlers and issues of separation, autonomy, mastery, and control
arise, parental differences with respect to authoritarian childrearing attitudes are
likely to become increasingly important. Infant sex also did not directly affect
parenting alliance development. It may, however, have a moderating effect in
conjunction with other variables (see McHale [1995] for the differential effect of
child sex in the coparenting interactions of maritally distressed couples).
The Parenting Alliance in a Changing Context

Changing post-birth factors were also associated with individual
differences in parenting alliance trajectories, in partial support of Hypothesis 4,
depending on the parameter being predicted. Once again, marital quality was the
most consistent predictor; but whereas pre-birth marital adjustment was related
only to parenting alliance level, post-birth marital adjustment was related to both
level and change. As expected, parents who were generally more satisfied with the
state of the marriage after the transition to parenthood also reported a stronger
parenting alliance. Contrary to expectations, however, change in marital
satisfaction and change in the parenting alliance were negatively related, so that

declines in marital quality were accompanied by increases in ratings of the

174

coparenting partnership and vice versa. This finding may indicate that one aspect
of the couple’s relationship is maintained at the expense of the other. Thus, a
couple could experience a more cooperative coparenting relationship, but the cost
might be less time together unaccompanied by the child. Conversely, the couple
may place a priority on maintaining the quality of their pre-birth marital
relationship, but devote less effort to coparenting endeavors; in this case, a tenable
hypothesis would be that there may be more discrepant individual parent-child
involvement, most likely with mothers expending greater effort in parenting than
in fathers. Given the suggestion that mothers are more pivotal in contributing to
variability in couples’ experiences of coparenting, additional research might also
explore whether mothers determine which aspect of the couple relationship
becomes a 9% priority.

Post hoc analyses conducted in an effort to delineate causal relations
between the parenting alliance and marital quality shed little light on this
unexpected association, as no significant negative cross-dimension correlations
emerged. Nonetheless, the cross-lag analyses did demonstrate processes through
which marital satisfaction affects the development of the parenting alliance. The
quality of the marriage pgo_r to the child’s birth appeared to determine whether the
parenting alliance was perceived as more or less successful. Among fathers, this
pre-birth influence was the primary factor in predicting subsequent coparenting

satisfaction, but for mothers, l-month marital adjustment continued to predict the

175

parenting alliance at 3 months. By 6 months, however, the marital quality rating
from the previous time was no longer directly associated with perceptions of the
parenting alliance.

It appears, therefore, that the quality of the marriage before the child was
born did indeed set the stage for the parenting alliance. Particularly for fathers,
how they feel about the couple’s relationship before the transition to parenthood is
a good indicator of how they will feel after the transition. For mothers, however,
reorganization of the marital system resulting fiom the entry of a new person
continues to reverberate, impacting perceptions of the parenting alliance beyond
the first month. By midway through the child’s first year, perceptions of the
coparenting relationship appear to have become independent of marital
fluctuations, instead arising primarily from prior evaluations. Although change in
the parenting partnership and change in marital satisfaction may happen
concurrently, it is not clearly attributable to causal relations. No significant cross-
parent effects were found; only the parent’s self-reported marital quality predicted
his/her own parenting alliance. A positive trend emerged, however, between
mothers’ mean post-birth marital satisfaction and fathers’ level of parenting
alliance. If replicated, this would indicate that fathers who were more satisfied
with coparenting had more happily married wives. This finding would then
suggest that fathers were not simply satisfied with the parenting partnership

because their wives shouldered most of the parenting burdens, which would most

176

likely be reflected'in wives’ lower ratings of the parenting alliance.

 

Violated expectations of the division of labor also emerged as an important
post-birth predictor of parenting alliance development, and manifested the
hypothesized stronger effects for childcare tasks than for housework. Each
parent’s own violated expectations for childcare were associated with the level of
their parenting alliance ratings, so that parents who did less childcare than they
had anticipated reported greater satisfaction with respect to the coparenting
partnership. It should be noted, however, that whereas both mothers’ and fathers’
expectations were violated, mothers typically did rpm and fathers le_ss_ than
anticipated. Both parents’ childcare responsibilities became increasingly
compatible with their pre-birth expectations over the course of the study; that is,
mothers did proportionally less childcare, while fathers did more. It follows, then,
that only for mothers was change in these violated expectations for childcare
related to change in the parenting alliance. As mothers’ expectations became less
violated, arfi as their husbands took on more of the childcare responsibilities, their
assessment of the parenting alliance improved. Fathers might be expected to then
perceive the parenting alliance in a worse light over time as they assume a greater
proportion of childcare, but this was not the case. Although fathers’ relative
contribution increased by 6 months, mothers were still responsible for the majority
of caregiving. Furthermore, fathers’ expectations were not violated in a negative

direction; rather they simply converged with the amount of work fathers initially

177

anticipated for themselves. Although logic dictates that changes in violated
expectations would causally affect changes in parenting alliance perceptions, the
cross-lag analyses did not identify significant causal paths between violated
expectations and the parenting alliance at any point. This lack is difficult to
explain, but suggests the presence of a third variable that covaries over time with
both violated expectations and the parenting alliance. Still, violated expectations
around the division of labor has been identified as a pivotal factor in Mal
decline over the transition to parenthood (Hackel & Ruble, 1992; Ruble etal.,
1988). The current study is the first to describe the role of childcare expectations
in coparenting development over the transition to parenthood.

As expected, violated expectations around housework were less relevant to
parenting alliance development than childcare. In fact, the only significant finding
was between change in mothers’ violated expectations for housework and change
in the parenting alliance. On average, mothers did more housework and fathers
less over time, in direct contrast to the changes in childcare. Thus, as mothers’
expectations for housework became increasingly violated in a negative direction,
their evaluations of the parenting alliance grew poorer. In families with young
children, the amount of housework increases as children become more active and
mobile; clothes are easily dirtied and toys become permanently scattered, and the
move to solid food requires more cooking and dishes. Women usually bear greater

reSponsibility for these types of household tasks, so even as the division of

178

ht

C0

l’lt

par

all

childcare becomes more equitable, the division of housework grows increasingly
traditional. Thus, it is not surprising that change in violated expectations for
housework and change in the parenting alliance were associated. Nonetheless,
parents appeared to associate childcare and housework with different aspects of
the couple relationship. The finding that childcare division of labor shows several
specific associations with the parenting alliance, while housework division of
labor is less related, provides additional evidence for the construct validity of the
coparenting partnership. Conversely, Ruble et al. (1988) demonstrated that
violated expectations for housework, but not childcare, were related to declines in
_rpari____tal_ satisfaction over the transition to parenthood. As with the childcare
variable, violated expectations for housework did not exhibit causal relations with
parenting alliance reports, and change in both may be the result of the influence of
a third variable.

The final post-birth factor related to parenting alliance development was
parenting efficacy. As expected, for both mothers and fathers, greater self-
confidence in one’s own parenting was associated with a more satisfactory
parenting alliance; in addition, fathers were more satisfied with the parenting

alliance when mothers felt efficacious. The results suggest that mothers were

 

responsible for guiding coparenting efforts. Although fathers’ parenting efficacy
increased over the 6-month period, it remained consistently lower than that of

mothers. Apparently, mothers were relatively unconcerned about, and may even

179

have expected, fathers’ comparatively tentative feelings about their caregiving
abilities. However, if mothers lack confidence in their own parenting, their
partners may feel directionless; fathers are aware of and react to mothers’
evaluations of their parenting abilities, and are more involved when they feel
mothers have confidence in them (McBride & Rane, 1998). Furthermore, if
mothers are insecure about parenting, greater weight falls on fathers’ parenting
abilities. Social attitudes about father involvement in parenting have become more
equitable, but mothers are still the primary caregivers, and fathers are likely to feel
resentful when mothers do not “do their job.” Although efficacy showed change
over time, this change was unrelated to change in the parenting alliance.

Of the hypothesized post-birth predictors, only perceptions of infant
temperament were unrelated to parenting alliance evaluations despite the fact that
infants were rated as progressively less difficult across the course of the study.
This is rather surprising, particularly in light of the significant effects for efficacy,
since parenting efficacy and temperament have been shown to be significantly
related, both in previous studies (Ahuja et al., 1999) and in the current
investigation. The single trend was in the expected direction, and indicated that
fathers who perceived their babies are more difficult tended to rate the parenting
alliance as poorer. Pending further investigation, this suggests that fathers may
feel less efficacious and desire greater support than is forthcoming from the

mother when faced with an irritable infant. On the other hand, in the context of a

180

difficult baby, mothers may seek greater-support from fathers, who may feel
impinged upon. Regardless, the lack of significance may indicate that relatively
few infants were fussy enough to cause high levels of stress.
Summag and Implications

The results of this study point toward four general conclusions that confirm
and extend the existing literature on the parenting partnership and the transition to
parenthood, and suggest a number of avenues for further research. First, the
parenting partnership and marital relationship are confirmed as related but
separate aspects of an overall relationship dimension. However, the coparenting
construct is broad and not yet well defined. The use of multiple measures is highly
recommended, therefore, to assess the intricate complexity of family relationships,
since both self reports and observational ratings add important information in
evaluating family functioning (Floyd & Zmich, 1991; Hayden et al., 1998). To
illustrate, the results of the construct validity analyses demonstrated that
alternative ways to conceptualize dyadic relationship variables can reveal major
group differences, in this case, between mothers and fathers; it remains to be seen
which domains are most critical for which processes and outcomes. Nonetheless,
the evidence presented in this study adds to the body of work indicating that one
can coparent successfully in the context of an unhappy marriage and vice versa
(Abidin, 1992; Cohen & Weissman, 1984).

Second, although the parenting partnership and the marriage are different

181

aspects of the relationship, the importance of the marital context for the
development of coparenting cannot be overestimated. The ways in which spouses
have organized their feelings about one another, attitudes regarding potentially
conflictual issues, need for affection and intimacy, and behaviors in dyadic
interactions set the context for whether they will be able to successfully organize
the new parenting relationship, or whether it will tend toward hostility,
resentment, and miscommunication. It was notable that even among these high-
functioning couples, problematic coparenting behavior patterns could be
prospectively identified from quite subtle variations in the pre-birth marital
relationship.

Third, the results of this study highlight the specific relationship
experiences of men and women, and identify pathways and correlates that
differentiate the development of coparenting for each. Whereas the pre-birth
relationship sets the stage for fathers’ perceptions of coparenting in a relatively
permanent manner, for mothers, the influence of marital quality persists after the
child’s birth. These findings support the research of many other investigators, who
find different pathways to child outcomes and parenting for mothers and fathers
(e.g., Katz & Gottrnan, 1996; Levy-Shiff, 1994; Thompson & Walker, 1989).

In addition, mothers distinguish between coparenting actions and the
marital relationship to a greater extent than do fathers. For example, what mothers

actually do in a family play situation bears little relation to how they feel about the

182

couple or behave when alone with their husband; for mothers, coparenting
behaviors comprise a relatively separate domain of family functioning. On the
other hand, fathers’ perceptions and behaviors across both relationship milieus are
more difficult to disentangle; if fathers feel positive about the marriage, they tend
to be satisfied with the parenting partnership and act supportively in both the
coparenting and marital contexts. Consideration of the paradigm used to elicit
coparenting behaviors encourages speculation regarding the basis for these
differences. The family play situation involved both parents on the floor with the
infant, directed to teach the infant how to complete a task. Within the home
environment, as they engage in the daily business of running a household, few
families are likely to participate in mutual play; more frequently, one parent is
responsible for monitoring the baby while the other completes household tasks.
Furthermore, most mothers are likely to have had greater responsibility and have
cared more frequently for their child as opposed to fathers. Thus, when placed in
the family play situation, mothers are more likely to drive the interaction. They
may focus on the infant and attempt to complete the teaching task, or they may sit
back and actively encourage fathers to interact with the baby, but for mothers,
c0parenting is in many ways an extension of individual parenting behavior, with
the focus on the infant. Fathers, on the other hand, center on their wife’s behavior;
c0parenting is more about the relationship with the wife than the infant. Thus, for

mothers, coparenting behaviors become distinct from other aspects of the couple

183

W2

b0

pat

H10

5‘95

tar:

relationship, but for fathers, marital, couple, and family time is more indistinct.

An attempt to identify subdimensions of the general alliance scale in
analyses not presented here suggest that mothers and fathers perceived
coparenting in distinctive ways. Both the number of factors and the items that
clustered together differed substantially across men and women. It is possible that
when men and women provide their perceptions of the parenting alliance, their
answers are based on quite disparate points of reference. In this study, the decision
was made to retain the full scale, which evidenced high internal consistency for
both parents, to measure the parenting alliance, but future work would be
enhanced by elaborating on the similarities and differences in maternal and
paternal perceptions of what comprise dimensions of coparenting in the early
months and over the course of parenthood.

The fourth and final point is that the couple relationship is an interacting
system of maternal and paternal attitudes, expectations, and behaviors. Fathers’
parenting alliance trajectories tended to be influenced by individual characteristics
of mothers, such as their anticipation of parenthood and their self-perceptions of
maternal parenting competence. These factors are likely to promote emotional
instability in the first months of parenting, which may be problematic for many
fathers, who are less accustomed to coping with the affective domain (Strazdins et
al., 1997). Mothers, on the other hand, were more affected by more paternal

factors that potentially could cause conflict over childrearing decisions, such as

184

differences in childrearing philosophy and their contribution to childcare, as well
as by fathers’ general feeling of marital well-being. However, it should be noted
that differences in childrearing philosophy, violated expectations for childcare,
and marital satisfaction are also dependent on maternal reports: childrearing
differences because both parents’ attitudes are taken into account, violated
expectations because as fathers’ expectations were positively violated, mothers’
were negatively violated, and the marital relationship due to the covariance
between mother and father perceptions. Thus, it can be hypothesized that mam;
characteristics are pivotal in parenting alliance development, affecting fathers’
naj ectories directly, while fathers’ characteristics affect the parenting alliance of
mothers only through interactions with maternal variables. This has critical
implications not only for the study of typical family development, but also raises
questions regarding the amplified role of maternal psychopathology on child
outcomes, as well as regarding the function of mothers in families in which
MEL] psychopathology (e. g., paternal alcoholism) places children at risk of
emotional and behavioral problems.

These research findings must be considered in light of the current social
construction of parenthood. LaRossa and colleagues note that “conventional
wisdom has it that America is in the midst of an unprecedented revolution in
men’s paternal role expectations—that popular attitudes about what fathers can and

should do are changing in ways not dreamed of before “ (LaRossa, Gordon,

185

Wilson, Bauran, & J aret, 1991, p. 994). Although fathers have increased their
involvement with children over the past decades (Pleck, 1997), couples
consistently fall into traditional patterns after the birth of the first child. Many of
the couples in this study were raised with the expectation that male and female
parenting is equivalent and planned to share all childrearing responsibilities.
However, most reverted to more traditional roles despite fi'equently explicit
intentions not to do so (Levy-Shiff, 1994; McBride & Rane, 1998). Thus, for this
generation, the transition to parenthood, although rewarding, is primed for
difficulties during the development of the parenting partnership.
Limitations and Future Directions

The exploratory nature of this study provided an initial description of
coparenting in the first months of the transition to parenthood, but raised many
more intriguing issues. Foremost is a problem that exists in most studies of the
transition to parenthood and in the family relationship literature in general: The
current sample was comprised of married, predominantly white, middle-class
volunteers for a 4-wave longitudinal study who received little compensation for
their participation; even among people who attended childbirth classes, highly
educated individuals were overrepresented. In addition, at no point did any
participant’s DAS score approach the clinical cutoff of 100; the lowest score
across four timepoints was 112. The composition of this sample has benefits: The

results are likely to be comparable to those of previous research in this domain,

186

which has, for the most part, been based on similar samples. Furthermore, the fact
that these high-functioning families demonstrated a moderate range of coparenting
attitudes and behaviors suggests that negative effects may be amplified in at-risk
families. Nonetheless, the limited generalizability of the study findings must be
considered when incorporated into theory building and research design. Different
mechanisms of parenting partnership development may well Operate in less
affluent, culturally diverse, or distressed groups, and research must be undertaken
to identify similarities and differences in coparenting in the population as a whole.

Furthermore, the current study spanned only a brief period, which, although
sufficient for the goal of examining the development of the parenting partnership
over the-potentially unstable first months of parenthood (Parke et al., 1994), may
have limited long-term relevance. Thus, follow-up research on this sample, or the
initiation of a long-term longitudinal study of coparenting is necessary to
investigate processes and outcomes over the entire period of family development.
Outcomes of particular interest center on the relation between coparenting and
child social adjustment, emotional regulation, and psychopathology, as well as the
identification of differences in coparenting interactions that can identify couples at
risk of separating.

The findings for perceptions of the parenting alliance beg the question as to
Whether systematic pre-birth and post-birth predictors of coparenting behavior can

also be identified, and if so, how they vary from self-reports. Since the 6-month

187

coparenting behavioral ratings showed relatively little relation to concurrent
parenting alliance reports, differences in the patterns of association are probable.
Still, investigation of this question is an important next step. Additionally, the
relationship between the parenting partnership and the marriage should be more
intensively explored. The present study described interrelations between the two
in a rudimentary fashion, but an important advance would be to specify
trajectories for marital quality as well as the parenting partnership, and to
determine whether the pre-birth and post-birth predictor variables differentially
relate to the two relationship trajectories. If separate models were required to
predict each dimension of the couple relationship, evidence for the construct
validity of the parenting partnership would be even more compelling.

The issue of dyadic versus whole-family interaction must be considered

(Fivaz-Depeursinge & Corboz-Wamery, 1999; F ivaz-Depeursinge et al., 1996).

 

The coparenting relationship has been conceptualized as a triadic interaction
because it is manifested because of a third person, the child. In actuality, however,
coparenting as operationalized in this and many other studies is a Mg
relationship; one that occurs in the _cpn_te_xt_ of a third (or more) person, but a
dyadic relationship nonetheless. An empirical question is the relative importance
of dyadic interactions, including coparenting and individual parent-child relations,
in comparison to ratings of the whole-family environment in predicting outcomes.

Some researchers have attempted to measure whole-family variables (McHale et

188

al., 1996; Paley et al., 1999). However, the question becomes further complicated
because almost all studies that have explored relations among more than two
family members have limited their investigation to three members (for an
exception, see Kreppner [1988]). There is every reason to believe that the addition
of a second child (or more) will once again destabilize the parenting partnership.
Anecdotally, most parents, many of whom found the transition to one child more
difficult than anticipated, report being even more overwhelmed by the presence of
two children. This suggests the hypothesized effects of additional children on the
parenting partnership are multiplicative rather than additive or nonexistent.

On a broader note, as a fairly new research domain, coparenting is not yet
well nor consistently operationalized, a critical venture in promoting a common
language and useful application of disparate research findings (Sabatelli &
Waldron, 1995). Several issues must be addressed to solve this problems: What
comprises coparenting? Who can be coparents? How does c0parenting change
over developmental time? First, this study operationalized the parenting
partnership in two ways: as the subjective feelings of support and mutuality
encompassed in the parenting alliance, and as specific patterns of competitive or
cooperative behavior that might be manifested during family play. Yet a
comprehensive conception of coparenting involves much more, even without
taking into consideration variations that are certain to exist across cultures. It

includes the ways that couples go about parenting within their own homes,

189

 

exit

in

3151'

and

19‘.

A .
8.1.

.C.

51...». .

amongst the demands of family, friends, work, and individual needs. It is
evidenced when couples assume responsibilities for direct physical care and
feeding (in fact, the division of childcare could itself be considered one measure
of coparenting) as well as organizational functions such as ensuring
immunizations are current, buying clothing, and arranging daycare. Coparenting
also plays a role in more distal and abstract functions like teaching how to express
and regulate emotion and behave in socially appropriate ways (Katz & Gottrnan,
1996). C0parenting can even occur when a parent is alone with a child, as when
the representation of the other parent is invoked, whether in a positive,
threatening, or derogatory way (McHale, 1997; McHale & Rasmussen, 1998). As
defined here, coparenting is much more than simply co-caregiving, but is instead
an essential part of a child’s physical, emotional, and social security.

Who, then, can be coparents? The present sample was limited to
heterosexual married partners, but in fact, coparenting occurs whenever two
individuals form a bond in the service of a child’s needs. Thus, coparenting can
occur between divorced, unmarried, and/or gay partners as well (Loukas,
Twitchell, Piejak, Fitzgerald, & Zucker, 1998; Silverstein & Auerbach, 1999).
Single parents may coparent with a wide variety of partners, including romantic
partners, close fiiends, siblings, older children, and, particularly in the case of
adolescent mothers, their own mothers. Each of these coparenting configurations

may have different mechanisms and processes through which more or less

190

successful outcomes occur, and none have been sufficiently investigated.
Finally, do definitions of coparenting change during different
developmental periods? No research has yet attempted to investigate how
coparenting tasks and demands are altered as children progress through
developmental stages; all studies to date have focused on children of one age
group, usually toddlers. However, coparenting toddlers is considerably different
from coparenting infants. For infants, physical care and security are paramount,
but for toddlers, emerging needs for control and autonomy, as well as the
development of language, promote the initiation of discipline tactics, emotion
management strategies, and alternative methods of stimulation. Each age group
has its special characteristics, and without defining what continuities and
differences are demanded of coparenting across child development, the field is
likely to find it difficult to integrate disparate research findings into a coherent

picture of family development.

191

APPENDIX A

Newsletter

192

  

Hello again, everybody!

The MSU Parenthood Project Team
wants to congratulate our project families
once again on the birth of all their beautiful
babies. We’ve had a wonderful time meeting
all the new children, and we hope you’ve
enjoyed having another group of people to
show off to! We’re taking this opportunity to
introduce the team members, update you on
the project, and give you a little information
on some of the issues that might apply to your
baby at this age.

Project Team Members :

You may have met one or several of
our team members. Currently, we have 11
people on the MSU Parenthood Project Team:

Laurie Van Egeren, Principal
Investigator and head of the project. Laurie is
finishing her doctoral work in developmental
psychology at MSU. She has a 2-year-old
daughter, Tess, and a 7—m0nth-old daughter,
Theo (two families from the study were
giving birth at the same hospital and at the
same time as Lauriel).

We have several research assistants
returning from last semester: Beth Robinson,
Junna Mita, Dyane Hawkins (herself a
mother of two, ages 13 and 9), Rene Teper,
Marina Nelson, Rick Lower, Donna Jones,
Marcy Hoopingarner, and Heather
Fountain. You have probably met at least one
ofthese project staff before.

MSU Parenthood Project ©

 

 

You may have met Renee Verysee,
Emalee Fields, Kristine Kolarovic,
Adrienne Auten, Beth White, Kristen
Sturman, Sarah Davis, Alicia Dawson, or
Joan Cook, former project team members.
They truly enjoyed getting to know the
families that they visited, and were very sorry
they couldn’t come on the final visits to see
“their” babies one last time!

Project Update . . .

As of this mailing, the last baby has
just been born; in all, we have 103 families in
the project. 513 we really do have half boys,
half girls (well, 51 boys, 52 girls). Amazingly,
everyone has named their baby a unique
name; we haven’t had any repeats! Most of
our families are in the greater Lansing area,
but some are in Grand Rapids, Detroit, Ann
Arbor, and even Texas, New York,
Washington, DC, and the West Coast.

What’s Next?

In about a month, we’ll be contacting
you to set up the final visit, where we’ll ask
you about how life has changed since the
baby was born, as well as doing a brief play
session with you and the baby. Since Laurie
will be out of commission for a few months,
Dyane or Marina will probably be calling
you; the student who has been visiting with
you in the meantime may also want to come
along to say good-bye. We’ll see you soon!

193

 

..J‘---

.m - - - - -

Fun and Games

Some games your baby will love to play:

0 Sit and hit: Dangle a toy within baby’s
reach, so he/she can punch or grasp it.

0 Grab and shake: Offer rattles, dolls,
socks, keys (especially keys).

0 Kick toys: Attach balloons, rattles, or
bells to baby’s ankles; let him/her kick.

9 Rolling: Lay baby over a large ball and
roll it slowly back and forth (will help
with balance.

9 Pulling games: Hold a thin bar or your
fingers in front of baby’s chest. Let
him/her grab on and slowly pull up.

F ive-Month Milestones
At 5 months of age, you can look forward
to your baby beginning to:
0 Roll over
0 Reach for objects
9 Smile a great deal
9 Grasp objects

Immunization Schedule
We all know this isn’t fun, but...

4 months: DTP‘, OPV, Hibl
6 months: DTP, Hib‘

6 to 18 months: OPV, HBV‘
l-low Baby Changes

As your new family member approaches
his or her S-month birthday, here are some big
events that you can look forward to:

Sitting up (with some aid): Prop baby up
with pillows on all sides. They love it!

Teeth: usually emerge between 3 and 12
months. Common signs:

0 drooling
0 increased night waking
O - ‘face rash
9 low fever
9 irritability

 

Solid food: real-people food is generally
encouraged beginning between 4 and 9 months.

Here are some good choices for beginners:
Rice cereal, squash, carrots, green beans, peas,
yogurt, applesauce, peaches

Pediatricians commonly suggest that you
start with one new food at a time, serving it for a
whole week, in order to identify potential food
allergies.

Will We Ever Sleep Normally Again?
Babies who are around 5 months of age
sleep an average of 14 1/2 hours per day. However,
this 14 '/2 hours often does not happen as we
might prefer.
Some tips that might help your baby sleep
more during the night:
0 Keep daytime naps less than 3 hours
9 Make nighttime feedings boring and brief
0 Begin a nightly ritual (sing a lullaby, read
a story, give a bath before bed)
0 Put baby to bed awake so he/she learns to
soothe him/herself to sleep
And expect night wakings to increase as baby
approaches major milestones in development,
such as sitting up, beginning to crawl, and
teething.

For Computer-Buff Parents . . .
Check out these websites:
Parent Soup: http://www.parentsoup.com
Parent Q&A by pediatricians and psych ologists:
http://www.parenting-qa.com
National Parent Info Network:
1-800-583-4135 (okay, you don’t have to
have a computer for this one)

To Contact the MSU Parenthood Project:
Telephone: (517)353-8807
Email: vanegere@pilot.msu.edu
Address: Department of Psychology
Michigan State University
East Lansing, MI 48824
c/o Laurie Van Egeren

194

APPENDIX B

Informed Consent

195

Transition to Parenthood Study
Informed Consent Agreement

We are asking you to participate in a study to examine changes that take place in couples’
relationships as they become parents. Becoming a parent can be both exciting and stressful, and
we think it important to understand how couples negotiate this period of their lives. Information
that you provide through questionnaires, interviews, and play sessions will be used for this study.
Please carefully review the information below. If you have any questions at all, be sure to ask the
person who is requesting your participation so that you fully understand the procedures involved.
If you decide to participate, as a token of our appreciation, at the end of the project you and your
spouse will receive an infant t-shirt and a certificate for your baby, both of which will state that
he/she has been a member of the MSU Transition to Parenthood study. You also have the
Opportunity to win a savings bond in your baby’s name.

 

By signing this agreement, you are] indicating that you have been fiilly informed of the purposes
and procedures of the study in which you are agreeing to participate. You have been informed of
and understand each of the following points:

1. By signing this agreement, you are giving permission for the following activities to be

conducted:

a) During the third trimester of pregnancy, a member of the research team will interview you and
your spouse for approximately 30 minutes. You will also complete a set of questionnaires,
which will take approximately 90 minutes.

b) Whenyour baby is 1 month old, you will complete a set of questionnaires, which will take
approximately 30 minutes.

c) When your baby is 3 months old, you will complete a set of questionnaires, which will take
approximately 30 minutes.

d) When your baby is 6 months old, a member of the research team will again interview you and
your spouse for approximately 30 minutgs. You will also complete a set of questionnaires,
which will take approximately 30 minutes. Finally, you, your spouse, and your child will
participate in a 12-minute play session.

All information will be collected in your home (unless you prefer to come to the MSU campus).

At each of the four points, a research team member will telephone you to set up an appointment at

a time when you and your spouse will both be available. The team member will be available during

the V'lSlt to answer any questions and assist in childcare.

 

2. Upon completion of the study, you and your spouse will receive an infmt t-shig and a
certificate for your infant, both of which will indicate that the baby has been a member of the
MSU Transition to Parenthood study. You will also receive the MSU Transition to‘ Parenthood

nevi/gm. Six couples will win a lottery to receive a $100 or $50 sayings mud in their child‘s
name.

3. You have the right to decide not to participate in the study, or to decline to answer any specific
question(s) that you are asked, or to discontinue your participation at any time.

196

4. You understand that the interviews and play session will be videotaped. No one other than
members of the research team will have access to these tapes, which will be identified on the
outside by a code number only. The tapes will be kept in a locked file cabinet in the project
research office.

5. Any information you provide in the questionnaires will be transformed into numbers and
entered into a computer data file. The written record will then be destroyed. All tapes will be
labeled with a code number. You also understand that the researchers are obligated to report to
the appropriate authorities any information that you might provide that suggests that your child or
another minor is being subjected to or has been subjected to physical or sexual abuse.

6. Ifyou have any questions or concerns about the study or you would like to receive any
publications that might result from the study, you can contact Laurie Van Egeren at the
Department of Psychology at Michigan State University by calling (517) 353-8807, writing to
Department of Psychology, Michigan State University, East Lansing, MI 48824, or through
e-mail at vanegere@pilot.msu.edu.

 

 

Signature Date

197

APPENDIX C

Measures

198

Information Questionnaire (TI)

1. What is your sex: Male __ Female___
2. How old are you?

3. What is the highest level of education you have completed? (circle one)
Less than high school

High school graduate

Some college/associate’s degree/vocational degree or certificate
Bachelor’s degree

Some graduate school

Master’s degree

PhD, J.D., MD.

NP‘MPP’N!‘

4. What is your
occupation?

5. What is your combined (you and your spouse) annual income? (circle one)
Less than 35,000
35,000-39,999
310,000-319,999
320,000-329,999
330,000-339,999
340,000-349,999
350,000-359,999
360,000-369,999

9. 370,000-379,999
10. 380,000-389,999
11. 390,000-399,999
12. $100,000 or greater

WSQMPP’N?‘

6. What is your primary ethnic group?

 

7. How long have you been married?

 

8. When is your due date?

 

9. How far along are you now?

 

10. Did you plan this pregnancy?

Yes No No, but we weren’t trying to avoid it either

199

Information Questionnaire (1‘2, T3, T4)
1. What is your baby’s sex? Male Female

2. What percentage of time have you been breastfeeding versus bottlefeeding your baby?

Breastfeeding %
Bottlefeeding %
Total = 100 %

3. How many hours per week does your child spend with a caretaker besides you and
your spouse? (“0" if you and your spouse take care of your child full-time)

hours

4. What type of childcare are you using? (circle one)
1. None
2. Family member
3. Babysitter in my home
4. Babysitter in their home
5. Family daycare
6. Group daycare
7. Other

 

5. What is your current employment status?
1. Stay-at-home mom
2. Part-time (if part-time, indicate how many hours per week )
3. F ull-time

 

200

Parenting Alliance Questionnaire

This questionnaire asks for information concerning your parenting experience and about
how you and your partner work together as parents.

W Using the scale described below, giggle the letter that indicates how
much you agree with each of the following statements.

SD=Strongly Disagree D=Disagree A=Agree SA=Strongly Agree

1. Parenting has brought my spouse and me

closer together. SD D A SA
2. My spouse tries to have the last word in how

we raise our baby. SD D A SA
3. My spouse is a good parent. SD D A SA

4. My spouse and I feel that we are growing and
maturing together through our experiences
as parents. SD D A SA

5. My spouse is willing to make some personal
sacrifices in order to help with the parenting. SD D A SA

6. My spouse tries to make sure I get some time
for myself away from the baby. SD D A SA

7. When there is a crisis with the baby, my spouse
doesn‘t help me as much as I would like. SD D A SA

8. I have learned that if my baby needs something
important, I can rely on my spouse to help
provide it. SD D A SA

9. My spouse does not really enjoy being alone
with the baby . SD D A SA

10. After my spouse or I have handled a difficult
situation with the baby, we discuss it and
try to figure out what we could have done better. SD D A SA

1 1. My spouse likes to play with the baby, but
then leaves the dirty work to me. SD D A SA

201

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

SD=Strongly Disagree D=Disagree A=Agree SA=Strongly Agree

I appreciate how much my spouse tries to be a
good parent.

My spouse resents that I have to give so much
of my time to the baby.

My spouse and I like to imagine together what
our baby will be like when he/she grows up.

When the baby is sick, I can turn to my
spouse for support

My spouse and I agree on our ideas, guidelines,
and rules for raising our baby.

My spouse has a good feel for the baby and
what he/she might need.

My spouse does not live up to my idea of a
good parent

Parenting has given my spouse and me a
focus for the future.

My spouse helps out with the parenting
whenever possible.

My spouse and I work closely together
as parents.

My spouse makes too many demands on me
as a parent

Ioftenthinkmybabywouldbe betteroff
with one parent (me) than with the both of us.

I do not feel that parenting is as much ofa
sharing experience with my spouse as I hoped
it would be.

I feel closer to my baby than to my spouse.

My spouse and I agree on how much time we
should spend with the baby.

202

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

D

A

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

SD=Strongly Disagree D=Disagree A=Agree

Sometimes I feel like my spouse is another
child instead of my partner.

My spouse pays too little attention to
the baby.

My spouse still wants to "do his or her
own thing” instead of being a responsible
parent.

My spouse preaches a lot about how to be a
good parent, but rarely puts it into practice.

Having a baby has helped me to see positive

qualities in my spouse that I never noticed
before.

My spouse sees parenting as my responsibility.

I worry about the baby's safety when
he/she is alone with my spouse.

My spouse has unreasonable expectations
for the baby.

My spouse is too self-centered to be a
good parent

I feel over-burdened as a parent because my

spouse is alien too involved with other things
toearryafairshareofthe load.

When I feel at my wits end as a parent, my
spouse gives me the extra support I need.

My spouse and I often talk together about
what is best for our baby.

203

SA=Strongly Agree

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

D

A

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

Dyadic Adjustment Scale

M For the following questions, please indicate how often you and your spouse

do the following things.

How often would you say the following events occur between you and your spouse?

 

 

Once a
Day

More
Often

 

Less Than Once or Once or
Once a Twice Twice a
Never a Month Month Week
1. Have a
stimulating
exchange of
ideas __ __ __ _
2. Laugh
together __ _—
3. Calmly
discuss
something __ __ __ __
4 Work together
on a project __ _ __ __
5. In general, how often do you think that things between you and your spouse are going well?
All of the time Occasionally
Most of the time Rarely
More often than not Never
6. Do you confide in your spouse?
All of the time Occasionally
Most of the time Rarely
More often than not Never
7. How often do you and your spouse "get on each others' nerves?"
___All of the time _______Occasionally
Most of the time Rarely
More often than not Never
8. How often do you and your spouse quarrel?
All of the time Occasionally
Most of the time _ Rarely
More often than not Never
9. How often do you or your spouse leave the house after a fight?
All of the time __ Occasionally
Most of the time Rarely
More often than not Never

204

10. Do you ever regret that you married?

All of the time Occasionally
Most of the time Rarely
More often than not Never
11. How often do you discuss or have you considered divorce or separation?
All of the time Occasionally
Most of the time Rarely
More often than not Never

12. Do you kiss your spouse?

 

Every day Rarely
Almost every day Never
Occasionally
13. Do you and your spouse engage in outside interests together?
Every day Rarely
Almost every day Never
__ Occasionally

14. The dots (‘) on the following line represent difl‘erent degrees of happiness in your
relationship. The middle point, "happy”, represents the degree of happiness of most relationships.
Please circle the dot which best describes the degree of happiness, all things considered, of your
relationship.

""" E‘ saga]?'"Aiiii’."“a'.ppy'"""'v'..y'"‘ia;‘u;i;i;"i»;;f.;t"""
unhappy unhappy unhappy happy happy

15. Which of the following statements best describes how you feel about the future of your
relationship?

I want desperately for my relationship to succeed, and would go to almost any length to
seethatit does.

Iwantverymuchformyrelationshiptosucceed,andwilldoalllcantoseethatitdoes.

Iwantverymuchformyrelationshiptosucceed, andwilldomyfairsharetoseethatit
does.

Itwouldbeniceifmyrelationshipsucceededbmlcan‘tdomuchmorethanlamnowto
helpitsucceed.

Itwouldbeniceifitsucceeded,bmlrefitsetodoanymorethanlamdoingnowtokeep
therelationshipgoing.

Myrelationshipcanneversucceed, andthereisnomorethatlcan dotokeep the relationship
going

205

Most persons have disagreements in their relationships. Please indicate below the approximate extent
of agreement or disagreement between you and yarn spouse for each item on the following list.

l=Always agree 4=Frequently disagree
2=Almost always agree 5=Almost always disagree
3=Occasionally disagree 6=Always disagree

16. Handling family finances

l7. Matters of recreation

18. Religious matters —

19. Demonstrations of affection

20. Friends

21. Sex relations

22. Conventionality (correct or proper behavior)
23. Philosophy of life

24. Ways of dealing with parents _
25. Aims, goals, and things believed important __
26. Amount of time spent together

27. Making major decisions

28. Household tasks

29. Leisure time, interests and activities
30. Career decisions

These are some things about which couples sometimes agree and sometime disagree. Indicate if

either item below caused difl'erences of opinions or were problems in your relationship during the
past few weeks. '

 

 

 

YES NO
31. Spending too much money.
32. Being too tired for sex.
33. Not showing love.
34. Not sharing household chores.

 

206

Current Division of Household Tasks

W: Using the scale provided, please indicate what percentage of the following
tasks are sweetly done by m and W-

 

None 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

 

 

 

 

 

 

 

:You Your Spouse Total °/o
Laundry 100
Cleaning 100
Dishes 100
Cooking 100
Shopping 100

 

 

 

 

 

Expected Division of Household Tasks

Now, complete the same questions with regard to how you m tasks to be divided
alter the baby is born:

 

None 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

 

 

 

 

 

 

You Your Spouse Total %
Laundry 100
Cleaning 100
Dishes 100
Cooking 100
Shopping 100

 

 

 

 

 

 

207

Expected Division of Childcare Tasks

mm: Using the scale provided, please indicate what percentage of the following
tasks you m to be done by ygp or m.

 

None 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

 

 

 

 

 

You Your Spouse Total %
Feeding the 100
baby
Changing the 100
baby
Playing with 100
the baby
Soothing the 100
baby when
fussy

 

 

 

 

 

 

208

Family Situations Checklist

W Use the scale shown below to‘answer each statement in terms of your own
parents' (or parent's and stepparent's) relationship when you were growing up.

never true occasionally true fairly true very true
1 2 3 4
1. My parents argued with each other in front of me. 1 2 3 4
2. My parents held grudges for a long time when fighting. I 2 3 4
3. My parents fought verbally with each other. 1 2 3 4
4. My mother tried to get me to side with her when she fought with l 2 3 4
5. My parents fought about money and possessions. 1 2 3 4
6. My parents were able to work out disagreements fairly quickly. 1 2 3 4
7. My father tried to get me to side with him when he fought with my 1 2 3 4
mother.

8. I enjoyed being with my father. 1 2 3 4
9. My parents talked together about my future. 1 2 3 4
10. My mother said negative things about my father. 1 2 3 4
11. My parents were able to discuss and work out their disagreements. 1 2 3 4
12. My parents worked together as parents. 1 2 3 4
13. My father said negative things about my mother. 1 2 3 4
14. My parents fought about childrearing and discipline. 1 2 3 4
15. We did things together as a family. 1 2 3 4
16. My parents were both involved in taking care of the children. 1 2 3 4

17. How well do you think your parents got along with each other in their marital

relationship?
1 2 3 4 5
very well badly
18. How well do you think your parents got along with each other in their parenting
relationship?
1 2 3 4 5
very well badly
19. How ofien did your parents get angry with one another?
1 2 3 4 5
never all the time

209

Ideas About Parenting

These statements describe a number of different ideas about raising children. Please
indicate how much you agree with each statement by using the following scale:

1 2 3 4 5 6 7 8 9
very much moderately neither agree moderately very much
agree agree nor disagree disagree disagree

1. Too much cuddling spoils a child
2. A parent should not allow his or her mind to be changed by a child

3. There is a special bond between mothers and their infants which fathers may
even be jealous of.

4. Infants and children should have a daily schedule which they can count on.

5. Children, one year old or younger, don’t feel their parents are really “with them”
if the parents are doing something else at the same time

6. I would feel quite comfortable buying a doll for my son and a truck for my
daughter.
7. Fathers have a special lmack for raising sons. -

8. prarents are ready to go out for the evening and their baby reacts by crying and
screaming, it is best for them to cancel their plans.

9. Babies should be held when they are upset so they will be secure in their parents’
love for them.

10. Even the best parents feel some disgust when cleaning up their infant’s diapers.

11. It is just as important for a girl to be self-assertive and well educated as it is for
a boy.

12. It is best for a preschool child to be cared for almost entirely by his or her own

parents.
13. Babies wish that their parents would stop fussing over them so much

14. It is all right for the mother of an infant to work half-time or more as long as
she has good childcare.

__ 15. A child should not talk back to a parent

__ 16. New parents will feel uncomfortable the first time they wash their baby’s
genitals.

__ 17. Parents should try to avoid open conflict with a child

__ 18. I won’t insist that my child eat food that he or she really dislikes.

__ 19. When a child is called, he or she should come immediately.

_ 20. A mother just naturally knows when to pick up her crying baby.

_ 21. Little boys can be expected to cry as much as little girls.

 

 

210

1 2 3 4 5 6 7 8 9
very much moderately neither agree moderately very much

agree agree nor disagree disagree disagree

22. A child gets a sense of security from a strong father and a sense of comfort fi'om
a warm and loving mother. Ifa child doesn’t have a strong father or a warm and
loving mother, that child loses out on these basic feelings.

23. Babies are frequently so demanding that their parents have no time for anything
else.

24. A six-month-old baby can tell you exactly what he or she wants if you watch
and listen.

25 . Since new babies are fiagile and delicate, they must be handled with extreme
care

_ 26. Good parents should keep a firm hold on their child’s expression of angry
feelings.

__ 27. It is terribly fi'ustrating to take care of a newborn because the infant can’t let
you know what he or she needs. '

28. Breastfeeding is more satisfying to a child than bottlefeeding.

29. There are times in the lives of small children when they need to be with people
other than their mothers and fathers.

30. Taking care of a baby is much more work than pleasme.

31. Children should be encouraged to express their anger as well as their more
pleasant feelings.

32. I want to try to tell my child what to do as little as possible.

33. Young children are only as curious about the world as their parents permit them
to be.

__ 34. I like to see a child have opinions and express them, even to adults.

__ 35. It’s all right for parents to go on a vacation together and leave the baby with a
good sitter.

_ 36. Fathers have a special knack for raising daughters.

__ 37. Although it’s not always easy to do, the best way of handling a child’s temper
tantrum is to ignore it

__ 38. Parents can expect to see differences in a boy or girl child no matter how they
try to raise them alike.

__ 39. An infant’s needs must always come before the parents’ needs.

40. A child of three who is cared for by the parents at home is better off than one
who is sent to daycare.

41. I will try to anticipate situations which could cause my child pain or discomfort
and help my child avoid them if at all possible.

42. One of the joys of parenthood is encouraging a child’s natural curiosity.
43. Other people will probably think I am permissive with my child.

211

44. A child should be punished for breaking his or her own toys in a fit of anger.
45. A mature person seldom gets angry at a child

46. A certain amount of fi'ustration and upset is necessary for a child’s emotional
growth. Because of this, parents should not be too protective.

212

Sentence Completion (M)

Please complete the sentence for each item.

1. Crime and delinquency could be halted if

 

2. Men are lucky because

 

 

3. I just can’t stand pe0ple who

 

 

4. At times he worried about

 

 

5.Iam

 

6. A woman feels good when

 

7. My main problem is

 

8. A husband has a right to

 

 

9. The worst thing about being a man

 

 

10. A good mother

 

 

213

ll.WhenIamwithawoman

 

12. Sometimes he wished that

 

 

13. My father

 

14. HI can’t get what I want

 

15. Usually he felt that sex

 

16. For a woman a career is

 

17. My conscience bothers me if

 

18. A man should always

 

214

Personal Attitude Inventory
Instructions: Please answer each item by circling the appropriate answer.

SD = Strongly Disagree D = Disagree A = Agree SA=Strongly Agree

1. If I receive a lukewarm dish at a restaurant,

I make an attempt to let that be known. SD D A SA
2. I resent authority figures who try to tell me what to do. SD D A SA
3. I find that I often have to question authority. SD D A SA
4. I enjoy seeing someone else do something that
neither of us are supposed to do. SD D A SA
5. I have a strong desire to maintain my personal freedom. SD D A SA
6. I enjoy playing "devil's advocate" whenever I can. SD D A SA
7. In discussions, I am easily persuaded by others. SD D A SA
8. Nothing turns me on as much as a good argument. SD D A SA
9. It would be better to have more freedom to do
what I want on a job. SD D A SA
10. If I am told what to do, I ofien do the opposite. SD D A SA
1 1. I am sometimes afraid to disagree with others. SD D A SA
12. It really bothers me when police officers tell
people what to do. SD D A SA
13. It does not upset me to change my plans because
someone in the group wants to do something else. SD D A SA
14. I don't mind other people telling me what to do. SD D A SA
15 . I enjoy debates with other people. SD D A SA
16. If someone asks a favor of me, I will think twice
about what this person is really after. SD D A SA

215

l7.

l8.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

I am not very tolerant of others' attempts to change me.
I ofien follow the suggestions of others.
I am relatively opinionated.

It is important to me to be in a powerful position
relative to others.

I am very open to solutions to my problems from others.

I enjoy "showing up" people who think they are right.
I consider myself more competitive than cooperative.

I don‘t mind doing something for someone, even when
I don't know why I‘m doing it.

I usually go along with others' advice.

I feel it is better to stand up for what I believe
than to be silent.

I am very stubborn and set in my ways.

It‘s very important for me to get along well
with the people I work with. '

216

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

U

UUUU U

U

{>

> > > > >

11>

11>

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

Motivations About Childbearing

2313.!

M935: Please circle the answer that best represents your feelings about each of the
following consequences of having a child

:Iery Modeiately Slig3htly Not at all

desirable desirable desirable desirable
1. Feeling a baby move and kick inside my wife. 1 2 3 4
2. Having a child who will carry on my family traditions. l 2 3 4
3. Having my child be a success in life. 1 2 3 4
4. Feeling needed and useful through my baby. 1 2 3 4
5. Knowing that I am fertile. 1 2 3 4
6. Giving my wife the satisfaction of motherhood. 1 2 3 4
7. Being the center of a large, active family. 1 2 3 4
8. Playing with my child. 1 2 3 4
9. Having my child provide me with companionship and

support later in life. 1 2 3 4
10. Having my family and friends admire me with my baby. 1 2 3 4
l 1. Helping my wife give birth to a baby. 1 2 3 4
12. Strengthening our marriage through a child. 1 2 3 4
13. Having my child contribute to society. 1 2 3 4
14. Having a helpless baby to love and protect. 1 2 3 4
15. Having a son. 1 2 3 4
16. Bottlefeeding a baby. 1 2 3 4
r7. Fulfilling my religious feelings about family life. 1 2 3 4

217

l 2 3 4

Very Moderately Slightly Not at all
desirable desirable desirable desirable
l8. Guiding and teaching my child 1 2 3 4
19. Feeling more complete as a man through my baby. 1 2 3 4
20. Having a daughter. 1 2 3 4
21. Holding and cuddling a baby. 1 2 3 4
22. Providing my parents with a grandchild 1 2 3 4
23. Sharing childraising with my wife. 1 2 3 4
24. Living a fuller, more enriched life through my child 1 2 3 4
25. Devoting myself and much of my time to raising children
and being a father. 1 2 3 4
26. Fulfilling my potential to have children. 1 2 3 4
27 . Experiencing the special love and closeness that
a child provides. 1 2 3 4
£911]!
W: Please circle the answer that best represents your feelings about each of the
following consequences of having a child Also, please note that “1 " is now “Very
l 2 3 4
Very Moderately Slightly Not at all
rmdesirable undesirable undesirable undesirable

28. Seeing my wife experience the discomforts of pregnancy. 1 2 3 4

29. Having an unhappy and poorly adjusted child 1 2 3 4
30. Having my wife kept from her career or job by a baby. 1 2 3 4
31. Strainingourmarriagewithababy. 1 2 3 4
32. Seeing my wife experience the pain of childbirth. l 2 3 4

218

l 2 3 4

Very Moderately Slightly Not at all
undesirable undesirable rmdesirable undesirable
33. Worrying about the health and safety of my child. 1 2 3 4
34. Being responsible for a needy and demanding baby. 1 2 3 4
35. Having a baby who strains my wife’s health. 1 2 3 4
36. Having a baby who is born deformed 1 2 3 4
37 . Spending time and energy involved in childcare. 1 2 3 4
38. Having a child who is a burden to my wife. 1 2 3 4

39. Worrying about whether I am raising my child the right way. 1 2 3 4
40. Having to put up with the mess and noise that children make. 1 2 3 4

41. Having a child who makes it necessary for my wife
to have a job. 1 2 3 4

42. Having a child who embarrasses or disgraces the rest
of the family. 1 2 3 4

43. Taking care of a baby who is disagreeable and irritating. l 2 3 4

44. Feeling guilty or inadequate as a parent 1 2 3 4
45 . Burdening our family finances with a child 1 2 3 4
46. Taking care of a sick child 1 2 3 4

47. Having a baby who takes away my freedom to do
other things. 1 2 3 4

219

Infant Characteristies Questionnaire

W: What is your perception of your baby’s personality? Circle the number that

is most typical of your baby. “About average” means how you think the typical baby
would be scored

1. How easy or difficult is it for you to calm or soothe your baby when he/she is upset?

I 2 3 4 5 6 7
very easy - about average difficult

2. How easy or difficult is it for you to predict when your baby will go to sleep and wake
up?

1 2 3 4 5 6 7
very easy about average difficult

3. How easy or difficult is it for you to predict when your baby will become hungry?

I 2 3 4 5 6 7
very easy about average difficult
4. How easy or diflicult is it for you to know what’s bothering your baby when he/she
cries or fusses?
l 2 3 4 5 6 7
very easy about average dificult

5. How many times per day, on the average, does your baby get fussy and irritable-for
either short or long periods of time?

1 2 3 4 5 6 7
never 1-2 times 3-4 times 5-6 times 7-9 times 10-14 times more than
perday perday perday perday perday 15

6. How much does your baby cry and fuss in general?

1 2 3 4 . 5 6 7
very little; much average amount; about as a lot; much
less than the much as the average baby more than the
average baby average baby

220

7. How did your baby respond to his/her first bath?

1 2 3 4 5 6 7
very well- neither liked nor terribly-
baby loved it disliked it didn’t like it
8. How did your baby respond to his/her first solid food?
1 2 3 4 5 6 7
very favorably- neither liked nor very negatively-
liked it immediately disliked it didn’t like it at all
9. How does your baby typically respond to a new person?
1 2 3 4 S 6 7
almost always responds favorably about most always responds
responds favorably half the time negatively at first
10. How does your baby typically respond to being in a new place?
1 2 3 4 5 6 7
almost always responds favorably about almost always responds
responds favorably half the time negatively at first
1 1. How well does your baby adapt to things (such as in items 7-10) eventually?
1 2 3 4 5 6 7
very well, ends up liking it almost always
always likes it about half the time dislikes it
eventually in the end
12. How easily does your infant get upset?
I 2 3 . 4 5 6 7
very hard to upset- about average very easily upset by
even by things that things that wouldn’t
upset most babies bother most babies

13. When your baby gets upset (e.g., before feeding, during diapering, etc.), how
vigorously or loudly does he/she cry and fuss?

l 2 3 4 5 6 7
very mild intensity moderate intensity very loud or intense,
or loudness or loudness really cuts loose

221

14. How does your baby react when you are dressing him/her?

l 2 3 4 5 6 7
very well— about average— doesn’t like it
likes it doesn’t mind it at all

15. How active is your baby in general?

1 2 3 4 5 6 7
very calm average very active
and quiet and vigorous

16. How much does your baby smile and make happy sounds?

I 2 3 4 5 6 7
a great deal, an average amount very little
much more than much less than
most infants most infants

17. What kind of mood is your baby generally in?

l 2 3 4 5 6 7
very happy neither serious serious
and cheerful nor cheerful

18. How much does your baby enjoy playing little games with you?

1 2 3 4 5 6 7
a great deal, about average very little,
really loves it doesn’t like it
very much

19. How much does your baby want to be held?

1 2 3 4 5 6 7
wants to be free sometimes wants to be held; a great deal—
most of the time sometimes not wants to be held
almost all the time

20. How does your baby respond to disruptions and changes in everyday routine, such as
when you go to church or a meeting, on trips, etc.?

1 2 3 4 5 6 7
very favorably, about average very unfavorably,
doesn’t get upset gets quite upset

222

21. How easy is it for you to predict when your baby will need a diaper change?

1 2 3 4 5 6 7
very easy about average difficult

22. How changeable is your baby’s mood?

1 2 3 4 5 6 7
changes seldom, and about average changes often
changes slowly when and rapidly
he/she does change

23. How excited does your baby become when people play with or talk to him/her?

l 2 3 4 5 6 7
very excited about average not at all

24. Please rate the overall degree of difficulty your baby would present for the average
parent

1 2 3 4 5 6 7
super easy ordinary, some problems highlzedailfficrtrhlt
to WI

223

Parenting Efficacy Questionnaire

We’d like to get a general idea of how you usually handle different situations with yom
baby. We realize no one is always effective or always ineffective. We all do better in
some situations than others. We’d like you to think about some situations that all parents
encounter.

1. When your baby is upset, fussy, or crying, how good are you at soothing him or her?

1 2 3 4
not good not good good enough very good
at all enough

2. How good are you at understanding what your baby wants or needs. For example, do
you know when your baby needs to changed or wants to be fed?

1 2 3 4

I do not I understand I understand I understand
understand my baby some my baby most my baby all
my baby of the time of the time of the time

3. How good are you at making your baby understand what you want him/her to do? For
example, if you want your baby to eat dinner or play quietly, how good are you at
making him/her do that?

l 2 3 4
not good not good good enough very good
at all enough

4. How good are you at getting your baby to pay attention to you? For example, when you
want your baby to look at you, how good are you at making him or her do it?

1 2 3 4
not good not good good enough very good
at all enough

5. How good are you at getting your baby to have fun with you? For example, how good
are you at getting your baby to smile and laugh with you?

1 2 3 4
not good not good good enough very good
at all enough

224

6. How good are you at knowing what activities your baby will enjoy? For example, how
good are you at knowing what games and toys your baby will like to play with?

1 2 3 4
not good not good good enough very good
at all enough

7. How good are you at keeping your baby occupied when you need to do housework?
For example, how good are you at finding things for the baby when you need to do
dishes?

1 2 3 4
not good not good good enough very good
at all enough

8. How good do you feel you are at feeding, changing, and bathing your baby?

1 2 3 4
not good not good good enough very good
at all enough

9. How good are you at getting your baby to show off for visitors? For example, how
good are you at making your baby smile or laugh for people who visit?

1 2 3 4
not good not good good enough very good
at all enough

10. In general, how good a parent do you feel you are?

1 2 3 4
not good not good good enough very good
at all enough

225

APPENDIX D

Preliminary Analyses for Explanatory Model A: Stable Predictors

226

APPENDIX D

Preliminary Analyses for Explanatory Model A: Stable Predictors

To determine which variables best contributed to an explanatory model of
stable pre-birth predictors, sets of variables were tested separately to assess their
influence independent of their intercorrelations. Several aspects of this strategy
must be noted. First, assimilation, as measured by the PCM and NCC scales, was
tested with both scales entered for both parents simultaneously, but for ease of
presentation, the results are described in two tables, one for mother and father
PCM (Table F2) and one for mother and father NCC (Table F3). In addition,
family of origin variables were correlated within each parent. These variables were
therefore tested separately to determine which, if any, was most influential in
prediction (Tables F4 through F6). For self-reported marital adjustment and for ego
development, mother and father ratings were related and were tested separately,
but are presented in one table (Tables F 1 and F9) to facilitate comparison. Finally,
the two marital interaction scales were correlated across parents. Preliminary
analyses were conducted first assessing the contribution of the parent’s own

marital interactions (Table F10) and then the partner’s marital interactions (Table

F11)

227

Table F1

Prim' l t l:E vl t

 

 

Parameter Unstandardized 1000) Effect size 1:
coefficient
Mothers
Level
MEGO -.OO -.24 .02
FEGO .00 .47 .05
Rate
MEGO -.OO -.29 .03
FEGO -.OO -.82 .08
Curve
MEGO .00 .22 .02
FEGO .00 .39 .04
Fathers
Level
MEGO .01 1.89‘ .19
FEGO .01 1.86‘ .19
Rate
MEGO -.01 -l . lO .1 l
FEGO -.00 -1 .01 . 10
Curve
MEGO .00 l .25 . 13
FEGO 00 1.00 .10

 

Note. MEGO = Mother ego development. FEGO = Father ego development.

Boldface indicates variables retained for further investigation.

t12<.10.

Table F2

 

 

' 1111-4 _ .1.“ = u 01.71;. 'oitiv 1111...:.=__ :_ ' .1o1
Parameter Unstandardized t(100) Efl'ect size r
coefficient
Mothers
Level
MPCM .04 .54 .05
FPCM -.16 -2.00* .20
Rate
MPCM .08 2.15* .22
FPCM -.01 -.10 .01
Curve
MPCM .02 .28 .03
FPCM .09 1.43 .14
Fathers
Level
MPCM .05 .81 .08
FPCM .00 .Ol .00
Rate
MPCM -.OO -.13 .01
FPCM -.O4 -l.47 .15
Curve
MPCM .07 2.01* .20
FPCM .01 .28 .03

 

111mg. MPCM = Mother report of positive childbearing motivations. FPCM =
Father report of positive childbearing motivations. Boldface indicates variables
retained for further investigation.

*1; < .05.

229

Table F3

Prelr -11-; 91- . 0 me] :Neave hi ! rm.“ 1 9

 

 

Parameter Unstandardized 1(100) Efi‘ect size 1‘
coeflicient
Mothers
Level
MNCC -.21 -2.71** .27
FNCC -.03 -.39 .04
Rate
MNCC .06 1.50 .15
FNCC .05 1.21 .12
Curve
MNCC .07 1.07 .1 l
FNCC .03 .49 .05
Fathers
Level
MNCC -.06 -.94 .09
FNCC .05 .67 .07
Rate
MNCC .02 .67 .07
- FNCC .03 .96 .10
Curve
MNCC -.01 -.29 .03
FNCC O4 .86 .09

 

Mtg. MNCC = Mother report of negative childbearing concerns. FNCC = Father
report of negative childbearing concerns. Boldface indicates variables retained for
further investigation.

**p_ < .01.

230

 

Table F4

 

 

Prl' ' =1; 11. .1 1111-: "Tr-1:1- . ' ' 1111 of. '.
Parameter Unstandardized t( 100) Effect size 1’
coefficient
Mothers
Level
MTRIANG -.02 -.28 .03
FTRIANG -.08 -l.79‘ .18
Rate
MTRIANG -.03 -1.20 .12
FTRIANG .01 .38 .04
Curve
MTRIANG .02 .40 .04
FTRIANG .03 .80 .08
Fathers
Level
MTRIANG .03 .61 .06
FTRIANG -.07 -l.72‘ .17
Rate
MTRIANG .01 .81 .08
FTRIANG -.02 -1.01 .10
Curve
MTRIANG .02 .68 .07
FTRIANG .01 .63 O6

 

Egg. MTRIANG = Mother report of triangulation in family of origin. FTRIANG =

Father report of triangulation in family of origin. Boldface indicates variables
retained for firrther investigation.

tp<.10.

 

Table F5

Prl'1,._s E 111 ‘to MelA' Po iiv P n 1 'n 1.1 ‘l o_ 011:.

 

 

Parameter Unstandardized _t_( 100) Effect size 1:
coefficient
Mothers
Level
MPOSPAR -.04 -.86 .09
FPOSPAR .09 l .63 .16
Rate
MPOSPAR .00 . 12 .01
FPOSPAR -.02 -.77 .08
Curve
MPOSPAR .02 .42 .04
FPOSPAR -.01 -.24 .02
Fathers
Level
MPOSPAR .00 .04 .00
FPOSPAR .10 2.25* .23
Rate
MPOSPAR -.01 -.61 .06
FPOSPAR .Ol .41 .04
Curve
MPOSPAR .00 -.O3 .00
FPOSPAR -.01 -.47 .05

 

Note. MPOSPAR = Mother report of positive coparenting in family of origin.
FPOSPAR = Father report of positive coparenting in family of origin. Boldface
indicates variables retained for further investigation.

*2 < .05.

Table F6

 

 

 

Pr 1'
Parameter Unstandardized t( 100) Efl‘ect size 1:
coefficient
Mothers
Level
MPFIGHT -.Ol -.13 .01
FPFIGHT -.05 -l .09 .l 1
Rate
MPFIGHT -.03 -1 .1 l .11
FPFIGHT .02 .64 .06
Rate
MPFIGHT .03 .86 .09
FPFIGHT .04 .99 .10
Fathers
Level
MPFIGHT .Ol .24 .02
FPFIGHT -.07 -1.87‘ .19
Rate
MPFIGHT .OO .40 .04
FPFIGHT -.02 -l .00 .10
Curve
MPFIGHT .03 1.33 .13
FPFIGHT -.00 -.15 .02

 

11919. MPFIGHT = Mother report of marital conflict in family of origin. F PF IGHT
= Father report of marital conflict in family of origin. Boldface indicates variables
retained for further exploration.

‘ p < . 10.

233

Table F7

1'1 1

 

 

Parameter Unstandardized 1(100) Effect size 1:
coefficient
Mothers
Level
WEACT -.O3 -. 19 .02
FREACT -.14 -.83 .08
Rate
MREACT .Ol .09 .01
FREACT .00 .00 .00
Curve
MREACT .21 1.78‘ .18
FREACT .09 .69 .07
Fathers
Level
MREACT -.04 -.33 .03
FREACT .01 .10 .01
Rate
MREACT .02 .29 .03
FREACT -.O6 -.95 . 10
Curve
MREACT -.O3 -.37 .04
FREACT .02 .20 .02

 

13193. MREACT = Mother report of reactance. F REACT = Father report of

reactance.
‘p< .10.

Table F8

 

 

Prelirn' lanat Model A: m hi n V '
Parameter Unstandardized t( 100) Effect size 1:
coefficient
Mothers
Level
' Age .01 1.57 .16
SES .17 3.29*** .33
Rate
Age -.00 -.14 .01
SES -.03 -.88 .09
Curve
Age .01 .92 .09
SES -.07 -l.69‘ .17
Fathers
Level
Age -.01 -1.31 .13
SES .14 2.94“ .29
Rate
Age .00 1.19 .12
SES -.01 -.58 .06
Curve
Age -.00 -.25 .03
SES -.00 .02 .02

 

111mg. Parent ages were correlated .81, age is scored as the mean of both. SES is
computed as the mean of the standardized scores for education and occupation
scores for both parents and family income. Boldface indicates variables retained
for further investigation. ‘12 < .10. "p < .01. "*p < .001

235

Table F9

 

 

Preliminary Explanatogy Mgel A: Self-Reportfl Marital Adjustment
Parameter Unstandardized t(100) Effect size _1;
coefficient
Mothers
Level
MDAS .01 4.03*** .40
FDAS .01 3.58““ .36
Rate
MDAS -.00 -.19 .02
FDAS -.00 -.39 .04
Curve
MDAS -.00 -.21 .02
FDAS -.00 -.39 .04
Fathers
Level
MDAS .01 3.10“ .31
FDAS .02 5.82“" .58
Rate
MDAS .00 .08 .01
FDAS .00 1.23 .12
Curve
MDAS -.00 -.19 .02
FDAS -.00 -.64 06

 

 

Egg. MDAS = Mother Dyadic Adjustment Scale. FDAS = Father Dyadic
Adjustment Scale. MDAS and FDAS estimates are derived fiom separate analyses.
Boldface indicates variables retained for fiirther exploration.

”p < .01. "*p < .001.

236

Table F 10

 

 

Praglimin .X11. .0 M1117 : 1 r ’ _.. 1 _. ‘ ’ 0111 1. -.‘
lamina
Parameter Unstandardized t( 100) Effect size L
coefficient
Mothers
Level
MPMI .10 2.52* .25
MNMI -.21 -5.67*** .57
Rate
MPMI -.02 -.67 .07
MNMI .02 .82 .08
Curve
MPMI .00 .09 .01
MNMI .00 -.10 .01
Fathers
Level
FPMI .11 3.39*** .34
FNMI -.00 -.08 .01
Rate
FPMI -.00 -.11 .01
FNMI -.00 -.14 .01
Curve
FPMI .00 .00 .00
FNMI .01 .31 .03

 

m. MPMI = Mother positive marital interactions. MNMI = Mother negative
marital interactions. FPMI = Father positive marital interactions. FNMI = Father
negative marital interactions. Boldface indicates variables retained for further
investigation. *1; < .05. "*2 < .001.

237

Table Fl 1

 

 

Pre im' E l t M : er’s M ' Interactions
Parameter Unstandardized .t_(100) Efl'ect size r
coeflicient
Mothers
Level
FPMI .09 2.28* .23
FNMI -.06 -l .28 .13
Rate
FPMI .01 .27 .03
FNMI -.03 -1.22 .12
Curve
FPMI -.03 -.90 .09
FNMI -.04 -l .22 .12
Fathers
Level
MPMI .08 2.04 .20
MNMI -.09 -2.52* .25
Rate
MPMI .01 .52 .05
MNMI .01 .49 .05
Curve
MPMI -.02 -.87 .09
MNMI -.01 -.63 06

 

Note. MPMI = Mother positive marital interactions. MNMI = Mother negative
marital interactions. FPMI = Father positive marital interactions. FNMI = Father
negative marital interactions. Boldface indicates variables retained for further
investigation. *9 < .05

238

Table F 12

Prlim' 1.‘ Elm-.1; M111.- A:D’1ren in hilr ' _Phi101h

 

 

Parameter Unstand. coeff. t( 100) Effect size r
Mothers

Level AUTNDIFF .03 .77 .08
AUTVDIFF -.05 -1.05 .1 l
PERMDIFF .02 .35 .04

Rate AUTNDIFF -.00 -. 12 .01
AUTVDIFF .03 1.30 .13
PERMDIFF -.08 -3.08** .31

Curve AUTNDIFF -.00 -. 12 .01
AUTVDIFF .08 230* .23
PERMDIFF -.12 -3.l8** .32

Fathers

Level AUTNDIFF .01 .38 .04
AUTVDIFF -.06 -1.38 .14
PERMDIFF .02 .35 .04

Rate AUTNDIFF .02 l .49 . 15
AUTVDIFF .01 .65 .07
PERMDIFF .01 .50 .05

Curve AUTNDIFF .01 .75 .08
AUTVDIFF .01 .22 .02
PERMDIFF -.02 -.75 .08

 

Note. Unstand. coeff. = Unstandardized coefficient. AUTNDIFF = Differences in
authoritarian beliefs. AUTVDIF F = Differences in authoritative beliefs.
PERMDIFF = Difl'erences in perrnissiveness/protectiveness beliefs. Boldface
indicates variables retained for further analysis.

*2 < .05. "p < .01.

239

Table F13

Preliminm Explanatgg Model A: Infant Sex

 

 

Parameter Unstandard ized 1(100) Efl‘ect size L
coeflicient
Mothers
Level
Infant sex -.07 -1.07 .11
Rate
Infant sex .01 .22 .02
Curve
Infant sex -.01 -.22 .02
Fathers
Level
Infant sex .04 ' .68 .07
Rate
Infant sex .01 .35 .04
Curve
Infant sex .03 .73 .07

 

240

 

 

 

 

REFERENCES

241

REFERENCES

Abidin, R. R (1986). Parenting Stress Index . Charlottesville, VA: Pediatric
Psychology Press.

Abidin, R. R. (1992). The determinants of parenting behavior.J Wm
CW 407-412

Abidin, R. R, & Brunner, J. F. (1995). Development of a parenting alliance
inventory. lgyrnal gf Clinig Child Psychgldgy, 24, 31-40.

Agarwal, P., & Srivastava, S. (1989). Sex-role conception, ego development, and
marital harmony. Indi J of n P c l ' h 4 75-
79.

Ahrons, C. (1981). The continuing c0parental relationship between divorced
Spouses W 415-428

Abuja, P., O’Hearn, M., Fetner, J., Spellman, M., Tamis-LeMonda, C., &
Greenstein, B. (1999). t rnal ch 1 i functionin '

d t
temmment in relatidn to parenting intemtions in low income familig

Paper presented at the biannual meeting of the Society for Research in
Child Development, Albuquerque, NM, April, 1999.

Antonucci, T. C., & Mikus, K. (1988). The power of parenthood: Personality and
attitudinal changes during the transition to parenthood. In G. Y. Michaels &

W. A- Goldberg (Eds), WW
ml; (pp. 62-84). Cambridge: Cambridge University Press.

Barnett, R C. ”Raudenbush S. W. ,Brennan, R. T. ,Pleck, J. H. ,& Marshall, N. L.

(1995). Change 1n job and marital experiences and change 1n psychological
distress: A longitudinal study of dual-earner couples. W

W 839-850

Bates, J. E., Freeland, C. B., & Lounsbury, M. L. (1979). Measurement of infant
diflicultness. WW 794-803.

Baumrind, D. (1967). Child care practices anteceding three patterns of preschool
behavior. WM 43-88.

242

Bearss, K. E., & Eyberg, S. (1998). A test of the parenting alliance theory. Eflbl
Eddgation and ngeldpment, 9, 179-185.

Belsky, J. (1984). The determinants of parenting: A process model. Qhfld
ngglopment, d5, 83-96.

Belsky, J. (1985). Exploring individual differences in marital change across the
transition to parenthood: The role of violated expectations. W

Marriage and the Family, 1037-1044.

Belsky, J ., Cmic, K., & Gable, S. (1995). The determinants of coparenting in
families with toddler boys: Spousal differences and daily hassles. Child

Evelopment, 66, 629-642.

Belsky, 1., & Hsieh, K. H. (1998). Patterns of marital change during the early
childhood years: Parent personality, coparenting, and division-of-labor

correlates. f i1 01 1 511-528.

Belsky, J ., & Isabella, R. A. (1985). Marital and parent-child relationships in
family of origin and marital change following the birth of a baby: A

retrospective analysis. Wig, 342-349.

Belsky, J. ., Lang, M. E. ,& Rovine, M. (1985). Stability and change 1n marriage
across the transition to parenthood. A second study. Wage

andtheFamilx, 855-865

Belsky, J. ., & Pensky, E. (1988). Marital change across the transition to
parenthood MWdEamilxResz. 133-156

Belsky, J ., Putnam, S., & Cmic, K (1996). Coparenting, parenting, and early
emotional development. In J. P. McHale & P. A. Cowan (Eds),
1e . 1in_ how famil -level1 1 LL! -ie1r hildre1' 1-vl1 111 n'

W (pp. 45-55). San Francisco: Jossey-Bass.

Belsky, J. ., & Rovine, M. (1990). Patterns of marital change across the transition to
parenthood Pregnancy to three years postpartum laumalnfMam’ageand
the Family, 52, 5-19.

Belsky, J ., Rovine, M., & Fish, M. (1989). The developing family system. In M. R.

Gunnar & E. Thelen (Eds), W (V 01. 22, ). Hillsdale,
NJ: Lawrence Erlbaum Associates.

243

Belsky, J. ,Spanier, G. B. ,& Rovine, M. (1983). Stability and change 1n marriage

across the transition to parenthood MW
567-577.

Belsky, J. ., & Volling, B. L. (1987) Mothering, fathering, and marital interaction in
the family triad during infancy. In P. W. Berman & F. A. Pedersen (Eds. ),
116'11111. 1 11 n 111111n11111 111‘ 1-x.1 12.111

emrience (pp. 37-63). Hillsdale, NJ: Lawrence Erlbaum Associates.

Belsky, J. ., Youngblade, L. ,Rovine, M. ,& Volling, B. (1991). Patterns of marital

change and parent-child interaction. W
88, 487-498.

Bentler, P M. ,& Bonett, D. G. (1980). Significance tests and goodness of fit 1n
the analysis of covariance structures. W88, 588-606.

Bentler, L. ,Sandowicz, M. F,isher, D. ,& Albanese, A. (1996). Resistance in
psychotherapy: What conclusions are supported by research? mm;

W 77-86

Bielby, W. T. & Bielby, D. D. (1989). Family ties: Balancing commitments to

work and family 1n dual-eamer households. W
L4, 776-789.

Block, J. H., Block, J ., & Morrison, A. (1981). Parental agreement-disagreement
on child-rearing orientations and gender-related personality correlates in

children. WW 965-974.
Bosquet, M. (1999). -

 

‘11:! 1 .0 ,.-.1°10 mm. . :.1 111‘6-4 I 1m: 1 L1; en Paper
presented at the biannual meeting of the Society for Research rn Child
Development, Albuquerque, NM, April, 1999.

Bowen, M. (1978). W. New York: Jason Aronson.

Brehm, J. (1966). W. New York: Academy Press.

Brody, G. H., & Flor, D. L. (1996). Coparenting, family interactions, and
competence among Afiican American youths. In J. P. McHale & P. A.

Cowan (Eds.,) nde h wfami -l V

W (pp 77-91) San Francisco:1
Jossey-Bass.

244

Brody, G. H., Flor, D. L., & Neubaum, E. (1998). Coparenting processes and child
competence among rural African-American families. In M. Lewis, C.

Feiring, et al. (Eds), Earnings, risk, and compgtgnce (pp. 227-243).
Mahwah, NJ: Lawrence Erlbaum.

Brody, G. H., Pellegrini, A. D., & Sigel, I. E. (1986). Marital quality and mother-
child and father-child interactions with school-aged children.

nggldpmgntal Psychglggy, 22, 291-296.

Brody, G. H., Stoneman, Z. ,& Flor, D. (1995). Linking family processes ad
academic competence among rural Afiican American youths. W

Mamageandrlniaurflnfl, 567-579

Brody, G. H., Stoneman, Z., Flor, D., McCrary, C., Hastings, L., & Conyers, O.
(1994). Financial resources, parentpsychological firnctioning, parent co-
caregiving, and early adolescent competence in rural two-parent African-

American families. Child mvelgpmem.‘ 85, 590-605.

Browne, M. W. ,& Cudeck, R. (1993). Alternative ways of assessing model fit. In

K A Bollen&J S Long(EdS) MW.
Newbury Park, CA. Sage.

Bryk A S Raudenbush S & Congdon, R (1996) MW

111-111- 111 .1? )- -_Tal_1LI 1-1. @114} 11'91112 11 .Chicago:
Scientific Software Internaltional.

BryKA S & Raudenbush, S W (1992) MW
Anuhcatinnsanddaaanalxsiamethach Newbury Park, CA Sage

Buchanan, C. M., Maccoby, E. E., & Dombusch, S. M. (1991). Caught between

parents: Adolescents' experiences in divorced homes. W918,
82, 1008-1029.

Camara, K. A. ,& Resnick, G. (1989). Styles of conflict resolution and cooperation
between divorced parents: Efiects on child behavior and adj ustrnent.

WWW 560-575

Campbell, D. T., & Fiske, D. W. (1969). Convergent and discriminant validation
by the mulitrait-multimethod matrix. W 81-105.

245

Chavez, D., Corkery, L., Cowan, C., Cowan, P., Epperson, R, Gordon, D., Kline,
M., McHale, J., Soule, C., Sullivan, C., & Weinberg, G. (1987 ). Parents
and partneps: A preventive interventipn fer me pgente pf preeehppl-ggfl

ehildrep. Paper presented at the American Orthopsychiatric Association,
San Francisco, CA.

Cohen, J. (1988). Sm'gical ppwe; @515 fer the mhavipral miepees. (2nd ed.).
Hillsdale, NJ: Lawrence Erlbaum Associates.

Cohen, R. S ,& Weissman, S. H. (1984). The parenting alliance. In R S. Cohen,

B J Cohler & S H Weissman (EdS) W
W flap. 3349). New York: The Guilford Press.

Cohn, L. D. (1991). Sex differences 1n the course of personality development: A
meta-analysis MW 252-266

Cowan, C. P. ,& Cowan, P. A. (1988a). Who does what when partners become

parents: Implications for men, women, and marriage. MW
Review, 12, 105-131. ~

Cowan, C. P., & Cowan, P. A. (1992). MW
ehapgejopeppples. New York: Basic Books.

Cowan, C. P., Cowan, P. A., Coie, L., & Coie, J. D. (1978). Becoming a family:
The impact of a first child's birth on the couple's relationship. In W. B.

Miller & L. F. Newman (Eda). Wm (pp.
296-324). Chapel Hill, NC: Carolina Population Center.

Cowan, C. P. ,Cowan, P. A. ,Heming, G. ,Garrett, E. ,Coysh, W. S. ,C,urtis-Boles
H. ,& Boles, A. J. (1985). Transitions to parenthood: His, hers, and theirs.

WHMSI

Cowan, P. A. ,& Cowan, C. P. (1988b). Changes in marriage during the transition
to parenthood. Must we blame the baby? In G. Y. Michaels & W. A.

Goldberg (Eds.,) ll.‘ — _ -
(pp. 114—154). Cambridge: Cambridge University Press.

 

Cowan, P. A., & McHale, J. P. (1996). Coparenting in a family context: Emerging
achievements, current dilemmas, and future directions. In J. P. McHale & P.
A. Cowan (Eds. ), nde din how f 'l l-level d

W (pp 93-106) San

Francisco: Jossey-Bass.

246

 

Cox, M. J., Tresch-Owen, M., Lewis, J . M., & Henderson, V. K. (1989). Marriage,
adult adjustment, and early parenting. Wm 1015-1024.

Crohan, S. E. (1996). Marital quality and conflict across the transition to
parenthood in African American and White couples. We

and the Family, 58, 933-944.

Cummings, E. M. (1994). Marital conflict and children's functioning. 331;;
mm 16-36.

Davies, P. T., & Cummings, E. M. (1994). Marital conflict and child adjustment:
An emotional security hypothesis. Psychdlpgieal Bjdletin, 11d, 387-411.

De Luccie, M. F. (1995). Mothers as gatekeepers: A model of maternal mediators
of father involvement WWW 115-131

Deal, J. E., Halverson, C. F ., Jr., & Wampler, K. S. (1989). Parental agreement on
child-rearing orientations: Relations to parental, marital, family, and child

characteristics. Wag, 1025-1034.

Deutsch, F. M. ,Brooks-Gunn, J. ., Fleming, A. ,Ruble, D. N. ,& Stangor, C. (1988).
Infonnation-seeking and maternal self-definition during the transition to

motherhood WWW 420-431

Dowd, E., Milne, C. ,& Wise, S (1991). The Therapeutic Reactance Scale. A

measure of psychological reactance. Wand
Develppmenp d2, 541-545.

Dowd, E., & Wallbrown, F. (1993). Motivational components of client reactance.

MW 1533-538

Dowd, E. ”Wallbrown F, Sanders, D. ,& Yesenosky, J. (1994). Psychological
reactance and its relationship to normal personality variables. ngnthe

WW 601-612

Dozier, B. S., Sollie, D. L., Stack, S. J ., & Smith, T. A. (1993). The effects of
postdivorce attachment on coparenting relationships. W

W19, 109-123.

247

Duncan, T. E., Duncan, S. C., & Hops, H. (1996). The role of parents and older
siblings in predicting adolescent substance use: Modeling development via
structual equation latent growth methodology. J of F 11

Esyehology, 10, 158-172.

Duncan, T. E. ”Duncan S. C. ,Strycker, L. A. ,Li, F. ,&A1pert,A. (1999). An
1111111 tiont1late tv b11111c -m-1111112'1;1- . 11. :1
appneatma. Mahwah, NJ. Lawrence Erlbaum Associates, Inc.

Duxbury, L. A. ,& Higgins, C. A. (1991). Gender differences in work-family
conflictJ W 60-74

Dyer, E. D. (1963). Parenthood as crisis: A re-study. ' e an 1
196-201.

Easterbrooks, M. A., & Emde, R N. (1988). Marital and parent-child relationships:
The role of affect in the family system. In R. A. Hinde & J. Stevenson-

Hinde (Eds), 'nhi wi i f '1i :M ' 11 (pp. 83-
103). Oxford: Clarendon Press.

Engfer, A. (1988). The interrelatedness of marriage and the mother-child
relationship. In R. A. Hinde & J. Stevenson-Hinde (Eds), Relatipnsln'ps

W (pp. 104-1 18). Oxford: Clarendon Press.

Erel, 0., & Burman, B. (1995). Interrelatedness of marital relations and parent-

child relations: A meta-analytic review. W22, 108-
132.

Feldman, R, Greenbaum, C. W., Mayes, L. C., & Erlich, S. H. (1997). Change in
mother-infant interactive behavior: Relations to change in the mother, the

infant,andthesocialcontext. 'r dDe l n 2 151-
163.

Feldman, S. S., Nash, S. C., & Aschenbrenner, B. G. (1983). Antecedents of
fathering. MW 1268-1363.

F erketich, S. L., & Mercer, R T. (1995). Predictors of role competence for
experienced and inexperienced fathers. W 89-95.

248

F ivaz-Depeursinge, E., Burgin, D., Corboz-Warnery, A., Lebovci, 8., Stem, D.,
Byng-Hall, J., & Lamour, M. (1994). The dynamics of interfaces: Seven
authors in search of encounters across levels of description of an event

involving a mother, father, and baby. Inf t n H J 1 69-

89.

F ivaz-Depeursinge, E., & Corboz-Wamery, A..(l999). W
W. New York: Basic Books Perseus,
Harper/Collins.

Fivaz-Depeursinge, E. F rascarolo F. ,& Corboz-Warnery, A. (1996). Assessing
the triadic alliance between fathers, mothers, and infants at play. In J. P.

McHale & P. A. Cowan (Eds) understanding hdw family-level dmajnies
1131‘ ch1-‘n 1v1111171: 1-11111 1 tw111. 1t 11. 1111:. (pp. 27-44).

San Francisco: Jossey-Bass.

Floyd, F. J ., Gilliom, L. A., & Costigan, C. L. (1998). Marriage and the parenting
alliance: Longitudinal prediction of change in parenting perceptions and
behavior. W62, 1461-1479.

Floyd, F. J., & Zmich, D. E. (1991). Marriage and the parenting partnership:
Perceptions and interactions of parents with mentally retarded and typically

deve10ping children. W 1434-1448.

Frank, S., Jacobson, 8., Hole, C. B., Justkowski, R, & Huyck, M. (1986).
Psychological predictors of parents' sense of confidence and control and

self- versus child-focused gratifications. W22,
348-355.

Frank, S. J. (1988). The Family Sim'pns Qheeklist. Unpublished manuscript.
Michigan State University, East Lansing, MI.

Frank,S. J. ,&Burke, L. (1992) I-id' =' z. '
gepli 1ieat ens and ext ensipns of earlier findings Paper presented at the

biannual meeting of the Society for Research on Adolescence, Washington,
DC.

 

Frank, S. J., Jackson-Walker, 8., Marks, M., Van Egeren, L. A., Loop, K., &
Olson, K. (1998). From the laboratory to the hospital, adults to adolescents,
and disorders to personality: The case of psychological reactance. M

W 361-381

249

Frank, S. J., Jacobson, S. ,.&Avery,C (1988). W
W. Unpublished manuscript, Department of Psychology, East

Lansing, MI.

Frank, S. J., Olmsted, C. L., Wagner, A. E., Laub, C. C., Freeark, K., Breitzer, G.
M., & Peters, J. M. (1991 ). Child illness, the parenting alliance, and

Parenting stress. WW 361-3711
Frank, S. J., Poorman, M. O., & Charles, M. (1994). W

111er11r 1 ' 1111. 11-1- - 11.111ia‘11wi11te 111oles e t r 1 111th
pmsses . Unpublished manuscript, Department of Psychology, Michigan
State University.
Frank, S111,& Tucr, M1 (1988). MW Paper
presented at the annual meeting of the American Psychological Association,
Atlanta, GA.

Frascarolo, F. ,Fivaz-Depeursinge, E., & Corboz-Warnery, A. (1999). 0A
1'11. 1.1. 111111111 111 .1111». 1".1111r11 111
interaetipn? Paper presented at the biannual meeting of the Society for
Research m Child Development, Albuquerque, NM, April, 1999.

Frosch, C. A., Mangelsdorf, S. C., & McHale, J. L. (1998). Correlates of marital

behavior at 6 months postpartum. WW 1438-
1449.

Fuller, S. ,& Swensen, C. H. (1992). Marital quality and quality of life among

cancer Patients and their Spouses W
41-56.

Gable, S. ,,Belsky J, & Cmic, K. (1992). Marriage, parenting, and child

development Progress and Prospects WW 276-
294.

Gable, S. ,Belsky, J. ., & Crnic, K. (1995). Coparenting during the child's 2nd year.
A descriptive account ImmatofMamagsandthsEamilxél. 609-616

Goldberg, W. A., & Easterbrooks, M. A. (1984). Role of marital quality in toddler
deve10pment WW 504-5141

250

Gorvine, B. J.(1999).P en on 01 - 1 sitiv afi‘ aren 'n

h '1 i in fam'l Paper presented at the biannual meeting
of the Society for Research in Child Development, Albuquerque, NM,
April, 1 999.

Gottrnan, J. M. (1994). What prediejs divprce? The relatipnehip mtwgn n1an'_tal_
preeeseee and marital duteprnes. Hillsdale, NJ: Lawrence Erlbaum.

Grych, J. H., & Fincham, F. D. (1990). Marital conflict and children's adjustment:
A cognitive-contextual fi’amework. Pacholpgifl Bnllgtin, 198, 267-290.

Hackel, L. S. ,& Ruble, D. N. (1992). Changes in the marital relationship afier the
first baby 15 born: Predicting the impact of expectancy disconfirmation.

mmwmmmfl, 944-957

Haley, J. (1976). Problem-golfing Lumpy. San Francisco: Jossey-Bass.

Han'ison, M. J. (1990). A comparison of parental interactions with term and
Preterm infants WW 173-179

Hauser, S. T. (1978). Ego development and interpersonal style in adolescence.
W 333-352-

Hauser, S. T., Powers, S. I., Noam, G. G., Jacobson, A. M., Weiss, B., &
F ollansbee, D. J. (1984). Familial contexts of adolescent ego development.

CW, 195-213-

Hayden, L. C. ,S,chiller M. ,,Dickstein S, Seifer, R. Sarnerofi‘, A. J., Miller, I.,
Keitner, G., & Rasmussen, S. (1998). Levels of family assessment 1.

Family, marital, and parent-child interaction. Wm
12, 7-22.

Heinicke, C. M., Diskin, S. D., Ramsey-Klee, D. M., & Given, K. (1983). Pre-birth
parent characteristics and family development in the first year of life. Child
Wit, 194-2081

Heinicke, C. M., & Guthrie, D. (1992). Stability and change in husband-wife
adaptation and the development of the positive parent-child relationship.
WM 109-127.

Heinicke, C. M., & Guthrie, D. (1996). Prebirth marital interactions and postbirth
marital development. W 140-151.

251

Hellman, C, & McMillin, W. (1997). The relationship between psychological
reactance and self-esteem. W21, 135-138.

Helson, R, & Wink, P. ( 1987). Two conceptions of maturity examined 1n the

findings of a longitudinal studyo 1W
Psychdlpgy, 53, 531-541.

Heming, G, Cowan, P. C. ,& Cowan, C. P. (1990). Ideas about parenting. In J.

Touliatos, P. F. Perlmutter, & M. A. Strauss (Eds.,) Handbppkeflamfly
megurement teehniguee (pp. 362-363). Thousand Oaks, CA: Sage.

Hinshaw, S. P., Zupan, B. A., Simmel, C., Nigg, J. T., & Melnick, S. (1997). Peer
status in boys with and without attention-deficit hyperactivity disorder:
Predictions fi'om overt and covert antisocial behavior, social isolation, and

authoritative parenting beliefs. anldmnelpprnengfifi, 880-896.

Hobbs, D. F ., Jr. (1965). Parenthood as crisis: A third study. 1W
andflrsfiamilxgl, 367-3721

Hobbs, D. F. (1968). Transition to parenthood: A replication and an extension.
fM ' e theF 11 413-417.

Hobbs, D. F. ,& Cole, S. P. (1976). Transition to parenthood. A decade replication.
MW 723-731

Hobbs, D. F., & Wimbish, J. M. (1977). Transition to parenthood by black couples.
0 ofM ' e d F 1 677-689.

Hockenberry, S. L., & Billingham, R. E. (1993). Psychological reactance and
violence within dating relationships. WM 1203-1208.

Hoffman, L. W. (1978). Efl‘ects of the first child on the woman's role. In W. B.

Miller & L F- Newman (Ede), WW (PP-
340-367). Chapel Hill, NC: Carolina Population Center.

Hofl'man, L. W. (1983). Increased fathering: Effects on the mother. In M. Lamb &
A. Sagi (Eds), 1 2.: . , .
167-190). Hillsdale, NJ: Erlbaum.

 

Hong, S. (1990). Effects of sex and church attendance on psychological reactance.
won. 494

252

Hong, S. ,& F aedda, S. (1996). Refinement of the Hong Psychological Reactance
Scale WWM 173-182

Hong, S., & Page, S. (1989). A psychological reactance scale: Development, factor
structure, and reliability. Peychdldgical Rem, 64, 1323-1326.

Howes, P., & Markman, H. J. (1989). Marital quality and child functioning: A
longitudinal investigation. ChddDeyelppmanLfiQ, 1044-1051.

Hy, L. X., & Loevinger, J. (1996). W1 (2nd ed.).
Mahwah, NJ: Lawrence Erlbaum Associates.

Isabella, R A. ,& Belsky, J. (1985). Marital change during the transition to
parenthood and security of infant-parent attachment. W

ma, 505-522.

J acoby, A. P. (1969). Transition to parenthood: A reassessment. M
W 720-727

Joreskog, K., & Sorbom, D. (1998). I RE . 0 2 . Chicago:
Scientific Software International.

J oubert, C. (1990). Relationship among self-esteem, psychological reactance, and
other personality variables. 13W 1 147-1 151.

Jouriles, E. N., Farris, A. M., & McDonald, R. (1991a). Marital functioning and
child behavior: Measuring specific aspects of marital relationships. In J.

Vincent (F141), WWW
(Vol. 5, pp. 25-46). London: Kingsley.

Jouriles, E. N., Murphy, C. M., Farris, A. M., Smith, D. A., Richters, J. E., &
Waters, E. (199 l b). Marital adjustment, parental disagreements about child
rearing, and behavior problems in boys: Increasing the specificity of the

marital assessment. W 1424-1433.

Jouriles, E. N., Pfifl‘ner, L. J ., & O'Leary, S. G. (1988). Marital conflict, parenting,

and toddler conduct Problems. WW
1 97-206.

Julien, D., Markman, H. J ., & Lindahl, K. M. (1989). A comparison of a global and
a microanalytic coding system: Implications for future trends in studying

interactions. W 81400.

253

Kach, J. A., & McGhee, P. E. (1982). Adjustment of early parenthood: The role of

accuracy of preparenthood experiences. lg;u_'nal of Family Issues. 3, 375-
388.

Kalmuss, D., Davidson, A., & Cushman, L. (1992). Parenting expectations,
experiences, and adjustment to parenthood: A test of the violated

expectations framework. 19mg of Marriage and the Family, 54, 513-526.

Kamey, B. R, & Bradbury, T. N. (1995). Assessing longitudinal change in
marriage: An introduction to the analysis of growth curves. W

W 1091- 1108

Kamey, B. R, & Bradbury, T. N. (1997). Neuroticism, marital interaction, and the

trajectory of marital satisfaction WWW
12W 1075-1092

Katz, L. F., & Gottrnan, J. M. (1996). Spillover efl‘ects of marital conflict: In search
of parenting and coparenting mechanisms. In J. P. McHale & P. A. Cowan
(Eds. ), U11 6 din h w famil -level (1 h'

WWW (pp0 57-76) San Francisco

J ossey—Bass

Kerig, P., Cowan, P., & Cowan, C. (1993). Marital quality and gender differences
in parent-child interaction. WW 931-939.

Kerig, P. K. (1995). Triangles in the family circle: Effects of family structure on

marriage, parenting, and child adjustment. 0 of F 'l h
2(1), 28-43.

Klinetob, N. A. ,& Smith, D. A. (1996). Demand-withdraw communication in
marital interaction: Tests of interpersonal contingency and gender role

hYPOflloSCS MW 945-957

Kluwer, E. ,Heesink, J., & Van De Vliert, E. (1997). The marital dynamics of
conflict over the division of labor.
635-653.

Kolak, A. M- (1999). WM Paper Presented
at the biannual meeting of the Society for Research in Child Development,
Albuquerque, NM, April, 1999.

 

254

Kurdek, L. A. (1998). The nature and predictors of the trajectory of change in
marital quality over the first 4 years of marriage for first-married husbands

and wives. a1 11 P 1 494-510.

Lane, A. R, Wilcoxon, S. A., & Cecil, J. H. (1988). Family-of-origin experiences
and the transition to parenthood: Considerations for marital and family

therapists. Family Them, 15, 23-29.

LaRossa, R, Gordon, R J., Wilson, A., Bauran, A., & Jaret, C. (1991). The
fluctuating image of the 20th century American father. W
and the Family, 53, 987-997.

Laub C (1990) WWW

WEI; Unpublished master‘s thesis
Michigan State University, East Lansing, MI.

Leifer, M. (1977). Psychological changes accompanying pregnancy and
motherhood W 55-96

LeMasters, E. E. (1957). Parenthood as crisis. WW2,
352-355.

Lerner, R. M. (1993). A developmental contextual view of human development. In
S. C. Hayes, L. J. Hayes, H. W. Reese, &T. R. Sarbin(eds.,) m

W®p 301-316). Reno, NV: Context Press.

Levy-Shifl', R (1994). Individual and contextual correlates of marital change

across the transition to parenthood. MW 591-
604.

Lewis, J. M. (1988). The transition to parenthood: 11. Stability and change in
marital structure. MW 273-283.

Lim,K G &Clements M L (19r99) We

f aren chi] ender. rPaper
presented at the biannual meeting of the Society for Research on Child

Development, Albuquerque, NM, April, 1999.

Lindahl, K. M. ,,Clements M, & Markman, H. (1997). Predicting marital and
parent functioning in dyads and triads: A longitudinal investigation of

marital Processes Wax, 138-151

255

Lindahl, K. M., Clements, M., & Markman, H. (1998). The development of
marriage: A 9-year perspective. In T. Bradbury, et al. (Eds), Ihe

developmental eeyge of mmtal dysfunction (pp. 205-236). New York:
Cambridge University Press.

Lips, H. (1997). Sex ans; gender. (3rd. ed.). Mountain View, CA: Mayfield.
Loevinger, J. (1976). W- San Francisco: Jossey-Bass.

Loevinger, J. (1985). Revision of the Sentence Completion Test for ego
development WWW 420-427

Loevinger, 1., & Wessler, R (1970). W
W. San Francisco Jossey-Bass

Loevinger, J., Wessler, R, & Redmore, C. (1970). W
W San Francisco Jossey-Bass

Loukas, A., Twitchell, G. R, Piejak, L. A., Fitzgerald, H. E., & Zucker, R. A.
(1998). The family as a unit of interacting personalities. In L. L'Abate et al.,
(Ed), or ' u 1 0 I - 1 01014211: '1. 01-; _m. o t. 11-4. run. a: ioavi
(pp. 35-59). New York: Guilford Press.

Lytton, J ., Watts, D., & Dunn, B. E. (1987). Twin-singleton differences in verbal
ability: Where do they stem from? M 359-369.

Maccoby, E. E. (1984). Middle childhood in the context of the family. In W. A.

Collins (Ed) W Washington, DC:
National Academy.

Maccoby, E. E. ,,Depner C. ,& Mnookin, R (1990). Coparenting 1n the second
year after divorce NW 141-155

McArdle, J. J ., & Epstein, D. (1987). Latent growth curves within developmental
structural equation models. W 110-133.

McBride, B. A., & Rane, T. R (1998). Parenting alliance as a predictor of father
involvement: An exploratory study. Eamflyfielefigngfl 229-236.

McHale, J. P. (1995). Coparenting and triadic interactions during infancy: The

roles of marital distress and child gender. vel n h l 1
985-996.

256

McHale, J. P. (1997). Overt and covert coparenting processes in the family. Emily
M 183-201.

McHale, J. P. (1999). Pre-birth precursore te whole-mily gflijy ever the

WW Paper Presented at the biannual
meeting of the Society for Research in Child Development, Albuquerque,

NM, April, 1999.

McHale, J. P., & Cowan, P. A. (Eds). (1996). WWW
ami ec hi dren' d vel ment' S ies f tw arent 1’ .San
Francisco: Jossey-Bass.

McHale, J. P., Kuersten, R, & Lauretti, A. (1996). New directions in the study of
family-level dynamics during infancy and early childhood. In J. P. McHale
& P. A. Cowan (Eds. ), de din how i] -1ev l '
i v 1 m n i o o- t il' (pp. 5-26). San
Francisco: J ossey-Bass.

McHale, J. P., & Rasmussen, J. L. (1998). Coparental and family group-level
dynamics during infancy: Early family precursors of child and family

functioning during Preschool. WW 39-
59.

Mebert, C. J. (1991). Variability in the transition to parenthood experience. In K.

Pillemer & K- McCartney (Eds), WM (PP-
43-57). Hillsdale, NJ: Lawrence Erlbaum Associates.

Mercer, R T. ,& Ferketich, S. L. (1995). Experienced and inexperienced mothers'

maternal competence during infancy WW
333-343.

Miller, B., & Sollie, D. (1980). Normal stresses during the transition to
Parenthood. W 459-465.

Miller, W. B. (1981). W. Washington, DC: National

Technical Information Service.

Miller, W. B. (1995). Childbearing motivation and its measurement. W
W 473-487

257

Minuchin, P. (1988). Relationships within the family: A systems perspective on
development. In R. A. Hinde & J. Stevenson-Hinde (Eds), Remix;
within mines; Mutual influenees (pp. 7-25). Oxford: Clarendon Press.

Minuchin, S., & Fishman, H. C. (1981). F aegily therapy teehnigeg. Cambridge:
Harvard University Press.

Moss, P., Bolland, G., F oxman, R, & Owen, C. (1986). Marital relations during

the transition to Parenthood. W
W 57-67.

Nettles, E. J., & Loevinger, J. (1983). Sex role expectations and ego level 1n

relation to problem marriages mmmmmmmm
_s, 676-687.

Noam, G. G., Hauser, S. T., Santostefano, 8., Garrison, W., Jacobson, A. M.,
Powers, S. I., & Mead, M. (1984). Ego development and psychopathology:
A study of hospitalized adolescents. MW 184-194.

Nock, S. L. (1981). Family life-cycle transitions: Longitudinal effects on family
members Wail. 703-714

Noller, P., & Fitzpatrick, M. A. (199.1). Marital communication. In A. Booth (Ed),
- - - .- .Minneapolis:

 

National Council on F am1ly Relations. 9 I

Novy, D. M., & Francis, D. J. (1992). Psychometric properties of the Washington

University Sentence Completion Test. WW
W52, 1029-1039.

O'Brien,M.,Peyton, V. D. &Roy,C. AM (1999), w.|. l v . ti .
ilM-I'C 11°11“ cox-L 1 o: ‘gr‘hgr‘ i1 i‘,0g1.'i0m06m

gusfigctien? Paper presented at the biannual meeting of the Society for
Research in Child Development, Albuquerque, NM, April, 1999.

Oliver, R, Mattson, D., & Moore, J. (1993). Psychological reactance in marital
enrichment training. W 3-10.

Owen, M. T., & Cox, M. J. (1997). Marital conflict and the development of infant-
parent attachment relationships. MW 152-165.

258

Paley, B. C,ox, M. ,Kanoy, K., Harter, K. ,& Margand, N. (1999).Ad1;l_
«1.; n ‘1 ‘ 41m. -- 1h - 1111reo 1f mlef 11
m teractiegs Paper presented at the biannual meeting of the Society for
Research in Child Development, Albuquerque, NM, April, 1999.

Palkovitz, R. (1985). F athers' birth attendance, early contact, and extended contact
with their newborns: A critical review. W 392-406.

Parke, R. D., Omstein, P. A., Rieser, J. J ., & Zahn-Waxler, C. (1994). The past as
prologue: An overview of a century of developmental psychology. In R. D.
Parke, P. A. Omstein, J. J. Rieser, & C. Zahn-Waxler (Eds), A eengery ef

dexeloprnentalmhology (PP- 1-70): Westview Press.

Pirsch,L.(l990). 1:111". 2411-6411'31- - 1f1.:_ 3144; 'V11 ' 1‘ 11 r..- r-
child integeegegs Unpublished master’s thesis, Michigan State University,
East Lansing, MI.

Pleck, J. H. (1997). Father involvement: Levels, sources, and consequences. In M.

E LambCEd) Warm PP
147-179). New York. Wiley.

Raudenbush, S. W., Brennan, R. T., & Barnett, R C. (1995). A multivariate
hierarchical model for studying psychological change within married

cooples. MW 161-174.

Redmore, C., & Waldman, K. (1975) Reliability of a sentence completion measure
of ego deveIOPment W32, 236-243

Reid, W., & Crisafulli, A. (1990). Marital discord and child behavior problems: A
meta-analysis. MW 105-117.

Rogosa, D. (1979). Causal models in longitudinal research: Rationale, formulation,
and interpretation. In J. R Nesselroade & P. B. Baltes (Eds), Lengimdinal

MW (PP 263-302) New
York: Academic Press.

Rossi, A. S. (1968). Transition to parenthood. Willi.
332, 26-39.

Rothbart, M. K. (1981). Measurement of temperament in infancy. Child
Dexelonmentjz, 569-578.

259

Ruble, D. N., Fleming, A. S., Hackel, L. S., & Stangor, C. (1988). Changes in the
marital relationship during the transition to first time motherhood: Efl‘ects of
violated expectations concerning division of household labor. M

Pegeenality gt! Smiel Embelegy, 55, 78—87.

Russell, A., & Russell, G. (1994). Coparenting early school-age children: An
examination of mother-father interdependence within families.

Qevelepmental Psychelegy, 30, 757-770.

Russell, C. S. (1974). Transition to parenthood: Problems and gratifications
W 294-301.

Rustia, J. G., & Abbott, D. (1993). Father involvement in infant care: Two
longitudinal studies. WWW 467-476.

Sabatelli, R M. ,& Waldron, R. J. (1995). Measurement issues in the assessment

of the experiences of parenthood MangaMMMfl.
969-980.

Sanchez, L., & Thomson, E. (1997). Becoming mothers and fathers: Parenthood,
gender, and the division of labor. WM 747-772.

Schoppe, S. J ., Mangelsdorf, S. C., Charlton, T., Dorris, M., & Frosch, C. (1999).
1 112- nin_ - df i1 - lian 2‘ 1141- '1- ith‘it‘ - iz°L 1 111‘
fiem 3 te 4 ym. Paper presented at the biannual meeting of the Society for
Research in Child Development, Albuquerque, NM, April, 1999.

Shap1roAF(1999)lr1l1f-1=-151-11.1421 1f1 11_1--_1._-
Meet Paper presented at the biannual meeting of the Society for
Research m Child Development, Albuquerque, NM, April, 1999.

Sheeber, L. B. ,& Johnson, J. H. (1992). Child temperament, maternal adjustment,

and changes In family life style WWW
178-1 85.

Shelton, B. A., & John, D. (1993). Does marital status make a difl‘erence?
Housework among married and cohabiting men and women. Male!

Eamilxlssuesfi. 401-420.

Silverstein, L. B. ,& Auerbach, C. F. (1999). Deconstructing the essential father.
MW 397-407

260

Sirignano, S. W., & Lachman, M. E. (1985). Personality change during the
transition to parenthood: The role of perceived infant temperament.

Evelepmegtal Psychelggy, 21, 558-567.

Smith, L. B., & Thelen, E. (1993). From the dynamics of motor skill to the
dynamics of development. In L. B. Smith & E. Thelen (Eds), my;

WWW (PP. 1-1 1). Cambridge, MA:
MIT Press.

Snyder, D. K., Klein, M. A., Gdowski, C. L., F aulstich, C., & LaCombe, J. (1988).
Generalized dysfunction in clinic and nonclinic families: A comparative

analysis. 0 A hil ch lo 1 97-109.

Spanier, G. (1976). Measuring dyadic adjustment: New scales for assessing the

quality of marriage and similar dyads WM
38, 15-32

Steiger, J. H. (1990). Structural model evaluation and modification: An interval
estimation aPProach MW 173- 180

Stevens, G., & F eatherman, D. L. (1981). A revised socioeconomic index of
occupational status. $55111 Scienee Reegch, 10, 37-40.

Stoneman, Z. ,,Brody G. H. ,& Burke, M. (1989). Marital quality, depression, and
inconsistent parenting: Relationship with observed mother-child conflict.

W105- 117

Strazdins, L. M., Galligan, R F., & Scannell, E. D. (1997). Gender and depressive
symptoms: Parents' sharing of instrumental and expressive tasks when their

children are young. F 11 P l 11 222-233.

Swensen, C. H. ,,Eskew R W. ,& Kohlhepp, K. A (1981). Stage of fam1ly life

cycle, ego development, and the marriage relationship. W
W 841 -.853

Tanaka, J. S. ,& Huba, G. J. (1985). A fit index for covariance structure models

under arbitrary GLS estimation. W
W 233-239

Teti, D. M., & Gelfand, D. M. (1991). Behavioral competence among mothers of
infants in the first year: The mediational role of maternal self-efficacy. Child

Moments; 918-929.

261

Thelen, E. (1992). Development as a dynamic system. Peychelogical Seiehg, 1,
189-193.

Thelen, E. (1995). Motor development: A new synthesis. Amerim Paehelegiet,
5_Q, 79-95.

Thompson, L., & Walker, A. J. (1989). Gender in families: Women and men in

marriage, work, and parenthood. J l f ' e d F i
230-239.

Tomlinson, P. S. (1987). Spousal differences in marital satisfaction during
transition to parenthood. W 23 9-243.

Von Klitzing, K., Simoni, H., & Buergin, D. (1999a). Child development and early
triadic relationships. W80, 71-89.

Von Klitzing, K., Simoni, H., & Buergin, D. (1999b). MW
-14-.4 ..Ir‘ 11.41. -1 1 1 1 ‘-1-_ 1‘ 11'1111'1-1-11- 1 111 - - -- 1; L111”-

Paper presented at the biannual meeting of the Society for Research in
Child Development, Albuquerque, NM, April, 1999.

Vuchinich, S., Emery, R. E., & Cassidy, J. (1988). Family members as third parties
in dyadic family conflict: Strategies, alliances, and outcomes. Child
W 1293-1302.

Waldron, H. ,& Routh, D. K. (1981). The efl‘ect of the first child on the marital
relationship Will. 785-788

Wallace, P. M. ,& Gotlib, I. H. (1990). Marital adjustment during the transition to
parenthood. Stability and predictors of change. WW
Earn—HLL. 21-29

Weissman, S., & Cohen, R S. (1985). The parenting alliance and adolescence.
WM 24-45.

Westenburg, P. M. ,& Block, J. (1993). Ego development and individual
differences in personality. 1 P 1
792-800.

White, K. M., Houlihan, J., Costos, D., & Speisman, J. C. (1990)- Adult Q
development in individuals and relationships. mm
M 371-386.

262

Wilkie, C. F. ,& Ames, E. W. (1986). The relationship of infant crying to parental

stress in the transition to parenthood. MW
48 545-550.

Willett, J. B. ,& Sayer, A G. (1994). Using covariance structure analysis to detect
correlates and predictors of individual change over time. PsyehMeal

Bulletih, 116, 363-381.

Woollett, A., & Parr, M. (1997). Psychological tasks for women and men in the
post-parturn. r uctiv d P c l 1 159-183.

Wright, P. J ., Henggeler, S. W., & Craig, L. (1986). Problems in paradise? A
longitudinal examination of the transition to parenthood. W

v 1 n h l 7 277-291.

263