s If.
llarta'l)
.0... III.

2:

.a

1

U

V
{.51 .

1'

 

I"? ‘5. ill}:

luvi‘ll‘l ’.

.1 l‘i ‘n!
. .hhm.fr:.?

IN.
sci-F
‘lux: «

.19.
.;0t’
Ila alert .1
7 ...\a.;.!.hnr
1...! Al}.- I
(A ntlilvllnn
30‘. ‘7 ﬁll!
. .Ma Iain. 1 0.

It.“ . I‘ll...
'1.

1'1:
‘ n 3.3.
(15%).. ..

'4',

.\|.. r...
7:.IIv _ .-
.Ollovl

. A III

11.....l-I4 .

III. .I u]: 92"!
6‘4 v oil!!-

.3:lu€~
{I}
It}??‘$l|l
it." novf‘ja‘l’
‘FO'ID' 1113" 1
I ll .3 ill). I.
. 31..-!)- ll}
'17 n, «1":I’
.l -I.

Lil: It

:1:

r 5’ '9’. 0 II

It'll... -

g

‘ I .16"
5‘

. . It'll-1....
1.7.1.... {it

D» .111! 1.
r O

 

rbnll.

Finn!!!
. . o
litillr 0g): I
. } l... A- Hv.3}... .41. .01. Li .i 2. w.
“Ni. .1:nhwu{i .....:..Ili£l’. .lll. Ix. t .

FHMﬁ‘ii-S

MICHIGAN STATE U IVERSITY

Will u I! All ﬁll/Ilium!

3 1293;00914 Mil/[7”!

 

 

 

 

 

ill/l Ill/ll

This is to certify that the

thesis entitled
THE RELATIONSHIP OF THE PARENTING ALLIANCE
To PARENTS' PROBLEM SOLVING BEHAVIORS

DURING A DISCUSSION TASK
presented by

Carol Cracchiolo Laub

has been accepted towards fulﬁllment
of the requirements for

M.A. Psychology

degree in

 

 

Majoﬁofessor

Date May 10, 1990

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

.ﬂ __ ,.__._ .__-___

 

F “-K
LIBRARY

Michigan State
‘ University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- m i ll

MSU Is An Affirmative ActioNEquel Opportunity Institution
chG—ot

 

 

 

 

THE RELATIONSHIP OF THE PARENTING ALLIANCE TO PARENTS’
PROBLEM SOLVING BEHAVIORS DURING A DISCUSSION TASK

By

Carol Cracchiolo Laub

A THESIS

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

MASTER OF ARTS

Department of Psychology

1990

\sﬁbv\ alt-,2. i. n\\\\ If.»

@49-e4e

ABSTRACT

THE RELATIONSHIP OF THE PARENTING ALLIANCE TO PARENTS’
PROBLEM SOLVING BEHAVIORS DURING A DISCUSSION TASK

BY

Carol Cracchiolo Laub

This study examined the relationship between parents’ perceptions of their
parenting alliance, the effectiveness of couple problem solving, and the problem
solving behaviors used by both parents during a discussion task of a discipline
problem. Subjects were 52 predominantly middle class couples, married an average
of 7 years, with at least one child between 3 and 4 years of age. Results revealed
that mothers’ problem solving behaviors were more predictive than fathers’ problem
solving behaviors of problem solving outcomes and of both parents’ perceptions of the
parenting alliance. The majority of the couples studied (63 96) adopted an egalitarian
power structure, which was associated with the most effective problem solving. The
remaining couples were categorized as either father- or mother- dominant systems.
Whereas father dominance was associated with the most positive perceptions of the
parenting alliance and fewer feelings of denigration, it was also associated with the
poorest problem solving outcomes. Mother dominance was associated. with the most
negative perceptions of the parenting alliance and greater feelings of denigration, but
the quality of problem solving outcomes among the mother dominant couples did not

differ from those of egalitarian couples.

This work is dedicated to my parents, who did not have the opportunity to pursue
higher education, yet impressed upon their children the importance of expanding
one’s intellectual horizons. It is through the inspiration of their life examples
and the emotional and instrumental support they have selﬂessly

given me, that I have been able to achieve what I have.

iii

ACKNOWLEDGMENTS

I would like to extend my appreciation to my thesis committee, Dr. Susan
Frank, Dr. Frank Floyd, and Dr. Kristine Freeark, for their willing assistance in
guiding this work to its completion. I would especially like to thank Dr. Susan
Frank, the chairperson of this committee, for her dedication, her high standards for
achievement, and for facilitating the cultivation of my research and clinical skills.

I would also like to thank Lisa Pirsch, for her tireless hours assisting in coding this
data, and for her unwavering optimism.

Above all, I would like to express my deepest gratitude to my best friend and
husband, Rick, for his unfailing support and dedication, his patience and
encouragement, and for his computer wizardry. Without his signiﬁcant personal

contributions, this work might never have come to fruition.

iv

TABLE OF CONTENTS

Page
LIST OF TABLES .................................... vii
LIST OF FIGURES .................................... vi
INTRODUCTION ..................................... 1
Literature Review ..................................... 4
Introduction .................................... 4
The Parenting Alliance .............................. 5
Operationalizing the Parenting Alliance .................... 9
Penman’s (1980) Manifest and Latent Coding Schemes .......... 10
Conﬂict Resolution in Marital and Family Systems ............ 19
The Role of Power in Conﬂict Resolution .................. 27
Conﬂict Resolution and the Parenting Alliance ............... 30
Hypotheses ......................................... 33
Method ........................................... 35
Subjects ....................................... 35
Procedures ..................................... 36
The Parenting Alliance .............................. 36
The Discipline Problems Task (DPT) ..................... 38

TABLE OF CONTENTS CONT ’D.

Page
Preparation and Coding of the DPT ...................... 39
The Quality of Parental Problem Solving .................. 41
Interrater Reliability ............................... 42
Results ............................................ 44
Discussion ......................................... 61
Appendices ................... ° ...................... 73
Appendix A .................................... 73
Appendix B .................................... 81
Appendix C .................................... 83
LIST OF REFERENCES ................................ 88

vi

LIST OF TABLES

Bag:
Fathers’ and Mothers’ Mean Proportion Scores for the
Interaction Variables ............................... 46
Univariate Correlations for Behavior Categories .............. 47
Univariate Correlations for Behaviors and Outcome Variables ...... 49
Partial Correlations for Behaivors and Problem Solving Variables . . . 50
Hierarchical Multiple Regression Analyses Comparing Fathers’
and Mothers’ Contributions to the Outcome Variables ........... 53
One-way Analyses of Variance (ANOVA) for Power Differences
and Outcome Variables ............................. 57

One-way Analyses of Variance (AN OVA) for Power Differences

and Parenting Alliance Variables ....................... 5 8

vii

LIST OF FIGURES

Page
Summary of Penman’s Manifest Level Scheme ............... 14
Summary of Penman’s Latent Level Sheme ................. 15
Integrated Manifest and Latent Level Schemes ............... 17

A Typology of Problem Solving Responses in Close Relationships . . . 21

Problem Solving Scores ............................. 42

viii

INTRODUCTION

Children who grow up in an environment in which there is parental discord
and conﬂict have been found to be at increased risk for the development of behavior
problems and other social, emotional, and psychological difﬁculties. Furthermore,
children from discordant families are more likely to experience dysfunctional parent-
child relationships. Negative parent-child interactions experienced as a child further
places these individuals, when adults, at greater risk for perpetuating negative parent-
child relationships with their own children (Belsky & Isabella, 1985; Engfer, 1988).
However, a growing body of literature suggests that marital discord of itself is not a
necessary nor sufﬁcient precursor of social and emotional difﬁculties in childhood.

A number of studies indicate that in non-clinic samples the relationship between
marital discord and child problems is much weaker than in clinic samples, suggesting
that some other ”third variable” may be inﬂating the association between marital and
child problems in clinic samples (Emery, 1982; Emery & O’Leary, 1984). Thus,
there is evidence that there are other factors which may mediate the association
between marital and child difﬁculties.

The parenting alliance is gaining attention in the literature as one potentially

mediating factor which is separate from, yet integrally related to the spousal

2
relationship. Cohen and Weissman (1984) and others (Benedek, 1959; Frank et a1. ,

1988) propose that the principal inﬂuence of the spousal system may be in its
restrictive or facilitative inﬂuence on the parenting partnership, especially for fathers
(Belsky, Youngblade, Rovine, & Volling, 1990). However, most research
investigating the effects of marital discord on parent-child outcomes has focused
solely on the marital relationship, often neglecting to consider the individual and
unique contributions of the parental subsystem.

This study examined the relationship between observations of parents’
interactions during a discipline conﬂict resolution task and parents’ self-reports of the
quality of their parenting partnership. The subjects in this study were drawn from a
sample of intact families recruited from the patient rosters of a pediatric clinic and
have at least one child between the ages of three and four. This sample allowed for
an examination of patterns of interparental behaviors in "normal” families at an early
developmental stage.

For many of these families the transition to parenthood was fairly recent and
many aspects of parenting attitudes and behaviors were still being negotiated between
spouses. In addition, the parenting alliance in this life cycle stage is less integrated
with and hence more distinct from the marital relationship than in later stages
(Belsky, 1979; Frank, Hole, Jacobson, Justkowski, & Huyck, 1986; Frank, Jacobson,
& Hole, 1988). For example, Frank, Jacobson, & Hole (1988) found that the
parenting experiences of parents with preschoolers were primarily attributable to the

parenting alliance alone, rather than to marital satisfaction; whereas in later family

3

life cycle stages marital satisfaction and the parenting alliance accounted for more
shared variance in parents’ experiences as parents.

This study sought to determine the extent to which parents’ behaviors as
parents are related to their perceptions of their parenting alliance and their
effectiveness in problem solving. Particular attention was given to the way in which
parents asserted power, were affectively involved, and were able to work effectively
on a task during a discussion of a parenting issue, all of which are integrally related

to and should be facilitative of a positive parenting alliance.

LITERATURE REVIEW
Wu

The marital relationship has been consistently implicated as having a direct
effect on parent-child relationships, and consequently on children’s psychosocial
development. The expansive body of literature linldng marital discord to child
behavior problems and less optimal adjustment has most clearly demonstrated the link
between the spousal relationship and child outcomes, and it is this literature which is
most often cited as supportive of the interdependencies of the subsystems within the
family. For example, marital discord has been more strongly associated with child
behavior problems (most prominently in boys) than other factors such as parental
psychopathology, parental cognitive factors, and parental stress (Block, Block, &
jSrde, 1986; Block, Block, & Morrison, 1981; Christensen, Phillips, Glasgow, &
Johnson, 1983; Emery & O’Leary, 1982; O’Leary & Emery, 1984).

However, there is increasing evidence which suggests that it is the parental
relationship, rather than the marital relationship per se, which has the greatest
inﬂuence on parent-child relationships and child outcomes. In addition, numerous
studies which do not directly investigate aspects of the parenting partnership or
alliance lend support to its importance as a mediating factor between the marital and
parental subsystems, especially for families in the early life cycle stages. For
example, Goldberg and Easterbrooks (1984) found stronger links between measures of
marital quality and adaptive parenting than between measures of marital quality and

child adjustment in families with toddlers. In addition, in Easterbrooks and Emde’s

5

(1988) extension of this work, parents’ abilities to resolve conﬂict and their agreement
about child discipline were strongly associated with positive parent-child interactions,
whereas marital adjustment was not signiﬁcantly related to observational measures of
parent-child interaction. Goldberg and Easterbrooks (1984) suggest that some "third
variable” not directly examined in their study may contribute simultaneously to both
marital and parent—child interactions. More recent investigations have demonstrated
the importance of the parenting alliance as this "third variable" which makes separate
contributions to marital and parent—child relationships (e.g., Frank et al., 1988).
Willem

Since its introduction in 1984 by Cohen and Weissman, the concept of the
”parenting alliance” is proving increasingly to be one of the most important aspects of
the spousal system, especially for parents of preschool children. According to these
authors who work within the psychoanalytic tradition, the parenting alliance is that
component of the marital relationship which is distinct from the spouses’ sexual and
libidinal needs. It evolves as parents engage in child-rearing and helps to facilitate
and afﬁrm the performance of parenting skills. Cohen and Weissman (1984) propose
that even in the absence of marital intimacy, parents who have the capacity to
”acknowledge, respect, and value the parenting roles and tasks of the partner" (p. 35)
can sustain the psychological and behavioral stresses of parenting. These authors’
clinical observations and several ﬁndings in the literature indicate that marital
intimacy is neither necessary nor sufﬁcient for a good parenting partnership; and that

even after the dissolution of a marriage, some parents are able to remain connected

6
through the shared process of empathic parenting (Durst, Wedenmeyer, & Zurcher,

1985; Emery, 1982; Hess & Camara, 1979). The parenting alliance thus provides a
stable basis for the psychosocial development of the parents, with signiﬁcant
implications for the parent—child relationship. Weissman and Cohen (1985) hold that
as motherhood and fatherhood develop simultaneously, the parenting alliance provides
the psychological underpinnings for parent-child interactions.

Frank, Hole, Jacobson, Justkowski, & Huyck’s (1986) study of the parenting
alliance supported the notion of the spousal system as limiting or facilitating adaptive
parenting via its effects on the parenting partnership (Benedek, 1970; Cohen &
Weissman, 1984; 1985). The same study also demonstrated that the constructs of
marital satisfaction and a positive parenting alliance are not interchangeable. In Frank
et al.’s (1986) interview study, parents coded as having a strong and positive alliance
described a "mutually supportive relationship characterized by shared decision making
and respect for each other’s parenting abilities“ (Frank et al., 1988, p. 6). In
contrast, parents classiﬁed as having a weak or negative parenting alliance felt
deprecated, indicated a lack of support, described an unwillingness or inability to
agree on important parenting decisions, or were themselves critical of their spouse’s
parenting abilities (Frank et al., 1986). This operational deﬁnition of the parenting
alliance was signiﬁcantly, but not strongly, related to a measure of marital intimacy.
The authors suggest that the parenting alliance and marital intimacy may make
different contributions to more adaptive parenting.

In a continuation of this work, Frank et a1. (1988) devised a questionnaire (the

7

Family Experiences Questionnaire-FEQ) based on their interview data to assess the
parenting alliance. Results from this study indicate that the unique contribution of the
parenting alliance to parents’ reports of overinvolvement (overcontrol and narcissistic
investments in the child) and adaptive parenting experiences (conﬁdence, empathic
investments in the child, and low reports of undercontrol), was much greater than that
of marital satisfaction. In addition, the relationship of marital satisfaction to the
parents’ overinvolvement and adaptive parenting experiences was to a large extent due
to its relationship to the parenting alliance. Furthermore, the parenting experiences of
parents with preschoolers were mostly attributable to the parenting alliance alone,
while parents’ experiences in later developmental stages were predicted by m the
parenting alliance alone and by its combined effect with marital satisfaction.

In sum, Frank et al.’s (1988) ﬁndings suggest that the parenting alliance
mediates the relationship between marital satisfaction and parents’ feelings and
attitudes about parenting. In addition, their data suggest that marital and parental
roles may become more closely integrated as children grow older. These results
underscore the importance of the parenting alliance, particularly to parents of
preschool children, as it is less integrated with the marital relationship at this stage
and has the greatest inﬂuence on more adaptive parenting experiences.

A number of related studies are consistent with Cohen & Weissman’s (1984).
observations and Frank et al.’s (1988) ﬁndings in that they suggest that the parenting
alliance makes independent contributions toward adaptive parenting experiences and

positive child and family outcomes, and that marital intimacy and/or satisfaction is not

8
necessary for a positive parenting alliance. For example, in a study examining family
relationships in families with a mentally retarded child, Floyd (1988) investigated the
role of the parenting alliance and its relationship to the marital and parental
subsystems. Subjects in this study were families with a mentally retarded child
between the ages of 6 and 18 years old who was living in the home and attending
special classes for trainable or edueable mentally impaired children. Results indicate
that marital quality and the parenting alliance contributed independently to family
functioning. In addition, the parenting alliance had both direct and indirect links to
family functioning by buffering stress and enhancing marital quality.

Studies of families with divorced parents underscore the importance of
distinguishing between the marital and parental systems. Durst, Wedenmeyer, and
Zurcher (1985) and Hess & Camara (1979) found that a cooperative coparent system
is the most adaptive family conﬁguration after divorce, with more favorable outcomes
including: reintegration of the family, better parent-child relationships, greater
consensus on parenting roles, less conﬂict and more support between parents, and
clear but ﬂexible boundaries between subsystems. Parents in this cooperative family
type are considered full partners in parenting, making joint decisions, and having
mutual respect for each other as parents. The relationship between spouses becomes
redeﬁned into a cooperative one which concerns only parental functioning, and

supports the needs of the children after divorce.

D . 1.2. l E . 511°

As deﬁned by Cohen and Weissman (1984), the core function of the parenting
alliance is to enable parents to sustain the psychological and behavioral stresses
associated with parenting. Recent studies support the stress-buffering hypothesis of
the parenting alliance, and demonstrate differential effects of the parenting alliance for
fathers and mothers. Floyd’s (1988) study of families with a mentally retarded child
(discussed above) demonstrated that the parenting alliance was directly and indirectly
linked to optimal family functioning by reducing stress and improving marital quality.
Likewise, in a study investigating child illness, the parenting alliance, and parent
stress, Frank, Wagner, Olmsted, Laub, Freeark, Breitzer, & Peters (1989) found that
parents’ perceptions of the parenting alliance were signiﬁeantly correlated with
stresses associated with perceived child characteristics (e. g. , demandingness,
moodiness, distractibility) for both mothers and fathers, and stresses associated with
parent characteristics (e. g. , depression, social isolation, parenting conﬁdence) for
fathers.

While these studies conﬁrm the stress-reducing function of the parenting
alliance, little is known about the actual processes by which the parenting alliance is
able to reduce parental stress. Furthermore, the parenting alliance and parental stress
are both subjective experiences and hence are appropriately assessed by self-report
measures. Thus, these conﬁrmatory studies have demonstrated a strong relation
between two subjective experiences: perceptions of the parenting alliance and

perceptions of parental stress. What remains to be determined is to what degree

10

parents’ observable interactions as parents are related to their perceptions of their
parenting alliance. Speciﬁcally, the question to be answered in this study was to what
extent parents’ abilities to resolve conﬂict about parenting issues and to effectively
deal with parenting difﬁculties as a dyad are related to their perceptions of the
parenting alliance. In particular, this study examined the relationship between
parents’ assertions of interindividual power, affective involvement, and problem
solving abilities in their relationship as. parents.

’ if

In this study it was hypothesized that parents who are mutually supportive and
actively involved in their interactions as parents would report a strong parenting
alliance. Moreover, parental interactions which involve shared positive affect and an
equitable distribution of power should foster a positive parenting alliance and promote
more effective problem solving. On the other hand, it is believed that parents who
are less actively involved in decision-making and more emotionally removed from
each other and the communication process, should report a negative alliance which is
characterized by a lack of respect for each other’s parenting abilities and feelings of
denigration. In addition, parental interactions which include more hostile and
aggressive responses in their attempts to resolve parenting issues are believed to be
particularly destructive to the parenting alliance and effective problem solving.
Robyn Penman’s (1980) dimensions of affective involvement and power offer a means
of testing these hypotheses. Penman devised a message classiﬁcation coding scheme

for use in marital dyads which conceptually relates the dimensions of power and

1 1
affective involvement. This scheme also has been used successfully with parent-child

dyads to identify a number of dyadic interaction styles (Hakim-Larson & Hobart,
1987).

The conceptual foundations of Penman’s theory of interpersonal
communication are rooted in systems theory, communication dreary, and rules theory.
Of these three bases, Penman considers systems theory to be the most far-reaching,
encompassing the latter two. The systems approach to the study of communication
processes is based on three fundamental tenets: emphasis on studying whole systems
rather than integral components in isolation, concern with the complex set of relations
within and between systems, and emphasis on the dynamic and self-directed activity
of living systems. According to systems theory, all open systems are characterized by
a dynamic homeostasis, moving toward increasing differentiation and organization. In
human groups, the exchange of information maintains the interrelations between
members and establishes structure. Furthermore, Penman (1980) holds that because
people are often unaware of why they behave as they do and inaccurate in reporting
their reasons for their behavior, an outside observers’ perspective is necessary to
understand the relationship and communieations between system members.

Based on this perspective of interpersonal communication theory, Penman
(1980) improved upon previous message category systems by developing one which is
of general relevance, has an extensive set of categories, and is designed speciﬁcally
for real groups (e.g. families, marital dyads, etc.). In addition, his scheme allows for

the simultaneous classiﬁcation at two different levels of inference, the manifest and

12

the latent. The inclusion of two levels makes it possible to identify incongruencies in
messages and therefore reveal some of the more enigmatic processes underlying
communication. The categories within each of these levels are based on ”mutually
exclusive and exhaustive criteria" and are intended for the classiﬁcation of all
communication behavior, with an emphasis on the voeal mode (Penman, 1980, p. 59).

Penman’s message classiﬁcation scheme conceptually interrelates the
dimensions of power and emotional involvement. The Power dimension is described
as “a speciﬁc instance of the differentiation tendency” , while the Involvement
dimension is described as "a speciﬁc instance of the integrative tendency” (Penman,
1980, p. 61). The Power dimension is deﬁned as interindividual inﬂuence. More
speciﬁcally, power will be deﬁned in this study as one’s attempts to inﬂuence the
other through power assertion. The Involvement dimension, which has historically
received relatively less attention in the literature than power, has been deﬁned in
terms of behaviors indicative of ”solidarity and afﬁliation” , and "accepting and
loving” attitudes; all of which reﬂect some degree of personal involvement with the
other (Penman, 1980, pp. 61-62). Involvement is a bidirectional dimension in which
the involvement may be either towards the other or away from the other, indicating
both type and degree of involvement.

Penman’s two coding schemes, manifest and latent, differ in the level of
inference necessary to classify each statement. The Manifest Level Scheme consists
of nine manifest message categories which vary from high to low power assertion,

and from negative to positive emotional involvement. At the manifest level, one is

13

primarily concerned with the explicit information expressed by the sender in terms of
its usual meaning, while ignoring extra-linguistic information as much as possible.
(See Figure 1).

In comparison, the Latent Level Scheme is more complex and the frame of
reference is more inferential and comprehensive. Both linguistic and extra-linguistic
information is to be considered, as well as the temporal context of the speech unit.
The Latent Level places the information from the manifest level into a relational
context. In this level, four positions are assigned to each dimension of Power and

Involvement, comprising a total of 16 Latent categories. (See Figure 2.).

14
Figure 1. Summary of Penman’s Manifest Level Classiﬁcation Scheme

AGGRESS ADVISE SUPPORT

 

 

 

 

' Asserts self Gives solutions Shows understanding i
E Shows aggression Gives guidance Reassures ;
P ; Justiﬁes behavior Gives explanation Shows Trust .1
1 .
o ; DISAGREE EXCHANGE AGREE
Q Differs Gives information Conﬁrms
I Corrects Gives suggestions Reconciles
W , Criticizes Asks for information Conciliates
i Contradicts Asks for suggestions
E I AVOID REQUEST CONCEDE
' Hesitates Asks for decision Passively accepts
I Withdraws Asks for approval Passively supports
R i Noncommittal Asks for evaluations Complies
g Shows uncertainty Asks for directions Acquiesces

INVOLVEMENT

From: Robyn Penman (1980). Wham Academic
Press, New York.

15

Figure 2. Summary of Penman’s latent Level Classiﬁcation Scheme

 

Discredits other
Denigrates

     
     

Takes over,
directs

i Stands for self
i while inviting
1 other

Joins forces

Confronts

Afﬁrms self &
other

 

Deﬁes, refuses
Defends self
Stands for self at
expense of other

EVADE
Vague, wordy
Does not respond
directly
Manoeuvers out
of situation

 

    

RESIST
Counteracts
Is cynical,
skeptical
Sets up obstacles

  
      
     
      

. OFFER
Tentatively
suggests

Informs other
‘ Is task oriented

 

CELL 4: OBLIGE l

   

COLLABORATE
Reciprocates
Cooperates
Expands on other

«4

OBLIGE
Accepts
Concurs w/other
Endorses other

 

Refuses to
participate

Ignores other

Dissociates self

 

Concedes defeat

Backs away

Abandons
previous
position

 

Defers to other

Gives
responsibility to
other

Takes path of
least
resistance

 

 

CIJNG
Seeks control by
other
Accepts any
directives
Mutually colludes

INVOLVEMENT

From: Robyn Penman (1980). Wattles Academic

Press, New York.

Note: Cell designations and labels were added by this author.

16

Superimposing the Manifest Scheme on the Latent Scheme, the domain
covered by any one manifest category corresponds with the domain of four latent
categories. (See Figure 3). Because the latent Level encompasses the Manifest
Level and conveys more information about the system relationships, it was the
classiﬁcation scheme used to analyze the observational data in this study.

The data base for the development of Penman’s (1980) classiﬁcation system
consisted of the tape recordings of 18 couples discussing 3 different items (i.e. ,
problem solving task, modiﬁed consensus inkblot technique, and discussing
characteristics of a happy couple) over a 1 hour period. Couples had been living
together for at least one year and had plans to continue to do so. The range of
relationship duration was from 1 to 33 years, with about two-thirds being together for
5 years or under. Participants’ ages ranged from 18-61 years, with approximately

two-thirds being 30 years or under.

17

Figure 3. Integrated Manifest and latent Level Schemes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

reject control initiate share
AGGRESS ADVISE SUPPORT
P
counter resist offer collaborate
O
DISl lGREE Ex" | GB AGRI :E
W
evade abstain seek oblige
E
AVOID REQUEST CONCEDE
R
remove relinquish submit cling

 

 

 

 

 

I N V O L V E M E N T

 

From: Robyn Penman (1980), WWW Academic
Press, New York.

 

18

Percentage frequencies of all manifest and latent categories were calculated for
each dyad. At the manifest level, Exchange was found to be the most frequently used
category for all dyads. Couples were similar at the manifest level in that Exchange,
Advise, and Concede were the most common types of message (accounting for at least
60% of all message units), and acts of giving support (Support) and seeking support
(Request) were the least common. In addition, at the latent level couples had similar
responses with Offer, Oblige, and Resist (accounting for more than 50% of all
message units) being the most common messages and Clinging, Sharing, Removing,
Relinquishing, and Rejecting being among the least common. Whereas Penman
(1980) does not directly address differences between men and women in their
communication styles, he does relate that there are few differences between members
of each dyad, and between couples as a group. In general, at both levels of
communication the central and most neutral category (Exchange and Offer,
respectively) is the most common communicative act in all couples analyzed. Penman
(1980) therefore views this type of act as the central point around which other acts
revolve and to which the communication process continually returns.

Penman’s (1980) classiﬁcation scheme can be reduced to its four major
quadrants to facilitate comparisons with other coding systems and to generate some
general hypotheses about the parenting alliance: Cell 1 (Aggress): high power
assertion-negative affective involvement (reject, control, counter, resist), Cell 2
(Initiate): high power assertion-positive affective involvement (initiate, share, offer,

collaborate), Cell 3 (Avoid): low power assertion-negative affective involvement

19

(evade, abstain, remove, relinquish), and Cell 4 (Oblige): low power assertion-
positive affective involvement (seek, oblige, submit, cling) (See Figure 2.). The
clearest hypothesis related to Penman’s cells is that the interactions falling within the
Aggress cell (1) would be most deleterious to the feelings of mutual respect and
positive regard for each other’s parenting abilities which characterize a positive
parenting alliance. Thus, behaviors falling within Cell 1 would be expected from
parents reporting negative alliances. In addition, as argued below, these interactions
should be related to poorer problem solving outcomes. On the other hand,
interactions which fall in the Initiate cell (2) would involve more active and
supportive involvement and more positive affect from both parents, thus fostering a
positive parenting alliance and more effective problem solving.
C ER 1' 'll'l 1E '15

While Penman (1980) has not related his coding system to measures of parental
harmony or marital adjustment, predictions as to more speciﬁc relationships between
the parenting alliance and his four behavioral quadrants can be formulated on the
basis of ﬁndings reported in the conﬂict resolution literature. Many studies
investigating conﬂict resolution styles include coding systems which are closely
related to the dimensions of power and affective involvement and can be
superimposed to varying degrees upon Penman’s classiﬁcation scheme (Billings, 1979;
Fineberg & Lowman, 1975; Markrnan & Notarius, 1987; Rusbult & Zembrolt, 1983).

The research on conﬂict resolution is of particular relevance to young families,

as parents in this life cycle stage are still negotiating family roles and functions; and it

20

is through the constructive resolution of inevitable conﬂict that the parental system
becomes deﬁned and integrated (Deutsh, 1969; Sprey, 1969). Productive conﬂict
resolution has been compared to the process of creative thinln'ng and has been found
to progress through phases from recognition of the problem, to efforts to solve the
problem, and ﬁnally to testing tentative solutions (Deutsch, 1969). One of the few
studies investigating the components of effective family problem solving revealed that
families who performed well across a variety of problem solving tasks were more
cooperatively and actively involved in the problem solving task, i.e. they: a)
attempted more problem solving strategies, b) deliberated longer over solutions, and
0) came to more satisfactory agreements (Blechman & McEnroe, 1985). On the other
hand, Deutsch (1969) reported that participants who resorted to strategies of power
assertion and relied on coercion and threat (interactions classiﬁed by Penman in the
high power assertion-negative affective involvement cell (1: aggress)), typically found
themselves caught up in a tycle of destructive conﬂict which led to less effective
resolutions to the problem. Destructive conﬂict presumably is caused by and causes
ineffective communication and misperceptions.

One way of conceptualizing the conﬂict resolution process that dovetails with
Penman’s (1980) system has been proposed by Rusbult & Zembrolt (1983). Their
multidimensional scales have been used in a number of studies examining differences
in the conﬂict resolution styles of distressed and non-distressed dating undergraduate
couples (Rusbult, Johnson, & Morrow, 1986). These authors identify four categories

of behavior which are based on a model of response to dissatisfaction. These

21

categories include: a.) "exit” ending the relationship or discussion (e.g. , breaking up,
physically abusing one’s partner), b.) "voice"-actively and constructively expressing
dissatisfaction (e.g. , suggesting solutions, discussing problems), c.) "loyalty”-
remaining passively loyal to the relationship and waiting for conditions to improve
(e.g., hoping things will improve, praying for improvement), and d.) ”neglect"-
passively allowing the relationship to atrophy (e.g. , ignoring one’s partner, refusing
to discuss a problem) (Rusbult & Zembrolt, 1983, p. 275—276). These four categories
differ along the two dimensions of constructiveness/destructiveness and
activity/passivity. Constructive categories include voice and loyalty, while more
destructive categories include exit and neglect. Exit and voice represent active

behaviors, whereas loyalty and neglect are more passive behaviors. (See Figure 4.).

"-r-' ‘ _ru9|' 0 .190 ‘u 0,i,' iuiuti.y '1 0 ; 1‘.e§01 {-

 

Destructive Constructive

NBCLBC'I' LOYALTY

 

Passive

 

From: Rusbult et a1. (1986). Wales): 5.0. 744-53.

This typology of problem solving can be compared to Penman’s Latent Level
classiﬁcation scheme, where the active/passive dimension is similar to Penman’s

power assertion dimension and the constructive/destructive dimension is similar to the

22

dimension of affective involvement. Behaviors in Penman’s high power assertion-
negative involvement cell (1: aggress) coincide with the Active/Destructive quadrant,
whereas behaviors in Penman’s low power assertion-negative affective involvement
cell (3: avoid) overlaps with the Passive/Destructive quadrant. Finally, Penman’s low
power assertion-positive involvement cell (4: oblige) corresponds with the

Passive! Constructive quadrant, and the high power assertion-positive involvement cell
(2: initiate) resembles the Active! Constructive quadrant.

Rusbult et a1. (1986) found that destructive problem solving behaviors (Neglect
and Exit) and partners’ perceptions of each other’s destructive problem solving
behaviors were strongly related to couple distress. In addition, destructive problem
solving behaviors (Neglect and Exit) were more predictive of couple functioning than
constructive problem solving behaviors, with Exit being the strongest predictor.
Whereas the tendency to respond with Voice was somewhat predictive of couple
functioning, Loyalty responses failed to predict couple distress/nondistress. In
addition, couple distress was associated with the tendency to reciprocate destructive
problem solving behaviors, and with the inability to respond constructively to
destructive responses from one’s partner. Thus, the tendency to react to relationship
difﬁculties with destructive behaviors has a signiﬁcant negative effect on the couple’s
adaptive functioning.

Gender differences were found in problem solving responses, with females
engaging in signiﬁcantly more Voice and Loyalty, and somewhat less Neglect

behaviors than men. There were no signiﬁeant gender differences in the tendency to

23
Exit. However, females’ Exit and Neglect responses were found to be more

deleterious to the relationship than their partners’ similar responses. Rusbult et a1.
(1986) conclude that in comparison with women, men are less likely to solve
problems through Loyalty or Voice, and are somewhat more likely to resort to
Neglect. Thus, when the female partner uses the destructive behaviors of Exit or
Neglect, the male is less likely than his partner in similar circumstances to ameliorate
the situation through Voice or Loyalty responses and avoidance of neglect. Women’s
destructive responses are therefore more salient and harmful to the relationship
because they are not part of her regular repertoire, and because they are not
compensated for by adaptive responses from her partner. The authors speculate that
since women generally utilize constructive responses, when they do issue destructive
behaviors, it is more indicative of serious difﬁculties within the relationship.

Rusbult et al.’s (1986) ﬁndings emphasize that women’s destructive responses
(especially Exit) are particularly damaging to the relationship. These authors imply
that it is the negative affective quality in women’s responses which is most predictive
of couple difﬁculties, since Exit and Neglect vary on the active/passive (or power
assertion) dimension. Yet, because these authors did not analyze within couple
differences, it remains unclear whether it is the negative affective quality, the
assertion of power, an interaction between the two and/or within couple differences in
these behaviors which are most predictive of couple dysfunction.

Additional conﬂict resolution studies which focus on the role of affect in the

marital subsystem coincide with Rusbult et al.’s (1986) results, pointing to the

24

destructive impact of negative affect and its reciprocity. These studies compare
distressed and non-distressed couples across a number of different interactional
paradigms and reveal three consistent patterns (Billings, 1979; Birchler, Weiss, &
Vincent, 1975; Gottman & Levenson, 1986; Margolin & Wampold, 1981; Vincent,
Weiss, & Birchler, 1975). First, distressed couples exhibit more negative affect than
non-distressed couples. Second, distressed couples engage in greater reciprocity of
negative affect. And third, distressed couples have more rigid, highly structured
interactions in which one spouse’s behavior is highly predictive of the other’s
subsequent behavior.

Gottman and Levenson (1986) have expanded the work on negative affective
reciprocity to include an escape conditioning model which accounts for gender
differences in conﬂict resolution styles and explains the process by which satisﬁed
couples (and perhaps those with positive parenting alliances) avoid negative and rigid
behavior patterns. This model is generated by physiological studies which indicate
that males show quicker, larger, and more prolonged autonomic nervous system
(ANS) responses to stress than females. These sex differences in ANS responses are
purportedly related to sex differences in the problem solving behavior of men and
women (Levenson & Gottman, 1983; 1985). Because chronic ANS activity is
considered unpleasant and harmful, these authors posit that men would tend to avoid
situations associated with high levels of ANS activation; Whereas negative affect has
been shown to produce higher levels of ANS activation in men (as measured by heart

rate and galvanic skin response), they may try to moderate the amount of negative

25
affect to which they are exposed by creating a rational rather than emotional climate

in relationships. These tendencies may take the form of more conciliatory, less
conﬂict-engaging, and more withdrawing behaviors in men.

The literature on sex differences in response to conﬂict conﬁrms these conﬂict-
avoidant behaviors in men. In addition, women have been shown to be less
conciliatory, more conﬂict-engaging, and less likely to withdraw from negative affect
(Kelley, Cunningham, Grisham, chebvre, Sink, & Yablon, 1978; Raush, Barry,
Hertel & Swain, 1974). Raush, Barry, Hertel, & Swain (1974) found similar
responses to conﬂict in marital couples problem solving in improvisational scenes.
Whereas men and women tended to behave in similar manners overall, there were
signiﬁcant differences between the two. Men made more attempts to resolve the
conﬂict and to reconcile their partners, whereas women responded with appeals,
personal attack, and twice as much coercion. These data appear to be somewhat
disparate from Rusbult et al.’s (1986) ﬁndings.

According to Gottman & Levenson’s (1986) escape conditioning model, these
sex differences in responses to conﬂict are expressed to varying degrees in satisﬁed
and dissatisﬁed couples. As found in this study and the research cited above,
dissatisﬁed couples respond with greater rigidity, stereotypy and predictability of
behavior. The escape conditioning model may explain both how marriages become
more and less satisfying. It links a reduction of physiological arousal to
reinforcement of the behavioral sequence. In explanation, when both spouses move

from a state of high ANS arousal (upset) to low levels of ANS arousal (calm), the

26
behavior patterns associated with the escape moments will become reinforced. In
sum:

“it seems likely that in unhappy marriages, there are many instances of

upset over unresolved issues, and thus there will be many conditioning

trials over which to strengthen the association that links a speciﬁc kind

of upset with a speciﬁc behavior that serves to reduce that upset. In

satisﬁed marriages, there will be fewer moments of upset, and thus

fewer conditioning trials to establish rigid response patterns” (Gottman

& Levenson, 1986, p. 46).

Therefore, in more satisﬁed couples, when supportive and empathic statements are
used to restore calm, this response is more likely to be integrated in the couple’s
repertoire in response to upset. It is expected that similar processes are at work in
the conﬂict resolution of parents with strong alliances. As these couples are likely to
respond to conﬂict with more positive affect, it is these same supportive responses
that should be reinforced as they help to reduce the stresses associated with resolving
parenting issues.

In sum, there is considerable evidence in the conﬂict resolution literature
which suggests that largely through their ability to avoid and to respond constructively
to negative affect, parents with a strong parenting alliance would be more effective in
resolving conﬂicts about parenting issues. Throughout the literature reviewed above,
negative affect has been more strongly associated than positive affect with measures
of marital satisfaction, communication satisfaction, and problem solving abilities
(Coombs, 1966; Haynes, Chavez, & Samuel, 1984; Wills, Weiss, & Patterson,
1974). Thus, the ability to respond more constructively to negative affect and the

avoidance of its reciprocation, should encourage more shared decision-making and

more mutual positive regard between parents, both essential characteristics of a

27

positive alliance. In sum, parents who are better able to avoid the reciprocation of
negative affect also avoid dissatisfaction in the communication process, exacerbation
of conﬂict, and less effective problem solving; thus fostering a more positive
parenting alliance (Menaghan, 1982; Birchler, Weiss, & Vincent, 1975; Gottman &
Levenson, 1986). In addition, parents who feel respected and valued by their partners
are more likely to have more positive communications and more effective problem
solving. This prediction is supported by Coomb’s (1966) study of dating couples
which found that individuals’ perceptions that their partners valued them were
signiﬁcantly related to ease in communication and satisfaction with the interaction.

II B l [E . [2 H' B l .

The above discussion of the conﬂict resolution literature has largely focused on
the importance of the quality of affective expression in the marital and parental dyad.
The role of power, while often alluded to and subsumed in behavioral classiﬁcation
schemes, has not been adequately deﬁned nor properly addressed in this research.
One of the clearest shortcomings of these studies is their failure to consider the
independent and shared contributions of power and affect to the conﬂict resolution
process. The power dimension must be more carefully addressed as it has important
implications for parental problem solving. In particular, there is some evidence that
independent of affect, an unequal distribution of power which favors the mother is
less adaptive for families than egalitarianism or father-dominance.

Research indicates that egalitarianism and husband-dominance are associated

with high marital satisfaction, while child and wife-dominant marriages are lowest in

28
marital satisfaction (Kolb & Straus, 1974; MacDonald, 1980). Low husband

dominance has also been associated with poor problem-solving ability and family
dysfunction (Jacob, Ritchey, Cvitkovic, & Blane, 1981; Kolb & Straus, 1974; Moos
& Moos, 1984). In his study of power distribution in ”normal” families, Jacob
(1974) found that in triadic interactions of parents and their adolescent sons, most
middle class and lower class families were father-dominant or egalitarian, as
measured by both process and outcome measures. However, previous studies
investigating power distribution and problem solving have suffered from diverse
methodological problems ranging from an over-reliance on outcome measures (e.g. ,
"win" scores in decision making), relative neglect of fathers’ reports, and
inconsistency between deﬁnitions and measures of power. In addition, previous
studies investigating power distribution and problem solving in couples have primarily
addressed conﬂicts and decisions which pertain to the family system, the marital
system, or to some other contrived scenarios not directly related to the individual
couple or family. Hence, the distribution of power in the parental system has not
been directly investigated during interactions involving conﬂict over pertinent
parenting issues.

There are some special considerations that need to be taken into account in
generalizing from the literature on power and the marital system to hypotheses about
power assertion and distribution in the parenting alliance. As childcare has often of
necessity been primarily the mother’s role, she traditionally has had the most

immediate inﬂuence in determining how children would be raised. For example, in

29
his 1977 study of young suburban families, Cohen found that due to the frequent

absence of fathers, mothers were principally responsible for making decisions relating
to their children. More important to this study are the ﬁndings that mothers typically
made the decisions related to child punishment, and that the only area in which
fathers took an active role in decision making was related to the formal education of
their child. Thus, it would seem that women may have more inﬂuential power in
making decisions pertaining to the discipline of their child. However, this data must
be scrutinized as to its applicability to the more contemporary couples in this study
sample and to parents with a strong alliance. Cohen (1977) indicated that the mothers
in his sample were quite dissatisﬁed with their husbands’ lack of involvement in
parenting, and they developed a supportive network in which they formed alliances
with other mothers to give them emotional and instrumental support in their roles as
parents. Whether this balance of power in favor of the mother is potentially
disruptive to effective problem solving by parents as a couple is as yet unclear.
Clearly, the relation of power assertion to the parenting alliance and effective
problem solving is complex. Whereas shared decision making and respect for each
other’s parenting abilities are associated with a positive parenting alliance, it follows
that parents who share more equally in making decisions about child-rearing, relying
on each other for both instrumental and emotional support, would have a positive as
opposed to a negative alliance. In addition, greater father involvement in parenting is
presumably associated with more positive parenting alliances rather than negative

alliances. Furthermore, numerous studies have demonstrated the importance of the

30

mother’s role in involving the father in active parenting by delegating power and
responsibility to the paternal role (Beitel, 1989; Belsky, 1979). However, it is
important to consider the interaction between power assertion and affect. Maternal
power assertion may not operate independently of affect, or negative affect may
increase the negative implications of maternal dominance. It is more likely that when
mothers assert an inordinate amount of power which is paired with less affective
involvement (i.e. , more aggressive, controlling behaviors), rather than eliciting the
father’s active involvement in parenting, this should cause fathers to feel denigrated,
to withdraw from parenting interactions, and to abdicate or defer child-rearing
responsibilities to their wives. This sequence obviously undermines the parenting
alliance. Alternatively, parents who demonstrate a more egalitarian or father-led
power distribution with more positive or neutral affect in problem solving interactions
should be able to maintain and facilitate a strong parenting alliance. Furthermore,
more equitable distributions of power assertion and positive affective involvement in
the parental dyad, should be linked to more effective problem solving.

CD'BI' ”12.5".

Th\er:search on conﬂict resolution has demonstrated that constructive and
effective wndrfretibhition involves a process in which members of a system (e.g. ,
the parental dyad in this study) work cooperatively to identify the problem, generate a
number of possible solutions to the problem, choose and agree upon a solution,
implement the solution, and then evaluate the solution (Bales & Strodtbeck, 1951;

Blechman & McEnroe, 1985 ; Deutsch, 1969). More effective problem solving is

31
expected to result in solutions which are likely to resolve the problem and are

developmentally appropriate for the age of the child.

As with the parenting alliance, Penman’s (1980) classiﬁcation scheme can be
used not only to assess this general hypothesis, but also to examine whether effective
parental problem solving is the outcome of somewhat different behaviors for fathers
and mothers. Clearly the least effective problem solving and the most negative
alliances are expected for both fathers and mothers whose problem solving styles
include high power assertion and negative affective involvement (Penman’s cell 1). A
similar yet somewhat weaker association with poor problem solving and a weak
alliance was expected for parents exhibiting low power assertion and negative
affective involvement (Penman’s cell 3). On the other hand, a position of high power
assertion and neutral or positive affect (Penman’s cell 2) should be strongly associated
with a positive alliance and effective problem solving, though (extrapolating from
Rusbult et al. ’s (1980) ﬁndings) this association should not be as strong as the above
two positions. Finally, interactions which indicate low power assertion and positive
affective involvement (Penman’s cell 4: obliging) are expected to have little (and if
anything a somewhat positive) association with the alliance and problem solving.
These behaviors can be described as system maintaining. In addition, when mothers
respond with more power assertion than fathers, combined with negative affective
involvement, it is expected to have more detrimental effects on the parenting alliance
and problem solving. In sum, an egalitarian distribution of power assertion or father

dominance, coupled with positive affective involvement of both parents should be

32

linked to the most positive alliances and most effective problem solving, whereas high
power assertion combined with negative affective involvement of both parents should

be associated with the most negative alliances and poorest problem solving.

HYPOTHESES
1) Individual behaviors characteristic of Penman’s cell 1 (aggress) of high power
assertion-negative affective involvement will be strongly and negatively associated
with the parenting alliance and quality of couple problem solving, and this will be
especially true for mothers.
2) Individual behaviors characteristic of Penman’s cell 3 (avoid) of low power
assertion-negative affective involvement will be strongly and negatively associated
with the parenting alliance and couple problem solving, and this will be especially
true for mothers.
3) Individual behaviors characteristic of Penman’s cell 2 (initiate) of high power
assertion-positive affective involvement will be moderately and positively associated
with the parenting alliance and couple problem solving. (Individual behaviors
characteristic of Penman’s cell 4 (oblige) of low power assertion-positive affective
involvement are not expected to be related to the parenting alliance nor couple
problem solving.)
4) The strongest predictor of the parenting alliance and problem solving will be the
behaviors in cell 1, the second best predictor will be the behaviors in cell 3, and the
third best predictor will be the behaviors in cell 2.
5) Mother dominance (as opposed to father dominance or egalitarianism) by itself will
not be negatively associated with the parenting alliance and couple problem solving.
However, when mother dominance is accompanied by mothers’ negative affective

involvement, this will be strongly and negatively related to the outcome variables.

33

34
8) The interaction of parents’ destructive (cell 1) problem solving behaviors will be

strongly and negatively associated with the outcome variables and will contribute
signiﬁcantly to the prediction of the outcome variables, above and beyond the
contributions of either mothers’ or fathers’ destructive (cell 1) behaviors.

9) The tendency to respond destructively (behaviors falling within cells 1 and 3)
rather than constructively (behaviors falling within cells 2 and 4) to partner’s negative
behaviors (cells 1 and 3) will be strongly and negatively associated with the parenting
alliance and couple problem solving, especially for mothers.

10) Fathers’ submission (cell 3 & 4 behaviors) to mothers’ aggressing behaviors (cell

1) will be associated with more negative outcomes.

METHOD

Snbimts

Subjects were 52 families participating in the Family Factors in Children’s
Health (Otitis Media) Study at Michigan State University (Frank et al., 1986), a
longitudinal study examining factors that may contribute to children’s health and
illnesses. Subjects were recruited from the Primary Care Clinic at Michigan State
University run by the Department of Pediatrics in the College of Osteopathic
Medicine. All families 1.) which had a child between the ages of 3 and 4 years old,
2.) which were intact at the time of contact, 3.) which had no more than 4 children,
4.) in which both parents had the equivalent of a high school education, 5.) in which
at least one parent was presently employed, 6.) in which the target child did not have
a serious chronic illness (i.e. , asthma, cystic ﬁbrosis, diabetes, cancer), were potential
candidates for this study. Potential candidates were ﬁrst identiﬁed by computer
printouts which list all patient contacts and diagnoses for children who were currently
between 3 and 4 years of age. Further screening and recruitment took place in a
phone interview. If both parents agreed to participate in the study they were then
visited in their home by project staff, who explained the study in greater detail,
recruited them into the project, conducted a brief interview, distributed
questionnaires, scheduled further contacts, and administered the Discipline Problems
Task (DPT). The couples in this sample were married an average of 7 years with a
range of 3 to 13 years. All families received $75 as monetary compensation for their

participation.

35

36
W

Each family participating in the study completed a number of questionnaires,
behavioral observation sessions, and interviews (Frank, Brietzer, & Freeark, 1986).
The data were collected throughout the course of an approximately 3 week contact
schedule. The data were collected both in the family’s home and at the Michigan
State University Psychological Clinic, where videotaping of structured interactional
tasks took place. Data were collected by a trained team of graduate and
undergraduate students. Data collectors were blind as to the target child’s health
status.

The instruments that are of particular relevance to this study examine parents’
views of their parenting behaviors and their parenting partnership with their spouse.
In addition, this study examined the processes of problem solving and conﬂict
resolution between parents during a behavioral interaction (the Discipline Problems
Task).

W. The parenting alliance was assessed with the Family
Experiences Questionnaire (FEQ) (Appendix A), a relatively new measure devised by
Frank, Jacobson, & Hole (1988). Frank et al. (1988) developed and tested the FEQ
on a sample of 760 parents with at least one child between the ages of 1 and 19 years
old living in their home. A pool of 106 items describing the parenting alliance was
generated from the interview protocols from 39 parents in the Frank et al. (1986)
study. Parents used a four point scale (1 =strongly disagree, 2= disagree, 3=agree,

4 =strongly agree) to indicate how much each of the items described their parenting

37
relationship with their spouse. Frequently omitted items and items with little

variability were then eliminated and the remaining 80 items were subjected to a
principal components analysis using oblique rotation. A scree test indicated that four
factors were useful to interpret. Items that had large loadings on one of the four
factors and smaller loadings on the other three factors were used.

The parenting alliance was evaluated with two scales from the EEO which
assess Cohen & Weissman’s (1984) description of the parenting alliance as
characterized by parents’ feelings of mutual respect and positive regard for each
others’ parenting abilities. Mothers’ and fathers’ scores will be examined
independently, as well as jointly in an average score for each couple.

The W (POSPA) scale includes 13 positively
and 18 negatively keyed items. A; high scores on this scale indicates that the
respondent 1) respects their spouse’s parenting abilities, 2) feels supported as a parent
by their spouse, 3) believes that their spouse does his or her share of the parenting,
and 4) feels that parenting has increased the extent of cohesiveness in the spousal
subsystem. Frank et al. ( 1988) report that this scale is internally consistent
(Chronbach’s alpha > .90); and test-retest reliability computed over a three week
interval for a sample of 30 mothers and fathers was .86 (Frank et al., 1988). Alphas
for the sample reported in Frank et al. (manuscript in preparation) (.96 for mothers
and .94 for fathers) are comparable to those reported by Frank et al. (1988).

The Denjgratedjmuse (DENIG) scale includes 6 positively and 4 negatively

keyed items (e.g. "My spouse makes me look like the ‘bad person’ in the eyes of our

38
children,” "My spouse makes me feel that I am the best possible parent for my

children”). Frank et al. (1988) report that Chronbach’s alpha for this scale was in the
range of .83 to .91.

Two recent studies have demonstrated the construct validity of the general
positive parenting alliance scale through its mediating effects on parent stress. In an
investigation of child illness and parent stress, Frank et al. (1989b) examined the
relationship between parents’ perceptions of the parenting alliance, parent related
stresses (e.g. , feeling trapped by parenting, depression, low parenting conﬁdence),
child-related stresses (e. g. , demandingness, moodiness, distractibility), and child
illness. Parents’ perceptions of the parenting alliance correlated signiﬁcantly with
parent—related stresses ([= -.65; p<.001) and child-related stresses (r= -.44; p,.011)
for fathers, but only with child-related stress ([= -.27; p < .05) for mothers. In
addition, Floyd’s (1988) study of families with mentally handicapped children linked
the parenting alliance directly and indirectly to optimal family functioning, via its
buffering of stress and enhancement of marital quality.

W. In a visit to the subject families’
homes, parents individually completed a brief questionnaire (Appendix B) to respond
to the question ”How difﬁcult is it for you and your spouse to agree on how and
when to discipline (child’s name)?” in part (A) of the DPT Questionnaire. Responses
range from "rarely a problem" to "almost always a problem" (rarely a problem =1,
sometimes a problem =2, often a problem =3, always a problem =4). Parents then

independently listed in order of difﬁculty (most to least difﬁcult) three examples

39
within the last few weeks when they had the most difﬁculty agreeing on how to

discipline their (target) child in part (B) of the DPT.

After parents completed the questionnaire, the interviewer selected a topic from
the discipline problems listed for them to discuss. The discipline problem selected
was that which was rated most difﬁcult by both parents (i.e. , that which had the
highest average severity rating). If there were no common discipline problems listed
by both the mother and father, then the discipline problem rated as most difﬁcult by
the mother was chosen. The parents were then told the following:

"Now I’d like the two of you to talk about how you might handle the following
better as a couple: (restatement of problem). You have ten minutes to come to
an W on how you will handle that problem in the future. I am going
to record your discussion but I will leave the room so that you can talk more
comfortably. "
The interviewer then turned on the tape recorder and left the room for 10 minutes
while the parents discussed the problem. In their study comparing direct and audio
recorded observations of parent-child interactions, Hansen, Tisdelle, & O’Dell (1985)
found little variance due to differences between the two methods of observation,
except for behavior categories which were primarily nonverbal. Thus, audio recorded
observation proved cost-effective, reliable, and comparable to direct observation.
WW (Appendix C). Each of the DPT
discussions were transcribed from the audiotapes. Transcriptions were then

segmented according to Penman’s (1980) deﬁnition of coding units as follows: ”there

40

may be more than one unit per utterance, a change in unit being signiﬁed by a change
of (a) information or meaning, (b) intonation, and/or (c) syntax" (p. 136).

Once the speech units were divided on the transcripts, one of two raters
assigned each speech unit to a latent category. A few adaptations were made to
Penman’s system in accordance with the suggestions made by Hakim-larson and
Hobart (1987), and in order to better suit our purposes in this study. First, a default
category was used (utterance-UT) for those segmented utterances which were not
meaningfully scorable according to Penman’s categories (e.g., 'Hmmmm...well").
This also includes all utterances which coders were unable to transcribe with
conﬁdence, due to mumbling, background noise, interruptions, etc. Second, a
"laugh” (LA) category was created to include all utterances which consist of laughing
without any other codable utterance. This included nonreciprocated laughter and
nervous laughter. This category excludes any instances of shared laughter which was
coded as "collaborate" as speciﬁed by Penman’s (1980) coding manual. Both default
categories were excluded from all but the sequential data analyses. Finally, to
alleviate confusion in category assignment when dual classiﬁcation seemed possible, a
prioritized ordering of the categories was used so that the negative categories received
the highest priority followed by the positive, and then the more neutral categories
(Hakim-Iarson & Hobart, 1987). Because coding from both the tape and the
transcript has been found to be a more reliable procedure (Hakim-[arson & Hobart,
1987; Penman, 1980), the raters independently coded the segmented transcripts that

they read while listening to the tapes.

41
Proportions based on frequencies of a particular category of response divided

by total number of utterances were calculated for each latent category for each
individual. In addition, proportion frequencies were calculated for each of Penman’s
four cells. Each category or type of response was also weighted for power assertion
and affective involvement, based on its relative position along the dimension axes.
For example, "rej ect" received a power assertion score of 4 and an involvement score
of 1, whereas "share” received a power assertion score of 4 and an involvement score
of 4. All weighted categories were then summed and divided by the individual’s total
number of utterances to get a total weighted power assertion score and a total
weighted affective involvement score. These separate scores allowed for the analysis
of the unique and shared contributions of power assertion and affective involvement to
the parenting alliance and problem solving.

WWW. Two problem solving scores were
then assigned to the protocols after reading them through thoroughly. These scores
included: a) type of outcome and b) quality of the solution. Criteria for each rating
are shown in Figure 5. The type of outcome score is based on the distinctions made
by Vuchinich, Emery, & Cassidy (in press). The quality of solution score was
developed for this study, based in part on clinical observation and on the problem
solving literature. This score required that a the rater make a judgement of how
likely the solution was to solve the problem, and of how developmentally appropriate

the solution was, given the age and psychosocial maturity of the child.

i Pr 1 lvin r

 

a) Type of Outcome
3 = Compromise: Each parent gives in a little to accept a position that falls
between the extremes in the conﬂict.
2= Submission: One parent gives in to the other.
1= Standoff: There is no resolution to the conﬂict.
l= Withdrawal: One parent refuses to continue to talk or withdraws from the
discussion (physically or emotionally).

b) Quality of the Solution
4 = Solution likely to resolve the problem and developmentally appropriate.
3 = Solution somewhat likely to resolve or alleviate the problem and mostly
developmentally appropriate.
2 = Solution somewhat unlikely to resolve problem, e. g. , a solution which

proposes to ”let child grow out of it” , and/or not very appropriate.
1= No solution proposed.

 

Finally, a problem solving total score was assigned by adding the two scores above
(i.e., total= outcome + quality).

Inter-rater Reliability

Segmenting: The investigator divided all 52 transcribed protocols into spwch
units based on Penman’s (1980) criteria. An undergraduate psychology major divided
a randomly selected subset of 15 % of the protocols into speech units to check the
reliability of the primary segmenter. Segmenting reliability was computed using the
following formula suggested by Hakim-Larson and Hobart (1987): number of perfect
agreements/one-half the sum of Rater A’s total segments plus Rater B’s total
segments. Inter-rater agreement on this subset of 8 protocols was 85 %.

Mg: The investigator and a second psychology graduate student used

42

Penman’s (1980) latent level coding system to code all speech units. A randomly
selected subset of protocols was used for training purposes until reliability reached
adequate levels (Cohen’s, 1960, Kappa > .70), as suggested by Dorsey, Nelson, and
Hayes (1986) and Landis & Koch (1977). The two raters then independently coded
another randomly selected subset of 6 protocols. Inter-rater agreement on this subset
was 75 96 (Kappa= .71). After establishing adequate reliability, the investigator
coded all of the remaining protocols, while the second graduate student coded every
ﬁfth protocol (10% of all protocols) to check the reliability of the ﬁrst coder.
Periodic meetings were held after coding each set of 5 reliability protocols to prevent
observer drift. Inter-rater agreement on this subset of 10 protocols was 80%
(Kappa= -77)-

Wm: The investigator also rated all subjects
on a scale of 1 to 3 for the type of outcome and on a scale of 1 to 4 for quality of the
solution arrived at by each couple at the end of the discussion period. In addition, the
second rater above scored a randomly selected subset of 18 protocols on these
dimensions to check the reliability of the ﬁrst coder. Correlations indicative of inter-
rater agreement were .74 (p< .001) for problem solving outcome and .73 (p< .001)

for quality of the solution.

43

Results
ﬁencraLﬁndings

As in Frank et a1. ’s (unpublished manuscript) prior sample, paired t-tests
revealed signiﬁcant differences in fathers’ and mothers’ perceptions of the parenting
alliance, with fathers reporting more positive alliances than mothers (1(104) =5. 18;

p < .001). Fathers and mothers did not differ in their feelings of denigration, nor did
they differ signiﬁcantly in their use of each category of problem solving behavior. In
addition, there were no signiﬁcant differences between fathers and mothers in their
overall power assertion or affective involvement (see Table l).

The problem solving quality and outcome scores were moderately correlated
(r=.52, p< .001). In addition, the problem solving outcome score was correlated
with mothers’ perceptions of the parenting alliance (r=.30, p<.05). No other
signiﬁcant relationships were found between the parenting alliance and the problem
solving outcome variables. Pearson correlations also demonstrated several signiﬁcant
associations among the behavior categories both within persons and couples (see Table
2). Positive correlations between fathers’ aggressing (cell 1) and mothers’ aggressing
(r:=.54, p< .001), fathers’ aggressing and mothers’ avoiding (cell 3)(r=.49, p< .01),
and fathers’ avoiding and mothers’ aggressing (r=.39, p< .01) lend some support to

the reciprocity of negative affect hypothesis. Fathets’ avoiding and mothers’ avoiding

is.

behaviors were not signiﬁcantly correlated however. An additional pattern was that

.w -""
vC- "'
”w

when one partner responded negatively, the other was also less positive. In

44

45
particular, mothers’ aggressing behaviors were negatively correlated with fathers’

obliging (cell 4) behaviors (r=-.35, p< .05). In addition, fathers’ aggressing and
avoiding behaviors were both negatively correlated with mothers’ initiating (cell 2)
behaviors (r=-.54, p< .05 and r=-.33, p< .05, respectively). However, mothers’
negative behaviors were not signiﬁcantly correlated with any of fathers’ initiating
behaviors. Finally, mothers’ and fathers’ overall affective involvement was
signiﬁcantly correlated (r(52)= .31, p < .05), whereas parents’ overall power

assertion was not related.

46

 

 

Table 1.
’ r’M Pr in rfrhnrainV'l
Interaction Variables Fathers Mothers
(N =52) (N =52)

Aggress (Cell 1)

M .08 09

SD .14 .13
Initiate (Cell 2)

M .56 56

SD .19 .19
Avoid (Cell 3)

M .03 01

SD .07 .03
Oblige (Cell 4)

M .21 21

SD .14 .13
Total Involvement

M 2.80 2.71

SD .50 .42
Total Power

M 2.71 2.67

SD .71 .44

47

 

 

Table 2. Univariate Correlations for Behavior Categories
Fathers (N=52) Mothers (N =52)

 

Aggress Initiate Avoid Oblige Aggress Initiate Avoid Oblige

Fathers

Aggress -.68*** .38" -.36** .54*** -.34* .49" -.04
Initiate -.39** -.l8 -.19 .14 -.24 -.09
Avoid -.24 .39" -.33* .12 .10
Oblige -.35* .23 -.10 .21
Mather:

Aggress -.54*** .24 -.23
Initiate -.29* -.48***
Avoid .04
Oblige

 

Note: Aggress= Cell 1 behaviors, Initiate= Cell 2 behaviors, Avoid= Cell 3
behaviors, Oblige= Cell 4 behaviors.

*p<.05; “p<.01; ***p<.001, two-tailed tests.

 

Correlations showing the relationships between fathers’ and mothers’ problem
solving behaviors on the one hand, and the problem solving and parenting alliance
outcome variables on the other hand, are shown in Table 3. For fathers, aggressing,
avoiding, and obliging behaviors were signiﬁcantly correlated with the problem
solving sum and type of problem solving outcome variables. In addition, fathers’
avoiding and obliging behaviors were related to the quality of the solution. For
mothers, aggressing and initiating were signiﬁcantly correlated with each of the
problem solving outcome variables.

Fathers’ initiating behaviors correlated signiﬁcantly with only one of the
parenting alliance variables (a negative correlation with overall denigration), and
mothers’ aggressing behaviors were signiﬁcantly and negatively correlated with the
combined general parenting alliance score. In addition, both mothers’ aggressing and
initiating behaviors correlated signiﬁcantly with mothers’ feelings about the alliance.

Overall, signiﬁcant correlations between the various types of problem solving
behaviors and the outcome variables were very close in magnitude. Hence, the
ordering hypothesis was not supported. For this reason the hierarchical analysis
which was planned to assess the contribution of weaker relationships after controlling
for stronger ones was deemed inappropriate. Rather, partial correlations, performed
separately for mothers and fathers, assessed whether each of the signiﬁcant correlates
independently related to the outcome variables after controlling for all other

signiﬁcant correlates (see Table 4).

49

Table 3. [Jory mat: Corrolations for Bohaviors mo Qotoomo Vao'ablos
’ h vi r Mothers’ Behaviors

 

Aggress Initiate Avoid Oblige Aggress Initiate Avoid Oblige
W
Total -.42*"‘ .21 -.46*** .43“ -.57*** .58*** -.14 .07
Outcome -.47*** .26 -.43*** .41** -.58*** .60*** -.14 .00

Quality -.25 .11 -.36** .34* -.4l** .42** -.11 .12

Parenting Alliaoco Sooros

Gen PA -.22 .24 -.05 -.06 -.39** .23 -. 12 .04
Denig .09 -.27* .05 .24 .14 -.03 .15 .05
F Gen PA -.13 .18 -.05 -.04 -.21 .03 .09 .08
M Gen PA -.22 .21 -.03 -.05 -.41** .33* -.26 -.01
F Denig .12 -.18 .10 .17 .17 .01 .04 .00
M Denig .00 -.24 -.05 .19 .02 —.05 .19 .08

 

Note: Problem Solving Total= Mean (Outcome + Quality of Solution)
Gen PA= Mean General Parenting Alliance Score
Denig = Mean Denigration Score
F Gen PA= Father’s General Parenting Alliance Score
M Gen PA= Mother’s General Parenting Alliance Score
F Denig= Father’s Denigration Score
M Denig= Mother’s Denigration Score

N=52 for all pairwise correlations.
*p< .05; **p< .01; ***p< .001, two-tailed tests.

 

Table 4.

__Eattrr.ts’_Behaxinrs_

Aggress Avoid Oblige Aggress Initiate
Wm:
Tom] '.21. -'34*b .31*c -'37**d .40**c
Outcome -.30*, -.30*., .27c -.38**,. .42“,
Quality -.05, -.28., .25c '34.: .26c

 

Note: , = controlling for Avoid & Oblige
b= controlling for Aggress & Oblige
,= controlling for Aggress & Avoid
a = controlling for Initiate
,= controlling for Aggress

N= 52 for all pairwise correlations

*p<.os; **p<.01; **“p<.001,two-tailed tests.

51

For fathers, avoiding behaviors were negatively and obliging behaviors were
positively and independently correlated with the problem solving total score. In
addition, fathers’ avoiding and aggressing behaviors were independently and
negatively associated with the problem solving outcome. However, neither avoiding
nor obliging behaviors correlated independently with the quality of the solution after
partialling out the effects due to each other.

For mothers, the partial correlations indicated that aggressing and initiating
behaviors both accounted for signiﬁcant variance in the problem solving scores,
whereas only mothers’ aggressing behaviors accounted for signiﬁcant variance in
mothers’ perceptions of the parenting alliance. Both aggressing and initiating
behaviors proved to independently correlate with both the problem solving total and
type of outcome. Whereas aggressing behaviors had more negative problem solving
implications, initiating behaviors had more positive problem solving implications.
Neither aggressing nor initiating behaviors continued to make independent
contributions to the quality of the solution after controlling for the effects due to each
other. Finally, after partialling out the effects of initiating behaviors, aggressing
behaviors were still independently associated with less positive reports of mothers’
parenting alliance. However, initiating behaviors no longer signiﬁcantly correlated
with mothers’ parenting alliance after controlling for aggressing behaviors.

In order to determine the relative contributions that fathers’ versus mothers’
behaviors made to the outcome measures, two sets of hierarchical multiple regression

analyses were performed. In the ﬁrst regression analysis, all of the fathers’ behaviors

52

were entered simultaneously in the ﬁrst step. Then, all of the mothers’ behaviors
were entered simultaneously in the second step of the analysis. In the second
regression analysis, all of the mothers’ behaviors were entered in the ﬁrst step,
followed by all of the fathers’ behaviors in the second step. As can be seen in Table
5 , after controlling for the variance due to fathers’ behaviors, mothers’ behaviors still
predicted a signiﬁcant amount of variance in all of the problem solving scores and in
mothers’ perceptions of the parenting alliance. However, after removing the variance
due to mothers’ behaviors, fathers’ behaviors failed to signiﬁcantly account for any
additional variance in the problem solving variables. Fathers’ behaviors however,
were associated with mothers’ feelings of denigration (which mothers’ behaviors did

not predict).

53
Table 5.

 

Outcomes (R2/ch)

 

5.191121951191915 mmwmm

 

 

1. Fathers’ behaviors 35*" 35*" .20* .07 .12
2. Mothers’ behaviors .25*""" 23*" .17* .24 *.13
1. Mothers’ behaviors 51*“ .49*** .31" .21* .04
2. Fathers’ behaviors .09 .09 .06 .10 .21*
Note:

PS TOT= Problem Solving Total Score (Outcome + Quality of Solution)

PS OUT= Type of outcome

PS QUAL = Quality of the solution

M Gen PA= Mothers’ general parenting alliance score

M Denig = Mothers’ denigration score
Note: Neither fathers’ nor mothers’ behaviors signiﬁcantly predicted the variance in
any of the other parenting alliance variables (general parenting alliance, general

denigration, fathers’ parenting alliance, fathers’ denigration).
*p<.05; **p<.01; ***p<.001

54
ff f m t mal minanc

Multiple regression analyses were also used to test the hypothesis that mother
dominance would be particularly detrimental when combined with negative affective
involvement. Differences in overall power assertion within couples were calculated
from mothers’ and fathers’ weighted power scores (i.e., mother’s minus father’s
weighted power score). Multiple regressions were then performed in which the
power difference scores, mothers’ weighted overall involvement score, and the
interaction of the two were regressed onto the problem solving and parenting alliance
scores. The hypothesis was not consistent with the data in that none of the interaction
effects were statistically signiﬁcant. There were some notable main effects however.
Firstly, greater overall affective involvement on the part of the mother was associated
with better problem solving outcomes (betas for problem solving sum, type of
outcome, and quality of the solution were .72, .64, and .60, respectively; p’s <
001.). Secondly, mother dominance tended to be associated with parents’ reports of
being more denigrated by each other (B= .30, p=.058).

Because the relationships between mother-father power differences and the
outcome variables could be curvilinear in nature, the distribution of power difference
scores was divided into three groups. A power difference score of zero was
considered to be indicative of an egalitarian power structure (the mean power
difference score for this sample was -.04 and the standard deviation was .57). The
distribution of power difference scores was thus divided into the following three

groups : father dominant (difference scores < -.50), egalitarian (-.50 < difference

55

scores < +.50), and mother dominant (difference scores > +.50). Approximately
63% (N=33) of all subjects fell into the egalitarian group, while the rest of the
subjects were about equally distributed in the upper and lower quartiles (father
dominant; N=9 and mother dominant; N=10). A series of one-way analyses of
variance were then executed to test for differences between the three groups in the
outcome measures. Although most of the F-values were not signiﬁcant at
conventional levels, a number were signiﬁcant at at least the p< .100 level. Because
between-group differences were expected from a priori hypotheses, it was deemed
reasonable to interpret these between-group differences when signiﬁcant at the p< .05
level, even though the overall F-test might not be signiﬁcant.

According to these criteria, signiﬁcant between-group differences were found
for each of the problem solving variables and for the denigration variable (individual
scores for mothers’ and fathers’ perceptions of denigration were in the same direction)
(see Tables 6 & 7). The results are partly consistent with the hypotheses: egalitarian
power distributions were associated with the most positive problem solving outcomes
and father dominant distributions were associated with the poorest problem solving
outcomes. Outcomes associated with mother dominant patterns were in between and
did not differ from the other two. In addition, the greater A mothers’ power relative to
fathers’, the more denigrated bot_h parents reported feeling in the parenting alliance,
whereas the greater fathers’ relative power, the less denigrated both parents reported
feeling. Feelings of denigration in egalitarian couples were in between those of

mother- and father—dominant couples and were not signiﬁcantly different from the

56
two. This pattern is similar for both parents’ reports of denigration. Taken together,
these results suggest a competing process in which assertive behaviors on the part of
the father are associated with less effective couple problem solving (especially in
comparison to an egalitarian pattern), but also with lesser feelings of denigration for
both parents (especially in comparison to a pattern of mother dominance).
Ir' ".2!" a. . moi ’ -...r' .41 .w-r. 1_' rt. '

A series of regression analyses was performed in which the effects of the
interaction (cross products) between mothers’ and fathers’ aggressing behaviors on the
problem solving outcomes and parenting alliance variables were tested after
controlling for the main effects. These analyses resulted in non-signiﬁcant interaction
terms. Hence, the hypothesis that the cumulative effect of parents’ aggressing

behaviors would be especially detrimental to the outcome variables (beyond the

individual effects of either parent’s aggressing behaviors) was not supported.

57

 

 

 

Table 6.
Ono-way Analysos of Vgo'anoo (ANOVA) for Power Difforoncos
ano Problem Solving Variables
Problem Solving Variables
Total. Outcome Quality
Eowor Djfforonoo N M SD M SD M SQ
Father > Mother 9 -1.21" 1.40 2.11" .78 2.33" .87
Father = Mother 33 .34‘ 1.58 2.64' .65 3.09‘ .84
Father < Mother 10 -.02"" 2.17 2.60"" .70 2.80"" 1.2
F 2.99‘ 2.13 2.43‘

 

Note: Column means with different supersCripts (a,b,c) are signiﬁcantly different

from one another at the .05 level.

t=trend at p<.05.

*p< .05

58

 

 

 

Table 7.
Ono-way Analysos of Variance (ANOVA) for Powor Difforonoos
and Paronting Allianco Variables
Parenting Alliance Variables
£32m lleoig
Eowo; Difforonco N M SD M SD
Father > Mother 9 3.21 .25 1.58" .24
Father = Mother 33 3.20 .27 1.72"" .27
Father < Mother 10 3.03 .24 1.90‘ .20
F 1.67 3.95“

 

Note: Column means with different superscripts (a,b) are signiﬁcantly different from

one another at the .05 level.

t=trend at p<.05.

*p< .05

Gen PA= Mean general parenting alliance score.

Denig= Mean denigration score.

 

Reciprocation of a partner’s negative behaviors was expected to be strongly
and negatively correlated with positive problem solving outcomes and positive feelings
about the parenting alliance. It was additionally expected that when mothers
reciprocate fathers’ negative behaviors, this would have more detrimental effects on
the outcome variables than when fathers reciprocate mothers’ negative behaviors. To
test this hypothesis, all mothers’ and fathers’ behaviors were collapsed into two
categories: positive and negative. Positive behaviors consisted of initiating and
obliging (cell 2 and 4) behaviors, while negative behaviors included aggressing and
avoiding (cell 1 and 3) behaviors. Pairs of behavior codes (mother negative-father
negative; father negative-mother negative) were used to summarize the patterned
sequences obtained from the problem solving task. Sequential analyses produced
measures of lagged dependency (z scores) for each couple at the ﬁrst lag, which were
then correlated (using Pearson correlations) with the outcome variables. The 2
statistics indicate the inﬂuence of the ﬁrst behavior on the second, or stated
otherwise, ”the degree to which prediction of the second behavior is improved by
knowledge of the ﬁrst behavior” (Phelps & Slater, 1985, p. 689). Because of the
nature of the program used to analyze the sequential data (Bakeman, 1985), z-scores
are not assigned when either the antecedent or consequent behavior never occurs.

The data were partially consistent with the hypothesis. A pattern in which
mothers’ negative behaviors followed fathers’ negative behaviors was associated with

a poorer parenting alliance (as reported by both parents combined) ([(21)=-.44,

60
p< .05), and greater feelings of denigration (r(21)=.47, p< .05), (frndings for

fathers’ and mothers’ individual parenting alliance and denigration scores were in the
same direction as the combined scores). In contrast, fathers’ negative behaviors
following mothers’ negative behaviors (N=22) did not signiﬁcantly relate to any of
the parenting alliance outcome variables. Sequences of lagged behaviors were not
associated with any of the problem solving outcomes for either fathers or mothers.
To assess the effects of fathers’ submitting with avoiding or obliging behaviors
to mothers’ aggressing behaviors, fathers’ avoiding and obliging behaviors were
combined into one behavior category. A sequential analytic procedure (as explained
above) was employed to derive a z-score for lag one for the mother aggress-father
submit pair of behaviors for each couple. The 2 scores were then correlated with the
outcome variables using Pearson correlations. The hypothesis that fathers’ submitting
to mothers’ negative power assertions would be associated with poorer outcomes
proved inconsistent with the data. Although this analysis was based on 28 couples,
none of the correlations were signiﬁcant. Thus, fathers’ taking a less powerful role in
response to mothers’ negative power assertions did not prove detrimental nor

beneﬁcial in regard to the parenting alliance scores or problem solving outcomes.

Discussion

 

In general, the results of this study provide some support for the underlying
assumption that parents’ problem solving behaviors are related to the outcomes of
their interactions and to their perceptions of the parenting alliance. However,
univariate correlations indicated that the signiﬁcant correlations between the problem
solving behaviors and outcome variables were relatively similar. Hence, the ordering
hypothesis was not supported by the data. Partial correlations, performed to assess
the independent relationships of each problem solving behavior with the outcome
variables, demonstrated that the predictive power of each behavior category varied
with the sex of the parent and with the individual outcome variable.

For mothers, both initiating and aggressing behaviors were independently
correlated with the problem solving sum and outcome variables. However, only
mothers’ aggressing behaviors independent related to mothers’ perceptions of the
alliance. Both of these predictive behaviors have high power assertion in common.
Thus, it appears that the most important behaviors for mothers are those which
involve a signiﬁcant assertion of power.

For fathers, avoiding and obliging behaviors, and avoiding and aggressing
behaviors were independently correlated with the problem solving sum and outcome
scores, respectively. Fathers’ initiating behaviors however, failed to independently

relate to the outcome variables.

61

62
h r ’ v ’ h vi r

Overall, these ﬁndings support the general hypothesis that mothers’ behaviors
have more salient implications for parental problem solving outcomes and the
parenting alliance than fathers’ behaviors. Although most of the conﬂict resolution
literature stresses the importance of mothers’ negative problem solving behaviors, this
data indicates that in general mothers’ behaviors (both positive and negative) make
more independent contributions to problem solving and parental relationship outcomes
than fathers’ behaviors.

First, aﬁer controlling for all of fathers’ problem solving behaviors, mothers’
behaviors still predicted a signiﬁcant amount of the variance in all of the problem
solving outcome scores and in mothers’ perceptions of the parenting alliance.
However, after controlling for all of mothers’ problem solving behaviors, fathers’
behaviors did not account for any additional variance in the outcome variables, except
for mothers’ feelings of denigration. Second, the parenting alliance was mostly
related to mothers’ behaviors. Overall, mothers’ problem solving behaviors were
more strongly correlated with the parenting alliance than fathers’ behaviors. Third,
sequential analyses demonstrated that mothers’ reciprocation of fathers’ destructive
behaviors was signiﬁcantly related to the parents’ overall perceptions of the parenting
alliance and feelings of denigration. In contrast, fathers’ reciprocation of mothers’
negative problem solving behaviors failed to signiﬁcantly relate to any of the outcome
variables. Fourth, father dominance was associated with the poorest problem solving

outcomes, whereas the best outcomes were associated with egalitarianism.

63

Furthermore, mother dominance was associated with problem solving outcomes which
did not differ from and were more similar to those associated with egalitarianism.
Thus, it appears that mothers’ behaviors in general are more strongly related to
positive outcomes than fathers’ behaviors. The ﬁnding that mothers’ reciprocity of
fathers’ destructive behaviors was related to reports of a poorer alliance and greater
feelings of denigration is consistent with the results of Levenson & Gottman (1983;
1985) and Gottman & levenson (1986). These investigators found a pattern in which
a decline in marital satisfaction was associated with greater reciprocity of the
husband’s negative affect by the wife and less reciprocity of the wife’s negative affect
by the husband. They suggest that these results may be indicative of a vicious cycle
in which dissatisfaction causes husbands and wives to produce those behaviors which
lead to further dissatisfaction «'emotional withdrawal on the part of the husband,
making the wife less satisﬁed; increased affect and negative affect reciprocity on the
part of the wife, making the husband less satisﬁed” (Levenson & Gottman, 1985; p.
91). A similar cycle may be related to parents’ behaviors and their feelings about the
parenting alliance. When mothers reciprocate father’s negative behaviors, fathers are
more likely to withdraw from active involvement in parenting, which in turn is likely
to lead to greater negative reciprocity on the part of the mother, and greater mutual
feelings of denigration and reports of a poor alliance.

Alternatively these data are inconsistent with the hypothesis that negative
behaviors would be more strongly associated with the outcome variables than positive

behaviors, and that destructive behaviors would be especially detrimental when

64
utilized by mothers. This discrepancy may be due to differences in the samples

studied. This sample of parents is a normal (i.e. , non-clinical) one in which
destructive behaviors are relatively infrequent. In contrast, couples studied in the
conﬂict resolution investigations typically reported in the literature usually include
distressed as well as nondistressed couples. Hence, there is a greater range and
frequency of occurrence of negative affect in these samples.
Mcthcndaminancc

Failure to ﬁnd signiﬁcant maternal dominance X maternal affective
involvement interactions was unexpected. Inferences made from previous studies in
the conﬂict resolution literature suggest that it is not mother dominance in itself which
is necessarily related to negative outcomes, but rather mother dominance when paired
with negative affect. Perhaps it is more difﬁcult to demonstrate these effects in a
normal sample where both parents are relatively equal in power and positive affect is
far more prevalent than mother dominance and negative affect. What these data did
reveal however, was that egalitarianism was associated with better problem solving
outcomes, especially in comparison to father dominance. In addition, a pattern
emerged in which the greater fathers’ power relative to mothers’ , the more both
parents reported a positive alliance and less denigration. These results indicate that
for this sample, a tension exists in which a more traditional (husband dominant)
distribution of power was more facilitative of the positive valuing of each partner’s
parenting role and skills, but detrimental to effective problem solving. Hence, while

parents appear to be able to maintain more positive feelings about their alliance in

65

father dominant systems, the cost is less effective parental problem solving.

These data are only partly consistent with results obtained in Kolb & Straus’
(1974) study linking marital power and marital happiness with problem solving
ability. These authors found that husband donrinant families were highest in marital
happiness, with no differences in marital satisfaction between more egalitarian or wife
dominant couples. However, the association between problem solving ability and
mother dominance differs in Kolb & Straus’ (1974) and the current sample. In the
present study, mother dominance was associated with more positive problem solving
outcomes than father dominance, whereas wife dominant couples in the Kolb & Straus
sample had relatively poorer problem solving skills than husband dominant couples.
These differences in problem solving ability and power distribution may be in part
due to the nature of the problem being resolved. Perhaps it is especially important
for mothers to take a more active role in problem solving issues related to parenting,
such as in the discipline problem task utilized in this study. Because mothers still
spend more time with their children than fathers, it is most likely the mothers who
have more information related to and more positive suggestions for resolving
discipline problems. On the other hand, for problems related more to general family
functioning (e. g. , ﬁguring out how to play a game in the Kolb & Straus study), it
may be important for families to assume a more traditional (father dominant) power
structure. Kolb & Straus suggest a number of factors which may explain the greater
marital happiness associated with husband dominance, including the fact that ”despite

the evolving equalitarian family values which they have presumably adopted, most

66

Americans seem to have remained committed to a hierarchical (though democratic)
family system” (p. 765). The present data diverge from these results in that the
majority of couples in this sample were egalitarian rather than father dominant. This
may be due to numerous societal changes which have occurred in the 16 years since
the publication of the Kolb & Straus study (e.g. , more women working outside the
home and the continued struggle for equal rights). The interaction pattern
characteristic of these egalitarian families often took the form of mothers taking an
initiating role while fathers largely facilitated mothers’ discussion through a non-
avoiding and active listening role.

However, those couples that did adopt a father dominant system in this study
did have more positive feelings about the parenting alliance, but were also least
effective in problem solving. Thus, it appears that the price of maintaining a more
traditional (father dominant) power structure is less positive problem solving of
parenting issues. There are two alternative ways of interpreting these data in father
dominant families. First, mothers in a sense may be allowing fathers to take the lead,
as indicated by Beitel’s (1989) ”mother as gate-keeper” hypothesis. Similarly,
Clarke-Stewart (197 8) suggests that mothers may have the tendency to let fathers
"take center stage” in certain aspects of parent-child interactions (especially play). A
second interpretation of these ﬁndings is that those fathers who predominate the
discipline discussion may be fathers who are truly more actively involved in
parenting. This is congruent with the more positive parenting alliances reported by

both parents in father dominant couples.

67
Finally, the hypothesis that fathers’ submitting (obliging or avoiding) to

mothers’ aggressing behaviors would be associated with poorer outcomes was not
supported by the data. Thus, fathers’ submission to mothers’ aggressing proved
neither beneﬁcial nor detrimental to the outcomes. As discussed above, the overall
power difference between mothers and fathers is much more important to the
outcomes than this single response pattern. Altogether, these results indicate that
fathers’ assertive behaviors may serve as a double-edged sword, in that they are
associated with more positive feelings about the parenting alliance and less positive
problem solving. This also suggests that fathers may not be as effective when they do
lead in solving parenting problems, although both parents appear to be comfortable
with this more traditional arrangement. This highlights the importance of mothers’
taking an active and equally powerful role in problem solving, yet in a manner which
does not threaten fathers’ sense of leadership, and perhaps the couples’ sense of
security. It also underscores the importance of enhancing fathers’ problem solving
skills.

5 l 1 Eli . l l . l .

There were no signiﬁcant differences in the behaviors of mothers and fathers
during the problem solving task utilized in this study. In addition, mothers and
fathers did not differ in their overall affective involvement or assertion of power in
the discussion. Both mothers and fathers employed initiating (cell 2) behaviors most
frequently, followed by obliging (cell 4) behaviors. Behaviors which were used less

frequently were aggressing (cell 1) and avoiding (cell 3).

68
The similarity in behaviors exhibited by mothers and fathers in the discussion

task is consistent with Penman’s (1980) data. He found that initiating behaviors (cell
2) were most frequently used in all dyads, and that obliging behaviors (cell 4) also
comprised a large proportion of all spwch units. Penman (1980) also found few
intra- or inter-couple differences in communication styles. This similarity between
mothers’ and fathers’ use of problem solving behaviors is inconsistent with Rusbult et
al. ’s (1986) and Gottman & Levenson’s (1986) data which revealed clear gender
differences in communication styles. In the Rusbult at al. (1986) study, women had a
greater tendency to use more constructive problem solving behaviors than men (cells
2 and 4) and men used more ”neglectful" (cell 3) behaviors. Similarly, according to
Gottman & Levenson’s (1986) escape conditioning model, men avoid situations
associated with high autonomic nervous system arousal by withdrawing.

The discrepancy in gender differences found between the Rusbult et al. (1986)
and Gottman & Levenson (1986) data on the one hand, and Penman’s (1980) and the
present data on the other hand, may be due to differences in the samples studied and
in the modes of measurement utilized. First, the Rusbult et al. ( 1986) sample
consisted of college student dating couples, whereas Penman’s and the current sample
both consisted of couples who had a more extensive relationship history (in the
former, couples had lived together at least a year; in the latter, parents had been
married an average of 7 years). Gottman & Levenson (1986) fail to report on
marriage length in their sample (Levenson & Gottman, 1983; 1985). Thus, the lack

of gender differences in the behavior of Penman’s and the present nonclincial samples

69

may be due to a growing similarity in response style over time in the relationship, as
well as a greater tendency to reciprocate similar problem solving behaviors. The
signiﬁcant and positive intra-couple correlations in problem solving behaviors lends
support to this explanation. Belsky, Gilstrap, and Rovine (1984) found comparable
results in their sample of new parents, in that differences in parental behaviors
declined over time.

Another reason for the disparate results from the different samples may be due
to the fact that the Rusbult et al. (1986) data was gathered by self-report whereas
Penman’s and the present study used direct behavioral observation. Therefore, these
gender differences may in part be due to a tendency for subjects to report that they
behave in a manner that is inconsistent with the way they actually behave. In
addition, when completing self-report measures, subjects may be responding to
cultural stereotypes. In this case, males might feel greater liberty than women to
report that they behave negatively toward their partners, as women have historically
taken the more afﬁliative role.

Finally, Gottman & Levenson (1986) report on longitudinal data from a sample
of distressed and non-distressed couples followed for 3 years (Levenson & Gottman,
1983; 1985). Gender differences in this sample are primarily examined in terms of
declining marital satisfaction over time, rather than separately for satisﬁed and
dissatisﬁed couples. Therefore, it is likely that some of these differences are due to
the greater negative affect experienced by distressed couples, and the differential

responses that high levels of affective arousal elicit in men and women.

70
F. 11... IE 11' .

One obvious limitation of this study is its cross-sectional and correlational
design which render it unable to address the direction of causality between parents’
perceptions of their parenting alliance and the behaviors they exhibit in their
interactions together. It seems reasonable, however, to speculate that the relationships
between parents’ behaviors, the outcomes of their behaviors, and their feelings about
the parenting alliance are reciprocal. In addition, other factors in the etiology of
parental behavior patterns and perceptions of their parenting relationship remain to be
clariﬁed.

Another weakness of this data is that the raters of the problem solving scores
also assigned the behavioral codes to the speech units, and therefore may have been
somewhat biased by their lmowledge of the behavioral codes. Caution was taken
however, to assign the behavioral codes and problem solving scores at two different
points in time. In addition, the different patterns of correlations (e. g. , the greater
strength of mothers’ versus fathers’ behaviors in predicting the outcomes) are unlikely
to be merely the artifact of rater bias.

One ﬁnal possible limitation of this data is related to the sequential analytic
program used to determine the z-scores for lagged events (Bakeman, 1985). Where
either an antecedent or a consequent event never occur in the sequence of behaviors, a
missing value is assigned to the z-score for that sequence. Hence, this may have
reduced or in some other way altered the correlations between the existing z-scores

and the outcome variables by dropping participants exhibiting the antecedent but not

71

the consequent behavior (i.e. , those where the consequence has a 0 probability of
occurrence).

This work has a number of strengths also worth noting. First, it includes
multiple modes of measurement-both self-report and behavioral observation. The
behavioral interaction was one which was personally relevant to the subjects, and
hence fairly likely to evoke affective involvement. It was also conducted in the
privacy of the subjects’ homes, and thus less likely to be inﬂuenced by observer
effects. In addition, the behavioral data lend support to the construct validity of the
self-report measure used to assess the parenting alliance. For example, mothers’
destructive problem solving behaviors in general were negatively related to the
parenting alliance, whereas their constructive behaviors in general were positively
related to the parenting alliance. In addition, mothers’ reciprocity of fathers’
destructive behaviors was associated with weaker parenting alliances and with parents’
feelings of denigration.

Another strength of this study is the use of categorical and sequential data in
the analyses which permits the assessment of different aspects of the behavioral
interaction. In addition, the sample is one in which the parenting alliance is
especially salient and more differentiated from the marital relationship than in older
couples. Thus, self-report measures assessing perceptions of the parenting alliance
rather than the marital relationship may be more relevant when studying a discussion
involving parenting issues. However, future studies need to compare parents’

perceptions of the parenting alliance and marital adjustment with observed

72
interactions.

Future research should also include longitudinal studies aimed at examining
more closely the directions of causality between parents’ perceptions and their
behaviors. In addition, longitudinal data need to address how a positive parenting
alliance develops early in the family life cycle, and how it is maintained. The
research should also be extended to include more heterogeneous samples of parents
including those with a lower socioeconomic background as well as clinic populations
with more "dysfunctional” parenting alliances so that the interparental behaviors
distinguishing a more from a less positive alliance may be more clearly deﬁned. This
will lead us toward a greater understanding of how one might prevent difﬁculties in

the alliance and what interventions might be effective in enhancing weaker alliances.

APPENDICES

AKPPElﬂ3EKlA
THE FAMILY EXPERIENCES QUESTIONNAIRE

INSTRUCTIONS: Using the scale described below circle the ietter(s)

that indicate how much you agree or disagree with each of the
following statements.
r

SD- Strongly Disagree D- Disagree A- Agree SA- Strongly Agree

Example item:
Ex: My spouse and I like to play baseball. SD D A 35

The circle around the D shows that you DISAGRBE with this _
statement. If you STRONGLY DISAGREB you would circle the SD, if you

AGREE you would circle the A. and if you STRONGLY AGREE you would
circle SA.

Now complete each of the following items in the same way:

1. My spouse and I are as well adjusted as any two SD D A SA
persons in this world can be.

2. I am not pleased with the personality . SD. D A SA
characteristics and the personal habits of my spouse.

3. I often overreact when my child misbehaves. SD D A SA
6. I live for my children. SD D A SA
5. Parenting has brought my spouse andum closer SD D A SA
together.

6. My spouse tries to have the last word in how we SD D A SA
raise our children. . '
7. I want my children to behave in public so that SD D A SA
people will know that I am a good parent.

8. I know that I am doing a good job as a parent. SD D. A SA
.9: Having children makes me feel like I am- SD D A SA

contributing to the future of society._

10. As a parent, I never stop enjoying seeing the SD D A SA
world through my children's eyes.

11. I try to give my children direction but mostly I SD D A SA

let them grow by themselves.
1

12. Being a'parent makes me feel more important because SD D A SA
I know that I am the center of someone's world.

13. My spouse and I understand each other completely. - SD D A SA

73

74

14. Being a parent turned out not to be as difficult
as I thought it would be.

15. My spouse thinks that I am a bad influence on the
children.

16. Parenting has taught me not to get too upset about
little frustrations.

17. I feel very good about how my spouse and I practice
our religious beliefs and values.

18. My spouse is a good parent.

19. My spouse appreciates how hard I work at being a
good parent.

20. Knowing that my children will carry on in my place

is the most important reward of being a parent that I
now.

21. My spouse backs me up as a parent.
22. Being a parent makes me feel drained and depleted.

23. My spouse and I feel we are growing and maturing
together through our experiences as parents.

24. I intend to push my children in order to make sure
that they achieve the things I never got to do myself.

25. My spouse is willing to make some personal
sacrifices in order to help with the parenting.

26. I am able to be consistent with my children so that
they do not have to wonder what I am going to do next.

27. My spouse tries to make sure I get some time for
myself away from the children.

28. I an overly protective of my children; it is better
to be safe than sorry.

29. I have the knowledge I need to be a good parent.

30. When there is a crisis with the children my spouse
doesn't help me as much as I would like. a

‘31. If my spouse has any faults I am not aware of them.

32. I have'learned that if my kids need something
important I can rely on my spouse to help provide it.

33. I am dissatisfied about the relationship my spouse
and I have with my parents, in-laws, and or friends.

SD

SD

SD

SD
SD

SD

SD

SD
SD

SD

SD

SD

SD

SD

'31)

SD

SD

SD

SD

SA

SA

SA

SA

SA
SA

SA

SA

SA
SA

SA

SA

SA

SA

SA

SA

SA

SA
SA

SA

75

34. My spouse does not reall ' '
t y an o bein alone with
the children. J y g

35. I get a feeling of pride from watching my children
accomplish a goal that they are proud of.

36. After my spouse or I have handled a difficult
situation with the children we discuss it and try to
figure out what we could have done better.

37. I am a very strict parent.

38. I should have read more books on parenting because
I often feel like I don't know what I am doing.

39. My children are reflections of myself.

40. My spouse likes to play with the children, but then
leaves the dirty work to me.

41. I appreciate how much my spouse tries to be a good
parent.

42. My spouse completely understands and sympathizes
with my every mood.

43. My children get on my nerves.

44. I am very happy with how my spouse and I handle
role responsibilities in our marriage.

45. Every new thing I have learned about my spouse has
pleased me.

46. One of the things I like most about being a
parent is that my children are so tuned in to what
I do and say.

47. If I could do it over again I would raise my
children the same way I am raising them now.

48. What I find most satisfying about being a parent
is showing my children the difference between right and
wrong.

49. I am not satisfied with the way my spouse and I
each handle our responsibilities as parents. ‘

50. My spouse resents that I have to give so much of
my time to the children.

51. I often worry that I am letting my children down.

SD

SD

SD

SD
SD

SD
SD

SD

SD

SD
SD

SD

SD

SD

SD

SD

SD

SD

SA

SA

SA

SA

SA

SA
SA

SA

SA

SA
SA

SA

SA

SA

'SA

SA

SA

SA

76

52. My spouse and I like to imagine together what our
children will be like when they grow up.

53. My spouse makes me look like the "bad person" in
the eyes of our children.

54. My kids are always trying my patience.

55. Hhen the children are sick I can turn to my spouse
for support.

56. Nhat I most enjoy about being a parent is

watching my children grow and change in ways that
I never imagined.

57. I see to it that my children are only exposed to
things that I want them exposed to.

58. I am going to make sure that my children accomplish
the things in life that are important to me.

59. Whenever I start feeling comfortable as a parent

something goes wrong and the doubts start all over
again.

60. My spouse and I agree on our ideas, guidelines. and
rules for raising our children.

61. My spouse forgets that kids are kids, not little
adults.

62. My spouse has a good feel for the kids and what
they might need.

63. Because my children are a part of me, I find it
difficult to let them be independent.

64. I worry that I am not doing the right thing as
a parent.

65. As a parent I really enjoy the feeling that I am
.mblding another human being.

66. I did not know how much anger I had inside of me
until I became a parent.

SD

SD

SD
SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

'31)

67. I like watching my children's personalities develop SD

even when they turn out differently from what I expected.

38. My.spouse and I do not agree on when to punish and
how much to'punish.

69. My spouse does not live up to my idea of a good
parent.

SD

'SD

SA

SA

SA
SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

SA

"SA

77

70. My spouse and I have conflicts about how much we SD
should do for our children.

71. I have learned to accept that I cannot shelter my SD
children from everything I do not like.

72. Parenting has given my spouse andme a focus for the SD
future.

73. I want my children to do the same things I did when SD
I was a child.

74. When I make a mistake with the kids I can talk it SD
over with my spouse.

75. I try not to box my children in with too many rules. SD

76. My spouse helps out with the parenting whenever SD
possible.

77. No matter how hard I try, I never seem to be a good SD
enough parent.

78. My spouse enjoys me both as a parent and a lover. . SD

79. When I get short with my children, I usually can SD
catch myself before I do something I regret.

80. As a parent, I cannot seem to do anything right in SD
my spouse's eyes.

81. My spouse and I work closely together as parents. SD

82. I am very pleased with how my spouse and I express SD
affection and relate sexually.

83. I get a great deal of pleasure out of shaping and SD
molding my children so that they grow up to be the
kinds of people I want them to be.

84. My spouse makes too many demands on me as a parent. SD

85. There are times when I do not feel a great deal of .SD
love and affection for my spouse. -

86. I often worry that I don't know enough to be a SD
good parent.

‘87. What I most enjoy about being a parent is that my SD
children make it possible for me to get a new a
perspective on the world and myself.

88. I feel my spouse does not understand me and I am .SD
not happy with our communication. '

SA

SA

SA

SA

SA

SA
SA

SA

SA
SA

SA

SA
-SA

SA

SA
'SA

SA

SA

'SA

78

89. I feel too ashamed about my mishaps with the
children to talk them over with my spouse.

90. I want my children to be interested in the things
I was interested in as a child.

91. I have to be on guard with my children all the
time to keep them from getting into trouble.

92. I get a thrill watching my children discover new
things all by themselves.

93. I am afraid of my spouse's anger when I do
something wrong with the kids.

94. I often think my children would be better off with
one parent (mg) than with the both of us.

95. I-do not feel that parenting is as much of a

sharing experience with my spouse as I hoped it
would be.

96. Compared to most parents I know, I seem to have
less difficulty disciplining my children.

97. I feel closer to my children than to my spouse.

98. My spouse and I agree on how much time we each
should spend with the children.

99. I often feel guilty about neglecting my children.
100. My marital relationship is not a perfect success.

101. One of the things I most enjoy about parenting is
seeing myself in my child.

102. My spouse and I get on each others' nerves when
the children are difficult or act up.

103. I work hard at shaping my children's lives rather
. than just letting them grow up as they would.

104. I bite my lip when my spouse disciplines the
children because if I say what I think it causes too
much tension.

105. I am very happy with how my spouse and I handle
our leisure activities and the time we spend together.

106. I often feel torn between my loyalties to my
spouse and my loyalties to my children.

107. I do not mind that being a parent makes my life
less orderly. - .

SD

SD

SD

SD

SD

SD

SD
SD

SD
SD
SD

SD

SD

'SD

SD

SD

SD

SA

SA

SA

SA

SA

SA

SA

SA

SA
SA

SA
SA
SA

SA

SA

SA

SA

SA

SA

.0

79

108. Sometimes I feel like my spouse is one of the
children instead of my partner.

109. I don't think any couple could live together with
greater harmony than my spouse and I.

110. My spouse overreacts when the children act up.

111. I am unhappy with our financial position and the
way my spouse and I make financial decisions.

112. My spouse pays too little attention to the
children.

113. I find it difficult to find the right balance
between discipline and love in raising my children.

114. I don't think anyone could possibly be happier
than my spouse and I when we are with one another.

115. I am very happy with the way my spouse and I make
decisions and resolve conflicts.

116. My spouse still wants to "do his or her own thing"
instead of being a responsible parent.

117. When I am around my children, I usually find
myself thinking "Why do they have to be so difficult?"

118. My spouse does not trust my abilities as a parent.

119. I have some needs that are not being met by my
relationship with my spouse.

120. Juggling all the responsibilities of being a
parent is one of my talents.

121. There are times when my spouse does things that
make me unhappy.

122. If every person in the world of the opposite sex
had been available and willing to marry me, I could not
have made a better choice. -

123. When my children show their will, I make sure they
know who is boss.

124. I have never regretted my relationship with my
spouse not even for a moment.

125. My spouse preaches alot about how to be a good
parent but rarely puts it into practice.

126. When I tell my children to do something, they will

SD

SD

SD
SD

SD

SD

SD

SD

SD

SD

SD
SD

SD

SD

SD

SD

SD

.31)

SD

SA

SA

SA
SA

SA

SA

SA

SA

SA

SA

SA
SA

SA

SA

SA

SA

SA

SA

SA

80

do it. no "ifs", "ands", or "buts".

127. Having children has helped me to see positive
qualities in my spouse that I never noticed before.

128. Parenting means a lot of responsiblities and
problems, but I always feel that I can cope with

the difficulties that come along.

129. My spouse has a lot of patience with the children.

130. I often feel that I have no control over my
children.

131. My spouse sees parenting as my responsibility.

132. When my kids do something I do not like I blow up
first and ask questions later.

133. I worry about the children's safety when they are
alone with my spouse.

134. My spouse expects too much from the children.

135. My spouse is too self-centered to be a good
parent.

136. When there is a crisis with the children, I know
that I will do what needs to be done.

137. I feel over-burdened as a parent because my spouse
is often too involved with other things to carry a fair
share of the load.

138. My marriage could be happier than it is.

139. When I feel at my wits end as a parent my spouse
gives me the extra support I need.

140. I have learned to accept that sometimes my kids
will not do what I want no matter how hard I try.

141. My spouse and I often talk together about what is
best for our children. '

142. My spouse makes me feel that I am the best
possible parent for our children.

‘143. When my child misbehaves or breaks a rule I try
to find out the reasons why.

b

SD

SD

SD

SD

SD
SD

SD

SD
SD

SD

SD

SD

SD

SD

SD

SD

SD

SA

SA

SA

SA

SA
SA

SA

SA
SA

SA

SA

SA
SA

SA

'SA

SA

SA

APPENDIX B

DISCIPLINE PROBLEMS TASK

We would like to learn more about the kinds of decisions and difﬁculties that you as
parents deal with in disciplining

 

First we would like you to complete this brief questionnaire without consulting your
spouse.

A. How difﬁcult is it for you and your spouse to agree on how and when to
discipline 9

 

Rarely a Sometimes Often a Almost always
problem a problem problem a problem
1 2 3 4

B. Now think about speciﬁc times within the last few weeks when you and your
spouse had the most difﬁculty agreeing on how and when to discipline

. Pick m examples that were especially difﬁcult and describe
them brieﬂy in the spaces below. List them with the most difﬁcult example ﬁrst and
the least difﬁcult last.

 

81

82
DISCIPLINE PROBLEMS TASK (CONT’D.)
For the interviewer:
Pick a problem both report. If more than one, pick the one with the highest average

rating for difﬁculty. If no problems are reported by both parents, pick mother’s
highest rated problem.

Say: Now I’d like the two of you to talk about how you might handle the following
problem better as a couple.

(Restate the problem.)

You have ten minutes to come to an W on how you will handle that problem
better in the future. I am going to record your discussion, but I will leave the room
so that you can talk more comfortably.

(TURN ON THE TAPERECORDER BEFORE YOU LEAVE. RETURN IN TEN
MINUTES AND TURN IT OFF.)

APPENDIX C

LATENT LEVEL CLASSIFICATION SCHEME
ABBREVIATED CODING MANUAL

(Adapted from: Robyn Penman (1980). WWW.
N.Y.: Academic Press.)

Offer (OF):

11351331913: Actor sustains own position and tentatively approaches the other.

Intent: to explore the situation without fully committing self to involvement with the
other.

1am: offers tentative proposals or suggestions; informs other; acknowledges
information from other; oriented toward task rather than other; hospitable but not
committed.

Examples: Mother suggests, "Perhaps that’s something that we both could do.”
Father informs mother, ”Well, she didn’t take her nap that day. "

Seek (SK):

Indjcgtnzs: Actor allows other to approach and plays down self by temporarily
ignoring own options.

Intent: to obtain assistance from the other.

Tactics: requests for action/information/clariﬁcation; allows other to start, seeks
conﬁrmation of own activity; indicates other has something to offer.

Enamplcs: Father begins discussion with, ”Well, what do you think we should do
about it?"

Mother asks father, ”How much did she eat for her snack?”

Abstain (AB):

135nm: Actor ignores the possibility of exerting self and avoids involvement in
current exchange.

Inmnt: to temporarily avoid a move which will commit self; or to discourage other
from continuing in present direction.

lam: actor shows indecision at point in interaction; actor reserves decision;
delaying tactics may be used, such as inﬂected ”hmm?”, unﬁnished sentences, or
joking to avoid commitment.

W: Mother asks father what he thinks they should do when the kids are dirty
and father replies, ”Hose ’em down.”

Father makes suggestion and mother replies, ”Hmm, could be...I don’t know.”

83

84

Reskt (RS):

Indigtms: Actor temporarily blocks other from movement towards him/her and
maintains own position.

Intentzto counteract position/ statement of other to establish or maintain own position.
Iaeties: arguing for own case without discrediting other or imposing self; is cynical
about other’s position; shows unwillingness to accept other’s suggestion; querying as
a way of indicating nonacceptance; setting up obstacles for other.

Fmemptes: Father offers suggestion and mother says, "I can’t imagine how that would
work. "

Mother gives a possible solution and father replies, ”Why would that work any
better?”

Control (CN): '

Indicators: Actor asserts self and temporarily blocks move towards greater
involvement.

Intent: to take over and manipulate interaction for elevation of self without
commitment of self.

Iaetics: direct, controlling ordering of other; indirect manoeuvering to gain control of
situation; challenging other to defend position; questioning without commitment;
taking charge of situation without reference to other and not allowing other to
interact.

Examples: After clear lack of agreement father states, ”So, we’re done then, we’re
going to ...(gives own solution)."

Mother repeats her position after father has made other suggestions, "Well, I still say
my way would work. "

Initiate (IN):

Innieatens: Actor afﬁrms self and approaches other in move towards further
involvement.

Intent: to lead without control; to offer possibilities for involvement.

Iaenes: suggests actions; constructively challenges other to move towards; makes a
stand without control or imposition of self.

Examples: Mother suggests, ”I think we both need to be more consistent in how we
handle that. "

Father states, "So we have to talk about what we’re going to do about it tonight.”

Share (SH):

Indieatens: Mutual afﬁrmation of self and other in a total involvement.

Intent: To commit self to joint enterprise of mutual beneﬁt.

Iaetjes: Shows acceptance of self and other; engages other; positive open
confrontation; reciprocated honest amusement; displays of undemanding affection.
Examples: Mother states, "I think we just worked together beautifully on that one.”
Father states, "I really appreciated that you had stepped in then. "

Couple laughs together about mutual joke.

85

Collaborate (CB):

mtljeettns: Actor sustains own position while joining the other in action.

Intent: to contribute positively to interaction.

IaetIes: strong afﬁrmation of other’s proposal; reciprocation of other’s move; adds
to/expands on/ other’s action; consents to cooperate.

Examples: Father makes suggestion and mother replies, ”Yeah, that’s a good
suggestion, and then we should explain to her why we are upset. "

Father responds to mother, "You’re right, that really bothered me too. "

Oblige (OB):

Indicates: Actor plays down self position or ignores other options and merges with
the other.

Intent: to go along with other rather than impose own position

Iaet'tes: willingly accepts other’s suggestions; concurs with move of other; acquiesces
to other.

Examples: Mother and father discuss options and mother states, "OK we’ll do it your
way. "

Father states, "I think you’re right about that.”

Cling (CL):

httljeam: Actor nihilates self and merges into other in parasitic manner.

Intent: to go along with anything other wants in order to remain involved in the
relationship.

Iaetles: seeks directives from other; indicates he! she cannot move without guidance
from other; willingly accepts any directives from other.

Examples: Father states sincerely, "I’ll do whatever you want me to. "

Mother proposes solution father disagrees with and then accepts father’s suggestion
with, "No, I don’t think it will work either."

Submit (SB):

Intljtatets: Actor nihilates own framework and allows other to approach.

Intent: to indicate to other that actor is willing to abdicate own position to prevent
other from moving away.

Iaetles: acts suggesting actor wants support; gives responsibility for moving to other;
defers to other; takes path of least resistance.

Examples: Mother suggests, ”I think it would be better if you started.”

Mother proposes view that father disagrees with, so mother says, "I guess you’re
right, that was in the back of my mind.”

 

86

Relinquish (RL):

Intljeatets: Actor negates self options but avoids other and keeps at a distance.
Intent: to abandon position that actor can no longer defend.

Iacties: concedes defeat; unwillingly allows other to take over; backs away form
confrontation.

Wes: After father restates his position, mother says, "OK then, have it your
way. "

Remove (RM):

Indicettn‘s: Actor separates self from other and negates self by non-participation.
Intent: to completely dissociate self from the current exchange and other.

Iaetjes: refusal to participate in exchange as is, not replying to direct questions or
directives; distracted by events outside the current exchange.

Exmples: Father responds "I don’t know" to a simple question from mother.
Mother does not like the way conversation is going so states "I nwd to go check on
the kids. "

Evade (EV):

Innieatets: Actor separates self from other and ignores possibility of exerting power
over other.

Intent: to manoeuver out of present situation or to discontinue it.

Iaetjes: doesn’t respond to other; attempts to change subject after uncomfortable
exchange; indicates to other that he/ she is not listening or involved; vague wordy
abstractions.

Examples: Mother asks father what he thinks they should do and father replies, "Why
ask for my opinion now, it never mattered to you before. "

Counter (CT):

Indjenttns: Actor severs relationship temporarily by refusing to accept other and
attempts to maintain own position.

Intent: to distance self from other or de-emphasize effect of other in order to stand for
self.

Iaen'txs: defense of own position at expense of other; rationalization or justiﬁcation of
own position; defying or refusing to accept other’s position.

mm: Mother rejects father’s statement and father responds, "She would too
respond to that if you’d approach it that way. "

After agreement to cooperate, mother maintains, ”But I still think my ﬁrst suggestion
would have worked best. "

87

Reject (RJ):

Ineieattn's: Actor severs relationship and afﬁrms self at other’s expense.

Intent: to assert self and to hurt/discredit/reject other and to completely sever the
relationship at that point.

Iaetjes: engages in aggressive acts with destructive consequences for other; dcnigrates
task as way of discrediting other or way of displacing anger at other; shows hostility
towards other.

Examples: Father says to mother, "I already told you I don’t think it’s a problem-
you’re just trying to ﬁnd a problem for this study.“

After disagreement, mother states ”You always have to ruin it when things are going
OK. "

LIST OF REFERENCES

LIST OF REFERENCES

Bakeman, R. (1986). PC ELAQ Versien 4. Atlanta, GA.

Bales, R.F., & Strodtbeck, F.L. (1951). Phases in group problem-solving. Ionmal
9W. i6, 485-495.

Beitel, A. (1989). Toward a reconceptualization of paternal involvement in infancy:
The role of maternal gatekeeping. SRCD, Kansas City, April, 1989.
Manuscript submitted for presentation.

Belsky, J. ( 1979) The interrelationship of parental and spousal behavior during
infancy in traditional nuclear families: An exploratory analysis. Ionnnalef

WM 41 749 755

Belsky, J. & Isabella, R.A. (1985). Marital and parent-child relationships in the
family of origin and marital change following the birth of a baby: A

retrospective analysis. W. 56, 342-349.

Belsky, J., Youngblade, L., Rovine, M., & Volling, B. (1990). Patterns of change
and parent-child Interaction. Unpublished Manuscript.

Benedek, T. (1959). Parenthood as a developmental phase. IpynnaleLtltLAmencan
PW, 2, 389-417.

Billings, A. (1979). Conﬂict resolution in distressed and nondistressed married
couples. f n lin 'n' l h l , ﬂ(2), 368-376.

Birchler, G.R., Weiss, R.L., & Vincent, LP. (1975). Multimethod analysis of
social reinforcement exchange between maritally distressed and nondistressed

spouse and stranger dyads. lenmal ef Persenality and Smim Esyehelegy,
11(2), 349-360.

Blechman, E.A. & McEnroe, M.J. (1985). Effective family problem solving. Child
mm, 5.6. 429-437.

Block, J.H., Block, J., & Gjerde, RF. (1986). The personality of children prior to
divorce: A prospective study. Child Qevelepment, 51, 827-840.

88

89

Block, J.H., Block, J., & Morrison, A. (1981). Parental agreement on child-rearing
orientations and gender-related personality correlates in children. Child

W. 52. 965-974-

Christensen, A., Phillips, 8., Glasgow, R.E., & Johnson, SM. (1983). Parental
characteristics and interactional dysfunction in families with child behavior

problems: A preliminary investigation. MW
thnlaex. 11(1). 153-166-

Cohen, G. (1977). Absentee husbands in spiralist families. InnmalefManjagemd
My 329 595'604'

Cohen, J. (1960). A coefﬁcient of agreement for nominal scales. EdueatjenaLand
W .2_Q (1). 37- 46

Cohen, R. S. & Weissman, S. H. (1984). The parenting alliance. In R. S. Cohen, B. J.

Cohler & S H Weissman (Eds ), WW
(pp. 33-49). N. Y.: Guilford Press.

Coombs, R. (1966). Value consensus and partner satisfaction among dating couples.
W 28, 166-173

Deutsch, M. (1969). Conﬂicts: Productive and destructive. Jetn'naleLStxziaIJssues,
25(1), 7-41.

Dorsey, B.L., Nelson, R. O., & Hayes, S. C. (1986). The effects of code
complexity and of behavioral frequency on observer accuracy and interobserver

agreement. W, 8. 349-363.

Durst, P.L., Wedemeyer, N .V., & Zurcher, LA. (1985). Parenting partnerships
after divorce: Implications for practice. M. 30(5), 423-428.

Easterbrooks, M.A. & Emde, RN. (1988). Marital and parent-child relationships:
The role of affect in the family system. In R.A. Hinde & J. Stevenson-Hinde

(Eds), l ' hi ihin ili ° ' n (pp. 83-103). N.Y.:
Oxford University Press.

Emery, R. E. (1982). Interparental conﬂict and the children of discord and divorce.
W 22(2). 310-330

90

Emery, R. E. & O’ Leary, K. D. (1982). Children’s perceptions of marital discord and

behavior problems of boys and girls MmalmtAbnarmalﬂhtlithplczx
19(1), 11-24.

Emery, R. E. & O’ Leary, K. D. (1984). Marital discord and child behavior problems

in a uonolinio samme MMnnammlsLPsxsbalch 12(3). 411-
420.

Engfer, A. (1988). The interrelatedness of marriage and the mother-child
relationship. In R. A. Hinde & J. Stevenson-Hinde (Eds. ), Relatjenshjpe

WW. 104—1 18). N. Y.: Oxford University
Press.

Fineberg, B. L. & Lowman, J. (1975). Affect and status dimensions of marital
adjustment MmaLQLMaWMLEamﬂx 31. 155- 160

Floyd, F. (1988, August). ° ' ' V .. ' ' .. ' '
temded_t‘,_udten. Paper presented at the I. A. S. S. M. D. meeting, Dublin,

 

Ireland.

Frank, S., Breitzer, G., & Freeark, K. (1986). Family factors in otitis media. Grant
application.

Frank, S., Freeark, K., Lopez, M. &Brietzer, G. (1989). W:
All in 1110111.".411111 11100-0“ :‘I'vl"!
W Paper presented at the Florida Conference on Child
Health Psychology.

Frank, S., Hole, C.B., Jacobson, S., Justkowski, R. J., & Huyck, M. (1986).
Psychological predictors of parents’ sense of conﬁdence and control and self-

versus child-focused gratiﬁcations. Wm, 22(3), 348-
355.

Frank S Jacobson S &Hole CB (1988) Watts
We Unpublished manuscript

Frank S &Tuer S (1988) MW
patentladeleseemadjnstmem. Manuscript submitted for publication.

Frank, S., Wagner, A., Olmsted, C., Laub, C., Freeark, K. ,Breitzer, G. & Peters,

1 (1989) WWW. Abstract
submitted to APA for 1989 conference.

91

Goldberg, W.A. 81 Easterbrooks, M.A. (1984). Role of marital quality in toddler
development. WW. 211(3). 504-514.

Gottman, J. M. & Levenson, R. W. (1986). Assessing the role of emotion in
maniage Went 8. 31-48

Gottman, J., Notations, C., Markrnan, H., Bank, 8., Yoppi, B. & Rubin, M. E.
(1976). Behavior exchange theory and marital decision making. Ienmalef

W 3.40). 14-23

Hakim-Larson, J. & Hobart, C.J. (1987). Maternal regulation and adolescent
autonomy: mother-daughter resolution of story conﬂicts. InnmaIpfleuthm
Adeleseenee.16(2). 153-166.

Hansen, D. J., Tisdelle, D. A. ,& O’Dell, S. L. (1985). Audio recorded and directly
observed parent-child interactions: A comparison of observation methods.
W 1. 389- 399.

Haynes, S. N., Chavez, R. E. ,& Samuel, V. (1984). Assessment of marital
communication and distress. W, 5, 315-321.

Hess, R. D. & Camara, K. A. (1979). Post-divorce family relationships as mediating
factors in the consequences of divorce for children. Wales,
35(4), 79-96.

Jacob, T. (1974). Patterns of family conﬂict and dominance as a function of child
age and social class. WW. 19(1). 1-12.

Jacob, T. Ritchey, D., Cvitkovic, J. & Blane, H. (1981). Communication styles of
alcoholic and nonalcoholic families when drinking and not drinking. Ienmal

eLSmdieemAMhel 1(5). 466-432

Kelley, H., Cunningham, J., Grisham, J., Lefebvre, L., Sink, C., 81 Yablon, G.
(1978). Sex differences in comments made during conﬂict within close
heterosexual pairs. W. 5(4), 473-492.

Kolb, T. M., & Straus, M. A. (1974). Marital power and marital happiness 1n relation
to problem-solving ability Will. 3.6 756-766

Landis, J. R. & Koch, 6.6. (1977). The measure of observer agreement for
eategorical data. Elemenjes, 31, 159-174.

92

Levenson, R. & Gottman, J. (1983). Marital Interaction: Physiological linkage and

affective exchange MmaLetPetsenahnLaniSeeialPsichelesx 45(3). 5 87-
597.

Levenson, R., & Gottman, J. (1985). Physiological and affective predictors of

change In relationship satisfaction WM
281911919831. 59(1). 85-94

Margolin, G. 81 Wampold, B. (1981). Sequential analysis of conﬂict and accord 1n

distressed and nondistressed marital partners. IenmaLefgensnltingjnd
Watchdog! _2(3). 554-567

Markrnan, H. J., 81 Notarious, C. I. (1987). Coding marital and family interaction:

current status In T Jacob (Ed ). WW (pp
329- -.390) N. Y.: Plenum Press.

McDonald, G. W. (1980). Family power: The assessment of a decade of theory and
research 1970-1979 WWII—1119.133911]! .42. 841 854

Menaghan, E. (1982). Measuring coping effectiveness: A panel analysis of marital

problems and o0ping efforts MW. 23. 220-
234.

Moos, R. & Moos, B. (1984). The process of recovery from alcoholism. III.
Comparing functioning 1n families of alcoholics and matched control families.
MmaleLStueimenAleehel. .50). 111- 118

Penman. R- (1980). WWW N-Y-= Academic

Press.

Raush, H., Barry, W. ,,Hertel R., &Swain,M..A (1974). W
W. San Francisco: Jossey-Bass Publishers.

Rusbult, C. B., Johnson, D. J., & Morrow, G. D. (1986). Impact of couple patterns of
problem solving on distress and nondistress 1n dating relationships. .19an
Wow. 10(4), 744-753.

Rusbult, C. E. & Zembrolt, I M (1983). Responses to dissatisfaction 1n romantic
involvements: A multidimensional scaling analysis. Wm
Watchmen. 12, 274-293.

Sprey, J. (1969). The family as a system in conﬂict. JeumaleLMamageandthe
Pamilx. 3.1. 699-706-

93

Vincent, J. P. ”Weiss R. L., & Birchler, G. R. (1975). A behavioral analysis of
problem solving 1n distressed and nondistressed married and stranger dyads.

Behamerﬂempx 6. 475-487

Vuchinich S., Emery, R.E., & Cassidy, J. (in press). Family members as third
parties in dyadic conﬂict: Strategies, alliances, and outcomes. Child
Development.

Weissman, S. & Cohen, R.S. (1985). The parenting alliance and adolescence.
AdeleseentPsxsbianu. 12. 24-45.

Wills, T. ,Weiss, R. ,& Patterson, G. (1974). A behavioral analysis of the

determinants of marital satisfaction hemaleLQensnlﬁnundﬂinieal
Psxcbelegx 42(6). 802- 811

"11111111111111ls