THESIS

Z

NIVRI

llllilllllllllll linilnmlflififll L

3 1293 01570 6249

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

THE RELATIONSHIP BETWEEN RECURRENT OI‘ITIS MEDIA
AND PARENT CHIDD INTERACTIONS
presented by

MCI—REL LEE IDPEZ

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Psychology

 

or professor

Datei/‘g‘l/7 }

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

 

LIBRARY

Michlgan State
University

 

 

 

 

 

 

 

PLACE ll RETURN BOX to runovo this checkout from your record.
TO AVOID FINES rolum on or Moro dot. duo.

    

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU In An Afflrmotlvo Action/Equal Opportunlty Institulon
Wanna-9.1

THE RELATIONSHIP BETWEEN RECURRENT OTITIS MEDIA

AND PARENT-CHILD INTERACTIONS

By

Michael Lee Lopez

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1997

ABSTRACT

THE RELATIONSHIP BETWEEN RECURRENT OTITIS MEDIA
AND PARENT-CHILD INTERACTIONS

By

Michael Lee Lopez

The purpose of the present study was to assess relationships among ROM, child
behavior and parent-child interactions for a group of families with children (n = 56) who
had varying degrees of recurrent otitis media (ROM) during the first three years of life. It
was anticipated that the quality of parent—child interactions would play a role in
moderating the deleterious effects of ROM on children's behavioral functioning. Data
analyses tested two alternative models postulating relationships between the severity of
children's ROM during the first three years of life, style of parental interactive behavior
(parental warmth and parental control), and children's behavioral adjustment at three to
four years of age.

The first model, the "independent influence” model, postulates that ROM severity
and style of parental interactive behavior (parental warmth and control) independently
influence children's behavioral adjustment, with ROM having a direct and indirect efi‘ect on
adjustment. The second model, the "buffering” model, argues that an effective style of
parental interactive behavior buffers the deleterious effects of ROM on children’s
behavior.

In general, the findings provided little support for the “independent influence”

model. A history of ROM predicted neither children’s negative behavior, nor parental

warmth or control during the structured, laboratory parent-child interaction task, with a
single exception. In contrast, some support was found for the “buffering” model in that
the results suggest style of parental interactive behavior may moderate the relationship
between ROM and child behavior problems; however the findings were not always in the
predicted direction. In particular, 1) ROM severity interacted with maternal warmth, and
2) ROM severity interacted with maternal control and child sex in predicting child
negative behavior during the structured, laboratory play task. The findings from the
present study indicate that health care providers need to become more aware of the
multiple factors that determine the impact an illness such as ROM may have on children's
emotional and behavioral adjustment, especially given that parents may be in the unique

position to moderate the effects of illness on their children’s psychosocial adjustment.

ACKNOWLEDGMENTS

I would like to acknowledge the input and guidance fiom my committee members
(Drs. Susan Frank, Hiram Fitzgerald, Bob Zucker and John McKinney), as well as their
patience and willingness to see me through to the end of this lengthy journey. I also have
to recognize the many different contributions from Suzy Pavick, who has never failed with
her wealth of institutional knowledge, skillful problem-solving and cheerful personality.
Of course, to the many friends and colleagues (Nick Ialongo, Peter Jensen, Miriam
Martinez, Leon Puttler, Louisa Tarullo and many more) who provided periodic, but quite
necessary prodding and words of support, if not actual technical input, I owe you many
thanks. I would also like thank my parents for providing the basic, early foundation, that
made this all possible. Finally, I wish to express my deepest gratitude to my wife, Shelly,
and my two wonderfiJl children, Jessica and Koji, for all of their support, tolerance and
love over the years, without whom I never could have realized this dream. I only hope, as
Jessica and Koji set out upon their own personal journeys, that my accomplishment will
serve as an inspiration and help them realize that despite all of life’s challenges, with

perseverance and hard work anything is possible.

ii
TABLE OF CONTENTS

LIST OF TABLES ................................................................................... iii
LIST OF FIGURES ................................................................................. iv
Introduction ............................................................................................ 1
Hypothesis 1 ............................................................................................................... 11
Hypothesis 2 ............................................................................................................... ll
Hypothesis 3 ............................................................................................................... 12
Literature Review .......................................................................................................... 13
Otitis media: Epidemiology ......................................................................................... 13
Research on recurrent otitis media (ROM) .................................................................. 15
Socialization influences. .............................................................................................. 19
Dimensions of parenting ............................................................................................. 22
Research examining dimensions of parenting ............................................................... 27
Limitations of Socialization Literature ........................................................................ 31
Limitations of current research on ROM and its relationship to the current study. ....... 47
Study Hypotheses .......................................................................................................... 52
Hypothesis 1 ............................................................................................................... 54
Hypothesis 2 ............................................................................................................... 54
Hypothesis 3 ............................................................................................................... 54
Methods ........................................................................................................................ 55
Participants ................................................................................................................. 55
Procedures ................................................................................................................. 56
Clinic observational sessions ....................................................................................... 57
Dependent Measures .................................................................................................. 59
Taping and coding procedures. ................................................................................... 64
Results ........................................................................................................................... 66
Measurement Issues. .................................................................................................. 66
Computation of recurrent otitis media (ROM) severity ................................................ 71
Demographic Relationships ........................................................................................ 71
Within task variation in child and parent behavior. ...................................................... 72
Tests of the Research Hypotheses ............................................................................... 74
Discussion ..................................................................................................................... 83
Hypothesis 1: ............................................................................................................. 84
Hypothesis 2: ............................................................................................................. 87
Hypothesis 3: ............................................................................................................. 88
Limitations of the Present Study ................................................................................. 94
Summary .................................................................................................................... 99
APPENDD( A ............................................................................................................. 102
APPENDIX B ............................................................................................................. 104
APPENDIX C ............................................................................................................. 114

References ................................................................................................................... 136

iii

LIST OF TABLES

Table 1. Examples of parent & child behaviors coded with the Dyadic Parent-Child
Interaction Coding System ............................................................................ 60

Table 2. A priori clusters of parent and child behaviors .............................................. 63

Table 3. Interrater reliabilities for frequency ratings of individual sets of coded
videotapes ............................................................................. 102

Table 4. Internal consistency of parent & child behavior scales ................................. 104
Table 5. Number of items and internal consistency of parent & child behavior scales .. 69
Table 6. Interrater reliability coefficients (Pearson rs) for summary variables .............. 70
Table 7. Correlations among child and parent variables ............................................ 114

Table 8. Average rates (fi'equencies per second) of child and parent behaviors within
observational tasks ....................................................................................... 73

Table 9. Correlations between ROM severity and Parental Warmth and Control .......... 76

Table 10. Beta coefficients resulting from regressing child negative behavior on parental
warmth or control ......................................................................................... 78

Table 11. Beta coefficients resulting from regressing child negative behavior on ROM
severity, parental warmth, and parental control within observational
segments ................................................................................. 79

Table 12. Summary of the characteristics of participants and design of recent ROM
studies ................................................................................... 86

iv

LIST OF FIGURES

Figure 1. Two alternative models suggesting the relationship between (1) ROM severity,
(b) style of parenting, and (c) children’s negative behavior ............................. 10

Figure 2. Relationship between (a) ROM severity, (b) maternal warmth, and (0) child
negative behavio ............................................................................................ 81

Figure 3. Relationship between (a) ROM severity, (b) maternal control, and (c) children’ 3
negative behavior. . ..82

Introduction

In recent years, researchers have begun to focus their attention on the negative
impact that childhood illnesses may have on the psychosocial development and
adjustment of young children and their families. The negative impact of illness on a
child's development can be especially severe during the earliest years of life, when the
child is undergoing development at a rapid pace (Perrin & Ferrity, 1984; Travis, 1976).

Otitis media is one of the most commonly treated illnesses in preschool-age
children (Koch, 1974; Schappert, 1992; Stool et al., 1994; Teele, Klein & Rosner, 1983).
Recurrent otitis media (ROM) has been defined as an incidence of two or more episodes
of acute otitis media occurring within a six month period, or three or more episodes
occurring within a twelve month period of time (F orgays, Hasazi & Wasserman, 1992;
Vernon-Feagans, Manlove & Volling, 1986; Wallace, Gravel, Schwartz & Ruben, 1996).
The illness typically involves the inflammation of the eustachian tube and middle ear,
with pus filling the middle ear cavity, resulting in what is commonly referred to as
effusion. In time, the inflammation will subside by itself, but often it is treated with an
antimicrobial agent. Once the inflammation subsides, the fluid in the middle ear either
drains through the eustachian tube, or is reabsorbed by the surrounding tissue (Paradise,
1981)

In 1982 alone, otitis media accounted for an estimated total of 21 million patient
visits to health professionals (National Drug Therapeutic Index, IMS America, New
York). It has been estimated that by the age of three, more than two-thirds of all children

will have had at least one episode of otitis media, one-third will have had at least three

2
episodes, and an even smaller percentage will experience episodes on a regular basis for

the first few years of life (Daly, 1991; Teele, Klein, & Rosner, 1989). Given the high
prevalence of recurrent otitis media (ROM) in young children, it is important to
investigate whether or not a severe history of the illness contributes to disruptions in
normal social development.

Otitis media, itself, often results in secondary effects of irritability,
inattentiveness, and/or fluctuating hearing loss (Bluestone et a1. 1983; F orgays, Hasazi, &
Wasserman, 1992; Roberts et al., 1995). Any one of these secondary effects would likely
have a direct influence on a child's temperament and/or behavior, and an indirect impact
upon a parent's ability to effectively interact with the child.

It has also been suggested that ROM may have a negative impact upon a child's
speech and language development, due to the fluctuating and/or permanent hearing loss
that is frequently experienced (e.g., Bluestone et al., 1983; Chan, Logan, & McBean,
1967, Downs, 1983; Fria, Cantekin & Eichler, 1985; Roberts et al., 1995). If a child's
speech and language development is delayed, such a delay would further contribute to
disruptions in parent-child interactions by increasing communication difficulties between
a parent and child, and by disrupting the normal development of important self-regulatory
mechanisms within the child. Thus, it appears that ROM may have a direct negative
impact upon children's behavior and/or social and emotional adjustment, as well as an
indirect negative impact upon parent-child interactive behavior. The relationships
probably are bidirectional, with respect to their direction of influence, especially when
considering the role that stress and/or certain parenting practices might have on a child’s

susceptibility to illnesses, in general. However, given the relative importance of

3
evaluating the impact of an illness on children's adjustment and related parenting

behaviors, the focus of the present study was limited to an examination of the impact that
ROM may have upon both child and parent behaviors, rather than vice versa.

The potential negative impact of a childhood illness may be most significant
during the preschool years, when children are undergoing a process of rapid social and
emotional development (Perrin & Ferrity, 1984; Travis, 1976). During the second and
third years of life, children rapidly become more autonomous as they move through the
separation-individuation process (Erikson, 1963; Mahler, Pine, & Bergman, 1975).
Parents begin to provide children with the structure and contingencies necessary for the
development of self-regulatory mechanisms by making more demands for appropriate
behavior, and imposing more restrictions on behaviors that fall outside the range of
acceptable child behavior (Kopp, 1987; Parpal & Maccoby, 1985). It is also at this stage
of development that children become increasingly able to use language and
representational imitation to reproduce the appropriate responses that are expected by
their parents, control their own behavior, and communicate more clearly with others
(Bandura, 1971; Smetana, 1994; Vygotsky, 1962). Thus, a major focus of parenting at
this age is preparing children for appropriate involvement in social relationships as they
begin to function more independently from their parents and are expected to assume more
control over the regulation of their own behavior (Kochanska et al., 1987).

Within the socialization literature, two clusters or dimensions of parenting
behavior consistently have been cited as essential components of effective parenting: (1)
parental warmth or responsiveness, and (2) parental control (Baumrind, 1967, 1968,

1971, 1988; Martin, 1987; Maccoby, 1983, 1992; Maccoby & Martin, 1983; Smetana,

4
1994). The dimension of parental warmth generally refers to a cluster of behaviors that

contain a strong affective component, indicating either approval or disapproval of
children's behavior. Examples of behaviors reflecting parental warmth include smiles,
praise, approval, encouragement, physical affection, and responsiveness to children's
needs (MacDonald, 1994; Straus & Tallman, 1971). On the other end of the continuum,
hostility would be manifested by such behaviors as criticism, sarcasm, negative
evaluations, corrections, and disapproval.

The second dimension, parental control, often refers to those behaviors which
reflect a parent's efforts to influence or otherwise exert control over the behavior of their
children. However, one of the major limitations of the socialization literature has been
the various ways that parental control has been defined. Too often, researchers have
combined negative and/or ineffective control techniques, such as punishment, yelling,
threats, vague directives, and deprivation of privileges, with more positive power-
assertive control behaviors, namely clearly stated commands and directives (e.g., Darling
& Steinberg, 1993; Maccoby & Martin, 1983; Rollins & Thomas, 1979; Steinmetz,
1979).

There are clear distinctions between these two types of parental control attempts.
The first group of behaviors reflect parental punitiveness, coerciveness, or ineffectiveness
and can be viewed as "functionally superfluous" control methods which may be
accompanied by strong negative affect (Lepper, 1981; Maccoby & Martin, 1983;
Smetana, 1994). On the other hand, clear parental commands and directives represent
consistent discipline techniques often used to effectively influence children's behavior

(e.g., Baumrind, 1971, 1988; Darling & Steinberg, 1993; Forehand & McMahon, 1981;

5
Patterson, 1982). Thus, it is important to distinguish between such positive parental

directives and the more negative, punitive, or ineffective control attempts.

In general, studies have shown that children's behavioral and social adjustment is
maximized when parents adopt a style of parenting that balances high levels of parental
warmth and positive control (Ainsworth, 1979; Baumrind, 1971; Lay, Waters & Park,
1989; Lemanek, Stone & Fishel, 1993; Rocisano, Slade, & Lynch, 1987; Rothbaum,
Rosen, Pott & Beatty, 1995; Schaffer & Crook, 1980; Wakschlag & Hans, 1997). In
summarizing the research on the dimensions of parenting, Martin (1987) concluded that
"the processes involved in warmth and control are most profitably viewed in interactive
terms. Harmonious interactions between parent and child are more likely to result when
parental control efforts are sensitive to and moderated by the infant's or child's state, in
other words, when accompanied by many of the qualities summarized as warm " (1987,
p.183). Similarly, both Maccoby (1983, 1992; Maccoby & Martin, 1983) and Kochanska
(1997) have argued that the socialization process involves inducting the child into a
reciprocal orientation or system of reciprocity, in which the development of a mutually
responsive relationship between parent and child, over time, results in a decreased need
for parental control behaviors. Thus, it is through the development of this mutually
responsive relationship that children gradually internalize their parent’s goals and values,
thereby reducing the need for parental coercion or control.

It is important to note that related studies have shown that more ineffective styles
of parenting can directly and/or indirectly result in an increase in the level of stress
experienced by children (Elder, Caspi, & Downey, 1988; Jouriles, Barling, & O'Leary,

1987; Patterson, 1982). Furthermore, it has also been suggested that exposure to high

6
levels of stress can substantially increase children's susceptibility to various illness

processes (Stein & Jessop, 1984; Boyce, Jensen, Cassel et al., 1977). However, for
heuristic purposes, the focus of the present study was limited to an examination of the
potential impact that a particular childhood illness has on children's behavior and parent-
child interactive behavior, rather than the potential effect of stress leading to childhood
illness.

Certain stressors, such as childhood illnesses, may impinge upon the early
socialization process by disrupting the normal maturation and/or attainment of
developmental milestones (Cerreto, 1986). Such a disruption to the child's socialization
likely may occur if the stressor directly impacts the child's temperament, and/or indirectly
influences the child's interactions with his/her parent(s).

It has been argued that over an extended period of time, a child with a "difficult"
temperament may gradually become an aversive stimulus for a parent, through the
process of conditioning (Lamb, 1978). Such conditioning, might then lead the parent to
become less warm, responsive, and sensitive to the child's needs, thereby placing the
child in a vulnerable position for the development of interpersonal and developmental
difficulties (Donovan, Leavitt, & Balling, 1978). Thus, a childhood illness, such as
ROM, not only may have a direct negative effect upon a child's temperament and/or
development, but also may indirectly affect parent-child interactive behavior.

Previous studies examining the impact of ROM on children's development have
found that parents report more behavior problems for children with histories of ROM,
compared to children without histories of ROM (Casey, 1983; F eagans & Proctor, 1994;

Forgays, Hasazi & Wasserman, 1992; Gottleib, Zinkus, & Thompson, 1979; Roberts,

7
Burchinal & Campbell, 1994; Vernon-Feagans, Manlove & Volling, 1996). Researchers

have thus inferred that a severe history of otitis media may have a direct negative impact
upon children's behavior and an indirect impact upon their parent's behavior, thereby
resulting in potentially dysfunctional parent-child interactions (Black et al., 1988; Black
& Sonnenschein, 1993; Casey, 1983; F orgays, Hasazi & Wasserman, 1992; Freeark et al.,
1992; Roberts et al., 1995; Wallace, Gravel, Schwartz & Ruben, 1996).

However, only recently have investigators begun to move beyond the reliance on
self—report data by including actual assessments of the relationship between recurrent
otitis media and parent-child interactions (Black et al., 1988; Black & Sonnenschein,
1993; Casey, 1983; Forgays, Hasazi & Wasserman, 1992; Freeark et al., 1992; Roberts et
al., 1995; Wallace, Gravel, Schwartz & Ruben, 1996). In one study, Black et a1.
(1988) found that mothers of one to two year old children with histories of recurrent otitis
media were less warm and sociable when interacting with their young children, than were
mothers of children with less severe histories of otitis media. F reeark et a1. (1992) in
examining the impact of recurrent otitis media on children’s verbal abilities, found that
parental verbal stimulation moderated the effects of otitis media on language
development in a sample of three to four year old children. Similarly, Wallace et a1.
(1996) investigated the association between early otitis media, style of parent interaction
behavior and children’s language development. The results of the study suggest that high
maternal verbal stimulation was associated with better language skills, whereas high
levels of observed maternal control behavior, during a laboratory interaction task, was
associated with low scores on measures of expressive language development. Thus,

maternal control had a negative, rather than positive association with the child outcome of

interest, expressive language development.

In a related prospective longitudinal study of 61 infants, Roberts et a1. (1995)
found an association between otitis media and subsequent hearing loss, language and
cognitive outcomes, but only when parenting style and quality of the home environments
were taken into account. This latter study supported a multifactorial model of
development within the context of exposure to early otitis media. In combination with
other recent studies, it underscores the important moderating role that family factors, such
as parent interaction style, play in minimizing the impact of otitis media on children’s
development. These recent studies also emphasize the likely transactional nature of the
relationship between early exposure to otitis media and the quality of parenting behaviors
and/or the home environment, as critical determinants of children’s subsequent cognitive,
language and socioemotional development. Based upon this review of the existing
literature, it is clear that additional research is needed in order to confirm and extend the
findings from these previous studies. In particular, research to date suggests several

hypotheses that will be investigated within the context of the present study.

Overview of the Hypotheses

The purpose of the present study was to assess the relationships among ROM,
child behavior and parent-child interactions, for a group of families with children who
had varying degrees of ROM during the first three years of life. It was anticipated that
the quality of parent-child interactions would play a role in moderating the deleterious
effects of ROM on children's behavioral functioning. Two alternative models of the
relationship between the severity of children's ROM during the first three years of life,
style of parent interactive behavior (parental warmth and parental control), and children's
behavioral adjustment at three to four years of age were tested (See Figure 1).

The first model, the "independent influence” model, postulates that an early
history of ROM and style of parent interactive behavior (parental warmth and control)
independently influence children's behavioral adjustment. According to this model,
recurrent otitis media and parenting independently contribute to children’s behavior so
that children who have suffered from severe histories of ROM will be better adjusted if
they are exposed to effective parenting behavior (both high parental warmth and positive
parental control), but would not be as adjusted as children without early histories of ROM
who also are exposed to effective parenting behavior.

The second model, the "buffering model", argues that an effective style of parent
interactive behavior buffers the deleterious effects of ROM on child behavior. Thus,
those children with severe histories of ROM exposed to less effective parent interactive
behavior will exhibit behavioral adjustment problems. However, children with histories

of ROM who were exposed to effective parent interactive behavior will be as well

10

"Independent Influence Model"

 

1
0.8
0.6
0.4

i

0.2

.................. N..I...-..........-.........-..-............................ ...._.....-......................-.....

\ \ \ High ROM

 

 

0

Child Negative Behavior

 

l

 

 

-0.2

 

Low Warmth/Positive Control High Warmth/Positive Control

 

“-Low ROM
- - Hg: ROM

 

 

 

"Buffering Model"

 

1
0.8
0.6
0.4

t

0.2

 

 

0

Child Negative Behavior

-0.2

 

 

 

- l

 

 

Low Warmth/Positive Control High Warmth/Positive Control

 

 

 

-Low ROM --High ROM

 

Figure 1: Two alternative models suggesting the relationship between (a) ROM severity,
(b) style of parenting, and (c) children's negative behavior.

l I
adjusted as children raised in similar environments, but without severe histories of ROM.

Hypotheses 1 and 2 tested components of the “independent influence model”, and
Hypothesis 3 tested the “buffering model”. Data on both child and parent behavior were
collected during a behavioral observation task that included three segments varying in
structure from less to more structure. These tasks involved child-directed play, parent-

- directed play and parent initiated clean-up. In general, it was expected that there would be
greater support for one or the other hypothesis as the structure and situational demands of
the observed parent-child interaction task increased. Given that parent-child interactions
were being assessed during increasingly demanding tasks, designed to elicit more negative
behavior and demands for greater parental control, it was during these increasingly
demanding tasks that the deleterious effects of ROM were most likely to be observed.
Hypothesis 1.

A severe history of ROM during the first three years of life will be positively
associated with a greater number of child behavior problems (as observed during parent-
child interactions in the laboratory), when compared to children who had experienced
fewer and less severe episodes of otitis media.

Emmi-L2

A history of severe ROM also will have an indirect impact upon parenting
behaviors; ie. a history of ROM will be positively associated with observations of less
effective parent-interactive behaviors (both lower levels of parental warmth and positive
parental control). Similarly, less effective parenting behavior will be positively associated
with a greater number of child behavior problems during the observed parent-child

interactions.

12
Exam;

Finally, more effective parenting behaviors, defined by greater parental warmth and
positive parental control behavior, will moderate the potentially deleterious effects of
ROM on children's behavioral problems, as evidenced by fewer observed child behavior

problems during the parent-child interactions.

Literature Review
Otitis media: Epidemiology

Otitis media with effusion is one of the most common illnesses experienced by
young children (Koch, 1974; Schappert, 1992; Stool et al., 1994; Teele, Klein & Rosner,
1983). In one early study, 85 percent of the 7-1/2 to 13-1/2 year old children who had
been followed in a private clinic for a minimum of five years, had at least one episode of
otitis media at some point in their lives (Brownlee, DeLoache, Cowan, & Jackson, 1969).
In another epidemiological study, 93 percent of the children who had been followed from
birth to age 7 years, experienced at least one episode of otitis media (Teele, et al, 1989).

In the year 1982 alone, it was estimated that otitis media accounted for a total of
21 million patient visits to health professionals (National Drug Therapeutic Index, [MS
America, New York). Other estimates calculate that otitis media is diagnosed in
approximately 1/3 of all visits to physicians for illness (Bluestone, et al., 1986; Teele et al.,
1983)

During the first year of life, the incidence rate for otitis media has been found to be
as low as 14 percent (Brownlee, DeLoache, Cowan, & Jackson, 1969) and 23 percent
(Teele et al., 1983) in some studies, and as high as 51 percent (Pukander, Sipila, & Karma,
1984), 73 percent (Marchant, Shurin, Turczyk, et al., 1984) and 91 percent (Gravel,
McCarton & Ruben, 1988), in other studies. By the age of three, it has been estimated
that more than two-thirds of all children will have had at least one episode of otitis media
with effiision, one-third will have had at least three episodes, and an even smaller
percentage will experience considerably more episodes (Daly, 1991; Teele, Klein, &

Rosner, 1989). It is this last group of children that often will have their first episode
13

14
several months after birth, and will continue to have recurring episodes on a regular basis

over the first three years of life.

Recurrent otitis media (ROM) has been defined as an incidence of two or more
episodes of acute otitis media occurring within a six month period, or three or more
episodes occurring within a twelve month period of time (Forgays, Hasazi & Wasserman,
1992; Vernon-Feagans, Manlove & Volling, 1986; Wallace, Gravel, Schwartz & Ruben,
1996). The illness typically involves the inflammation of the eustachian tube and middle
ear, with pus filling the middle ear cavity, resulting in what is commonly referred to as
effusion. In time, the inflammation will subside by itself, but often it is treated with an
antimicrobial agent. Once the inflammation subsides, the fluid in the middle ear either
drains through the eustachian tube, or is reabsorbed by the surrounding tissue (Paradise,
1981)

The actual duration of an episode of otitis media can vary anywhere from a few
days to a year, or longer. However, the average duration of an episode ranges from one
to three months (Hartsen, Prellner & Heldrep, 1989; Marchant, et al., 1984; Paradise,
1981). With respect to the duration of the illness, it is important to distinguish between
the length of the acute infection, which often can resolve in 7-10 days, and the duration of
the effusion, which is more likely to persist long after the infection has cleared. It is the
effusion that has an average duration of one to three months.

Secondary symptoms often associated with episodes of otitis media include: acute
pain, decreased hearing, malaise, fussiness, and irritability (Bluestone et al., 1983;
Forgays, Hasazi, & Wasserman, 1992; Roberts, et al., 1995). During the acute phase of

the illness, children also may have trouble with concentration, and with sleeping at night.

15
A temporary conductive hearing loss accompanying otitis media with effirsion is

quite common and can vary anywhere from 20 to 40 decibels in magnitude (Bluestone et
al., 1983; Chan, Logan, & McBean, 1967, Downs, 1983; Fria, Cantekin & Eichler, 1985;
Roberts et al., 1995). Hearing losses of 15 to 30 decibels are typically considered mild,
while losses between 31 to 50 decibels are considered moderate (Northern & Downs,
1974). With a mild to moderate loss of hearing, a child will be able to understand speech
only under optimal conditions (Holm & Kunze, 1969). That is, the child will have
difficulty understanding what is being said by another person, unless that person is directly
in front of him/her and speaking in a loud tone of voice. As a final note on the
epidemiology of ROM, several studies have found that the age of the first episode of otitis
media is negatively correlated with the total number of episodes, the duration of an
episode, and/or the amount of hearing loss experienced (Casey, 1983; Daly, 1991; Teele et
al., 1989). These findings suggests that the earlier the onset of ROM, the greater the
potential for the development of subsequent problems related to otitis media.

Research on recurrent otitis media (ROM)

Various studies have shown that ROM occurring during the first one to three years
of life may be associated with learning difficulties (F eagans, Sanyal, Henderson, et al,
1987; Howie, 1980, Teele et al., 1990; Zinkus & Gottlieb, 1980), speech and language
impairments (Friel-Patti & Finitzo, 1990; Gottlieb, Zinkus, & Thompson, 1979; Lehmann,
Charron, Kummer, et al, 1979; Teele et al., 1990; Ventry, 1980), and various behavioral
problems (Casey, 1983; Feagans & Proctor, 1994; Forgays, Hasazi & Wasserman, 1992;
Gottleib, Zinkus, & Thompson, 1979; Roberts, Burchinal & Campbell, 1994; Vemon—

Feagans, Manlove & Volling, 1996). Iverson (1987) found significant correlations

16
between a history of ear infections and subsequent communication deficits (r=.35),

attentional deficits (r=.32), tantrums (r=.22), and "detachment" (r=.21) in a sample of 122
children between the ages of three and 12-1/2 years old. It should be noted, however, that
this data was primarily based on retrospective parental reports, and thus must be
interpreted with caution.

Other studies have emphasized the importance of assessing the impact of the illness
itself on the family's ability to cope with the stress of having a child suffering from a
chronic illness, such as ROM (Casey, 1983; Forgays, Hasazi & Wasserman, 1992;
Tavorrnina, Boll, Dunn, Luscomb, & Taylor, 1981). However, only a few of the
previously mentioned studies have actually attempted to examine the complex interaction
between ROM and different aspects of family functioning (Black et al., 1988; Black &
Sonnenschein, 1993; Casey, 1983; Forgays, Hasazi & Wasserman, 1992; Freeark et al.,
1992)

In one such study, Casey (1983) investigated the impact of ROM on the cognitive
abilities, verbal abilities, reported behavioral problems, and parent's perceptions of stress
among 60 children between the ages of four and five. ROM was defined by the following:
(1) the children must have had their first episode of otitis media in their first year of life,
and (2) they must have had no fewer than three episodes of otitis media during each of the
first three years of life. The results indicated that ROM did not have an impact upon
cognitive abilities, yet was related to a higher number of reported behavior problems and
significantly greater amounts of parentally reported stress.

One of the major limitations of the study was the sole reliance on parental reports

for information regarding the frequency and duration of otitis media episodes. A second

l7
limitation of the study involved the operational definition of otitis media. The participants

were selected based solely on parental reports of the frequency of otitis media episodes,
without consideration for the variability in duration of the infection itself or the effusion
that often accompanies the infection (Hartsen, Prellner & Heldrep, 1989; Marchant, et al.,
1984; Paradise, 1981). Such inclusion criterion may mask important firnctional deficits
that result from the multiplicative effects of the frequency, intensity, and duration of otitis
media episodes, rather than from any one aspect alone.

In another related study (F orgays, Hasaazi & Wasserman, 1992), the investigators
found a similar relationship between a history of recurrent otitis media during the first
three years of life and mother’s reports of more child demandingness and maternal
depression, as well as less positive perceptions of their own competency as parents. It
also must be noted, that both of these studies (Casey, 1983; Forgays, Hasaazi &
Wasserman, 1992), as well as a number or other early studies on ROM, consistently have
failed to utilize observational data to directly assess child behavior, parent behavior, or
parent-child interactions. Rather, a majority of the studies have relied solely upon parental
reports of child behavior problems as indicators of child adjustment.

Only recently have investigators begun to expand the scope of their studies to
include assessments of the relationship between recurrent otitis media and parent-child
interactions (Black et al., 1988; Black & Sonnenschein, 1993; Casey, 1983; Forgays,
Hasazi & Wasserman, 1992; Freeark et al., 1992; Roberts et al., 1995; Wallace, Gravel,
Schwartz & Ruben, 1996). In one study, Black et a1. (1988) found that mothers of the
one to two year old children with histories of otitis media that they studied were less warm

and sociable when interacting with their young children. F reeark et al. (1992) examined

18
the impact of recurrent otitis media on children’s verbal abilities and the findings indicate

that parental verbal stimulation appeared to moderate the effects of otitis media on
language development in a sample of three to four year old children.

Similarly, Wallace et al (1996) investigated the association between early otitis
media, style of parent interaction behavior and children’s language development. The
results of the study suggest that high maternal verbal stimulation was associated with
better language skills, whereas high levels of observed maternal control behavior during a
laboratory interaction task was associated with low scores on measures of expressive
language development. It should be noted that maternal control was defined in this study
as a combination of several different types of both positive and negative control behaviors,
including regulatory language, language seeking, information language, nonjudgmental
language, and judgmental language.

In a related prospective longitudinal study of 61 infants, Roberts et a1. (1995)
found an association between otitis media and subsequent hearing loss, language and
cognitive outcomes, but only when parenting style and quality of the home environments
were taken into account. This latter study supported a multifactorial model of
development within the context of exposure to early otitis media, and in combination with
the other studies cited, underscores the importance of including such family factors as
parent interaction style, among others, in fiiture studies examining the impact of otitis
media on children’s development. These recent studies also emphasize the likely
transactional nature of the relationship between early histories of recurrent otitis media
and the quality of parenting behaviors and/or the home environment, as critical

determinants of children’s subsequent cognitive, language and socioemotional

l 9
development.

Socialization influences.

For children, the first three years of life are a time of rapid grth and
development. During these early years, development becomes increasingly influenced by
complex interactions with adults and other children. It also is during the second and third
years of life that children become more autonomous as they move through the separation-
individuation process (Erikson, 1963). With significant advancements in locomotive and
cognitive abilities, children at this stage begin to consolidate their distinct sense of
separateness from their parents and organize their self-identity and self-concept (Mahler,
Pine, & Bergman, 1975). However, as children move toward greater independence, there
also are increasing socialization demands that are placed upon the parents.

Dubin and Dubin (1963) have referred to the time between birth and six years of
age as the "authority inception period". Throughout this period, one of the major tasks
for parents is to prepare children for appropriate involvement in social relationships
outside the home. Similarly, Baumrind (1975) has asserted that the early socialization
process is a time in which the child, "through education, training, and imitation, acquires
his culture as well as the habits and values congruent with adaptation to that culture"
(1975, p.2).

The child's preparation for appropriate participation in social relationships is
accomplished primarily through the imposition of parental authority. Parents must provide
a sufficient amount of structure to allow the child to experience and develop an enduring
perception of the range of acceptable behaviors. However, parents are also confronted

with a difficult challenge of balancing the need for the child's conformity to external

20
standards with the simultaneous need for the encouragement and development of the

child's autonomous behavior (Dubin & Dubin, 1963). This apparent conflict is resolved as
the child consistently experiences parentally imposed boundaries, within which he/she is
able to freely express his/her individuality.

Once a child is able to understand and conform to parentally imposed standards of
normative behavior, the next task of socialization is for the child to begin to assume
responsibility for the self-regulation of his or her own behavior (Kopp, 1987). This self-
regulation is assumed to follow directly from the structure and organization that the
parents have previously provided, as well as the child’s own increasing ability to use
language and representational imitation to reproduce the appropriate responses that are
expected by his/her parents (Bandura, 1971).

Vygotsky (1962) asserted that the critical factor responsible for the development
of self-regulatory behavior is the internalization of speech. In normal children, the
development of self-regulatory behavior is believed to occur in three-stages (Luria, 1961,
1969). At the first stage of development, the child's behaviors are brought under the overt
control of external agents, primarily the parents. At this stage the initiation and/or
inhibition of behavior is clearly controlled by the parent's speech and related actions. In
the second stage of development, around three to four years of age, parents continue to
provide the necessary demands, restrictions, and predictable contingencies for the child's
behavior, but the child begins to internalize parental standards as his/her own and starts to
control his/her own behavior with the use of overt speech. This often is seen when a
toddler "talks to him/herself“ while playing. In the final stage, which occurs around 5-6

years of age, the child begins to utilize internal speech to control his/her own behavior.

21
With firrther increases in age, there are corresponding increases in the use of internalized

speech as a primary means of controlling one's own behavior.

Baumrind (1983, 1988) has argued that the development of children's self-
regulation occurs when the social control by parents is consistent and direct. Additionally,
it has been suggested that children's identification with their parents further fosters the
development of self-regulatory behavior (Steinmetz, 1979). Within the context of a strong
affectional relationship between a parent and child, the child's imitation of parental
behaviors or the perception of the similarity between him/herself and the parent will itself
become intrinsically rewarding (MacDonald, 1992 Sears, Maccoby, & Levin, 1957).
Through such identification with the parent, the child is then able to adopt the parental
standards for appropriate behavior as his/her own. Thus, the development of a child's self-
regulatory behavior is both a function of the structure and contingencies provided by the
parents, as well as the child's identification with nurturing parents.

In summarizing the crucial role that parents play in the socialization of their young
children, Kochanska, Kuczynski, Radke-Yarrow, & Welsh (1987) asserted that parents
"must preserve their ability to effectively control their children's behavior, but they must
also adopt strategies that support their children's developing autonomy" (1987, p.442). In
order to understand the specific mechanisms that allow parents to effectively balance the
control of their child's behavior with the support of their increasing independence and
strivings for control of their own behavior, we must first examine the various models of

parenting that have been proposed.

22
Dimensions of parenting

Symonds (193 9) first argued that the early socialization process largely is
determined by the multiplicative effects of two dimensions of parenting: acceptance-
rejection and dominance-submission. He argued that the two dimensional space created
by these dimensions could describe most styles of parenting. Other researchers have since
proposed similar dimensions and/or models of parenting.

Schaefer's (1959) circumplex model of maternal behavior was based on the
dimensions of autonomy versus control and love versus hostility. Freedom granting,
detachment, uninvolvement, and democratic behavior were placed toward one end of the
autonomy versus control axis, while possessiveness, authoritarianism, and
overprotectiveness were located toward the control end of the continuum. Along the
love versus hostility axis, acceptance, indulgence, and cooperation reflected maternal love,
whereas rejection, neglect, and antagonistic demands indicated hostility.

A three-dimensional model containing an affective dimension (warmth-hostility), a
control dimension (permissiveness-restrictiveness) and an intensity dimension (anxious
involvement-calm detachment) was described by Becker (1964). The affective dimension
was anchored on the warmth end by such behaviors as affection, approval, fiequent
explanations, positive responses to dependency behavior, use of reasoning, and praise,
while the hostility end of the continuum was defined by the opposite behaviors.
Restrictiveness included placing many restrictions on the child and strictly enforcing
demands in most areas of daily functioning, whereas permissiveness was defined by few
restrictions and lax enforcement of demands. The third dimension, anxious emotional

involvement versus calm-detachment was described on the anxious emotional involvement

23
end by high emotionality directed toward the child, overprotectiveness, babying, and

concern for the child's welfare, while the calm-detachment end was defined by the
opposite behaviors.

Subsequent work by Schaefer (1965) yielded a three-dimensional model of
parental behavior that was based on children's reports of parent behavior. The three
dimensions that emerged were acceptance versus rejection (formerly love versus hostility),
psychological control versus psychological autonomy, and firm control versus lax control.
The variables that defined the positive end of the acceptance versus rejection dimension
included positive evaluation, sharing, expression of affection, and equalitarian treatment.
The rejection end was defined by ignoring, rejection, and neglect. Psychological control
was described by intrusiveness, possessiveness, protectiveness, parental direction, and
control through the use of guilt, whereas psychological autonomy was defined by extreme
autonomy. This dimension was assumed to measure "covert, psychological methods of
controlling the child's activities and behaviors that would not permit the child to develop
as an individual apart from the parent" (1965, p. 555.). The third dimension, firm control
versus lax control, was best defined by punishment and strictness on the firm control end,
and lax discipline or extreme autonomy on the lax control side of the dimension. This
third dimension was argued to measure the extent to which parents establish rules, set
limits, and enforce the rules and limits.

Similarly, Roe and Siegelman (1963) outlined three dimensions of parental
behavior: loving-rejecting, casual-demanding and overt attention. These three dimensions
were somewhat related to Schaefer's (1965) three dimensions of parental behavior,

although they were not conceptualized within a three-dimensional circumplex model.

24
Rather, Roe and Siegelman argued for the importance of each dimension, separately.

Finally, Schutz (1966) put forward a different three-dimensional model that
included (1) an affective dimension, (2) a dominance dimension (control), and (3) an
inclusion dimension. The affective dimension was represented on the positive end by
approval, affection, and acceptance, and on the negative end by discouragement,
disapproval, rejection, and blame. The positive end of the control dimension was
described by those parental behaviors which reflected the promotion of independence in
the absence of unreasonable control. The other end of the control dimension was
comprised of such behaviors as restrictiveness, demands for obedience, coercive
suggestions, and numerous rules for appropriate child behavior. Inclusion ranged from
child-centeredness, parental concern for the child, indulgence, and frequent and/or intense
contact on the positive end of the dimension to neglect, understimulation, ignoring, and
reduced interactions on the negative end.

Central to the various models of parenting that have been proposed are the two
dimensions of parental warmth-hostility and parental control. The dimension of parental
warmth-hostility generally refers to a cluster of behaviors that contain a strong affective
component, indicating either approval or disapproval of the child's behavior and the degree
of emotional responsiveness to the child. Examples of behaviors reflecting parental
warmth would include smiles, praise, approval, encouragement, physical affection, and
responsiveness to the child's needs (Straus & Tallman, 1971). Hostility would be
manifested by such behaviors as criticism, sarcasm, disapproval, and harsh punishment.
The dimension of parental warmth-hostility is operationally defined as the summation of

the frequencies of the above behaviors, with behaviors reflecting parental hostility coded

25
in the negative direction (Rollins & Thomas, 1979).

The second dimension of parenting behavior, parental control, often refers to those
behaviors which reflect the parent's efforts to influence or otherwise exert control over the
behavior of their children. In one review of the parenting literature, Maccoby and Martin
(1983) identified two major types of control attempts that have been delineated: power-
assertive discipline and "love-oriented" discipline. Power-assertive control techniques
included physical punishment, yelling, shouting, fOI'CCfiJl commands, and threats, many
which were accompanied by strong negative affect or hostility. On the other hand, love-
oriented control behaviors consisted of showing disappointment, isolation, withdrawal of
love, praise, reasoning, and the contingent provision of affection. The most problematic
aspect of Maccoby and Martin's (1983) model is that it does not effectively address which
category would encompass the set of clearly stated parental directives that do not contain
a strong negative affect component. Rather, this category of clear parental directives, that
is well espoused within the parent-training literature (e. g., Eyberg & Robinson, 1982;
Forehand & McMahon, 1981; Hanf & Kling, 1973; Patterson, 1975), appears to have
been combined with those parental control techniques that contain a significant amount of
negative affect, such as hostility.

In another review article, Steinmetz (1979) divided parental control attempts into
three distinct types: love-oriented positive, love-oriented negative, and power-assertive
techniques. The love-oriented positive behaviors consisted of induction, praise, reasoning,
compromise, and mediation. Expressing disappointment, shaming, ridicule, isolating the
child, and the withdrawal of love fell in the love-oriented negative category. The last type

of parental control, power-assertive techniques, included yelling, shouting, physical

26
punishment, threats, deprivation of privileges, forcefiil commands, and tangible rewards.

As was seen with the Maccoby and Martin (1983) conceptualization of parental control
techniques, Steinmetz (1979) also appears to have allowed the dimension of parental
control to be severely confounded by the separate domain of warmth-hostility.

Finally, Rollins and Thomas (1979) concluded that the different types of parental
control attempts that have been investigated could best be separated into the categories of
induction, coercion, and love withdrawal. Inductive control attempts included such
behaviors as reasoning, which were used in an effort to influence the behavior of the child
without engaging in a confrontation. On the other hand, coercive parental behaviors
included threats, utilization of force, physical punishment, and withdrawal of material
objects or privileges. The last category, love withdrawal, consisted of parental rejection,
ignoring, isolating, disapproving, or other similar behaviors that would indicate
disappointment with the child's behavior. Both Rollins and Thomas (1979) and Steinmetz
(1979) appear to infer that the qualities of negotiation and/or avoidance of conflict are
essential aspects of those parental control techniques that fall within the "love-oriented
positive" or "inductive" control categories. As was also seen with Maccoby & Martin's
model (1983) and Steinmetz's model (1979), clearly stated parental directives (without a
strong negative affective component) do not appear to be accounted for in Rollins and
Thomas's (1979) description of types of parental control behaviors.

Despite the similarities in the way parental control attempts have been
conceptualized across studies, there are a variety of differences that have been somewhat
problematic, such as the confounding of different types of control attempts with the

dimension of warmth/hostility and the failure to differentiate between positive and

27
negative control techniques. A fiirther detailed discussion of the problems that have been

encountered in the definition of the parental control dimension will be presented later.

First, I will turn to research examining the joint effects of parental warmth and control.

Resegch examining dimensions of parenting

 

It was the influential work of Diana Baumrind (1967, 1968, 1971, 1988) that
firmly established the importance of the two dimensions of parental warmth and control.
Baumrind (1971, 1988) was interested in determining the parental antecedents of
competence in preschool-aged children. In her research, she found that parents who
displayed a pattern of firm control in combination with high parental responsiveness or
warmth tended to have children who were securely attached, altruistic, independent,
obedient, and high in self-esteem. This pattern of parental behavior has subsequently been
labeled "authoritative parenting" (Baumrind, 1967, 1988).

The authoritative parent was described as a parent who attempts "to direct the
child's activities but in a rational, issue-oriented manner. She (sic) encourages verbal give
and take, and shares with the child the reasoning behind her policy. She values both
expressive and instrumental attributes, both autonomous self-will and disciplined
conformity. Therefore, she exerts firm control at points of parent-child divergence, but
does not hem the child in with restrictions. She recognizes her own special rights as an
adult, but also the child's individual interests and special ways. The authoritative parent
affirms the child's present qualities, but also sets standards for fiiture conduct. She uses
reason as well as power to achieve her objectives. She does not base her decisions on
group consensus or the individual child's desires; but also does not regard herself as

infallible or divinely inspired" (Baumrind, 1968, p.261).

28
In contrast to the authoritative style of parenting, two other styles, "authoritarian"

and "permissive" parenting, were not as strongly related to the development of
competence in preschool age children (Baumrind, 1971). Authoritarian parents were
described as more demanding, controlling, rigid, and punitive, as well as less accepting of,
and responsive to, the needs and interests of the child, than were authoritative parents. On
the other extreme, permissive parents were unable or unwilling to provide the necessary
structure, make appropriate demands, or communicate the expectancies for mature
behavior on the part of the child. Thus, Baumrind (1983, 1988) argued that one of the
most effective styles of parenting balanced the need for firm control with the provision of
warmth and support to encourage the development of the child's independence and self-
regulatory behavior.

Subsequent studies have generally supported Baumrind's findings. Stayton,
Hogan, & Ainsworth (1971) found that early maternal responsiveness and sensitivity to
children's needs were inversely related to the subsequent emergence of such behaviors as
crying and/or noncompliance to maternal commands. It was also found that neither the
frequency of verbal commands nor the frequency of discipline-oriented physical
interventions, by themselves, were correlated with child compliance. These findings
suggest that a balance between parental warmth and control is a more important
determinant of child compliance than either one alone.

Lytton (1979) showed that when parental commands were combined with positive
behaviors, such as smiling or praise, the effectiveness in securing compliance was greatly
increased. On the other hand, commands that were combined with physical control or

other aversive parental behavior (e. g., criticism), reduced the effectiveness in obtaining

29
compliance.

In another study, Schaffer & Crook (1980) found that a mother's responsiveness
and sensitivity to her child's immediate attentional state significantly influenced whether or
not her requests were successfirl in obtaining compliance. Similarly, the degree to which
parents were able to follow their child's attentional state and synchronize their attempts to
direct or control the child's behavior predicted subsequent compliance to parental control
attempts (Rocisano, Slade, & Lynch, 1987).

More recent studies have provided additional evidence supporting the importance
of the role that both parental warmth and control play in early childhood socialization
(e.g., Lay, Waters & Park, 1989; Lemanek, Stone & Fishel, 1993; Rothbaum, Rosen, Pott
& Beatty, 1995; Wakschlag & Hans, 1997). For example, Lay, Waters & Park (1989)
found that maternal responsiveness was associated with increased child compliance in a
sample of four year old children. In another study, Lemanek, Stone & Fishel (1993) found
that preschool age children’s compliance was associated with parents use of both
instructions and cues, as well as reinforcement, whereas noncompliance was associated
with the parent’s use of structure and verbal/nonverbal attention-getting behaviors, in the
absence of reinforcement behaviors. In yet another recent study, Wakschlag & Hans
(1997) found a link between low maternal responsiveness in infancy and the subsequent
development of disruptive behavior disorders in a sample of high risk children prenatally
exposed to opioids.

In summarizing the research on the dimensions of parenting, Martin (1987)
concluded that "the processes involved in warmth and control are most profitably viewed

in interactive terms. Harmonious interactions between parent and child are more likely to

30
result when parental control efforts are sensitive to and moderated by the infant's or child's

state, in other words, when accompanied by many of the qualities summarized as warmth"
(1987,p.183)

Thus, it is the clearly communicated control attempts that explicate the parental
wishes for children’s behavior. Then, through repeated compliance to parental directives,
the child is able to observe him/herself behaving in accordance to parental standards, and
eventually intemalizes these standards of appropriate behavior as their own. However,
both the immediate attainment of compliance and the subsequent internalization of such
standards for appropriate behavior are maximized when the control attempts occur within
the context of a responsive, supportive, and nurturant relationship (Baumrind, 1983, 1988;
Maccoby, 1983, 1992; Maccoby & Martin, 1983).

When such a balance of warmth and control occurs it suggests that the child is
respected as an individual and the parent actively seeks to establish a legitimate base of
power without compromising the autonomous development of the child. Within the
context of the comfort and safety experienced in the presence of the parent, the child is
subsequently able to freely explore and experience new aspects of his/her environment,
including other social relationships.

Both Maccoby (1983, 1992; Maccoby & Martin, 1983) and Kochanska (1997)
have argued that the socialization process involves inducting the child into a reciprocal
orientation or system of reciprocity, in which the development of a mutually responsive
relationship between parent and child, over time, results in a decreased need for parental
control behaviors. Thus, it is through the development of this mutually responsive

relationship that children gradually internalize their parent’s goals and values, thereby

3 l
reducing the need for parental coercion or control.

Limitations of Socialization Literature

Parental Control: One of the major limitations of the socialization literature has
been the various ways that parental control attempts have been defined, particularly
"coercive" control attempts. Too often, researchers have combined ineffective negative
control techniques, such as punishment, yelling, threats, vague directives, and deprivation
of privileges, together with more effective power-assertive control behaviors, namely,
clearly stated commands and directives (see Maccoby & Martin, 1983, Rollins & Thomas,
1979, Steinmetz, 1979). However, there is an important distinction between these two
types of control attempts (Darling & Steinberg, 1993; Lee & Bates, 1985; Smetana,
1994)

The first group of behaviors often reflect parental punitiveness, coerciveness, and
hostility and can be viewed as "firnctionally superfluous" control methods (Lepper, 1981;
Maccoby & Martin, 1983). On the other hand, parental commands and directives
represent consistent discipline techniques often used to effectively influence children's
behavior (e. g., Forehand & McMahon, 1981; Patterson, 1982; Smetana, 1994). Although
the category of parental commands often ranges from subtle suggestions to forcefiil
directives, it nevertheless remains important to distinguish between clear parental
directives and the more ineffective, punitive, or hostile control attempts.

Within the socialization literature, few researchers (e. g., Baumrind, 1971, 1988;
Darling & Steinberg, 1993; Lee & Bates, 1985; Maccoby & Martin, 1983; Smetana, 1994)
appear to have made the clear distinction between more positive and more negative types

of parental control techniques. For example, Lee & Bates (1985) separated parental

32
control into a "positive control" factor and a "negative control" factor. The positive

control factor included suggesting, giving choices, and giving incentives for compliance,
whereas the negative control factor consisted of prohibiting, scolding, repeated
prohibiting, physical punishment, removing or restraining the child, and removing objects.

Upon examining those parental control attempts that typically have been
considered inductive control methods, it is clear that these statements often consist of a
clearly stated directive followed by explanations, reasons, or other statements that are
assumed to increase the probability of both compliance and subsequent internalization of
the particular parental standard. Therefore, it appears more usefiil to conceptualize
directive control attempts as falling within the dimension of inductive or "positive"
parental behaviors.

Baumrind (1971, 1988), in defining the dimension of "firm control", clearly
distinguished between appropriate or positive control techniques and negative control
behaviors. Firm control was defined as "firm enforcement of rules, effective resistance to
the child's coercive demands, and willingness to guide the child by regime and structured
interventions. It does not imply many rules or intrusive directiveness of the child's
activities" (Baumrind, 1971, p.98). Thus, firm control was distinguished fiom other
parental control attempts, such as punitiveness or unqualified power assertion, that might
be used to overprotect or over-restrict the child's behavior. In fact, Baumrind (1983,
1988) later contended that although authoritative parents were high in firm control they
were also low in the use of negative control techniques, such as unqualified power

assertion and punitiveness.

33
The importance of distinguishing between "positive" control techniques and those

punitive or coercive techniques which are less effective in obtaining compliance is firrther
highlighted by Bell & Harper's (1977) theory of upper and lower limit controls and
Lepper's (1981) notion of the "minimally sufiicient pressure". According to Bell &
Harper's (1977) homeostatic control theory of interaction, parents will attempt to maintain
a certain level of contact and control within a given interactional situation with their child.
If the level of interactive behavior moves outside the permissible range, either too high or
too low, the parent then will employ upper- or lower-limit control techniques to restore
the interaction to more desirable levels.

Lower—limit controls are used to increase the level of the child's interactive
behavior. For example, if a child is withdrawn, inactive, inhibited, or otherwise
performing at a level below what previously has been established as a baseline, a parent
likely would attempt to stimulate the child's behavior through the use of praise, requests,
suggestions, drawing attention to stimuli, helping, or other similar behaviors. Conversely,
upper-limit controls such as disapproval, distraction, prohibitions, and physical control or
punishment are utilized to reduce the amount of inappropriate or excessive behavior.

Extrapolating from Bell and Harper's (1977) theory, an argument can be made that
the group of upper-limit control behaviors more accurately reflects the coerciveness and
punitiveness typically attributed to the "functionally superfluous" categories of parental
control techniques, which also is consistent with Patterson's (1982) theory of coercive
family processes. Although many parents will inevitability use upper-limit control
behaviors, depending upon the demands of the situation, those parents who are less skilled

in using more directive control attempts would be expected to resort to upper-limit

34
control methods more fiequently.

In spite of the general utility of a homeostatic theory for explaining parental
control efforts with children's behavior that clearly falls outside of the range of normal
interactive behavior, it does not adequately address interactions that fall at different points
within the acceptable range of normal interactive behavior. For example, the theory does
not explain why a parent might allow their child to engage in mildly disruptive behavior in
one setting, yet place some minor restrictions on the child's behavior in another setting,
particularly if the only difference between the two situations was the type of setting and
not whether the child's behavior actually exceeded an objective threshold of parental
tolerance. Therefore, it would be usefirl to examine the effectiveness of other parental
control attempts that are utilized when the child's behavior falls within the range of
acceptable behavior.

Lepper (1981) suggested that the most effective socialization (internalization of
parental standards) occurs when parents utilize the least amount of control necessary to
obtain compliance. Therefore, within the range of acceptable child behavior, the most
effective parents would be expected to utilize more subtle control methods, such as
induction, reasoning, and clear commands. This position also is consistent with the more
recently advanced notion, mentioned earlier, that the socialization process involves
inducting the child into a reciprocal orientation or system of reciprocity, in which the
development of a mutually responsive relationship between parent and child over time
results in a decreased need for parental control behaviors (Kochanska, 1997; Maccoby,

1983, 1992; Maccoby & Martin, 1983).

35
In summary, one of the major limitations of the socialization literature has been the

failure to differentiate between punitive or coercive parental control attempts and more
positive or inductive control attempts. "Negative" parental control attempts are best
described by behaviors such as vague commands, no-opportunity commands, threats,
yelling, shouting, prohibitions, and physical restraint or punishment.

The group of parental behaviors that reflects high "positive" control includes
making clear demands or requests of the child, asking directive questions, and setting clear
and consistent contingencies for the child's behavior. On the other hand, low positive
parental control or permissiveness would be represented by few parental directives and lax
or inconsistent enforcement of rules.

Developmental Variations: A second limitation of the socialization literature has
been the tendency for researchers to ignore important developmental variations and
expectations with respect to child and parental behavior. First, many reviews of the
literature have attempted to generalize findings across a wide range of ages (Maccoby &
Martin, 1983; Rollins & Thomas, 1979; Steinmetz, 1979). Second, the conclusions that
have been reached by these reviewers typically have assumed that the especially efi‘ective
parenting conditions at one age will be equally effective at many other ages. Until
recently, the distinction that has received the most attention has been the differences in
parenting for young children and adolescents. Few researchers have examined parenting
differences that take into account developmental difi‘erences occurring within the
preschool years alone.

Only recently have researchers attempted to take a more developmental

perspective when investigating parental socialization efforts with preschool-aged children

36
(e.g., Baumrind, 1983, 1988; Dubin & Dubin, 1963; Kochanska, 1997; Kochanska et al.,

1987; Kuczynski, Kochanska, Radke-Yarrow, & Girnius-Brown, 1987; Lay, Waters &
Park, 1989; Lemanek, Stone & Fishel, 1993; Maccoby & Martin, 1983; Rothbaum,
Rosen, Pott & Beatty, 1995; Smetana, 1994; Vaughn, Kopp, & Krakow, 1984; Waschlag
& Hans, 1997). The most important developmental consideration is that younger children
necessarily require more control and directiveness than older children (Kochanska, 1997;
Kuczynski et al., 1987). As noted before, parents must initially provide a sufiicient
amount of structure to allow young children to experience and develop an enduring
perception of the range of parentally acceptable behaviors. Once a foundation has been
established, parents then are able to gradually relax the amount of control that is necessary
to regulate their child's behavior (Kochanska, 1997; Maccoby, 1983, 1992; Maccoby &
Martin, 1983). However, it remains that "parents typically confront children with rules,
sanctions, and norms long before the children are capable of cognitively appraising their
legitimacy or validity." (Baumrind, 1983, p.139).

Over time, as children begin to internalize parental standards of appropriate
behavior and regulate their own behavior, parents develop greater expectations for
compliant behavior (Kochanska, 1997 ; Kopp, 1982). As expected, increases in
compliance to maternal control attempts have been found to be age-related, even among
children between one to four years of age (Kopp, 1982; Schafi‘er & Crook, 1980;
Kochanska, 1997; Kochanska et al., 1987; Maccoby, 1992; Vaughn et al., 1984).
Similarly, parental control strategies have been shown to be age related, as well. In one
study examining developmental differences in maternal control strategies among children

between 15 and 44 months of age, parents of the older children were observed to utilize

37
fewer physical control methods and more indirect verbal strategies than did the parents of

the younger children, including more maternal explanations, bargaining, and reprimands
(Kuczynski et al., 1987). It was suggested by the researchers that the parents were
attempting to match their control techniques to the developmental level of the child, with
more direct control methods being necessary for younger children.

If the shift from direct to indirect control methods occurs at too early of an age,
negative consequences may occur. For example, Hatfield, Ferguson, and Alpert (1967)
and Yarrow, Campbell, and Burton, (1968) both found that the use of indirect control
techniques such as reasoning were positively associated with the aggressive behavior of
preschool-aged boys in the home. Thus, parents need to carefirlly match their parenting
style with the child's developmental level, being carefirl not to provide the child with too
much latitude too quickly. When the child is young (under three years of age), parents
necessarily tend to utilize more direct control attempts to clearly establish the boundaries
of acceptable behavior for the child, and lay a foundation for the development of the
child's self-regulatory mechanisms. Later, around the age of three to four years, as the
child gradually intemalizes the parental standards of behavior and develops greater self-
regulatory behavior, the parent can begin to adopt a less directive stance toward
controlling the child (Kochanska, 1997; Maccoby, 1983, 1992; Maccoby & Martin, 1983;
Smetana, 1994). Therefore, within the present study, it was expected that parents, in
general, would exhibit an even balance between direct and indirect control behaviors, as
appropriate for children between three to four years of age.

Nonclinical Populations: A final limitation of the socialization literature has been

the primary focus on nonclinical populations. Much of the developmental research has

38
been directed towards the examination of the socialization process, under normal

circumstances. However, there are certain stressors, such as childhood illness, which may
impinge upon this socialization process by disrupting the normal maturation and/or
attainment of important developmental milestones (Cerreto, 1986). Furthermore, such a
disruption to the child's socialization most likely would occur if the stressor directly
impacts upon the child's temperament, indirectly influences the child's interactions with
their parent(s), and/or indirectly has an impact upon the parent's perceptions, attitudes,
and/or behavior directed towards the child . Therefore, it is essential to consider the
temperamental state or clinical status of the child, as well as the potential immediate and
long-term effects these conditions may have on the child's behavior, the parent's behavior,
and the parent-child interactional system.

The concept of temperament typically has been defined as a constitutionally
determined behavioral style that is somewhat stable over time (Goldsmith & Campos,
1982; Rothbart & Derryberry, 1981). One of the most widely cited theories of infant
temperament originally was proposed by Thomas, Chess, Birch, Hertzig, & Korn (1963).
They argued that there are nine separate dimensions that constitute the notion of
temperament. The dimensions are: activity level, approach/withdrawl, regularity,
adaptibility, threshold, intensity, mood, distractibility, and attention span. From these nine
dimensions, Thomas et al. (1963) derived three distinct typological classifications of infant
temperament: "easy", "difficult", and "slow-to-warm-up". Other researchers since have

proposed similar theories of infant temperament (e. g., Buss & Plomin, 1975, Rothbart &

Derryberry, 1981; Brazelton, 1973).

39
Bates (1980) concluded that the most essential qualitative component of a

"difiicult temperament" was the parent's perception of the infant's fiequency and intensity
of negative affect. Similarly, it has been argued that, over an extended period of time, a
child with a "diflicult" temperament gradually may become an aversive stimulus for a
parent, through the process of conditioning (Lamb, 197 8). Such conditioning,
subsequently might lead the parent to become less warm, responsive, and sensitive to the
child's needs, thereby placing the child in a more vulnerable position (Donovan, Leavitt, &
Balling, 1978). This process might also be expected to result in the emergence of coercive
parent-child interactions and/or child behavioral difficulties (Patterson, 1982).

With respect to its impact on styles of parenting, maternal reports of early difficult
temperament (age 6-13 months) has been shown to be positively correlated with later
observations of resistance to maternal control attempts and parental use of power
assertion at 24 months (Lee & Bates, 1985). Children, rated by their mothers as
"temperamentally difficult" when they were between 6 and 13 months of age, were more
likely to approach breaking an established rule or damage something within the home at
24 months of age, than were "temperamentally easy" children. In response to their
children's behavior, mothers of the "difficult" children, compared to mothers of "easy"
children, were more likely to use intrusive control tactics, which were resisted more often
by the "difficult" children. These findings provide evidence that a child with a diffith
temperament in infancy may elicit negative parental behavior and may also be at risk for
the subsequent development of interactional difficulties and/or behavioral problems.

In another study, maternal reports of "diflicult" temperaments in their children

were positively related to both maternally reported behavior problems and observed

40
negative interactions between mothers and their three to five year-old children (W ebster-

Stratton & Eyberg, 1982). More specifically, the children described by their mothers as
active and inattentive were observed to respond to their mothers in a more negative,
nonaccepting, and noncompliant manner than normal children responded to their mothers.
Similarly, mothers who rated their children as temperamentally "difficult", when compared
to mothers who rated their children as "temperamentally compatible", were more negative
in affect, nonacceptance, and submissiveness. However, the findings fiom this study
must be interpreted cautiously as the families had been recruited to participate in a parent
training program designed to help parents manage child misbehavior, which may have
influenced the mother's response biases as well as their interactions with their children.
Additionally, the ratings of temperament and behavior problems were both based on
maternal reports which only were taken at the time of the study, rather than at an earlier
age. Nevertheless, these results are consistent with other studies reviewed, indicating that
children with difficult temperaments may exhibit more behavior problems, respond more
negatively and noncompliantly towards their mothers, and may elicit more negative
behaviors fi'om their mothers, than normal children.

The powerfirl effects that a child's characteristics and behavior can have on parent's
behavior is further illustrated by the work of Humphries, Kinsboume, & Swanson (1978)
and Barkley & Cunningham (1979) with hyperactive children and their mothers. Using a
double-blind, drug-placebo crossover design, Humphries, Kinsboume, & Swanson (1978)
examined the effects of methylphenidate on mother-child interactions during a structured
task situation in a clinic playroom setting. In the medication condition, compared to the

placebo condition, the hyperactive children exhibited increases in their verbal directions

41
given to their mothers and decreases in the amount of critical or negative statements.

Similarly, the mothers showed decreases in their directiveness, decreases in their
criticisms, and increases in their praise in the medication condition, compared to the
placebo condition.

Barkley and Cunningham (1979) used a similar triple-blind, drug-placebo
crossover design to examine changes in mother-child interactions, attributable to
methylphenidate administration. When the hyperactive children were given
methylphenidate, there were significant increases in the children's sustained attention and
compliance to maternal commands and a decrease in the amount of off-task behavior,
when compared to interactions during a placebo condition. In response, the mothers
displayed a significant decrease in the number of commands and negative responses, and
an increase in the amount of attention, praise, and other positive responses to the
children's increased compliance during the medication trials. The results of both studies
indicate that when hyperactive children were receiving medication, their behavior was
perceived by their mothers as more acceptable and thus did not require as much control as
when they were given a placebo. Additionally, the mothers of hyperactive children were
able to increase the amount of contingent postive responding when the children displayed
more appropriate behavior, following the administration of the methylphenidate. Overall,
these findings indicate that children can have a significant negative impact upon the quality
of parent-child interactions, by exhibiting deviant and noncompliant behavior, as well as by
eliciting certain types of behavior from their parents, but that this impact can be changed

through interventions with the child.

42
Sameroff and Chandler's (197 5) transactional model of development provides

another usefirl way of understanding how children's temperamental states can influence
expressed styles of parenting, as well as how parenting behaviors can have a significant
impact upon children's behavioral adjustment and their overall emotional development.
According to their model, developmental outcomes are determined by the continuous
dynamic interaction between child characteristics and environmental context (parenting),
over time. That is, at any given point in time, a child's developmental competency is a
result of the interaction between the effect of the child on the parents and the subsequent
effect of the parents on the child. The previously reviewed literature on the effects of
children's temperament upon parent's behavior illustrated the powerful direction of
influence from children to their parents. On the other hand, the socialization literature that
was reviewed (and the literature on specific clinical populations that will be reviewed
later) revealed the means by which parent's behaviors have a direct impact upon children's
behavioral and emotional adjustment.

Utilizing Sameroff and Chandler's (1975) transactional model, one can appreciate
the potential deleterious, synergistic effects that might occur when a child with a diflicult
temperament is exposed to an ineffective style of parenting, especially over time. The
resulting cycle of dysfimctional interactions is similar to Patterson's model of coercive
family interactions (Patterson, 1982). According to Patterson's theory of coercive family
interaction, once a pattern of dysfunctional interactions is initiated the likelihood of further
escalation is greatly increased due to the creation of a self-reinforcing negative feedback
loop of dysfunctional behaviors. Thus, it becomes clear that a difficult temperament

and/or an ineffective style of parenting can independently contribute to the development of

43
child behavioral difficulties and dysfirnctional parent-child interactions, over an extended

period of time. However, if both situations exist, there will be a significantly greater risk
for the development of child behavioral difficulties and dysfunctional parent-child
interactions, because the negative effect of a child's diflicult temperament will tend to
interact synergistically with the negative effect of an ineffective style of parenting.

In summary, the major limitations of the socialization literature have included (1)
combining positive parental control behavior with more punitive or coercive control
techniques to form a broad dimension of parental control, which then may be confounded
with the dimension of parental warmth/hostility, (2) ignoring important developmental
variations when considering early childhood socialization, and (3) the relative lack of
attention directed towards examining the potential effects of a difficult temperament on
parent behavior and parent-child interactions. Given these limitations, it would be USCfUl
to examine the clinical literature to explore the potential effects that child stressors may
place upon the parent-child relationship.

Resaarch on Clinical Populations. Various populations of clinic-referred children
and their parents have been studied to better understand how specific child behavior elicits
various parenting behaviors, and vice versa. Several of the more extensively studied
populations include conduct-disordered, hyperactive, and abused and/or neglected
children. Observations of these children's interactions with their parents have consistently
revealed a pattern of dysfiinctional parent-child interactions, which can be extended to
other groups of children with related characteristics.

Conduct-disordered children often have been described as displaying high rates of

aggressive, destructive, disobedient, manipulative, and other aversive behaviors (e.g.,

44
Bemal, Duryee, Pruett, & Burns, 1968; O'Leary, O'Leary, & Becker, 1967; Patterson,

1974; Peed, Forehand, & Roberts, 1977). Estimates indicate that conduct-disordered
children comprise 1/3 to 1/2 of the total number of children referred for child behavior
problems (Patterson, Reid, Jones, & Conger, 1975). Thus, this group of children
represents a substantial portion of the children treated in clinics each year.

In families with antisocial children, studies generally have found a greater
frequency of aversive interactions by both parents and children (e. g., Delfini, Bemal, &
Rosen, 1976; Johnson & Lobitz, 1974; Lobitz & Johnson, 1975; Patterson, 1976; Snyder,
1977 ; Wahler, Hughey, & Gordon, 1981). In one study, clinic-referred children displayed
more deviant behavior, while their parents displayed more commands and negative
behaviors, as compared to normal children and their parents (Lobitz & Johnson, 1975).
Other studies also have shown that conduct-disordered children are more deviant and
noncompliant than normal children (Delfini, Bemal, & Rosen, 1976; Forehand, King,
Reid, & Yoder, 1975; Griest, Forehand, McMahon, & Wells, 1980). Similarly, parents of
children with conduct disorders generally are more controlling, negative, and critical than
parents of normal children (Forehand, King, Peed, & Yoder, 1975; Green, Forehand, &
McMahon, 1979; Terdel, Jackson, & Garner, 1976).

Within the hyperactive population, mothers have been observed to be more critical,
directive, and controlling, yet less positive than mothers of normal children (Campbell,
1973; Campbell, 1975; Cohen, Sullivan, Minde, Novak, & Keens, 1983; Cunningham &
Barkley, 1979). In one study, the mothers of hyperactive children were more negative,
more directive, asked fewer questions, gave less praise, and initiated fewer interactions

during a structured task situation, than mothers of control children. At the same time, the

45
hyperactive children themselves were less compliant and responsive to parental behaviors

(Mash & Johnston, 1982). Similarly, Cunningham & Barkley (1979) found a greater
number of parental commands and negative responses, and fewer positive responses to the
social interactions, quiet play, and cooperative behavior of hyperactive children.

Within the third illustrative population of children, a sample of abusive and
neglectfirl mothers were shown to emit fewer positive behaviors (approval and support)
and more control attempts, disapproval, and rejection, as compared to non-abusive
mothers (Burgess & Conger, 1978). The abused and neglected children showed
comparably higher rates of aversive behaviors than well-reared children. Other
researchers have found similar levels of aversive behaviors in abusive families (Boshua &
Twentyman, 1984; Reid, Taplin, & Lorber, 1981).

In summary, the findings from observational studies of conduct-disordered,
hyperactive, and abused or neglected children and their parents, have yielded a general
pattern of dysfunctional interactions. For the clinic-referred children, the pattern that
emerged consisted of greater levels of aversive or deviant behavior and greater
noncompliance to parental requests, as compared to normal children. Similarly, the
parents of these children emitted more controlling, negative, and critical behaviors, as well
as fewer positive or supportive behaviors, than did the parents of the comparison children.

In comparison to the pattern of effective parenting that has been outlined within
the socialization literature, the parents examined above were much higher on the negative
control dimension and much lower on the dimension of warmth or responsiveness, relative
to parents of normal children. The extreme level of negative control attempts likely

reflects control strategies similar to Bell & Harper's (1977) upper-limit controls. As the

46
children's behavior moved beyond the limits of the range of parentally acceptable behavior,

upper-limit control behaviors were utilized by the parents to restore the interaction to
more desirable levels.

At the same time, the parents of the clinic-referred children displayed less warmth
or responsiveness, which likely resulted from an extended history of aversive interactions
with the child. Over time, the child may have become an aversive stimulus for the parent,
leading the parent to become less responsive to the child's interactions and needs
(Donovan, Leavitt, & Balling, 1978; Lamb, 1978).

These dysfiinctional relationships between parent and child behavior were firrther
exacerbated by increases in the structure or situational demands of the observed
interaction tasks, at least in some of the studies reviewed above (e. g., Campbell, 1975;
Cohen et a1. 1983; Cunningham & Barkley, 1979; Mash & Johnston; Robinson & Eyberg,
1981; Webster-Stratton, 1985). For example, Cohen et al. (1983) found a significant
increase in the overall frequency of negative maternal interactions with their hyperactive
children in a difficult task, as compared to an easier task situation. The observed pattern
of more suggestions, prohibitions, attentional commands, and disapproving comments
during an easy task was further elevated in the difficult task situation, when compared to
the interactions of mothers of non-clinic referred children.

Other researchers have found corresponding increases in child deviant behavior
and noncompliance to parental requests, as well as increases in parental control attempts
and critical behavior in structured task situations, when compared to free-play situations
(Robinson & Eyberg, 1981, Webster-Stratton, 1985). Overall, these findings indicate that

parent-child interactions have the potential to become more dysfirnctional as the structure

47
and situational demands increase, particularly if the parents are not able to balance their

control attempts with warmth or responsiveness. With increases in structure or situational
demands parents can become more controlling and critical, as well as less responsive.
Likewise, the children can become more noncompliant and display more deviant behavior.
These findings are particularly applicable to conduct-disordered, hyperactive, and abused
or neglected children and their parents, yet also may apply to other populations as well.

In summary, a review of the socialization literature emphasizes that effective
parenting reflects a balance between parental warmth or responsiveness and positive
control. It was also argued earlier that the multiplicative effects of parental warmth and
positive control are more important than either dimension alone, or the additive effects of
each dimension. Furthermore, the remaining developmental literature that was reviewed
underscored the importance of establishing a match between the style of parenting and the
child's temperament or developmental level.

In the review of the clinical literature it was shown that there are situations where
the normal socialization process has gone awry, resulting in dysfiJnctional parent-child
interactions. Childhood illnesses are another example of how stress can directly and/or
indirectly influence both a child's temperament and a parent's style of interacting with their
child. As noted earlier, the research on ROM illustrated how one particular childhood
illness may have such a negative impact upon parent-child interactions. Yet, this particular
area of research also has several limitations.

Limitation; of current research on ROM and its relationship to the current study.
One of the major limitations of the early studies on ROM was the noticeable absence of

observational data on parent-child interactions. Most of the early studies relied almost

48
exclusively upon parental reports of parent-child interactions and the child's behavior to

evaluate the child's adjustment to the illness (Casey, 1983; Hersher, 197 8; Gottleib,
Zinkus, & Thompson, 1979). It is clear that reliance on potentially biased data places a
researcher in a precarious position with respect to the validity and generalizability of the
results (Atkeson & Forehand, 1978; Hetherington & Martin, 1979; Schnelle, 1974). The
importance of utilizing multiple assessment methods to evaluate both parent-child
interactions and the presence and severity of behavioral problems in families has been well
espoused in the literature (Aragona & Eyberg, 1981, Cantwell & Baker, 1971; Eyberg,
1985; Lobitz & Johnson, 1975; Lytton, 1974; Webster-Stratton & Eyberg, 1982).

Only recently have investigators begun to expand the scope of their studies to
include assessments of the relationship between recurrent otitis media and parent-child
interactions (Black et al., 1988; Black & Sonnenschein, 1993; Casey, 1983; Forgays,
Hasazi & Wasserman, 1992; Freeark et al., 1992; Roberts et al., 1995; Wallace, Gravel,
Schwartz & Ruben, 1996). However, even these more recent studies have mainly focused
on examining the relationship between otitis media and such outcomes as subsequent
hearing loss, language and cognitive outcomes. Few, if any of the more recent studies
have utilized actual observations of parent-child interactions to explore the potential
impact of otitis media on children’s behavioral adjustment.

Despite this limitation, the more recent studies nevertheless have provided support
for a multifactorial model of development within the context of exposure to early otitis
media. In combination with the other studies cited, the recent studies on otitis media
underscore the importance of including such family factors as parent interaction style,

among others in fixture studies examining the impact of otitis media on children’s

49
development. These recent studies also emphasize the likely transactional nature of the

relationship between early exposure to otitis media and the quality of parenting behaviors
and/or the home environment, as critical determinants of children’s subsequent cognitive,
language and socioemotional development.

Regarding the scope of observational data, the review of the socialization literature
emphasized the importance of examing both negative and positive parent-child
interactions. Too often, the focus of a study will concentrate only on the negative
interactions, while ignoring the potential contribution of positive interactions. This point
was highlighted earlier, where the multiplicative effects of the dimensions of parental
warmth and control have been demonstrated to be more important than the contribution
made by either dimension separately. Researchers only recently have begun to emphasize
the role of such positive parent-child interactions, since they are a means through which
both parents and their children learn to be mutually reinforcing, cooperative, and sensitive
to each other needs (Dowdney, Mrazek, Quinton, & Rutter, 1984; Gardner, 1987;
Kochanska, 1997; Maccoby, 1983, 1992; Maccoby & Martin, 1983).

A second limitation, that extends beyond the existing chronic illness literature, is
the surprising lack of attention to the contributions of fathers to overall family functioning.
Most available studies have focused almost exclusively on mothers. Those few studies
within the general chronic illness literature that have included fathers typically have found
both important similarities and differences in parent's self-reported data.

For example, a study by Tavormina, Boll, Dunn, Luscomb, and Taylor (1981)
found that both mothers and fathers of hearing-impaired children reported a significantly

greater number of behavioral problems and behavior management concerns, when

50
compared to parents of asthmatic, diabetic, and cystic fibrotic children. However, despite

the agreement on overall levels of reported behavior problems, mothers and fathers
difi‘ered in their specific patterns of reported problem behaviors. Mothers reported more
problems with aggression and overactivity for boys, whereas the fathers indicated more
overactivity and sleep disturbances. Thus, it is important to examine both parent's
contributions to the child's functioning and adjustment.

In order to better understand how a specific illness, such as ROM, might have an
impact upon certain aspects of family firnctioning and vice versa, several factors must be
considered. First, as described earlier, an acute episode of otitis media may predispose a
child to be quite irritable and fussy, primarily due to the pain and discomfort of the actual
ear infection (Bluestone et al., 1983; Forgays, Hasazi, & Wasserman, 1992; Roberts et al.,
1995). Second, the effusion associated with the illness often results in a temporary, yet
sometimes substantial, hearing loss (Bluestone et al., 1983; Daly, 1991; Downs, 1983;
Fria, Cantekin & Eichler, 1985; Roberts et al., 1995). It may be the combination of these
two factors that, over time, directly places the child at risk for exhibiting communication
difficulties or a language delay (Cass & Kaplan, 1979; Friel-Patti & Finitzo, 1990;
Gottleib, Zinkus, & Thompson, 1979; Lehmann, Charron, Kummer et al., 1979; Teele et
al., 1990; Ventry, 1980), behavioral problems (Casey, 1983; Forgays, Hasazi &
Wasserman, 1992; Gottleib, Zinkus, & Thompson, 1979; Roberts, Burchinal & Campbell,
1994; Vernon-Feagans, Manlove & Volling, 1996), and/or indirectly places the child at
risk for parent-child interactional problems (Black et al., 1988; Black & Sonnenschein,

1993; Casey, 1983; F orgays, Hasazi & Wasserman, 1992; Freeark et al., 1992).

5 l
A parent who has a child with fluctuating hearing losses, accompanying irritability,

and behavior problems may begin to perceive their child as "difficult" or temperamental.
This negative perception of the child then would be exacerbated if the parent inadvertently
interprets observed behavioral problems and/or noncompliance as intentional, rather than
primarily due to recurrent ear infections and associated sequelae. The perception of a
"difficult" child in the context of continued interactional difficulties also would be
expected to contribute to an increase in the parent's fiustration and a reduced sense of
competency as a parent over time. In such a situation, a parent might be more likely to
respond to their child in a less positive and more negative or controlling manner, firrther
exacerbating any existing difficulties that the child is experiencing in regulating his/her
behavior that may have initially originated as an indirect result of ROM.

However, as was previously noted in the review of the socialization literature, if a
parent is sensitive to their child's needs, able to convey warmth, and able to synchronize
their attempts to direct or control the child's behavior with the child's temperamental state,
then the likelihood of obtaining compliance and maximizing the child's behavioral and
emotional development should be significantly increased (Baumrind, 1983, 1988;
Kochanska, 1997; Kochanska, Kuczynski, Radke-Yarrow & Welsh, 1987; Maccoby,
1983, 1992; Martin, 1987; Rocisano, Slade, & Lynch, 1987; Schaffer & Crook, 1980).
The present study investigated whether more effective parenting behaviors moderate the
relationship between recurrent otitis media and children's behavioral problems, among

children experiencing varying early histories of ROM.

Study Hypotheses

The purpose of the present study was to assess relationships among ROM, child
behavior and parent-child interactions for a group of families with children who had
varying degrees of ROM during the first three years of life. It was anticipated that the
quality of parent-child interactions would play a role in moderating the deleterious effects
of ROM on children's behavioral functioning. Two alternative models of relationships
between ROM severity during the first three years of life, style of parent interactive
behavior (parental warmth and parental control), and children's behavioral adjustment at
three to four years of age were tested within the context of the present study (See Figure
1).

The first model, the "independent influence” model, postulates that ROM severity
and style of parent interactive behavior (parental warmth and control) independently
influence children's behavioral adjustment, with ROM having a direct and indirect effect on
adjustment. According to this model, recurrent otitis media in the first 3 years and
parenting independently contribute to children’s behavior so that children with severe
histories of ROM will be better adjusted than children with similar histories if they are
exposed to effective parenting behavior (both high parental warmth and positive parental
control), but would not be as adjusted as children who also are exposed to effective
parenting behavior but do not have early histories of ROM.

The second model, the "buffering model", argues that an effective style of parent
interactive behavior buffers the deleterious effects of ROM on child behavior. Thus, those
children with severe histories of ROM exposed to less effective parent interactive behavior

will exhibit poor behavioral adjustment. However, children with early histories of ROM
52

53
exposed to effective parent interactive behavior will be as well adjusted as children raised

in similar environments, but without severe histories of ROM.

The present study sought to determine whether style of parent interactive behavior
and ROM history were independently associated with child behavior problems, as
suggested in the “independent influence model” and/or whether style of parent interactive
behavior would moderate the relationship between ROM and child behavior problems, as
suggested by the “buffering model”.

Data on both child and parent behavior were collected during a behavioral
observation task that included three segments varying in structure. These tasks involved
child-directed play (the least structured), parent-directed play and parent initiated clean-up
(the most structured). In general, it was expected that ROM effects would be more
observable, and hence, there would be greater support for one or the other hypothesis as
the structure and situational demands of the observed parent-child interaction task
increased. Because more structured tasks are likely to elicit more negative behavior and
push for greater parental control, it was during these increasingly demanding tasks that the
deleterious effects of ROM were most likely to be observed. Thus, a linear relationship
was expected to exist across tasks, with more negative child and/or parent behavior
occurring in (1) the parent-directed play segment compared to the child-directed play
segment and (2) the clean-up segment compared to both the parent-directed play and
child-directed play segments.

In testing each hypothesis, I began by examining whether child sex moderated the
predicted relationships among the variables of interest. When there were significant

interactions between a predictor variable and child sex, the analyses were subsequently

54
performed and reported separately for boys and girls. Otherwise, the reported results are

based on analyses using data for males and females combined.
Hypothesis 1

As a test of the direct effects of ROM on behavior, postulated by the independent
influence hypothesis, the association between the severity of ROM and child behavior
problems was examined first. It was hypothesized that a severe history of ROM during
the first three years of life will be positively associated with a greater number of child
behavior problems (as observed during parent-child interactions in the laboratory), when
compared to children who had experienced fewer and less severe episodes of otitis media.
Hypothesis 2

As a test of the indirect effects of ROM severity on behavior, also postulated by
the independent influence model, it was predicted that a severe history of ROM would be
associated with observations of less effective parent-interactive behaviors (both lower
levels of parental warmth and lower levels of positive parental control). Similarly, it was
expected that less effective parenting behavior would be positively associated with a
greater number of child behavior problems during the observed parent-child interactions.
Hypothesis 3

Finally, as a test of the “buffering hypothesis”, it was expected that more effective
parenting behaviors, defined by greater parental warmth and positive parental control
behavior, would moderate the potentially deleterious effects of ROM on children's
behavioral problems. Thus, it was expected that high levels of parental warmth and
positive parental control would be associated with fewer child behavior problems during

the observed parent-child interactions.

Methods

Participants

The participants for the present study included 56 children (28 males and 28
females) between three and four years of age, and their parents. Child participants ranged
in age from 35 to 51 months (M=42 months, SD. = 4.9) and were identified from the
patient roster of a university-based pediatrics clinic. In order to study intact families and
avoid the potentially confounding influences of poverty or other major life stresses on
child development, families were asked to participate only if (1) parents were married and
living in the home; (2) both parents had at least a high school education; (3) a minimum of
one parent was employed on a fiill-time basis and the family was not receiving public
assistance; (4) the family included not more than four children; and (5) the target child did
not suffer from any other serious medical condition, disability or history of chronic illness
other than ROM. Sixty-eight percent of the families initially identified from the clinic
rosters who met the inclusion criteria were located, contacted by phone and agreed to
participate. Unfortunately, limited information was available both on the children’s illness
histories and the demographics of families who could not be located and/or those families
that refirsed participation once contacted. Eighty-six percent of the participating families
were White; and 82% of mothers and 91% of fathers had at least some college education.
Seventy-seven percent of the fathers were employed in lower middle to upper middle class
occupations or were full-time graduate students. Approximately 70% of the mothers were
employed and 36% were working at least 21 hours per week. Families were provided a

cash remuneration for their participation in the study.

55

56
Participants were not selected on the basis of the child’s otitis media history, but

once a family joined the study the child’s history of otitis media was determined. Medical
records were obtained from each medical facility in which the child had been evaluated
during his/her first 3 years of life. From those records both the number of separate
episodes of otitis media and the total number of days of effusion over the first 3 years
were determined. As recommended by Teele et al. (1984), effusion was estimated to have
lasted 29 days (per episode) when it was impossible to make a more precise determination
from information contained in the medical records. The total number of separate episodes
of otitis media during the children's first three years of life ranged from 0 to l 1, with a
mean of 4.2 (SD = 2.7). The estimated total number of days with effusion ranged fiom 0
to 480 days, with a mean of 133 (SD. = 100). Fifty-three percent (11 = 29) of the children
had suffered from “recurrent” otitis media (defined as three or more separate episodes
within 12 months; Teele et al., 1984, 1990; Vemon-Feagans et al., 1996) at least once
between 0 and 3 years of age. In addition, twenty-five percent (n = 14) of the children had
suffered at least 4 or more episodes of otitis media (within a given 12 month time span) at
least one time during their first three years of life.
Procedures

Potential participants initially were identified from computer printouts of all patient
contacts and diagnoses for three to four year old children who had been seen at the
Pediatrics Clinic, College of Osteopathic Medicine at M.S.U. The computer listings
included dates of contact and specific diagnoses, making it possible to determine children's
histories of otitis media and other illnesses. All families initially identified as potential

participants through the screening process were sent a letter describing the nature and

57
purpose of the study. In the letter, parents were asked to consider participating in the

study and were informed that a member of the research team would call them as a follow-
up to the letter.

Shortly after families had been sent the initial recruitment letter they were
contacted by telephone. At that time, mothers were asked to participate in a brief (20 - 30
minute) phone interview. During the phone interview the interviewer carefully explained
the nature of the study, collected additional information on the child's health history
(including a detailed account of the mother's recollection of OM episodes), and
determined whether or not the family met the criteria for inclusion. If the family met all of
the inclusion criteria they were asked if they were willing to participate.

Following the telephone interview, one of the members of the research team made
a visit to the family's home. After obtaining the parent signatures on the informed consent
forms, the interviewer explained to the parents that they would be paid $75.00 for their
participation and interviewed both parents. Following the parent interview, the
interviewer secured signed release of information forms to obtain any additional medical
records on the child if they had been seen by a physician outside the MSU Clinical Center,
and left a set of questionnaires for the parents to fill out prior to their visit to the clinic for
the observational session.

Clinic observational sessions

Two 15-minute, semistructured dyadic parent-child interaction sessions were
conducted and videotaped in a clinic laboratory setting, one session was with the mother
and child and one was with the father and child. The order of the mother or father

interaction sessions was randomly counterbalanced to control for order effects. Visits for

58
the observational sessions were scheduled approximately two weeks apart and at times in

which the child was not experiencing any episodes of otitis media, according to the
parent’s reports. The observational sessions took place in an 11 by 16 foot laboratory
playroom, located within the MSU Psychological Clinic. The playroom was equipped
with two ceiling microphones, a one-way mirror, an adjacent observation room, and
videotape equipment. Two chairs and a table were placed in the center of the room,
facing the one-way mirror. One of two matched sets of toys was arranged in a specified
manner for each interactional session. The toys in the first toy set included (1) a wooden
truck with several zoo animals, (2) crayons and a pad of paper, (3) several plastic colored
rings on a post, and (4) a box of plastic "waffle" blocks. The second set of toys included
(1) a foam puzzle, (2) a toy doctor's kit, (3) a Mr. Potato Head set and (4) a box of Duplo
blocks.

Each parent and child was videotaped during the three standardized situations
(Eyberg & Robinson, 1983) that comprise the dyadic interactional task: (1) the Child-
Directed Interaction (5 minutes), (2) the Parent-Directed Interaction (5 minutes), and (3)
the Clean-up (1-10 minutes) Instructions for each of the three situations were provided
through a "bug-in-the—ear" microphone by an observer behind a one-way mirror, such that
only the parent was able to hear the instructions.

The first situation was the least structured, and during the Child-Directed
Interaction, the parent was instructed to tell the child to play with whatever the child
wanted to choose. The parent then was told to allow the child to choose an activity and
they were to follow the child's lead and play along with him or her. In the more-structured

Parent-Directed Interaction, the parent was instructed to inform the child that it was the

59
parent’s turn to choose an activity. The parent then was told to keep the child playing

with them according to the parent’s own rules. In the final and most structured situation,
the Clean-up Interaction, the parent was instructed to tell the child that they were finished
playing and the toys must be put away. The parent then was told to make sure they had
the child put the toys away. These three situations were specifically designed to vary in
the degree to which parental directiveness or control was required, in order to assess
whether increased structure would lead to a corresponding increase in observed child
behavior problems.

Dependent Measures

Summary categories. Nineteen categories of parent and child behaviors (see Table
1 for examples) were used to code the parent-child interactions during the three segments
of the observational task (Eyberg & Robinson, 1983). Discrete types of parental
behaviors were grouped on an a priori basis, by the present investigator, to form the two
dimensions of parenting that have been underscored in the socialization literature: parental
warmth or responsiveness and parental control. Based on the critical evaluation of the
definition of parental control within the developmental and clinical literature, an attempt
was made to further divide the parental control dimension into two sub-dimensions:
"positive" parental control and "negative" parental control.

Parental warmth or responsiveness was defined, by the present investigator, by
combining a group of behaviors contained within the Eyberg & Robinson (1983) Dyadic
Parent-Cth Interaction Coding System that were presumed to be indicative of parental
warmth, attentiveness, and positive responsiveness to the child. These included the

following behaviors: high levels of labeled and unlabeled praise, positive physical

60

Table 1. Examples of parent & child behaviors coded with the Dyadic Parent-Child

Interaction Coding System.

Labeled Praise:

Unlabeled Praise:

Positive Physical Contact:

Reflective Statement:

Descriptive Statement:

"1 like the way that you drew that house".

"That was a good job".

Parent pats the child's arm in an affectionate manner.
Child: "My teacher is taking us to the zoo next week".
Parent: "Oh, you're going to go to the zoo soon?".

"You're putting the cow in the barn".

Descriptive/Reflective Question: "Where does that piece go?"

Acknowledgement:

Critical Statement:

Direct Command:

Indirect Command:

Nagative Command:

Child: "Is this the right place, mommy?"

Parent: "Yes it is".

"You didn't do a very good job cleaning up in here, young
man".
"Put the blocks into the box".

"How about giving me that one?"

"Don't get up on the table! ".

Physical Negative Contact (Parent): (Parent grabs a toy away from the child after the

child won't start cleaning up.

bio-opportunity Command: "Put the toys away", (2 seconds) "Come on and put them

away, now! "

Table l (cont’d).

Irrelevant Verbalization:

 

£1.12
Smart Talk:

Destructive Action:

61

"I wonder what your sister is doing right now".

(Child gives an inarticulate utterance of distress for 5
seconds).

(Child says in a high-pitched, nasal voice) "But I don't want
to clean up now!"

(Child makes a loud scream for 3 seconds).

"Nol, you put the toys awayl".

(Child slams a toy on the table and breaks it into several

pieces).

Physical Nggative Contact (Child): (Child hits parent).

M29:

Noncompliance:

(Child obeys, begins to obey, or attempts to obey a direct or
indirect command).
(Child does not begin to obey a parental command within 5

seconds or actively begins an incompatible activity).

Note. From “Dyadic Parent-Child Interaction Coding System: A Manual,” by S. M.

Eyberg and E. A. Robinson, 1983. (Available from S. M. Eyberg, University of Florida,

Gainsville, FL).

62
contact, reflective statements, descriptive statements, descriptive and/or reflective

questions, and acknowledgements (see Table 2). Labeled praise included any statement
which expressed a positive evaluation of a child's specific activity, product, or personal
attributes, whereas unlabeled praise was a nonspecific positive evaluation of a child's
activity, product, or personal attributes. Physical positive behavior was coded when there
was any neutral or positive physical contact between parent and child. Reflective
statements were declarative phrases that followed a child's verbalization, which contained
the same basic content as the child's original statement. A descriptive statement was a
declarative statement that provided an account of the objects, people, or activities
occurring during the interaction. Descriptive/reflective questions were descriptive or
reflective statements phrased in the form of a question, which followed, rather than
attempted to direct, the child's activity. Finally, acknowledgments were brief responses to
children's verbalizations which contained no apparent content other than a simple yes or
no response. The parental warmth dimension was also defined by low levels of critical
statements which were assumed to reflect parental hostility toward the child. Critical
statements consisted of any negative evaluations of the child's activities, products, or
personal attributes.

The "positive" parental control dimension was defined in the present study by high
levels of direct and indirect commands, which reflected parent's attempts to control or
otherwise direct children's behavior in a clear and/or positive manner. Direct commands
were clearly stated directives that contained specific information as to what was expected
of the child. Indirect commands were directives which were nonspecific, implied, or

stated in the form of a question.

63

03“an Z Rumba
aouo< ozpoawmon
0:3. 58m

2:55

=o>

to

 

coSEom o>§moz Bomb—m
358800 bastonnoé Z

$58800 33%» Z

 

fire—GOO £o>mua oz:

€095 855: mucoEoufim REED
mucoaomvogo§u<

305030 o>too¢3<o>matomoa
358035 038539

358035 0380qu

88:00 186.93 o>Emom

 

muggoo 80:9: 835 @20315
358800 885 835 “S233
.235 gazes? 3

82223 220 23 82.3 no Boas—o coma < .N 033.

64

The "negative" control dimension was defined in the current study by high levels of
negative commands, no-opportunity commands, and physical negative behavior (parent).
These negative control techniques were assumed to represent punitive, coercive, and harsh
methods of attempting to obtain compliance. Negative commands were commands which
directed a child to stop or inhibit their behavior. No-opportunity commands were
commands that did not allow ample opportunity for compliance. Physical negative
behavior (parent) was coded when a parent initiated contact with a child which inflicted
pain, restrained the child, or accompanied a critical statement.

The various coded child negative behaviors included the following: cry, yell,
whine, smart talk, destructive action, and physical negative. The frequencies of all parent
and child behaviors were coded continuously during each S-minute videotaped segment,
yielding total frequency scores for each category of behavior. The various child deviant
behavior frequency scores (cry, whine, yell, smart talk, destructive action, and physical
negative) were summed together to yield a "total child deviant behavior" frequency score.
All three task segments were designed to last 5 minutes each, but the Clean-up segment
often was completed in less than the allotted amount of time. Hence, the various child and
parent behavior frequency scores were subsequently transformed into scores representing
rates of occurrence (frequency divided by time in seconds), to account for the variability in
time across the three standardized situations that comprised the dyadic interactional task.
T_ap_ir_rg_and codingprocedures.

Parent-Child interactions were recorded on videotape and subsequently coded at a

later date. Coding of the observational sessions was done by at least one of four trained

65
coders (three undergraduate students and one graduate student), all of whom were blind

to the specific ROM status of the children at the time of coding. Initially, the coders
received over 20 hours of extensive training in the Dyadic Parent-Child Interaction Coding
System (Eyberg & Robinson, 1983) and were required to maintain a 70-80% rate of
reliability (Cohen's kappa) with practice tapes before they were allowed to begin coding
study tapes. In order to further maximize the accuracy of the observational training a
criterion tape which had been coded by one of the authors of the DPICS was included in
the set of practice tapes. Throughout the course of the study weekly training and practice

sessions were also held to maintain coding accuracy.

Results

M_easurement Issues.
R_e_li_2_1bility of the behavioral observation waiting;

Interrater reliability for the observational data. Interrater reliability checks were
conducted by randomly selected pairs of coders throughout the process of coding the firll
set of videotapes. These interrater reliability checks were conducted on approximately
25% of the study videotapes (randomly selected). Reliability of the frequency data was
measured by Cohen’s kappa, where reliability coefficients were calculated for each
separate pair of reliability codings of the nineteen individual child and parent behaviors
(Jacob, Tennenbaum & Krahn, 1987). Cohen’s kappa coefficients, computed across tasks
and indicating levels of interrater reliabilities for the frequency ratings of individual sets of
nineteen child and parent behaviors (across tasks), ranged from .59 to .85 (M=.73).
Cohen’s kappa coefficients, computed within tasks, were as follows: 1) child-directed
segment (M= .76, range = .41 to .90); parent-directed segment (M= .74, range = .44 to
.83); and 3) clean-up segment (M= .69, range = .35 to .79). The Cohen’s kappa
coefficients for the firll set of frequency ratings of individual sets of coded videotapes,
across and within task segments, are summarized in Table 3 in Appendix A.

Reliability for the summm variables. Rate data initially were aggregated into
summary variables describing child negative behavior and mother or father warmth and
control, as described earlier; these variables were computed both across and within
observational segments or tasks (child-directed, parent-directed, clean-up). Internal
consistency of the summary variables was assessed by Chronbach’s Alpha, and inter-rater

reliability for these variables was assessed by Pearson r correlation coefficients, across all

66

67
pairs of reliability codings for each of the summary variables (across tasks and within each

of the individual segments).

Table 4 (see Appendix B) contains the final groupings of parent and child behavior
which yielded the highest coefficient alphas for the two separate scales of parent behavior
and single child negative behavior scale. Due to the strong relationship between labeled
and unlabeled praise, they were combined into a general category of parental praise. For
both mothers and fathers, the highest alpha for the parental warmth scale was obtained
with the same group of five of the seven parent behaviors (descriptive/reflective questions,
descriptive statements, reflective statements, praise, & acknowledgements). The
behaviors positive physical contact and critical statements were dropped from the parental
warmth scale due to both a low rate of occurrence and/or a low correlation with the scale.
It is important to note that although the parental warmth scale exhibited adequate overall
internal consistency, there was poor internal consistency in the parent-directed segment,
which improved only slightly in the clean-up segment.

For parental control, it did not appear that there were two separate sub-scales of
parental control ("positive" and "negative"), as had been expected. Rather, parental
control behaviors were part of a single dimension, which differed somewhat for mothers
and fathers. The single resulting parental control scales for mothers and fathers were
utilized in all subsequent analyses.

For mothers, the parental control scale (alpha= .67) was best defined by the
following behaviors: direct commands, direct commands/no opportunity, indirect
commands/no opportunity, negative commands, critical statements, and physical negative

behavior. Indirect commands were dropped from the parental control scale due to a low

68
correlation with other types of control behavior and with the overall scale. Although the

category of critical statements was originally placed within parental warmth scale
(inversely scored), it appeared to correlate more highly with the behaviors within the
parental control scale. Contrary to what had been expected, the categories of negative
commands, no-opportunity commands, critical statements, and physical negative behavior
were positively correlated with direct commands, rather than negatively correlated. For
fathers, the parental control scale (alpha=.65) was best defined by the same group of
behaviors as for mothers, minus the categories of critical statements and physical negative
behavior. These latter two behaviors were not included in the final Father Control scale
because they substantially lowered the internal consistency of the scale. The categories of
critical statements and physical negative behavior were not correlated with many of the
remaining items or the overall scale.

Table 5 shows the number of items and alpha coefficients for each of the summary
variables within and across segments (range = .05 to .74, M=.53) and Table 6 shows the
Pearson product-moment correlations, computed across the full set of paired reliability
coders, for each of the summary variables (across tasks and within each individual
segment) used in the data analyses (range = .72 to .99, M=.92). Alpha’s and interrater
reliability coefficients could not be calculated for child negative behaviors during the child
directed tasks with fathers or mothers because behaviors conprising these scales rarely
occurred during this observational segment. Alphas for the summary variables generally
were acceptable with the notable exception of parental warmth during the parent-directed

segment of the interaction task (especially the mother-directed

69
Table 5. Number of items and internal consistency of parent & child behavior scales.

 

 

No. of items Alpha
Mothers Fathers Mothers Fathers

Parent Warmth

Total 5 5 .50 .56

Child directed 5 5 .63 .74

Parent directed 5 5 .05 .22

Clean-up 5 5 .33 .46
Parent Control

Total 6 4 .67 .65

Child directed 6 4 .57 .50

Parent directed 6 4 .67 .55

Clean-up 6 4 .55 .48
Child Problem Behflior

Total 22 14 .66 .71

Child directed’ --- --- --- ---

Parent directed” 10 5 .70 .37

Clean-up" 9 8 .54 .70

' Child Problem behaviors rarely occurred so that alphas could not be computed.

b Child initiated crying and destructive behavior did not occur during the mother directed
task.

° Child initiated crying, yelling, whining, smart talk, destructive behavior, and physical
negative behaviors did not occur during the father directed task.

d Child initiated yelling, smart talk, and physical negative behavior did not occur during
clean up with mothers or fathers and child responding physical negative behavior did not
occur during clean up with fathers.

70

Table 6. Interrater reliability coefficients (Pearson rs) for the summary variables.

 

 

 

 

Mather F_ath_et
Obsermons Observations

Child Negative Behaviors

Total .94 .93

Child directed" -- --

Parent directed .94 .92

Clean up .93 .94
Parent Warmth

Total .98 .95

Child directed .98 .96

Parent directed .99 .85

Clean up .98 .98
Parent Control

Total .94 .92

Child directed .96 .82

Parent directed .81 .87

Clean up .94 .94

' Frequencies of child negative behavior during the child-directed segments were too low
to calculate interrater reliabilities.

71
segment) and to a lesser degree, parental warmth during the clean-up segment.

Nevertheless, interrater agreement on ratings of the summary variables generally was high.
Corrmutation of recurrent otitis media (ROM) severity.

The correlation between the two otitis variables (total days of effusion and total
number of episodes) was .73, p<.001. Thus, a single indicator of otitis severity (“ROM
severity”) was computed by standardizing and averaging these two variables.
Demographic Relationshipa

Sex differences. Analyses comparing males and females on each of the
standardized variables considered in this study identified relatively few differences between
males and females. Out of a total of 36 comparisons, only two were significant at the .05
level which is approximately what would be expected by chance: Across tasks, fathers
were warmer with their daughters (M = .89) than with their sons =-1.05; p<.05); sex
differences in father warmth were significant at the .01 level in the child-directed task, but
were not significant in the father-directed and father clean-up tasks.

Correlations with other demoggrphic vamea Correlations between the
variables considered in the research hypotheses and three other demographic variables
(child age in months, father and mother educational attainment) resulted in 6 significant
relationships. Because I tested 76 relationships in all, four would be expected by chance.
Three of the significant findings indicated that more educated fathers were warmer in their
interactions with their children (overall and during the father directed and clean-up tasks)
than less educated fathers (significant r’s ranged from .33 (p<.05) to .44 (p<.001). In
addition, younger children behaved more negatively with their fathers overall (r = -.28,

p<.05) and during the father directed task (r=-.38, p<.01). One other correlation between

72
mother education and child negative behavior during the father directed task (r = .37,

p<.01) was difficult to explain. It should be noted that the variable ROM severity was not
significantly correlated with any of the demographic variables. This finding is consistent
with at least one prospective, epidemiological study that found no relationship between
ROM and SES (Teele et al., 1990). Additional correlations among each of the variables
considered in the a priori hypotheses are summarized in Table 7 in Appendix C.

Within task variation in chcLa._nd_parent behavior.

Repeated measures analyses using a 3 (Type of Task) x 2 (Child Sex) design
assessed whether task variation (child-directed, parent-directed, or clean up) would elicit
different frequencies of child negative behavior and mother or father warmth or control
behaviors. Analysis of child negative behavior, using the mother-child interaction data,
resulted in a significant effect for Task (F (2,104) = 5.8, p<.01) and also in a significant
Sex x Type of Task interaction (F (2, 104) = 4.9, p<.01). Analysis of the father interaction
data resulted in a significant effect for Type of Task, but effects involving Sex were not
significant. As can be seen in Table 8, even though between-task comparisons were not
always statistically significant, the general pattern was for child negative behavior to
increase as the task situation became more structured (i.e., fi'om child-directed to parent-
directed to clean-up). An exception was that boys behaved most negatively in interactions
with their mother when the play task was mother (rather than child-) directed; in contrast,
girls behaved most negatively when their mother directed them to clean up. For both
sexes, child negative behaviors occurred at a very low frequency during the child-directed

segments.

73

Table 8. Average rates (frequencies per second) of child and parent behaviors within

observational tasks.

Child Negaaive Behavior
All
Males
Females
Parent Warmth
All
Males
Females
P_arent Control
All
Males

Females

d Row means for mothers with the same superscript do not differ significantly (p>.05)

‘ Row means for fathers with the same superscript do not differ significantly (p>.05)

Mother Interactions d

Child

Directed

.0006a
.0008'
.0003'

.1911'
.2086a
.1743

.0148a
.0158'
.0139‘

Parent
Directed

.0047”
.0073'
.0022'

.1669”
.1800‘
.1542

.0598”
.0648”
.0548”

Clean up

.0061”
.0034
.0089’

l373°

.1261”
.1485

.0633”
.0690”
.0575”

 

Father Interactions °

Child

.0001ll
.0000'
.0002'

.1925'I
.1580‘I
.2258‘I

.0143'I
.0130'
.0156‘I

Parent
Directed Directed

.0040”
.0047”
.0033'

.1790'
.1669'
.1907”

.0620”
.0588”
.0652”

Clean
UP

.0104”
.0088”
.0120”

.1422”
.1298”
.1542c

.0681”
.0677”
.0685”

74
Table 8 shows differences in parent warmth and control within each of the

observational segments. A repeated measures analysis using a 3 (Type of Task) x 2 (Child
Sex) design resulted in significant Task, as well as Task x Sex interactions for both mother
warmth (for Task, F(2,104) = 17.7, p<.001, and for the 2-way interaction, F(2,104)= 5.5,
p<.01) and father warmth (for Task, F(2, 106) = 13.8, p<.001, and for the 2-way
interaction, F(2,106) = 3 .3, p<.05). Only the Task repeated measures factor was
significant when parent control was compared within situations (for mothers, F (2,104):
46.0 and for fathers, F(2,106) = 67.3, p’s<.001). As can be seen in Table 8, mothers
became less warm as the task became more structured, but these differences were stronger
and only statistically significant when mothers interacted with their sons (rather than
daughters). In contrast, fathers became less warm in interactions with their daughters as
structure increased from one task segment to the next. However, with their sons, fathers
showed less warmth in the clean up as compared to the child or father directed situations,
but warmth during the child and father directed tasks did not differ significantly (and
actually was somewhat higher when the father rather than the child directed the play
interaction). Finally, task differences in parent control mostly resulted because both
mothers and fathers were less controlling during the child directed task as compared to the
parent directed or clean up tasks.
Tega of the Research Hypotheses

In examining each of the research predictions, 1 tested whether child sex would
moderate the predicted relationships by assessing the significance of relevant sex x
predictor variable interactions. Data are presented for males and females, combined,

except in those instances in which the sex x predictor interactions were statistically

75
significant (p< .05). Significant interactions were followed up by within sex analyses so

that results for these hypotheses could be presented separately for girls and boys.

Because of the large number of effects resulting from the various analyses I only
interpreted those effects that involved ROM severity. Tests of the hypotheses predicting
child negative behavior were performed using variables summarized across task segments
and variables within the parent-directed and clean-up segments. Because of the low
frequency of observed child negative behavior during the child-directed segment, I did not
separately consider data from this segment in the analyses.

Hypothesis 1: Relationships between ROM severity and child negative behavior.
Regression analyses in which child negative behavior across tasks was regressed on ROM
severity did not result in any significant relationships for interactions with mothers or
fathers. Similarly, ROM severity was unrelated to child negative behavior observed within
the child-directed, parent-directed, and clean-up segments of the parent-child interactions.

Hypothesis 2: Relationships between a) ROM severity and parent warmth and
control and b) parent warmth and control and child negative behavior. Regression
analyses predicting mothers’ warmth or control from ROM severity did not result in any
significant relationships across or within each of the interaction segments (univariate
correlations are shown in Table 9). These relationships also were mostly insignificant
when warmth and control referred to fathers; the one exception was a significant
interaction between child sex and ROM severity in predicting father warmth (beta =33,
p<.05). Follow up analyses indicated that ROM was associated with more father warmth

with sons (beta = .34, p<.09) and less warmth with daughters (beta = -31, p<.13), but

76
Table 9. Correlations between ROM severity and Parental Warmth and Control

Mothers Fathers
Warmth Control Warmth Control
Observational Segment r r r R
Total .18 .17 -.08 .08
Child directed .20 -.05 -.10 .06
Parent directed .21 .22 .04 .09
Clean-up .08 .08 -.O3 .01

' p<.05
b p<.01

° p< .001

77
neither of these relationships was significant at conventional levels.

Regression analyses predicting child negative behavior from parental warmth and
control identified a significant relationship between mother control behaviors and child
negative behaviors across tasks (beta = .51, p<.001), with more control associated with
more negative child behavior. Within tasks, there was a significant sex x mother control
interaction for the mother directed task (beta = .28, p<.05). Follow-up analyses indicated
that mother control was associated with more child behavior problems among boys (beta
=.72, p<.001), but was more weakly (and not significantly) related to negative behavior
among girls (beta= .24, p, us). As can be seen in Table 10, no other relationships
between parental warmth or control and child negative behavior were statistically
significant.

Hypothesis 3: The buffering implications of parent warmth and control for the
effects of ROM severity on child negative behaviors. Tests of Hypothesis 3 were based on
regression analyses examining the significance of a) interactions between ROM severity
and parental warmth as well as b) ROM severity and parental control in predicting child
negative behavior. Interaction terms were computed as cross-products and tested in a
stepwise fashion after controlling for the relevant main effects and simpler interaction
effects. As in the tests of the previous hypotheses, initial analyses assessed whether child
sex would moderate any of the predicted relationships and examined the relevant
relationships across, as well as within different types of tasks.

Interactions between ROM and father warmth or control were not statistically
significant regardless of the type of interaction task (see Table 11). Alternatively, mother

warmth interacted with ROM severity in predicting child negative behavior across

78
Table 10. Beta coefficients resulting from regressing child negative behavior on parental

warmth or control.

Mothers Fathers
Warmth Control Warmth Control
Observational Segment
Total .02 .49c -.02 .1 1
Child directed -- -- -- --
Parent directed .03 .66c -. 16 .20
Clean-up .08 .08 .24 .07
' p < .05
” p < .01

°p< .001

79

Table 11. Beta coefficients resulting from regressing child negative behavior on ROM

severity, parental warmth, and parental control within observational segments.

Predictor Variables Interaction Effects

ROM Warmth Control R x W R x C R x C x W

beta Beta beta beta beta beta
Mothers
Total . 14 -.04 48° -.29’ -- --
Child directed -- -- -- -- -- --
Parent directed .08 -.06 65° -- -- --
Clean-up . 1 l -.07 . l4 -- -- --
Fathers
Total -.04 -.O4 . 12 -- -- ..
Child directed -- -- -- -- -- --
Parent directed -. 12 .20 .24 -- -- --
Clean-up -.02 .09 .07 -- -- --
' p < .05
b p < .01

° p< .001

80
situations. As can be seen in Figure 2, when mothers showed greater warmth, children

with a severe history of ROM showed le_sa negative behavior whereas children without an
ROM history showed mpg negative behavior. In addition, Child Sex interacted with
Control and ROM in predicting child negative behavior during the mother-directed task
(beta = .29, p<.05). Figure 3 shows the nature of this interaction. In particular, greater
mother control was associated with las_s child negative behavior among girls with a history
of severe ROM and with r_no_re negative child behavior among girls without a severe ROM
history. Alternatively, more maternal control was associated with may; negative behavior
among boys with a history of severe ROM and had a slightly positive, but generally weak

relationship with negative behavior among boys without this history.

81

Child Negative Behavior

at Low and High Maternal Warmth

 

1

03 L; ............................
0.6 ...................................... Lap: ....................................................................................
0,4 .................................
0,2 .................................

 

0

-0.4
-05 ...............................

t

Child Negative Behavior

 

-02 , .................................
Low ROM

................................. \

     
 

.............................................................................................

 

 

Low M-afemal Warmth

High Maternal Warmth

 

L—Low ROM —-High ROM]

 

Figure 2: Relationship between (a) ROM severity, (b) maternal warmth, and (c) child

negative behavior.

82

Child Negative Behavior (Girls)

at Low and High Maternal Control

 

 

 

 

 

 

 

 

 

1
h 08 ..............................................................................................................
,g 0.6 Main; ................................. "Myron ..................................................................
(U \ \
g O4 ......................... T.‘..‘.....\..\....\. .............................................
g 0.2 ................ \\_\\\\\ .................
.m 0 a \ \ \\ \
g -02 ..................... LowROM \‘\\\‘
g _0.4 ............................ \ ......................................................................................................
8 -0.6 ....................................................................................................... . ..............................
-0.8 .....................................................................................................................................
- t
Low Maternal Control High Maternal Control

 

|-Low ROM --High ROM ]

 

(Mother-Directed Segment)

 

Child Negative Behavior (Boys)

at Low & High Maternal Control

 

1

 

 

 

 

 

 

 

5 06 ......................... . ................................................... HIQNROM/// ..................

> //

2 0.4 ...................................................................................... \y/ ......................................

(D //

m 02 /,/ .......

3 LowlROM 1//

g 0 - / - w.

2 _02 //// ........... .. .....................................

g _04 // .....................................................................................................

.C //

O -0.6 L/// ................................................ . ....................................................................
-0.8 /../.f ........................................... . ......................... _ ...........................................................

Low Maternal Control High Maternal Control

 

l—Low ROM --High ROMI

 

(Mother-Directed Segment)

Figure 3 : Relationship between (a) ROM severity, (b) maternal control, and (c) children’s
negative behavior.

Discussion

The purpose of the present study was to assess the potential impact of ROM on
child behavior and parent-child interactions in a group of families with children who had
varying degrees of ROM during the first three years of life. It was anticipated that the
quality of parent-child interactions would play a role in minimizing or moderating the
deleterious effects of ROM on children's behavioral fiinctioning. Data analyses tested two
alternative models (the “independent influence” and “buffering” models) postulating
relationships between the severity of children's ROM during the first three years of life,
style of parental interactive behavior (parental warmth and parental control), and children's
behavioral adjustment at three to four years of age.

The first model, the "independent influence” model, postulates that an early history
of ROM and positive parenting behaviors (warmth and control) independently influence
children's behavioral adjustment in opposite ways. According to this model, recurrent
otitis media and parenting independently contribute to children’s behavior so that children
who have suffered from severe histories of ROM will be better adjusted if they are
exposed to effective parenting behavior (both high parental warmth and positive parental
control), but would not be as adjusted as children without early histories of ROM who are
equally exposed to effective parenting behavior.

The second model, the "buffering” model, argues that an effective style of parental
interactive behavior buffers the deleterious effects of ROM on child behavior. Thus, those
children with severe histories of ROM exposed to less effective parental interactive
behavior will exhibit problems with their behavioral adjustment. However, children with

early histories of ROM who were exposed to effective parental interactive behavior will be

83

84
nearly, or just as well adjusted as children raised in similar environments, but without

severe histories of ROM. The present study sought to determine whether style of
parental interactive behavior and ROM history were independently associated with child
behavior problems, as suggested in the “independent influence” model or whether style of
parental interactive behavior would moderate the relationship between ROM and child
behavior problems, as suggested by the “buffering” model.

In general, the findings presented here provided little support for the “independent
influence” model. A history of ROM predicted neither children’s negative behavior, nor
parental warmth or control during the parent-child interaction task, with a single
exception. The one exception was a significant interaction between child sex and ROM
severity in predicting father warmth. The direction of the beta coefficients suggested that
ROM severity is associated with more father warmth with sons, and less warmth with
daughters. However, subsequent within child gender analyses indicated that neither of
these relationships was significant at conventional levels.

In contrast, some support was found for the “buffering” model in that the results
suggest style of parental interactive behavior may moderate the relationship between ROM
and child behavior problems; however the findings were not always in the predicted
direction. Specific findings for each of the a priori hypotheses formulated from the two
models are discussed in greater detail in the section which follows.

Hypothesis 1:

The hypothesis that severity of ROM would be positively associated with observed

child behavior problems was not supported by the present analyses. Previous studies

showing correlations between severity of ROM history and parental reports of behavior

85
problems implied that such a relationship should exist (e. g., Casey, 1983; Feagans &

Proctor, 1994; Forgays, Hasazi & Wasserman, 1992; Gottleib, Zinkus & Thompson,
1979; Iverson, 1987; Roberts, Burchinal & Campbell, 1994; Vernon-Feagans, Manlove &
Volling, 1996). However, many of these earlier findings were based on maternal
responses to a child behavior problem questionnaire rather than actual observations of
child behavior in a laboratory setting. One possibility is that there is no direct relationship
between severity of ROM history and observed child behavior problems, as indicated by
the findings fiom the present study. Alternatively, it may have been the case that the
relatively brief, single point-in-time, laboratory observation of child negative behavior,
within the context of the structured parent-child interaction task, was not a sensitive
enough measure to detect the behavioral patterns that were hypothesized to exist, as
compared to maternal reports which are based on maternal observations over a more
extensive period of time. It also is possible that a longer observational period and/or a
more naturalistic setting would allow for a more valid assessment of behavioral problems
during parent-child interactions.

The actual severity of ROM history was a second factor that may have influenced
the results of the present study. The mean number of episodes of ROM in the present
study was relatively low, 4.18 episodes across the first three years of life. This is in
contrast to the severity of ROM found in samples of participants used in other studies.
Table 12 contains a summary of the characteristics of participants, including ROM
severity, and the design features of several recent studies examining the relationship

between ROM and children’s behavior and/or firnctioning. For example, there were 5.4

86

.83» N. 88 $388888 a: 330:8 :5 8:: 8 3:33.. 0888 368808023 cows: a mo 3: 288.8

.8808 86% 33 38:3 68.88 :0 988m 08 c8 own :38 o: :1 $80 macaw 20M: .8 38638 own 838

A83» N $08.3 98m
aC3.» 38 088 :0 {com
A83: m 8863 on...”
E3.» m 888V 00:.
A83» N 86.63 vww

A83: m $863 8.2

 

3:880 20%

.838 £80 6883 mama“: me 38:83: 82> 08: 68 :0 $8 63— 8 838 8:8 83 38:3

.8808 vm :O owfiozw 3 88 3368888 a: 326:8 :3 2808 N: own 8 3:208 2:88

85 32
826-288

88 888
0838-32 63.62%
826-288
826-288

mmm

.83.» h 88 >3888882 a: 330:8 3e 8:: 8 38:62 2:88 a

83> w 88 b3883882 a: 326:8 88 :85 8 3:98»: 2888

was N - 85
..88 6m 1 S
we.» a l 85
.8 w r 85
.888 EN
.8585 2.8

9334

83-338 8808 o :83 8088338 8 own :38

38:82
@6232
8838.2
8853882
3338.886
Ecomuoowémoh.

mailman

emu:
on":
SN":
_ou=
on":
own:
688%Haw

go— ..3 3 03:85

cam: in 3 meeweoméofio>
com: in 8 63:.

m8: .8 3 33:3:

82 ..a a 886":

$2 .888

Am “80:83.4

3:83 208 882 mo :83: :5 3586888 mo 8338880 08. :O 3883 .m: 23:.

87
episodes across the first M9 years of life in the Wallace et al. (1996) study; 8.84 episodes

in the first t_w_o years of life in the experimental group of the F orgays et al. (1992) study; 7
episodes across the first t_l1r_e§ years of life in the Roberts et al. (1994) study; and 15.63 in
Casey's (1983) experimental group of children. Only Teele et al. (1990) studied children
with fewer than 4 episodes of otitis media during the first three years of life (mean number
of episodes of ROM was 3.79), but their results nevertheless did reveal a relationship
between ROM severity and various child cognitive, speech and language and school
performance outcomes. Given that the children in the present study experienced fewer
episodes of ROM than the participants in many of the earlier related studies, the potential
negative effects of ROM on behavioral adjustment might have been less severe and
thereby less observable, than was otherwise expected. Similarly, there may be a minimum
threshold of ROM severity, below which direct negative effects on children’s behavior
would not be expected to exist.

Hypothesis 2:

The present study also did not yield much support for the hypothesis that a more
severe history of ROM would have an indirect impact on child behavior via an effect on
parenting behaviors; in particular, a more severe history of ROM was expected to be
associated with observations of less effective parental-interactive behaviors (both lower
levels of parental warmth and lower levels of positive parental control). As noted earlier,
the single exception found was a significant interaction between child sex and ROM
severity in predicting father warmth. Although the subsequent examination of this
interaction did not yield statistically significant relationships between the variables of

interest, the direction of the beta coefficient suggested that ROM severity might have been

88
associated with greater father warmth with sons and less warmth with daughters. As

noted above, the overall lack of an association between severity of ROM history and
parenting behaviors may have been attributable to the relatively low mean number of
episodes of ROM experienced by children in the present study, as compared to other
related studies (e. g., Casey, 1983; Forgays, Hasazi & Wasserman, 1992; Roberts et al.,
1994; Teele et al., 1990; Wallace et al., 1996).

Hypothesis 3:

In many ways the most interesting findings came from the multiple regression
analyses that were utilized to examine whether style of parental interactive behavior
(parental warmth and control) would moderate the potentially deleterious relationship
between ROM severity and children’s behavioral problems. Maternal (but not paternal)
warmth interacted with ROM severity in predicting child negative behavior across the
different types of tasks. For children with a more severe ROM history, greater maternal
warmth was associated with @ child negative behavior. In contrast, for those children
with a less severe ROM history, greater maternal warmth was associated with m child
negative behavior.

For those children with more severe ROM histories, the increased maternal
warmth observed in the present study may represent mothers’ attempts to compensate for
the negative effects of ROM on children’s behavioral firnctioning, by providing increased
instructions, cues and reinforcement. This finding is not surprising given that the sample
consisted primarily of well-educated, middle-class families, so that these mothers might be
more likely to display such compensatory behavior. These findings also are consistent

with earlier related research that has documented a general link between increased

89
maternal warmth and responsiveness and subsequent decreases in child behavior problems

and noncompliance (e.g., Lay, Waters & Park, 1989; Lemanek, Stone & Fishel, 1993;
Martin, 1987; Rocisano, Slade & Lynch, 1987; Schaffer & Crook, 1980; Stayton, Hogan
& Ainsworth, 1971; Wakschlag & Hans, 1997).

On the other hand, for those children without a history of severe ROM, greater
maternal warmth was associated with mo_re_ child negative behavior, which is difficult to
explain at first glance. However, examination of the relative levels of child negative
behavior at different levels of both ROM severity and maternal warmth (see Figure 2),
indicates that low ROM severity was associated with relatively lower levels of child
behavior problems, regardless of the level of maternal warmth. Thus, despite the positive
association between maternal warmth and child negative behavior, the observed pattern of
child negative behavior for children with less severe ROM histories remained well below
the level observed for children with more severe ROM histories, and may even reflect an
entirely different underlying dynamic in the parent-child relationship (e. g., greater
tolerance for mildly deviant behavior by warmer parents).

With respect to parental control, there was no interaction between father’s control
behavior and ROM severity in predicting child negative behavior. On the other hand,
there was a significant interaction among ROM severity, maternal control, and child sex,
in predicting child negative behavior during the mother-directed task. For girls with a
more severe history of ROM, greater maternal control was associated with leg negative
child behavior, whereas for girls with a less severe history of ROM, greater maternal
control was associated with m negative child behavior. Conversely, for boys with a

more severe history of ROM, greater maternal control was associated with more negative

 

90
behavior, whereas for boys with a less severe history of ROM, there was a slightly

positive, but generally weak relationship between greater maternal control and child
negative behavior.

Thus, it is clear that maternal control was interacting quite differently with ROM
severity for boys and girls; however the reasons for such differences are not clear. For
children with low ROM severity, boys appeared to be relatively unaffected by different
levels of maternal control, whereas girls exhibited more negative behavior in the context of
greater maternal control. Within this context of low ROM severity, increased maternal
control behavior may have been experienced by girls as somewhat aversive, as suggested
by the corresponding increase in child negative behavior. However, this latter finding is
somewhat inconsistent with the findings from other studies that generally have found
greater levels of compliance and shared positive affect between mothers and daughters, as
compared to mothers and sons (Kochanska, 1997 ; Maccoby, 1990). However, it is
possible that maternal control behavior may have been less normative and experienced as
unfamiliar and therefore aversive by girls, as compared to boys.

In contrast, children with more severe ROM histories displayed a very different
pattern of behavior within the context of different levels of maternal control. Boys with
more severe ROM histories may have experienced higher levels of maternal control as
somewhat aversive, as suggested by the corresponding increase in child negative behavior.
In contrast, given that girls with more severe ROM histories exhibited fewer negative
behaviors in relationship to higher maternal control behavior, it may have been the case
that increased maternal control provided an appropriate level of structure that offset the

negative effects of a severe history of ROM and/or is reflective of a greater level of

91
internalization of parental socialization attempts (Kochanska, 1997). However, for boys

there may have been other factors involved. Other research typically has found higher
levels of a range of maternal control behaviors with boys, as compared to girls, and higher
levels of child negative behavior and/or noncompliance, again for boys as compared to
girls (e.g., Kochanska, 1997; Lytton & Romney, 1991, Smetana, 1989). Thus, the results
of the present study for boys with more severe ROM histories are consistent with findings
from other related studies.

Given the cross-sectional design of the present study, it is difiicult, if not
impossible to determine actual causality or direction of influence of the parent and/or child
behavior. Thus, it is not possible to accurately assess whether for boys (with severe ROM
histories) higher levels of observed maternal control behavior represented antecedents or
consequences to the child negative behavior exhibited during the parent-child interaction
task. Nevertheless, the fact that the parent-directed segment of the interaction task was
specifically designed to have parents initiating the shift in focus from a child-directed to
parent-directed focus, suggests that the higher levels of observed maternal control
behavior with boys may have preceded the increase in child negative behavior. Similarly,
Maccoby (1992) has argued that although the normal socialization process involves an
evolving system of reciprocity, mothers nevertheless are the more powerfiil agents in the
early stages of establishing this system of reciprocity. This argument supports the notion
that maternal control behavior preceded the child negative behavior observed during the
parent-chfld interaction task of the present study. However, more refined measures of
sequential parent-child interaction would have been necessary to accurately assess the

timing and precise nature of this relationship (Kochanska, 1997; Kuczynski et al., 1987;

92
Rocissano, Slade & Lynch, 1987). Regardless of the key mechanism of influence, the

results clearly suggest that mothers of boys with more severe ROM histories were not able
to effectively match their style of interaction to the child’s attentional or emotional state in
such a manner that would result in a more harmonious interaction (Baumrind, 1988;
Kochanska, 1997; Kuczynski et al., 1987; Maccoby, 1983, 1992).

It is interesting to note that the overall set of findings for father-child interactions
were very different from mother-child interactions and generally not consistent with the
hypotheses. The paucity of significant findings for father-child interactions is especially
interesting given the popularly held notion that fathers may be becoming more involved in
the lives of their children (Furstenberg, 1988; Marsiglio, 1995). However, contemporary
research examining the various complexities of paternal involvement, beyond the more
proximal sphere of parent-child interactions, have begun to identify and study a wide
variety of types of paternal involvement (Lamb, 1997; Lamb, Pleck, Charnov & Levine,
1987; Palkovitz, 1997; Pleck, Lamb & Levine, 1986). This recent research suggests that
mothers continue to play a more central role in the lives of their children, especially with
respect to the more proximal parent-child relationship issues (Bloom-Feshbach, 1981;
Lamb, 1997; Lamb et al., 1987; Maccoby, 1992; Pleck et al., 1986), and as such are more
likely to be directly affected by the negative impact of illnesses such as recurrent ROM.

In a review of the literature, Lamb et al. (1987) found that although fathers do appear to
be spending considerably more time interacting with their children when mothers are
employed, the proportionate increase in the amount of time spent in interactions was
largely accounted for by a decrease in the amount of actual time that mothers were

available to interact with their children. Thus, it is clear that future research in this area is

93
greatly needed and may help to explicate the heretofore elusive processes underlying the

popularly held notion of increased involvement of fathers in the lives of their children.

In summary, the results provided little support for the “independent influence”
model. In general, a history of ROM predicted neither children’s negative behavior, nor
parental warmth or control during the parent-child interaction task. There was, however,
some support found for the “buffering” model in that the results suggest style of parental
interactive behavior may moderate the relationship between ROM and child behavior
problems; however the findings were not always in the predicted direction. In particular,
1) ROM severity interacted with maternal warmth, and 2) ROM severity interacted with
maternal control and child sex in predicting child negative behavior during the structured,
laboratory play task. These results add to the findings from several other recent studies on
otitis media, which have documented the key role that parent behavior plays in moderating
the negative effects of ROM on children’s development (Roberts et al., 1995; Wallace,
Gravel, Schwartz & Ruben, 1996).

Wallace et al. (1996) investigated the association between early otitis media, style
of parental interaction behavior and children’s language development. The results of the
study show that high maternal verbal stimulation was associated with better language
skills, whereas high levels of observed maternal control behavior during a laboratory
interaction task were associated with low scores on measures of expressive language
development. In a related prospective longitudinal study of 61 infants, Roberts et al.
(1995) found an association between otitis media and subsequent hearing loss, language
and cognitive outcomes, but only when parenting style and quality of the home

environments were taken into account. Both of these earlier studies, as well as the present

94
study, underscore the important moderating role that family factors, such as parental

interaction style, play in minimizing the impact of otitis media on children’s development.

On the one hand, the findings suggest that increased maternal warmth may have
been utilized by mothers to offset the negative effects of a history of ROM by providing
additional supportive guidance, prior to the emergence of any child negative behavior or in
response to such deviant behavior. On the other hand, higher levels of maternal control
may have been utilized as an attempt to regulate the play activity (which was more
successful for girls versus boys with more severe ROM histories), either before and/or
after the emergence of child negative behavior. It is clear from the results of the present
study that future studies must continue to explore the potential dynamic between parental
warmth and control as a buffering mechanism against the deleterious effects of recurrent
otitis media. However, firture research also will need to address several of the key
limitations to the present study.
Lthatimrsgf the Present Study

Several notes of caution should be offered regarding the results of the present
study. First, the sample size was relatively small (n=56), thereby limiting the power to
detect anything but large effects. Cell sizes associated with analyses attempting to
examine the three-way interaction between ROM severity and parental warmth and
control obviously were especially small. Future research efforts should utilize much larger
samples of children, with representative sampling of all age groups between the ages of
birth to at least 6-years, in order to better support these types of statistical analyses. This
need to conduct studies with larger and more representative samples is particularly

important given that other researchers (e. g., Casey, 1983; Daly, 1991; Paradise, 1981;

95
Roberts et al., 1994; Teele et al, 1989) have documented that the negative effects of ROM

may be greatest during the first few years of life.

Second, as was mentioned earlier, the actual severity of ROM history was a
second factor that may have influenced the results of the present study. The mean number
of episodes of ROM in the present study was relatively low (4.18 episodes across the first
three years of life), compared to the levels found in most other study samples. The fact
that other related studies have included children with more severe early histories of ROM
could explain why they found strong relationships between ROM severity and parent
and/or child behavioral difficulties. Conversely, given that the children in the present
study experienced fewer episodes of ROM than the participants in many of the earlier
related studies, the potential negative effects of ROM on behavioral adjustment might
have been less severe and thereby less observable, than was otherwise expected.

Third, the sample consisted of children from well-educated, middle-class families.
Other studies (e. g., Roberts et al., 1995; Wallace, Gravel, Schwartz & Ruben, 1996),
utilized samples that either were more varied on SES or included a higher number of
lower income families. Higher educated, middle class parents are likely to have more
available resources and hence better parenting skills, than the general population, which
also translates into reduced vulnerability to stresses associated with childhood illnesses.
Given that the participants were representative of a higher functioning group of families,
who regularly seek medical care for their children's physical ailments, the potential
negative impact of illnesses on their children's emotional and behavioral functioning likely
was minimized. Future studies should include larger samples with enough variability in

SES to allow for greater generalizability of the findings.

96
Fourth, the cross-sectional design of the present study severely limits the ability to

determine causality among the variables of interest. More specifically, the study did not
utilize a prospective, longitudinal design that would allow for a more precise examination
of the multiple effects of ROM on children’s developmental trajectories over time, as have
other related studies (e.g., Roberts et al., 1995; Vernon-Feagans, Manlove & Volling,
1996; Wallace, Gravel, Schwartz & Ruben, 1996). Similarly, the cross-sectional design of
the study did not allow for the observation of whether parental behavior fluctuated in
response to the effects of ROM episodes on their children’s behavior, over time. Thus, it
would be important for firture studies to include longitudinal designs to examine
prospectively the impact of ROM on childhood development over a significant period of
time.

F ifih, the actual documentation of ROM history in the present study was based
primarily on medical record review, rather than on a more direct assessment over time, via
otoscopy, tympanometry, auditory brain stem assessments, or other calibrated physical
assessment approaches (e. g., Roberts et al., 1995; Teele et al., 1990; Vernon-Feagans,
Manlove & Volling, 1996; Wallace, Gravel, Schwartz & Ruben, 1996). Many of these
other recent studies relied on one or more of the more direct assessment methods and
often conducted the assessments on a regular basis over the course of the study. Thus, the
resulting ROM severity index used in the present study, likely was much less precise than
those used in other studies, and thereby less likely to allow the detection of small effects.

Sixth, the participants were not evaluated for the potential presence of conductive
hearing loss, which has been shown to be related to both a history of ROM and increased

behavioral difficulties (e. g., Baldwin, 1993; Bluestone et al., 1983; Gottlieb, Zinkus, &

97
Thompson, 1979; Mattison, Cantwell & Baker, 1980; Roberts et al, 1995). Although the

parent-child interaction sessions were conducted when children were not experiencing an
acute episode of otitis media, actual hearing assessments were not conducted prior to their
participation in the observational sessions. Thus, it was not possible to determine whether
any of the children were experiencing any actual hearing loss at the time the observed
parent-child interaction sessions were conducted. Although the presence of conductive
hearing loss at the time of observation might be expected to result in more negative child
behavior, it nevertheless presents a potential confounding factor, especially if the hearing
loss were to occur in children with less severe ROM histories. Thus, the optimal situation
would be to ensure that children being observed were not experiencing any temporary
hearing loss at the time of observation.

A seventh limitation of the present study was that a comprehensive evaluation of
other potential situational stressors was not conducted due to the necessarily limited scope
of the present study. It is acknowledged that behavioral and interactional problems are
often multiply determined, and as such, the reported findings may have been influenced by
additional variables, such as the parent's level of perceived stress, marital difficulties, and
financial difficulties, among others. Therefore, in future studies it will be necessary to
replicate the present findings while considering the influence of other potential stressors on
the parent-child relationship.

The eighth limitation of the present investigation was that the particular interaction
task used may not have been long enough or naturalistic enough to elicit an adequate
I‘ange of behavior, including more significant dysfimctional parent-child interactions as

Well as more responsive and nurturing interactions. The relatively low frequency of

98
parent and child behaviors in general, and negative behavior in particular, in each of the

individual interaction segments made it more diflicult to examine the complexity of
relationships among parental warmth and control and child negative behavior, especially as
a function of varying ROM histories. Although the findings provided partial support for
several of the hypotheses, and represent a contribution to the existing literature on the
impact of ROM on children's development, it is possible that the results may have been
more robust with a longer and more varied interaction task.

A ninth limitation of the present study was related to the internal consistency of the
parental warmth scale. For both mothers and fathers, although the overall parental
warmth scale exhibited adequate internal consistency, there was poor internal consistency
in the parent-directed segment, which improved only slightly in the clean-up segment.

This poor internal consistency likely was a result of the low rate of occurrence of the
behaviors within the parental warmth scale during the parent-directed segment.
Alternatively, within the different tasks, the meaning and structure of the scale may change
as a function of the unique context of the situation. In either case, this limitation reduced
the level of confidence that the underlying construct of parental warmth actually was being
measured.

A final limitation was the way in which parental control was defined in the present
study. Although the initial intent was to draw upon the socialization literature and define
two distinct sub-scales of parental control behaviors, a “positive” control scale and a

“negative” control scale, preliminary analyses did not support this approach. Rather, as
noted earlier, individual parental control behaviors were best subsumed under a single

parental control scale, which differed somewhat for mothers and fathers. Careful

99
examination of the categories of behavior that were included in the parental control scale

further indicated that the resulting scale used in the present study may have been more
heavily weighted towards the "negative" side of parental control, especially for mothers.
This assertion is based on the observation that a majority of the parental behaviors
included in the control scale (e. g., negative commands, direct and indirect commands with
no opportunity to comply, and for mothers but not fathers, critical statements, and
physically negative behavior) all were initially subsumed under the "negative" control
dimension based on a review of the clinical literature. Thus, it was not possible to
examine whether a combination of parental warmth and “positive” parental control would
provide the most effective buffering mechanism against the deleterious effects of ROM on

children’s adjustment.

Summgy

In summary, there were several important findings from the present study, which
represent contributions to the existing literature on the impact of ROM on children and
their families. The most interesting and consistent findings in some ways support the
“buffering” model in that the results suggest style of parental interactive behavior may
moderate the relationship between ROM and child behavior problems. In particular, the
interactions between 1) ROM severity and maternal warmth, and 2) ROM severity,
maternal control and child sex, in predicting child negative behavior during the structured,
laboratory play task support the assertion that parents are in the unique position to

moderate the potentially deleterious effects of ROM on a child's emotional and behavioral

adjustment.

100
Given that certain illnesses such as ROM have been found to be associated with

increases in child behavior problems and disruptions in parent-child relationships, there are
several important implications for the current set of findings. The most important
implication is that it may be most fiuitfiil for fixture research efforts to examine what
specific factors enable some parents of children with severe ROM histories to successfiilly
adjust their interaction style to best match variations in situational demands and their
children's temperament.

As was shown in the earlier review of the literature, there are clear advantages to
utilizing different styles of parenting, depending upon certain child characteristics and
situational circumstances. If a parent is sensitive to the child's needs, able to convey
warmth, and able to synchronize attempts to direct or control the child's behavior with the
child's age and temperamental state, then the likelihood of obtaining compliance and
maximizing the child's behavioral and emotional development should be significantly
increased (Kochanska, 1997; Maccoby, 1983, 1992; Marfo, 1990; Martin, 1987;
Rocisano, Slade, & Lynch, 1987; Schaffer & Crook, 1980). While the results of the
present study did not specifically support the importance of the combination of parental
warmth and control, the results nevertheless did suggest that certain aspects of parental
interactive behavior may moderate the relationship between ROM and child behavior
problems.

More generally, this and other studies indicate that health care providers will need
to become more aware of the multiple factors that determine the impact an illness such as
ROM may have on a children's emotional and behavioral adjustment, especially given that

parents’ qualities may moderate effects of illness on the child’s psychosocial adjustment.

101
Thus, these providers are coming to recognize that they can no longer can assess and/or

treat the impact of ROM in isolation and need to consider such issues as the availability of
family resources, overall family functioning, and parent’s abilities to compensate for the

effects of the illness over time.

APPENDIX A

 

102

Table 3. Interrater reliabilities for frequency ratings of individual sets of coded
videotapes‘.

 

 

Olegll Child-directed fluent-directed Clean-up
l. K= .77 = .62 = .83 K: .56
2. = .85 = .73 = .76 = .56
3. = .72 =60 = .71 =63
4. = .77 = .73 = .62 = .67
5. = .65 = .71 = .44 = .41
6. =63 . = .41 = .49 = .50
7. =65 = .69 =62 = .53
8. = .75 = .65 = .74 = .55
9. = .73 =68 = .69 = .65
10. = .68 = .59 =63 = .35
11. = .72 = .63 = .55 =69
12. = .76 = .60 = .66 = .67
13. = .77 = .59 = .70 = .79
14. = .64 =63 = .55 = .72
15. = .59 = .65 K = .48 =65

‘ Interrater reliabilities were calculated on 29 sets of frequency ratings, with each set of
ratings conducted by a randomly selected pair of two of four trained coders. Each set of
frequency ratings was conducted across the three separate task segments using the
nineteen individual child and parent behaviors and the kappas presented are for the overall

interrater reliabilities, as well as separate reliabilities for each task segment.

 

103
Table 3 (cont’d).

16. = .78 K = .90 K = .82 K = .53
17. =82 K = .75 K = .79 K = .69
18. =.82 K=.68 K=.81 K=.7O
19. = .73 K = .56 K = .80 K = .62
20. = .78 K = .89 K = .76 K = .76
21. =.66 K=.67 K=.6l K=.66
22. = .69 K = .78 K = .70 K = .60
23. = .83 K =85 K = .80 = .60
24. = .82 = .74 = .74 = .74
25. = .75 = .74 =68 = .68
26. = .79 = .67 = .72 = .79
27. = .70 = .57 = .69 = .63
28. = .75 = .73 = .78 = .60
29. = .72 K = .75 K = .72 K = .44

Mean = .73 Mean = .76 Mean = .74 Mean = .69

APPENDIX B

104
Table 4. Internal consistency of parent & child behavior scales.

 

 

 

 

 

 

Mothers Fathers
Item- Alpha Item- Alpha
Total If Item Total If Item
TotaL- fluent Warmth Correlation Deleted Correlation Deleted
Descriptive/Reflective .35 .37 .40 .52
Descriptive Statements .49 .24 .46 .40
Reflective Statements .27 .50 .47 .55
Acknowledgements . l 7 .52 .40 .47
Praise .32 .45 .35 .53
Alpha = .50 Alpha = .56
Child directed - Parent Warmth
Descriptive/Reflective Questions .38 .59 .49 .76
Descriptive Statements .72 .37 .80 .58
Reflective Statements .72 .37 .80 .58
Acknowledgements .01 .73 .30 .76
Praise .26 .65 .36 .76

Alpha = .63 Alpha = .74

105
Table 4. (cont’d).

 

 

 

 

 

 

Mothers Fathers
Item- Alpha Item- Alpha
Total If Item Total IfItem
Brent directed - Parent Wmth Correlation Deleted Correlation Deleted
Descriptive/Reflective Questions .02 .03 .08 .29
Descriptive Statements .01 .05 .10 .20
Reflective Statements -.05 .06 .23 .20
Acknowledgements -.04 .11 .23 .10
Praise .18 -.07 .18 .18
Alpha = .05 Alpha = .22
Ql_ean up — Par_ent Warmth
Descriptive/Reflective Questions .30 .17 .31 .35
Descriptive Statements .33 .11 .31 .36
Reflective Statements .15 .33 .27 .47
Acknowledgements .08 .33 .25 .40
Praise .12 .31 .30 .40

Alpha = .33 Alpha = .46

106
Table 4. (cont’d).

 

 

 

 

 

Mothers Fathers
Item— Alpha Item- Alpha
Total If Item Total If Item
Tatai- Parent Control Conmtion Deleted Corr_dation Deleted
Critical Statements .44 .03 -- --
Physical Negative .21 .05 -- --
Direct Commands .42 .62 .46 .56
Direct Commands/
No opportunity .55 .60 .59 .49
Indirect Commands -
No opportunity .42 .62 .37 .65
Negative Commands .57 .55 .38 .62
Alpha = .67 Alpha = .65

Table 4. (cont’d).

Child directed - Parent Control
Critical Statements
Physical Negative
Direct Commands
Direct Commands/

No opportunity
Indirect Commands -
No opportunity

Negative Commands

107

 

Mothers
Item- Alpha
Total If Item

Correlation Deleted

.48 .49
.OO .60
.34 .59
.52 .45
.34 .52
.41 .51
Alpha = .57

 

Item-
Total

Corralation Deleted

.40

.37

.31

.31

Alpha

If Item

.35

.42

.42

.47

Alpha = .50

108
Table 4. (cont’d).

 

 

 

 

 

Mothers; Fathefi
Item- Alpha Item— Alpha
Total If Item Total If Item
Parent directed - Parent Control Correlation Deleted Correlation Deleted
Critical Statements .31 .66 -- --
Physical Negative .50 .69 -- --
Direct Commands .49 .60 .30 .52
Direct Commands -
No opportunity .62 .59 .46 .39
Indirect Commands -
No opportunity .53 .62 .30 .55
Negative Commands .51 .61 .48 .47

Alpha = .67 Alpha = .55

Table 4. (cont’d).

Claan up - Pakrent Control
Critical Statements
Physical Negative
Direct Commands
Direct Commands/

No opportunity
Indirect Commands -

No opportunity

Negative Commands

109

 

Mothers
Item- Alpha
Total If Item

Correl_ation Deleted

.30
.11

.41

.27

.32

.49

Alpha = .5

.54

.57

.46

.52

.50

.40

5

 

Item-

Total

Correlation Deleted

Alpha

If Item

 

.34 .34
.36 .36
.26 .46
.26 .46
Alpha = .48

l 10
Table 4. (cont’d).

 

 

 

 

 

Mothers Fathers
Item- Alpha Item— Alpha
Total If Item Total If Item
Total; Child Negative Behavior Corralation Deleted Conelflrn Deleted

Cry - Ignored .40 .55 .40 .56
Cry - Responded to .47 .49 .68 .53
Yell - Ignored .07 .59 .00 .62
Yell - Responded .22 .58 .54 .56
Whine - Ignored .37 .57 .58 .57
Whine - Responded .56 .45 .40 .57
Smart talk - Ignored -.07 .59 .00 .62
Smart talk - Responded .20 .63 .30 .65
Destructive Action - Ignored .49 .58 .63 .60
Destructive Action - Responded .51 .49 .87 .59
Child Physical Negative - Ignored .07 .59 .00 .62
Child Physical Negative - Responded .21 .58 -.O6 .62

Alpha = .58 Alpha = .61

l 1 1
Table 4. (cont’d).

 

 

Mothers Fathers
Item- Alpha Item- Alpha
Total If Item Total If Item

Child directed - Child Neg. Behayior‘ Correlation Deleted Correlation Deleted
Cry - Ignored -- -- -- --
Cry - Responded to -- -- -- -_
Yell - Ignored -- -- -- ..
Yell - Responded -- -- -- --
Whine - Ignored -- -- -- --
Whine - Responded -— -- .. --
Smart talk - Ignored -- .. .. --
Smart talk - Responded -- -- .. ..
Destructive Action - Ignored -- -- -- ..
Destructive Action - Responded -- -- -- --
Child Physical Negative - Ignored -- -- -- --
Child Physical Negative - Responded -- -- -- --

Alpha = -- Alpha = --

' Child Problem behaviors rarely occurred so that alphas could not be computed.

 

l 12
Table 4. (cont’d).

 

 

Mothers Faghers
Item- Alpha Item- Alpha
Total If Item Total If Item

P_arent directed - Child Neg. Behavior“ Corrfition Deleted Correlation Deleted

 

Cry - Ignored .00 .69 .00 .38
Cry - Responded .88 .59 .04 .38 F
Yell - Ignored .06 .69 .00 .38
Yell - Responded .51 .67 .64 .33
Whine - Ignored .06 .69 .00 .38 l;
Whine - Responded .74 .58 .28 .27
Smart talk - Ignored -.05 .69 .00 .38
Smart talk - Responded .49 .63 .44 .16
Destructive Action - Ignored .00 .69 .00 .38
Destructive Action - Responded .70 .57 .61 .28
Child Physical Negative - Ignored .06 .69 .00 .38
Child Physical Negative - Responded .06 .69 -.06 .38
Alpha = .69 Alpha = .37

’ Child initiated crying and destructive behavior did not occur during the mother directed
task.

° Child initiated crying, yelling, whining, smart talk, destructive behavior, and physical
negative behaviors did not occur during the father directed task.

Table 4. (cont’d).

Clean up - Child Negative BehaLviord

113

 

 

Mothers Fathers
Item- Alpha Item- Alpha
Total If Item Total If Item

Correlation Deleted Correlation Deleted

Cry - Ignored .67 .40 .56 .60
Cry - Responded to .49 .40 .83 .56
Yell - Ignored .00 .53 .00 .67
Yell - Responded -.02 .53 .40 .64
Whine - Ignored .68 .48 .73 .60
Whine - Responded .47 .41 .40 .63
Smart talk - Ignored .00 .53 .00 .67
Smart talk - Responded .07 .62 .23 .69
Destructive Action - Ignored .7 5 .50 .77 .65
Destructive Action - Responded .21 .51 .77 .65
Child Physical Negative - Ignored .00 .53 .00 .67
Child Physical Negative - Responded .42 .51 .00 .67
Alpha = .52 Alpha = .67

 

d Child initiated yelling, smart talk, and physical negative behavior did not occur during
clean up with mothers or fathers and child responding physical negative behavior did not
occur during clean up with fathers.

APPENDIX C

114

an n 5
88:

oom.ua
:8 u 5
82.0

8:. "a
5 u 5
~83

mow.na
:3 u 5
£25

on“. "8
Gm u 5
82.0

m8. "a
Gm u 5
88.0.

g;

In. a .- .5.er tiff. « . .

com. "a
an n 5
ES

5 u 5
88:

48. “8
:mm H 5
82¢

mmv.ua
an n5
:5

83.. an
A8 u 5
803

Now. ”8
am u 5
38.0-

208

8:. "8
Gm u 5
32.0

38. “a
an n 5
82.0

. "a
:8 n 5
88:

coo. "8
Gm n 5
8:3

8m. ”8
Gm u 5
88¢

m8. "8
Gm u 5
22.0-

.9: MEGS:

8m. "8
an n 5
3:5

mmv.ua
Ammn5
:.o

80. "8
Gm u 5
8:3

. "a
Gm n 5
88:

mmm.na
Gm u 5
82¢

4m».ua
Gm n 5
83¢.

.Qm Mg<m

3N. "a
5 n 5
483

94. "a
5 u 5
82¢

3m. "a
:8 u 5
35.0

mmm.ua
Gm n 5
82.0

. "a
Gm u 5
88.:

N5... an
em u 5
82¢

Xmm QED

$4. "a
5 n 5
885.

Now. "a
A? u 5
:85-

8N. "8
Gm u 5
2.2.?

$5. "a
Gm u 5
83.0-

$4. "8
Gm n 5
82.0

. "a
6n n 5
88.:

.mOS: m0<

3%;

20m

.9: MEGS:

.Qm MEIR/E

XMm QED

.mOS: m0<

33283 823 :3 .226 muons 86820800 .N. 23:.

115

Sn. L
A3 u 5
68¢.

em. "a
Gm u 5
£85

mm. L
5 u 5
836

So. "a
an n 5
:86

So. "a
g n 5
8:6

80. "a
5 u 5
:53

So. "a
5 u 5
82¢

g;

2m. um
E n 5
885

OS. "a
am n 5
2:3

as. "a
an n 5
385.

man. "a
am u 5
mm: .o

ham.ua
:mu5
”ES

5. "a
am u 5
ago

«2. ”a
am n 5
23¢

20%

So. "a
3 n 5
88¢

85. "a
5 u 5
:36

mom.ua
5. "5
ES

mwm.ua
an n 5
28.0

5. "a
g u 5
32.0

NS. um
Gm u 5
836

:w. "a
an n 5
«.86

.Dm flaw—.02

E. "a
Ax u 5
835

N3. "a
an n 5
38.0

8m. nu
an n 5
32.?

mg. ”a
Gm u 5
28¢.

mm. "a
A: n 5
$26

we. "a
Gm u 5
a»;

08. um
Gm u 5
95¢

.Qm dab—Rm

SN. um
A: u 5
32 .o

fix. "a
Gm u 5
~33

mom. "a
an n 5
823

v2. "a
an n 5
803

5. na
2% u 5
$2.?

mmo.uq
an n 5
38.0

as. "a
an n 5
33

Xmm QED

MS. "a
A: u 5
88.0

so. L
5 u 5
88.0

aoh.uq
Gm u 5
38.0.

:m. "a
an n 5
38.0

0%. "a
A: u 5
:03-

$6 an
AR u 5
83¢.

am. "a
Gm u 5
RB?

.mOE m0<

30482002

MQHHZOUE

QUHHZOUE

ACME/502

MODE;

mg;

MQUEBE

3.285 .5 23¢

oNo.no
A? n 5
momoo-

won no
5 n 5
ofioo

:Vm. no
an n 5
ovooo-

ovo.no
A? n 5
moooo-

116

So. no
Gm n 5
Sooo

o3. no
A? n 5

SS.
ooo. no
Go n 5
goo-

g;

 

ox. no
so n 5
«mooo

go. no
an n 5
:moo

oom.no
am n 5
ooooo

o3. no
an n 5
omooo-

woo.no
an n 5
osoo

E. no
so n 5
too.

«on no
Q n 5
goo-

20M

mom. no
an n 5
Gooo

we. no
Go n 5
Sooo

so. no
Go n 5
onto

mom.no
Gm n 5
ooooo

So. no
an n 5
woto

no. no
an n 5
voooo-

vow. no
so n 5
mmooo

.Dm M502

ovo. no
Go n 5
38.?

So. no
Go n 5
mooo

coo. no
5 n 5
oooo

So. no
A? n 5
mono

So. no
am n 5
88o

mom.no
Gm n 5
23o

ooo. no
so n 5
goo

.Dm agafl

o3. no
an n 5
moooo-

go. no
5 n 5
goo-

so. no
Gm n 5
moooo-

o2. no
an n 5
82o-

So. no
an n 5
omooo-

moo. no
Go n 5
go.

So. no
Go n 5
«homo-

XMm QED

ooo. no
so n 5
onto-

mom. no
5 n 5
$3.?

an. no
so n 5
32o-

moo.no
an n 5
oSoo

33. no
an n 5
Sooo-

ooh no
so n 5
Nmmoo-

moo. no
5 n 5
woooo

.mOE m0<

MOAHZOUm

QUAHZOUm

AOMHZOUm

gvégm

mead/«am

M90332:

Siam

«no. no
Go n 5
woooo-

who. no
Go n 5
Eoo-

moo. no
so n 5
ooooo

so. no
5 n 5
Noooo

117

woo. no
Go n 5
Noooo

mow.no
5 n 5
ooooo

omo.no
5 n 5
oooo-

Egg—2

 

so. no
an n 5
38o

o3. no
an n 5
amoo-

Nov. no
so n5
«So

we. no
an n 5
voooo

so. no
so n 5
ooRo

on. no
5 n 5
oflmo

N8. no
an n 5
mooo

20M

noo.no
Gm n 5
83o-

so. no
an n 5
23

5. no
A: n 5
Sooo

o2. no
Go n 5
New;

woo.no
an n 5
ammo

oo_. no
Gm n 5
vomo

NNN.no
Gm n 5
ES

.Qm MEG;

as. no
an n 5
88o

mmo.no
Gm n 5
oomoo-

ovo.no
cum n 5
ovvoo-

go. no
Go n 5
~33

an. no
Go n 5
82o

5.. no
Go n 5
moooo

woo. no
so. n 5
ooooo

.Qm ME];

5. no
an n 5
82o-

mow.no
no n 5
amoo

woo. no
3 n 5
oomoo-

ooo.no
an n 5
28o

omm.no
Gm n 5
So

So. no
Go n 5
Naoo-

2o. no
an n 5
zoo

Xmm QED

o3. no
so n 5
Eoo

ovo. no
Go n 5
22o-

own. no
A: n 5
omooo-

moo. no
an n 5
E3.

M3. no
5 n 5
32 .o-

an no
Go n 5
goo-

no. no
so n 5
ovoo-

.mOE mD<

MQUmmZU

MmmZU

MDUEmZU

MAG/575

Inflow/520

MEQZU

DUAHZOUH

ooo.no
Gm n 5
ES

So. no
Go n 5
Sooo

N2. no
Gm n 5
SoS

So. no
an n 5
oomoo

118

MQHHZOUE

ooo.no
an n 5
88o-

omo.no

Gm n 5
ooooo-

Egg

mom. no
so n 5
ES

oom.no
an n 5
omNS-

mom.no
Gm n 5
88o

ooo.no
Go n 5
Smoo-

QUHHZOU—Z

omo.no
am n 5
ES.

5.. no

so n 5
$2.?

20%

now. no
an n 5
goo

moo.no
an n 5
goo-

we. no
an n 5
82 .o

:o. no
an n 5
ES

AOMHZOUE

So. no
Go n 5
£8o

ooo.no
A? n 5
”mono

.Dm MED:

o2. no
A: n 5
82 .o

mm~.no
3. n 5
ES

5. no
A: n 5
SS.

9%. no
@m n 5
Eoo .o-

MODE;

Nmo.no
Gm n 5
moooo-

oom. no

Go n 5
moooo

.Qm Mamba;

moo. no
an n 5
38o

”2. no
Go n 5
52o

moo.no
an n 5
ES

So. no
Go n 5
oovoo-

Mg;

am. no
5 n 5
ooooo

:o. no

Go n 5
Bo

Xmm QED

:ono

an n 5
RS

So. no
5 n 5
ozoo

ooo. no
so n 5
58o

mom. no
an n 5
28o-

maugkfiz

o2. no
an n 5
$3.?

woo. no

so n 5
«mono-

.mOE m0<

.Qm MEGS

.Qm Mg<m

Xmm QED

.mOE m0<

50595

MEmZU

ooo. no
Go n 5
ooao

ooo. no
Go n 5
85o

moo. no
9% n 5
3S.

ooo. no
an n 5
ooooo

119

ooo.no
an n 5
«zoo-

o3. no
an n 5
ooooo

o2. no
an n 5
3:3

man—HZOUE

. no
Go n 5
oooo._

moo. no
5 n 5
Soto

ooo. no
3 n 5
SMS-

no. no
Go n 5
SMS.

So. no
Go n 5
SSS

5. no
an n 5
wovoo

E. no
so n 5
vomoo-

QUAHZOUE

P

ooo. no
an n 5
Soto

. no
Go n 5
884

Ev. no
2% n 5
SS.

o; no
so n 5
RES.

ooo.no
A? n 5
S ooo-

So. no
an n 5
zooo

Nam. no
so n 5
mm: .o

ACME/HOD:

ooo. no
9% n 5
mono-

5.. no
2% n 5
SS.

A: n 5
ooooo

So. no
A: n 5
23o

ooo. no
3 n5
mono

ooo. no
3 n 5
mofio

ooo.no
cm n 5
ES

M30232

ooo.no
an n 5
5S-

ooo. no
an n 5
83o-

ooo. no
9% n 5
23o

5 n 5
ooooo

No. no
an n 5
Somo

ooo.no
5 n 5
SS

5. no
so n5
88o

many—<32

So. no
Go n 5
oomoo

mg. no
an n 5
2 8o.

ooo. no
9% n 5
mono

No. no
Go n 5
ooomo

Gm n 5
oooo._

ooo. no
so n 5
oooS

NS. no
so n 5
mNoS

M90555:

QUAHZOUE

AOMHZOUE

gov/2552

Mg;

MOD—213‘;

Egg:

20M

:o. no
Go n 5
23o

ooo. no
5 n 5
ovooo-

o8. no
so n 5
32o-

So. no
A? n 5
63S-

120

ooo. no
an n 5
Romo-

ooo.no
2% n 5
Soto

. no
an n 5
oooo._

MDHHZOOE

o

.

E
. .I 3.3.! 1 .1.

o2. no
A? n5
.223

So. no
an n 5
voooo

:o. no
Go n 5
ommo-

NR. no
A? n 5
woo-

N2. no
Go n5
:oS-

an. no
A: n 5
«ES

ooo. no
an n 5
ooao

QUAHZOOE

ooo. no
an n 5
ooo.no

no». no
Go n 5
omoo-

ooo. no
Go n 5
RS-

ooo. no
Go n 5
28o-

o:. no
Go n5
58o-

ooo. no
9% n 5
Eooo

ooo. no
5 n 5
85o

AOMHZOUZ

ooo. no
Go n 5
$S-

ooo.no
an n 5
mmooo-

me. no
an n 5
ammo

ooo. no
So u 5
$2 .o-

2o. no
5 n 5
momoo-

ooo. no
9% n 5
moooo-

ooo.no
3m n 5
ooooo

Mela/‘32

33. no
5 n 5
oovoo

so. no
so n 5
goo-

EH. no
so n 5
32o-

ooo. no
Go n 5
moooo

at: no
so n 5
82.?

3%. no
9% n 5
.52 .o-

:o. no
Go n 5
vzoo-

MQUEMSSE

ACME/Moon

MDUEE

Mggm

Know/EASE

Eégm

DORE/H002

MQHHZOUE

121

ooo.. no
9% n 5
moooo

ooo. no
Go n 5
oomoo

E. no
Go n 5
omomo

So. no
A? n 5
33o

So. no
5 n 5
ES

So. no
so n 5
SS

ooo. no
Go n 5
88o-

MQHHZOQZ

ooo. no
2% n 5
33.?

«3. no
5 n 5
oooo

mom. no
Go n 5
SMS

now. no
an n 5
RES.

ooo. no
Go n 5
38o

ooo.no
Gm n 5
ommoo

ooo.no
an n 5
oovoo-

QOAHZO02

3. no
A: n 5
82o

ooo. no
A? n 5
voooo

So. no
5 n 5
83o

ooo. no
am n 5
owovo

ooo. no
5 n 5
Sumo

ooo. no
an n 5
oommo

moo. no
Go n 5
~agnoo.

AOEZO02

So. no
A: n 5
ooooo

so. no
A: n 5
ES

:3. no
3 n 5
ooooo

ooo. no
A3. n 5
«ooo-

o2. no
3 n 5
85 .o-

So. no
9% n 5
«ooo-

ooo. no
9% n 5
flooo

M302M<§

ooo. no
3 n 5
moo

5. no
so n 5
moooo

ooo.no
Go n 5
ES

ooo.no
Gm n 5
ovooo

So. no
an n 5
SES-

ooo. no
Go n 5
Smoo-

ooo. no
5 n 5
$2.?

Mg;

ooo. no
3 n 5
ooooo-

o2. no
5 n 5
$2.?

ooo. no
Go n 5
goo

now. no
Go n 5
ooooo-

om. no
Go n 5
ovooo

03. "a
Go n 5
mmood

moo. no
so n 5
EoS

MQUEMdS/E

MSU—>520

M§m20

MQOEmZ0

§m20

D0§O0m

MDdZO0m

Q0§O0m

122

so. no
Go n 5
Noooo-

ooo. no
5 n 5
ES-

ACME-200m

ooo.no
Gm n 5
88.?

om. no
an n 5
Swoo-

oom. no
an n 5
$2.?

omo. no
Go n 5
voooo

MQnE-ZO0E

 

o2. no
5 n 5
82.9

ooo. no
an n 5
23o

g0§<3m

3.. no
an n 5
SoS-

ooo. no
Go n 5
SS

ooo. no
Go n 5
ES-

So. no
so n 5
58.?

9013-2002

so. no
so n 5
82o-

33. no
5 n 5
ooooo-

mag-fl

ooo. no
so n 5
ES-

So. no
Go n 5
Smoo-

oom. no
so n 5
SES-

ooo. no
A? n 5
mmooo

ACME-200E

moo. no
Gm. n 5
ES-

ooo. no
Go n 5
Smoo-

MQ02M<3m

ooo. no
:5 n 5
Gooo-

ooo. no
3 n 5
ago-

ooo. no
3 n 5
88.?

New. no
9% n 5
oooo .o-

g0§<§

moo. no
Go n 5
NES-

moo. no
A?- n 5
ooooo

Egg-m

5. no
5 n 5
£2 .o

ooo. no
Go n 5
MES-

ooo. no
Go n 5
ommo-

ooo. no
Go n 5
28o

Mg;

So. no
9% n 5
«SS

moo. no
so n 5
88o

DOAHZO0E

5. no
an n 5
EMS-

fim. no
so n 5
moo-S-

So. no
A? n 5
RES.

ooo. no
Go n 5
omto-

MQO§<§

Xmm Qua-m0

.mOZ m0<

g0mm20

MEmZU

MQOmmZ0

123

vow. no
5 n 5
5 S-

woo. no
5 n 5
3; .o-

go. no
5 n 5
oovoo

5. no
5 n 5
Swoo-

oom. no
5 n 5
ooooo

ooo. no
5 n 5
52o

ooo. no
5 n 5
oooo

AOMHZOPm

 

S. no
5 n 5
vovoo

ooo. no
5 n 5
58.?

so. no
5 n 5
ovoo-

So. no
5 n 5
moooo-

5». no
5 n 5
58o-

ooo.no
5 n 5
ooooo

So. no
5 n 5
55o

M00324:

5. no
5 n5
52o

ooo. no
5 n 5
5.8

E. no
5 n 5
5S-

So. no
5 n 5
Sooo

woo. no
5 n 5
oovoo

SN. no

5 n 5
Soto

so. no
5 n 5
82o

5. no
5 n 5
ongo-

omo. no
5 n 5
32 .o-

ooo.no
5 n 5
moooo

o3. no
5 n 5
SS-

mow. no
5 n 5
22.?

So. no
5 n 5
$8.?

mom. no
5 n 5
Bio

Mg: MO0SE<E

So. no
5 n 5
ago

So. no
5 n 5
momoo-

5. no
5 n 5
woo-S-

ooo. no
5 n 5
goo-

5. no
5 n 5
Swoo-

5. no
5 n 5
58o

ooo. no
5 n 5
5S

BEE

 

5.. no
5 n 5
Soo-

ooo. no
5 n 5
«Soo-

Sé. no
5 n 5
$2 .o-

ooo. no
5 n 5
smoo-

5. no
5 n 5
ooooo

ooo. no
5 n 5
ooooo

E. no
5 n 5
ago

002002

M302M<§

mg;

#2005435:

32-35232

20M

.Qm man-02

.Qm— Mg<m

2o. no
5 n5
2:.o

moo.no
5 n 5
ooooo

Ev. no
5 n 5
oooo

ooo. no
5 n 5
25o

124

:o. no
5 n5
Ego

ooo. no
5 n5
88o

ooo. no
5 n 5
55o

AOEZOUm

ooo. no
5 n 5
53o

ooo. no
5 n 5
$53

ooo. no
5 n 5
3o

05. no
5 n 5
ooooo-

95. no
5 n 5
55o-

So. no
5 n 5
voooo

5. no
5 n 5
omoo-

MDUEMAVE

. no
5 n 5
oooo._

So. no
5 n 5
ovomo

ooo. no
5 n 5
oo :o

now. no
5 n 5
moooo-

ooo. no
5 n 5
22o-

:o. no
5 n 5
omoo-
ooo. no
5 n 5
zoo-

Mg?

 

 

So. no
5 n 5
953

5 n 5
oooo._

ooo. no
5 n 5
:93

50. no
5 n 5
58o-

So. no
5 n 5
55o-

ooo. no
5 n 5
ooo-

omo. no
5 n 5
53o-

MODE:

ooo. no
5 n 5
o:o.o

ooo. no
5 n 5
:23

5 n 5
oooo._

ooo. no
5 n 5
88o-

_o_. no
5 n 5
55o-

ooo. no
5 n 5
22.9

o:. no
5 n5
3:3-

E;

E. no
5 n 5
moooo-

one. no
5 n 5
58o-

ooo. no
5 n 5
88o-

. no
5 n 5
885

moo. no
5 n 5
Soto

an. no
5 n 5
~23

ooo. no
5 n 5
«Boo

DUAHZOUE

 

Mme/$2.3m

MQUEBR

23m

DUHHZOUE

MQHHZOOE

Qua—H.200:

AOMHZOUE

55m.no
5 n 5
5o

ooo.no
5 n 5
55o

ooo. no
5 n 5
£3

ooo. no
5 n 5
voooo

125

ooo. no
5 n 5
Evo

. no
5 n 5
885

ooo. no
5 n 5
58o

AOMHZOUm

o5. no
5 n 5
983

So. no
5 n 5
ago

:5. no
5 n5
ooo

o5. no
5 n 5
«53.

So. no
5 n 5
ES

ooo. no

5 n 5
58o

5 n 5
oooo._

Evian

o3. no
5 n 5
«So

omo.no
5 n 5
58o

5. no
5 n 5
88o

now. no
5 n 5
ovooo

ooo.no
5 n 5
ooooo

”S. no
5 n5
2:.o

ooo. no
5 n 5
53o

4"ng

ooo. no
5 n 5
momoo

5. no
5 n 5
ooooo-

2v. no
5 n 5
8:?

«5. no
5 n 5
55o-

moo.no
5 n 5
958

So. no
5 n 5
ooooo

ooo.no
5 n 5
$53

MQUEMFSE

ooo. no
5 n 5
58o

5v. no
5 n 5
52o

woo.no
5 n 5
ooooo

5. no
5 n 5
oovoo-

:o. no
5 n5
:Xo

ooo.no
5 n 5
oooo

ooo. no
5 n 5
zoo

Egan

ooo.no
5 n 5
ooomo

ooo. no
5 n 5
Ego

«8. no
5 n 5
ovooo

ooo.no
5 n 5
5: .o

5. no
5 n 5
38o

So. no
5 n 5
25o

on». no
5 n 5
$8.?

DUHHZOUZ

MOUSE/HO

M3520

DUHHZOUHH

Mast/50m

OUHHZOUR

ACME/50m

goék’m

So. no
5 n 5
ooooo

2;. no
5 n5
ooooo

5. no
5 n 5
58o-

126

ooo. no
5 n 5
:3

ooo. no
5 n 5
vooo

so. no
5 n 5
53

AOMHZOUm

So. no
5 n 5
58o

So. no
5 n 5
moooo-

ooo.no
5 n 5
ooooo

So. no
5 n 5
52o

moo. no
5 n 5
55o

9%. no
5 n 5
92o

MDUék/m

E. no
5 n 5
oomoo

ooo. no
5 n 5
32o-

mom.no
5 n 5
58o

ooo. no
5 n 5
ago.

5. no
5 n 5

mooo-
:o. no
5 n 5
58o-

mag

 

mom. no
5 n 5
oooo-

ooo. no
5 n 5
58o

5o. no
5 n 5
28o

mom.no
5 n 5
«too.

ooo.no
5 n 5
28o

ooo. no
5 n 5
«to-

MQUEE

so. no
5 n 5
58o

ooo. no
5 n 5
:5?

ooo. no
5 n 5
ooooo

moo. no
5 n 5
:8?

5. no
5 n 5
:2o

5. no
5 n 5
moooo-

Egan

Em. no
5 n 5
Nooo-

o5. no
5 n 5
58o-

ooo.no
5 n 5
oooo

ooo. no
5 n 5
58o-

05. no
5 n 5
ooooo

ooo. no
5 n 5
55o

DUEOUE

gone/HO

Mama/HO

MQUmmZU

MmmZU

MDUEMZU

M93520

127

one. no
5 n 5
woooo-

5. no
5 n 5
58o

5. no
5 n 5
58o

o2. no
5 n 5
N53

52. no
5 n 5
52o

So. no
5 n 5
$53

moo. no
5 n 5
52o-

dezmzo

 

vow. no
5 n 5
moooo

ooo. no
5 n 5
53o

30. HQ
5 n 5
ooooo

owo. no
5 n 5
55o

o5. no
5 n 5
82o

o5. no
5 n 5
So

5. no
5 n 5
52o-

MOUSEZU

 

mom.no
5 n 5
ooooo-

5. no
5 n 5
ooooo

o2. no
5 n 5
53

o2. no
5 n 5
goo

:5. no
5 n 5
moooo

So. no
5 n 5
58o-

5. no
5 n 5
Sooo-

MEMZU

5. no
5 n 5
o5o.o

55. no
5 n 5
Goo-

ooo. no
5 n 5
mooo

ooo. no
5 n 5
52o

5.. no
5 n 5
88o

ooo. no
5 n 5
zoo

5. no
5 n 5
ovoo-

DUAHZOUm

55. no
5 n 5
58o

55. no
5 n 5
88o-

5. no
5 n 5
55o

mom. no
5 n 5
53

o5. no
5 n 5
ooooo-

ooo. no
5 n 5
moooo-

ooo. no
5 n 5
52o-

MGHHZOUH

So. no
5 n 5
52o

5. no
5 n 5
88o

So. no
5 n 5
Eco

5o. no
5 n 5
:53

3%. no
5 n 5
52o

55. no
5 n 5
58o-

moo. no
5 n 5
$2 .o-

OOH—.209.“

M90392;

mg;

20M

.Qm MEOE

.Qm Mmmhkm

Xmm 91:30

.mOE mG<

Q

128

5. no
5 n 5
585-

ooo.no
5 n 5
5.55

ooo. no
5 n 5
35¢

5o. no
5 n 5
wood

ooo.no
5 n 5
355

So. no
5 n5
:85

5. no

5 n 5
58¢

MAM—2520

 

woo. no
5 n 5
585

Re. no
5 n 5
gone

So. no
5 n5
58.0

5m. no
5 n 5
225

So. no
5 n 5
855

5. no
5 n 5
$8.0

So. no
5 n 5
5:3

many—>520

 

3.. no
5 n 5
58:.

«8. no
5 n 5
53

so. no
5 n 5
538

mom.no
5 n 5
885.

ooo. no
5 n 5
£35

wwo.no
5 n 5
89¢.
wmm.no
5 n 5
$85

M3520

5. no
5 n 5
$85

«.8. no
5 n 5
$3.0

m8. no
5 n 5
58

who. no
5 n 5
85.0

:5. no
5 n 5
25.0

5. no
5 n 5
85.?

En. no
5 n 5
52.0-

DUAPZOUm

E. no
5 n 5
83.0-

o5. no
5 n 5
5: .o

:0. no
5 n 5
83

as. no
5 n 5
58.0

08. no
5 n 5
855

5. no
5 n 5
$3.

80. no
5 n 5
50¢-

MQHHZOUm

:o. no
5 n 5
:55

5. no
5 n 5
3:;

go. no
5 n 5
5o?

5. no
5 n 5
£35-

80. no
5 n 5
38.0.

moo. no
5 n 5
285

one. no
5 n 5
52.0.

QUHHZOUm

Egam

DUHHZOUE

MORE/MODE

OOH—.2002

ACME/502

#3033232

4mg;

ms. um
am "5
N23

3. "a
Gm n5
5:5

as. "5
Gm u 5
885

so. "a
Gm u 5
$35

129

9%. nu
Gm u 5
~33

:o. "a
Gm u 5
$8.?

com. "a
Gm u 5
N26-

Mme/520

E. "a
Gm n 5
33°

3. ”a
3m u 5
326

am“. "a
Gm n 5
33.0-

mm. "a
Gm n 5
Ed

on. "a
an n 5
£25

a? "a
Gm u 5
83

OR. "a
Gm n 5
88¢

mQUZMZU

 

E

:5. "a
Gm u 5
ago

80qu
an n 5
88.?

mmo.nq
5 u 5
28¢.

8o. "a
Gm n 5
23$

35. "a
Gm u 5
$3

N3. "a
fin n 5
$85

a. nu
5 u 5
88¢.

M3520

. um
5 u 5
803

ooo. "a
Gm n 5
mad

©m~.nq
5 n 5
MS;

ooo. "a
5 n 5
23

em. "a
Gm. u 5
8.0

03. "a
Gm n 5
88.0

3. "a
an n 5
8:5-

DUHHZOUm

So. "a
an n 5
3.00

am u 5
803

woo. "a
Gm u 5
33

So. "a
Gm u 5
323

am. "a
an n 5
$2.0-

93. "a
Gm u 5
$85

3%. L
Gm n 5
:85-

MQAHZOUm

08. "a
Gm u 5
22.0

”8. "a
Gm u 5
$3

. um
Gm u 5
oooo._

So. "a
Gm n 5
Sta

SN. L
an n 5
ES

o8. L
Gm u 5
885

So. "a
an n 5
£26

QUEOUm

DORE/50m

MDHHZODm

QUEOUM

AOMHZOUm

60%:

mg:

MOUéE

EN. L
an n 5
ES

:8. L
Gm u 5
88.0-

«8. L
an n 5
383

E. L
9% "5
:83

130

an n 5
883

98. L
Am J 5
£88

88. L
Am J 5
858

#293520

 

 

 

JJ

 

a. L
an n 5
22.8

as. L
an n 5
88.0-

mg. L
Gm u 5
888

ME. L
9% n 5
888.

88. L
an n 5
£88

. L
Gm. n 5
88._

E. L
an "5
2:3

Mun—03520

_mh.uq
5 u 5
88.8

85. L
an n 5
88.8-

:8. L
Gm n 5
89.0

ooo.ua
:8 u 5
:28

08. L
Gm u 5
88.8

5. L
am "5
238

. L
5 u 5
88._

M3520

”S. L
Gm u 5
8.8

8m. L
5 u 5
228.

m8. L
an n 5
28.8

80. L
:8 n 5
88.8

ms. L
cm H 5
8:8

2. L
Gm n 5
88.8

:8. L
Gm u 5
89.8

D0200m

ov_.ua
an n 5
588

am. L
5 u 5
32.8.

NS. L
Gm u 5
88.8

08. L
9% u 5
82 .o-

3.. L
Gm n5
5:.

82. L
A2 "5
32.0

83L
Gm u 5
88.8-

MQHHZOUHH

9;. L
Gm u 5
83

Ms. L
an n 5
8.8

8h. L
cm H 5
38.8-

5. L
3 u 5
82.8.

Na. L
Gm n 5

88.8
86 L
Gm u 5
88.8-

$0. L
Q n 5
£88-

90200m

mem20

MQUmmZU

E20

M30220

#29420

#263520

M220

131

v8. L
Gm u 5
82 .o

MQEm20

oma. L
am u 5
28.8-

mg. L
5 n 5
:88

Nm8.uq
an n 5
«~88-

:8. L
Gm n 5
88.0

82. L
Gm u 5
82.8.

Mb0mm20

mmo.ua
an n 5
888

MQ02m20

5. L
an n 5
$2.8-

80. L
an n 5
~83

oom.ua
A3 u 5
88.8

:o. L
Gm u 5

8.0
88. L
an n 5
£88-

Mem20

o8. L
Gm u 5
38.0

M3520

 

8». L
an n 5
38.8

E. L
Gm u 5
88.8-

mm. L
Gm n 5
88.8

Rm. L
Gm u 5
82.8-

one.ua
an n 5
88.8

MQOMm20

08. L
Gm. n 5
~88

30200m

«8. L
an n 5
38.8-

8N. L
5 u 5
83

mm”. L
Gm u 5
88.8-

mm». L
an n 5
88.8

88. L
Gm n 5
£38..

Mmm20

ME. L
Gm u 5
68.8.

MOHHZO0M

NS. L
:mu5
v2 _ .o

5. L
g n 5
58.8

wvu.na
A: u 5
88.8-

8o. L
g u 5
888.

m8. L
9;. u 5
88.8-

M30220

2”. L
Gm n 5
32.8-

Q0200M

EOM

.Qm MEGS

.Dm— ME<M

Xmm Q§0

.mOMZ m0<

M39220

132

88. L
Gm u 5
88.8-

«8. L
5 n 5
N83-

mow.ua
an n 5
:88-

oov.ng
A: u 5
58.8.

5.. L
Am J 5
383

:m. L
an n 5
82.8-

nub.uq
Gm u 5
38.0-

M30mm20

 

08. L
Q n 5
888..

38. L
5 u 5
SS

8:. L
an n 5
88.8-

88. L
:8 u 5
888-

08. L
5 u 5
8:8-

E. L
5 u 5
82.8-

omo.na
Q n 5
88.8-

Mem20

an. L
Gm u 5
82.8-

an. L
Gm n 5
8:8-

bw~.na
Gm u 5
S: .8.

:8. L
9% u 5
88.8-

80. L
an n 5
8.3-

38. L
5 u 5
38.0-

88. L
an n 5
88.8-

M30mm20

88. L
Am J 5
88.0

2”. L
Gm n 5
38.8-

omo.nq
5 n 5
28.8

9%. L
9% n 5
82.8-

8». L
5 u 5
88.8

83. L
Gm u 5
8:8-

mum.na
Gm u 5
MES-

M2220

8v. L
:8 n 5
$88

88.. L
a n 5
838.

Sn. L
:8 u 5
82 .8

08. L
@m n 5
88.8

8:. L
A: n 5
8.8

08. L
2% n 5
88.8-

8o. L
A: u 5
88.8

M30220

 

MQAH200E

30122002

3052002

M30535:

Mg;

MDOEM<§

Egg

133

28. L
an n 5
8828-

88. L
38 u 5
8888

88. L
an n 5
888

E. L
Gm u 5
888

$8. L
an n 5
888-

E. L
$8 n 5
888

:8. L
:8 u 5
~88-

M30mm20

 

L.

8:. L
Gm u 5
828

8:. L
38 n 5
828

888. L
5 u 5
888-

888. L
Gm u 5
$28-

88>.uq
Gm u 5
$88

88. L
Gm u 5
:88-

88. L
A: u 5
8888-

M9220

28. L
an n 5
8888

888.na
5 u 5
888-

88¢.u8
Gm u. 5
$28

888.na
an n 5
588

88. L
Gm u 5
:88

88. L
Gm n 5
88888

hvw.na
:8 u 5
$88

M00mm20

88. L
38 u 5
388-

88. L
5. u 5
S _ .8

EN. L
$8 u 5
828

~8~.uq
an n 5
£88-

mhm.na
5 u 5
:88-

858.u8
an n 5
:88-

mg. L
:8 u 5
£88-

5. L

98 u 5
8888-

888. L
98 u 5
888

8.2. L
:8 u 5
88888

83. L
0.8 r. 5
888-

“58.ng
:8 n 5
288

38. L
A: u 5
:28

omm.ua
$8n5
8_88A8

M30220

001220083

40MM200M

M302M<E

M02M<E

M002M<E

E2M<E

302002

134

 

 

War.

888. L 888. L
5. n 5 5 n 5
-~88 $88
-8. L 8~8. L
$8 u 5 A3. u 5
8888 8288-
v-. L _m~. L
Am J 5 38 u 5
82 .8 ~58
88. L 88. L
Gm n 5 5 u 5
888 882 .8
888. L 5. L
Gm n 5 5 u 5
$38 8988
88. L 88. L
an n 5 5 u 5
~$88 88
82.. L 8:. L
Gm n 5 an n 5
$88- £8~8
88820 «Emzo

88. L
38 u 5
$88

88. L
:8 u 5
8:88-

:8. L
an n 5
888-

8:. L
an n 5
8888-

8~hL
Gm u 5
388-

8v~.uq
an n5
~58-

~m~.u8
an n 5
82 .8-

M00m20

38 u 5
88.8

38. L
A: u 5
8-~8

888. L
an n 5
vw-8

888.ua
38 n 5
3.88

88. L
Gm n 5
889.8

888.na
5 u 5
8:88

~88. L
38 u 5
$88

Mmm20

38. L
A: n 5
8-~8

:8 u 5
8888. _

82. L
3 "5
3:8

8:. L
9.8 u 5
888-

888. L
as. u 5
£88

888.n8
:8 u 5
888

8.8. L
9;. u 5
882 .8-

M30220

M8320

M30220

M0220

M00220

M2m20

30200m

M0200m

135

am n 5
88._

~88. L
98 u 5
8888

88. L
5 n 5
8888-

M30mm20

 

~88.uu
Gm u 5
8888

:8 u 5

88._
8%. L
Gm u 5
E88-

M92320

 

88. L
Am J 5
888-

-8.n8
an n 5
R88-

. L
an n 5
883

M00mm20

ooo.na
5 u 5
-~88

888.L
Am J 5
$88

~88. L
98 n 5
$88

Mmm20

-8.u8
:8 n 5
888

8~8. L
98 u 5
8288-

888.ua
9% n 5
:88-

M30220

M30mm20

M820

M00mm20

 

LIST OF REFERENCES

LIST OF REFERENCES

Aragona, J. A. & Eyberg, S. M. (1981). Neglected Children: Mothers' report of
child behavior problems and observed verbal behavior. Child Development. 52, 596-602.

 

Atkeson, B. M., & Forehand, R. (1978). Parent behavioral training for problem
children: An examination of studies using multiple outcome measures. Journal of

 

 

 

 

Abnormal Child Psychology, 6, 49-460.

Baldwin, R. L (1993). Effects of otitis media on child development. 115 9:“
American Journal of Otology, fi(6), 601-604. :

El

Bandura, A. (1977). Social learning theou. Englewood Cliffs, NJ: Prentice-Hall.

Barkley, R. A., and Cunningham, C. E. (1979). The effects of methylphenidate on i
the mother-child interactions of hyperactive children. Archives of General Psychiatg, 3_6_, 1
201-208.

 

Bates, J. E. (1980). The concept of difficult temperament. Merrill-Palmer
Quarterly, 26, 299-319.

Baumrind, D. (1988). Rearing competent children. In W. Damon (ed. ), Child
Development TodaLand Tomorrow. San Francisco: Jossey-Bass, 1988.

Baumrind, D. (1983). Rejoinder to Lewis's reinterpretation of parental firm

control effects: Are authoritative families really harmonious? Psychological Bulletin, 24
(1), 132-142.

Baumrind, D. (1975). Early socialization and the discipline controversy.
Morristown, NJ: General Learning Press.

Baumrind, D. (1971). Current patterns of parental authority. Developmental
Psychology Monographs, fl, (1, Part 2).

Baumrind, D. (1968). Authoritarian versus authoritative parental control.
Adolescence. ;, 255-272.

Baumrind, D. (1967). Child care practices anteceding three patterns of preschool
behavior. Genetic Psychology Monographs, 2_5_, 43-88.

Becker, W. C. (1964). Consequences of different kinds of parental discipline. In
M. L. Hoffman & L. W. Hoffman (Eds), Review of child development resear_ch (Vol.1).
New York: Russell Sage Foundation.

136

137

Bell, R. Q., Harper, L. V. (1977). Child effects on Adults. Hillsdale, N.J.:
Erlbaum.

Berman, S. (1981). Otitis media with effusion: Its relationship to language
development, intellectual fimctioning, and academic performance. Advgrces in Bglrafioral
Pediatrics, 2, 129-140.

Black, M. M., Gerson, L. F., Freeland, C. A. B., Nair, P., Rubin, J. S., &
Hutcheson, J. J. (1988). Language screening for infants prone to otitis media. Journal of
Pediatric Psychology, Q, 423-433.

Black, , M. M. & Sonnenschein, S. (1993). Early exposure to otitis media: A
preliminary investigation of behavioral outcomes. Journal of Develormiental and
Behavioral Pediatrics. 14, 150-155.

Bloom-Feshbach, J. (1981). Historical perspectives on the fathers’ role. In M. F.
Lamb (Ed), The role of the father in child development (pp. 71-112). New York: Wiley.

 

Bluestone, C. D., Fria, T. J ., Arjona, S. K., et al. (1986). Controversies in
screening for middle ear disease and hearing loss in children. Pediatrics, 71, 57-70.

Bluestone, C. D., et al. (1983). Special article: Workshop on effects of otitis
media on the child. Pediatrics, 11, No. 4, (April), 639-652.

Boshua, D. M., Twentyman, C. T. (1984). Mother-child interactional style in
abuse, neglect, and control groups: Naturalistic observation in the home. Journal of
Abnormal Psychology, 93 1, 106-114.

—,

Boyce, W. T., Jensen, E. W., Cassel, J. C., et al. (1977). Influence of life events
and family routines on childhood respiratory tract illness. Pediatrics, 6_0:609-615.

Brazelton, T. B. (1973). Neonatal Behavioral Assessment Scale. National
Spastics Society Monograph. Philadelphia: Lippincott.

Brownlee, R. C., DeLoache, W. R., Cowan, C. C., & Jackson, H. P. (1969).
Otitis media in children: Incidence, treatment, and prognosis in pediatric practice. Journal
of Pediatrics, 15, No.4, 636-642.

Budd, K. S., Green, D. R., & Baer, D. M. (1976). An analysis of multiple
misplaced parental social contingencies. Journal of Applied Behavior Analysis, 2, 459-
470.

Buss, A. H. & Plomin, R. A. (1975). A temperament theory of personalin
development. New York: Wiley.

138

Burgess, R. L. & Conger, R. D. (1978). Family interaction in abusive, neglectfirl,
and normal families. Child Development. 49, 1163-1173.

Campbell, S. B. (1973). Mother-child interaction in reflective, impulsive, and
hyperactive children. Developmental Psychology, 3, 341-349.

Campbell, S. B. (1975). Mother-child interaction: A comparison of hyperactive,
learning-disabled, and normal boys. American Journal of Orthopsychialry, 4;, 51-57.

Cantwell, D. P. & Baker, L. (1971). Psychiatric disorder in children with speech
and language retardation: A critical review. Archives of General Psychim, 3444 583-591.

Casey, B. (1983). Recurrent Serious Otitis Media: Its impact on the cognitive
performance and social_adjustment of pre-school children. Unpublished doctoral
dissertation, University of Virginia, Institute of Clinical Psychology.

-19 at 21! m3!
5'

Cass, R. & Kaplan, P. (1979). Middle ear disease and learning problems: A
school system's approach to early detection. Journal of School Hea1t_h, 557-560.

 

1 gun-l1.
'.
-:

Chan, J. 0, Logan, G. B., & McBean, J. (1967). Serous otitis media and allergy.
American J ourng of Diseased Children. 114. 684-690.

Chess, S. & Rosenberg, M. Clinical differentiation among children with initial
language complaints. Journal of AutisirLand Childhood Schizophreng. 1974, 4(2), 99-
109.

Cerreto, M. C. (1986). Developmental Issues in Chronic Illness: Implications and
Applications. Topics in Early Childhood Special Education. 5(4), 23-35.

Cohen, N. J ., Sullivan, J ., Minde, K., Novak, C., & Keens, S. (1983). Mother-
child interaction in hyperactive and normal kindergarten-aged children and the efl‘ect of
treatment. Child Psychiatgg and Human Development. 13(4), Summer, 213-224.

Cunningham, C. E., Reuler, E., Blackwell, J. and Deck, J. (1981). Behavioral and
linguistic developments in the interactions of normal and retarded children with their
mothers. Child Development. fl, 62-70.

Daly, K. A. (1991). Epidemiology of otitis media. Otolagagology Clinics of
North America, a, 775-786.

Darling, N. & Steinberg, L. (1993). Parenting Style as Context: An integrative
model. Psychological Bulletin. 113 (3), 487-496.

Delfini, L. F., Bemal, M. E., & Rosen, P. M. (1976). Comparison of deviant and

139

normal boys in home settings. In E. J. Mash, L. A. Hamerlynck, & L. C. Handy (Eds),
Behavior modification and families. New York: Brunner/Mazel.

Donovan, W. L., Leavitt, L. A., & Balling, J. D (1978). Maternal physiological
response to infant signals. Psychophysiology, 1;, 68-74.

Dowdney, L., Mrazek, D., Quinton, D., & Rutter, M. (1984). Observation of
parent-child interaction with two-to-three year olds. Jourpal of Child Psychology and
Psychiatg, 3, 379-409.

Downs, M. P. (1983). Audiologists overview of sequelae of early otitis media. In
Bluestone, Klein, et al. Pediatrics, 7_1, 639-652.

Dubin, E. R. & Dubin, R. (1963). The authority inception period in socialization.
Child Development. 34, 885-898.

Eisen, N. H. (1962). Some effects of early sensory deprivation on later behavior:
The quondam hard of hearing child. Journal of Abnormal Social Psycholcgy, Q, 338-
342.

Elder, G. H., Caspi, A., & Downey, G. (1988). Problem behavior and family
relationships: Life course and intergenerational themes. In A. Sorenson, F. Weineert, & L.
Sherrod (Eds), Human Development: Multidisciplipary Perspectives. Hillsdale, N.J.:
Elbaum.

 

Erikson, E. H. (1963). Childhood and society (2nd ed.). New York: Norton.

Eyberg, S. M. (1985). Behavioral assessment: Advancing methodology in
pediatric psychology. Journal of Pediatric Psychology, E, No. 2, 123-139.

Eyberg, S. M. & Matarazzo, R. G. (1980). Training parents as therapists: A
comparison between individual parent-child interaction training and parent group training.
Joumfi of Clinical Psychology, 3_6, No. 2, 492-499.

Eyberg, S. M. & Robinson, E. A. (1983). Dyadic Parent-Child Interaction Coding
System: A Manual. (Available from S. M. Eyberg, University of Florida, Gainsville, FL).

Eyberg, S. M. & Robinson, E. A. (1982). Parent-Child interaction training:
Effects on family functioning. Journal of Clinical Child Psychology, 11 No. 2, 130-137.

—,

Feagans, L. V. & Proctor, A. (1994). The effects of mild illness in infancy on
later development: The sample case of the effects of otitis media (middle ear effusion).
Develgimental Psycholpgy, 3_Q, 701-708.

Feagans, L., Sanyal, M., Henderson, F., et al. (1987). Relationship of middle ear

 

140

disease in early childhood to later narrative and attention skills. Journal of Pediatric
Psychology. L2, 581-594.

Forehand, R. L., King, H., Peed, S., & Yoder, P. (197 5). Mother-child
interactions: Comparison of a non-compliant clinic group and a non-clinic group.
Behavior Research and Therapy, 13, 79-84.

 

Forehand, R. L. & McMahon, R. J. (1981). Helping the noncompliant child: A
clinician's guide to parenting. New York: Guilford Press.

F orgays, D. K., Hasazi, J. E. & Wasserman, R. C. (1992). Recurrent otitis media

and parenting stress in mothers of two-year-old children. Developmental and Behavioral r-
Pediatrics, £6), 321-325.

Frank, S. J ., Freeark, K., & Breitzer, G. (1988). Research Proposal for Studying
ffanfilflactors in Otitis Medyia". Proposal submitted for review by Lilly Research
Laboratories, A division of Eli Lilly and Company.

 

Freeark, K., Frank, S. J ., Wagner, A. E., Lopez, M. L., Olmsted, C. & Girard, R. F.»—
(1992). Otitis media, language development, and parental verbal stimulation. Journal of
Pegatric Psychology, fl(2), 173-185.

Friel-Patti, S. & Finitzo, T. (1990). Language learning in a prospective study of
otitis media with effusion in the first two years of life. Journal of Speech and Hearing
Research, 22, 188-194.

Furstenberg, F. F. (1988). Good dads-bad dads: Two faces of fatherhood. In A.
J. Cherlin (Ed), The changing American family and public policy (pp. 193-218).
Washington, DC: Urban Institute.

Gardner, F. E. M. (1987). Positive interaction between mothers and conduct-
problem children: Is there training for harmony as well as fighting? Journal of Abnormal
Child Psychology, Q, No. 2, 283-293.

Goldsmith, H. H. & Campos, J. J. (1982). The concept of temperament in human
development. In R. N. Emde & R. J. Harmon (Eds), Development of attachment and
affiliative systems. New York: Plenum Press.

Gottleib, M. I., Zinkus, P. W., & Thompson, A. (1979). Chronic middle ear
disease and auditory perceptual deficits: Is there a link? Clinigl Pediatrics. 18, 725-732.

Gravel, J. S., McCarton, C. M. & Ruben, R. J. (1988). Otitis media in neonatal
intensive care unit graduates: A 1-year prospective study. Journal of Pediatrics, 82, 44-
49.

141

Green, K. D., Forehand, R., & McMahon, R. J. (1979). Parental manipulation of
compliance and noncompliance in normal and deviant children, Behavior Modification. 3,
245-266.

Guemey, B. (1964). Filial Therapy: Description and rationale. Journal of
Consultiag Psychology, 28, 4, 304-310.

Hanf, C. & Kling, F. (1973). Facilitating parent-child interaction: A two stage_
training model. Unpublished manuscript, University of Oregon Mdeical School.

Harsten, F., Prellner, K., & Heldrep, J. (1989). Recurrent otitis media: a
prospective study of children during the first three years of life. Acta Otolaryragology, '9‘"
M, 11-15.

Hatfield, J. S., Ferguson, L. R., & Alpert, R. (1967). Mother-child interaction
and the socialization process. Child Develgment. 18, 365-414.

 

Hefferman, A. (195 5). A psychiatric study of 50 pre-school children referred to 32
hospital for suspected deafiiess. In Caplan, J. (ed): Emotional Problems of Eaglg fa...-
Childhood, New York, Basic Books Inc., p. 269.

Hersher, L. (1978). Minimal brain dysfunction and otitis media. Perceptual Motor
Skills, 12, 723-726.

 

Hetherington, E. M. & Martin, B. (1979). Family interaction. In H. Quay & J.
Werry (Eds), Psychological disorders of childhood. New York: Wiley.

Holm, V. A. & Kunze, L. H. (1969). Effect of chronic otitis media on language
and speech development. Pediatrics, 43, 833-839.

 

Howie, V. M. (1980). Developmental sequelae of chronic otitis media: A review.
Journal of Developmental Behavioral Pediatrics, _1_, 34-38.

 

Humphries, T., Kinsboume, M., & Swanson, J. (1978). Stimulant effects on co-
operation and social interaction between hyperactive children and their mothers. Journal
of Child Psychology & Psychiatg 19 13-22.

 

Ingram, T. T. S. (1959). Specific developmental disorders of speech in children.
Brain, _8_2_, 450-467.

Iverson, T. (1987). The developmental effects of ea__r;and upper respiratog
problems on behavior. Paper presented at the Twenty-First Annual Meeting of the
Association for Advancement of Behavior Therapy, Boston, MA.

Jacob, T., Tennenbaum, D. L. & Krahn, G. (1987). Factors influencing the

142

reliability and validity of observation data. In Jacob (Ed) F_arnily Interaction and
Psychopathology: Theories. Methods a_n_d Finding; New York, Plenum Press.

Johnson, S. M., & Lobitz, C. K. (1974). The personal and marital adjustment of
parents as related to observed child deviance and parenting behavior. Journal of
Abnorma Child Psychology, 2 193-207.

..J

Jouriles, E. N., Barling, J ., & O'Leary, K. D. (1987). Predicting child behavior
problems in maritally violent families. Journal of Abnormal Child Psychology, Q, 165-
174.

Katz, J. (1978). The effects of conductive hearing loss on auditory function. Asha,
20, 879-886.

Koch, H. (1974). Dennison NJ: Office visits to Pediatricians. National Medical
Care Service, National Center for Health Statistics.

Kochanska, G. (1997). Mutually Responsive Orientation between Mothers and
Their Young Children: Implications for Early Socialization, Child Development. _6_§ (1),
94-112.

Kochanska, G., Kuczynski, L., Radke-Yarrow, M., Welsh, J. D. (1987).
Resolutions of control episodes between well and affectively ill mothers and their children.
Journal of Abnormal Child Psycholagy, 15(3 1, 441-456.

Kopp, C. B. (1987). The grth of self-regulation: caregivers and children. In

Eisenberg, N. (Ed) Contemporag topics in developmental psychology, New York: John
Wiley & Sons, 34-55.

Kuczynski, L., Kochanska, G., Radke-Yarrow, M., & Girnius-Brown, O. (1987).
A developmental interpretation of young children's noncompliance. Developmental
Psychology, 23(6), 799-806.

Lamb, M. E. (1997). The role of the father in child development. New York:
John Wiley & Sons.

Lamb, M. E. (1978). Social interaction in infancy and the development of
personality. In M. E. Lamb (Ed), So®& personalig development. New York: Holt,
Rinehart, & Winston.

Lamb, M. E., Pleck, J. H., Charnov, E. L. & Levine, J. A. (1987). A biosocial
perspective on parental behavior and involvement. In Lancaster, J. B., Altman, J ., Rossi,
A. S. & Sherrod, L. R. (Eds), Parenting across the lifespan: Biosocial dimensions (pp.
111-142). New York: Aldine de Gruyter.

 

143

Lay, K. L., Waters, E., & Park, K.A. (1989). Maternal Responsiveness and Child
Compliance: The role of Mood as a Mediator. Child Development. @, 1405-1411.

Lee, C. L. & Bates, J. E. (1985). Mother-child interaction at age two years and
perceived difficult temperament. Child Development. E, 1314-1325.

Lehmann, M. D., Charron, K., Kummer, A., Keith, R.W. (1979). The effects of
chronic middle ear efiusion on speech and language development: a descriptive study.
International Journal of Pediatric Otorhinolmgalpgy, 1979, 1, 137-144.

Lemanek, K. L., Stone, W. L. & F ishel, P. T. (1993). Parent-child interactions in
handicapped preschoolers: The relation between parent behaviors and compliance. n.
Journi of Clinical Child Psychology, 2(1), 68-77.

Lepper, M. (1981). Intrinsic and extrinsic motivation in children: Detrimental
effects of superfluous social controls. In W.A. Collins (Ed), Aspects of the development .
of conaetence: The Minnesota symposium on child psychology, Vol. 14, Hillsdale, N.J.: ,
Erlbaum.

 

Lesser, S. R. & Easser, B. R. (1972). Personality differences in the perceptually
handicapped. Joumaly of Child Psychiatg, 12, 458-465.

Lobitz, G. K. & Johnson, S. M. (1975). Normal versus deviant children. Journal
of Abnormal Child Psycholagy, 2(4), 353-374.

Lytton, H. & Romney, D. M. (1991). Parents’ differential socialization of boys
and girls: A meta-analysis. Psychological Bulletin. 109. 267-296.

Lytton, H. (1979). Disciplinary encounters between young boys and their mothers:
Is there a contingency system? Developmental Psychology, Q, 256-268.

Lytton, H. (1974). Comparative yield of three data sources in the study of parent-
child interaction. Merrill-Palmer Quarterly, 53-64.

Luria, A. (1961). The role of speech in the regalation of normal and abnormal
behaviors. New York: Liveright.

Luria, A. (1969). Speech and formation of mental processes. In M. Cole & I.
Maltzman (Eds), A handbook of contrrnporary Soviet psychology. New York: Basic
Books.

Maccoby, E. E. (1992). The role of parents in the socialization of children: An
historical overview. Developmflal Psychology, 28 1006-1017.

—,

Maccoby, E. E. (1990). Gender and relationships. American Psychologiat, 515,

144
513-520.

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family:
Parent-child interactions. In E. M. Hetherington (Ed), findbook of child psychology;
Vol. 4. SociaLization, personality and social development (pp. 1-101). New York: Wiley.

MacDonald, K. (1992). Warmth as a Developmental Construct: An Evolutionary

Analysis. Child Development. Q, 753-773.

Mahler, M. S., Pine, F., & Bergman, A. (1975). The psychological birth of the
human infant. New York: Basic Books.

Marchant, C. D., Shurin, P. A., Turczyk, V. A., et al. (1984). Course and
outcome of otitis media in early infancy: A prospective study. Joumaflf Pediatrics. 104.
826-831.

Marsiglio, W. (1995). Fathers’ diverse life course patterns and roles: Theory and
social interventions. In W. Marsiglio (Ed), firtherhood: Contemporaryatheory. research.
and social policy. Thousand Oaks, CA: Sage.

Martin, B. (1987). Developmental Perspectives on Family Theory and
Psychopathology. In T. Jacob (Ed), F_amily Interaction and Psychopathology: Theories.
Methods. and Finding; New York: Plenum.

Mash, E. J. and Johnston, C. (1982). A comparison of the mother-child
interactions of younger and older hyperactive and normal children. Child Development.
52, 1371-1381.

Mattison, R. E., Cantwell, D. P., & Baker, L. (1980). Dimensions of behavior in
children with children and language disorders. Journal of Abnoml Child Psychology,
1980, 8(3), 323-338.

McCarthy, D. McCarthy SQes of Children's Abilities. New York: The
Psychological Corporation, 1970.

Menyuk, P. (1977). Effects of hearing loss on language acquisition in the
babbling stage. In B. Jaffe (ed) Hearing Loss in Children. Baltimore: University Park
Press.

Needleman, H. (1977). Effects of hearing loss from early recurrent otitis media on
speech and language development. In Jaffe, B. (ed), Hearing loss in Children. Baltimore,
MD, University Park Press, 640-649.

Northern, J. L. & Downs, M. P. (1974). Hearing in children. Baltimore: Williams
and Wilkins.

II. it.

'5z4M 4.:-

 

145

Palkovitz, R. (1997). Reconstructing “involvement”: Expanding
conceptualizations of men’s caring in contemporary families. In A. J. Hawkins & D. C.
Dollahite (Eds), Generative fathering: Beyond deficitaperspectives. Thousand Oaks, CA:
Sage.

Paradise, J. L. (1981). Otitis media during early life: how hazardous to
development? A critical review of the evidence. Pediatrics, 68(6), 869-873.

Parpal, M. & Maccoby, E. E. (1985). Maternal responsiveness and subsequent
child compliance. Child Development. 5_6, 1326-1334.

v1." .5 A“. J
I

Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castillia Press.

Patterson, G. R. (1976). The aggressive child: Victim and architect of a coercive
system. In E. J. Mash, L. A. Hamerlynck, & L. C. Handy (Eds) Behavior modification
and families: Vol. 1. Theog and research. New York: Brunner/Mazel.

 

Patterson, G. R. (1975). A three-stage functional analysis for children's coercive pr
behavior: A tactic for developing a performance theory. In B. C. Etzel, J. M. LeBlanc, &
D. M. Baer (Eds), New developments in bepayioral research: Theory. methodsyaad
applications. Hillsdale, N.J.: Lawrence Erlbaum.

Peed, S., Roberts, M., & Forehand, R. (1977). Evaluation of the effectiveness of
a standardized parent training program in altering the interaction of mothers and their
noncompliant children. Behavior Modification. 1(3), 323-349.

Perrin, E. C. & Gerrity, P. S. (1984). Development of children with a chronic
illness. Pecfltric Clinics of North America. 3_1(1), 19-31.

Pleck, J. H., Lamb, M. E., & Levine, J. A. (1986). Epilog: Facilitating future
change in men’s family roles. Marriage and the Family Review. _9_ (3/4), 11-16.

Pukander, H. Sipila, M. & Karma, P. (1984). Occurrence of and risk factors in
acute otitis media. In: Lim, D. J ., Bluestone, C. D., Klein, J. 0., & Nelson, J. D. (eds)
Recent advances in otitis media with effusifl. Philadelphia: BC Decker, Inc., pp. 9-13.

Reed, J. B., Taplin, P. S., & Lorber, R. (1981). A social interactional approach to
the treatment of abusive families. In R. B. Stuart (Ed), Violent behavior: Social learning
approaches to prediction. management, and treatment. New York: Brunner/Mazel.

Roberts, J. E., Burchinal, M. R. & Campbell, F. (1994). Otitis media in early
childhood and patterns of intellectual development and later academic performance.
Joranal of Pediatric Psychology, Q (3), 347-367.

146
Roberts, J. E., Burchinal, M. R., Medley, L. P., Zeisel, S. A., Mundy, M., Roush,
J ., Hooper, 8., Bryant, D., & Henderson, F.W. (1995). Otitis media, hearing sensitivity,
and maternal responsiveness in relation to language during infancy. The Journal of
Pediatrics, 126(3), 481-489.

Roberts, M. W., McMahon, R. J ., Forehand, R., & Humphreys, L. (1978). The
effect of parental instruction-giving on child compliance. Behavior Therapy, 2, 793-798.

Rocissano, L., Slade, A., Lynch, V. (1987). Dyadic synchrony and toddler
compliance. Developmental Psychology, 23(5), 698-704.

Roe, A. & Siegelman, M. (1963). A parent-child relations questionnaire. Child
Development. 21, 355-369.

Rothbart, M. K. & Derryberry, D. (1981). Development of individual differences
in temperament. In M. E. Lamb & A. L. Brown (Eds), Advances in developmenta
psychology (Vol. 1). Hillsdale, NJ. :Lawrence Erlbaum Associates.

Rothbaum, F., Schneider, Rosen, K., Pott, M., & Beatty, M. (1995). Early
Parent-child relationships and later problem behavior: A longitudinal study. Merrill
Palmer Quarterly. 41, 133-151.

Sameroff, A. J ., & Chandler, M. J. (1975). Reproductive risk and the continuum
of caretaking casualty. In Horowitz, F. D., Hetherington, M., Scarr-Salapatek, S., &
Siegel, G. (Eds), Review of child development resear_ch (Vol. 4), Chicago: University of
Chicago Press.

Schaefer, E. S. (1959). A circumplex model for maternal behavior, Journal of
Abnormal and Social Psycholrgy, 3, 226-235.

Schaefer, E. S. (1965). A configurational analysis of children's reports of parent
behavior, Journal of Consulting Psychology, 2(6), 552-557.

Schafl‘er, H. R. & Crook, C. K. (1980). Child compliance and maternal control
techniques. Developmental Psychology, 16 54-61.

—)

Schappert, S. M. (1992). Office visits for otitis media: United States, 1975-90.
Advance data from Vital and Health Statistics of the Centers for Disease Control, 214.
Hyattsville, MD, US. Department of Health and Human Services, Public Health Services,
Centers for Disease Control.

Schnelle, J. F. (1974). A brief report on the invalidity of parent evaluations of
behavior change. Journal of Applied Behavior Analysis, 1, 341-343.

 

Schutz, W. C. (1966). The interpersonal underworld: FIRO-B. Palo Alto, CA:

 

 

147
Science and Behavior Books.

Sears, R., Maccoby, E., & Levin, H. (1957). Patterns of Child Rearing. Evanston,
111.: Row, Peterson.

Smetana, J. G. (1994). Parenting Styles and Beliefs About Parental Authority,
New Directions for Child Development. 6_6, 21-36.

Smetana, J . (1989). Toddlers’ social interactions in the context of moral and
conventional transgressions in the home. Developmental Psychology, 2_5_, 499-508.

Snyder, J. J. (1977). A reinforcement analysis of intervention in problem and
nonproblem children. Journal of Abnormal Psychology, &(5), 528-535.

Stayton, D. J ., Hogan, R., & Ainsworth, M. D. S. (1971). Infant obedience and
maternal behavior: The origins of socialization reconsidered. Child Development. 12,
1057-1069.

Stein, R. E. K. & Jessop, D. J. (1984). Relationship between health status and
psychological sdjustment among children with chronic conditions. Pediatrics, 73(2), 169-
174.

 

Steinmetz, S. K. (1979). Disciplinary techniques and their relationship to
aggressiveness, dependency, and conscience. In Burr, W. R., Hill, R., Nye, F. I., & Reiss,
I. L. (Eds) Contemporary theories about the family. (Vol. 1), Free Press.

Stevenson, J. & Richman, N. (1978). Behavior, language, and development in
three year old children. Journal] of Autism and Childhood Schizophrenia, 8(3), 299-313.

Stool, S. E., Berg, A. O., Berman, S., Carney, C. J ., Cooley, J. R., Culpepper, L.,
Eavey, R. D., F eagans, L. V., Finitzo, T., Friedman, E. M., et al. (1994). Otitis media
with efl‘usion in young children: Clinical practice guideline, number 12 (AHCPR
Publication no. 94-0622). Rockville, MD: US. Department of Health and Human
Services, Agency for Health Care Policy and Research, Public Health Service.

Straus, M. A. & Tallman, I. (1971). SIMFAM: A technique for observational
measurement and experimental study of families. In Adous, J ., Condon, T., Hill, R.,
Straus, M., and Tallman, I. (Eds), family problem solving. Hinsdale, 111.: Dryden Press.

Symonds, P. M. (1939). The psychology of parent-child relationships. New
York: Appleton-Century.

Tavormina, J. B., Boll, T. J., Dunn, N. J., Luscomb, R. L. & Taylor, J. R. (1981).
Psychosocial effects on parents of raising a physically handicapped child. Journal of
Abnormal Child Psychology, _9_, No.1, pp. 121-131.

 

148

Teele, D. W., Klein, J. 0., Chase, C., Menyuk, P., Rosner, B. A., eta]. (1990).
Otitis media in infancy and intellectual ability, school achievement, speech and language at
age 7 years. The Journal of Infectious Diseases, 162, 685-694.

Teele, D. W., Klein, J. O., Rosner, B. A., eta]. (1989). Epidemiology of otitis
media during the first seven years of life in children in greater Boston: A prospective
study. The Journaflf Infectious Diseas_e_s, 160(1), 83-94.

Teele, D. W., Klein, J. O., Rosner, B. A., et.al. (1984). Otitis media with effusion
during the first three years of life and development of speech and language. Pediatrics, E,
No. 2, (August) 282-287.

Teele, D. W., Klein, J. O., Rosner, B. A. (1980). Epidemiology of otitis media in
children. Annals of Otolgy. Rhinology, & Larmgology, fl, (Supplement 89), 5-6.

Terdel, L. G., Jackson, R. J ., & Garner, A. M. (1976). Mother-child interactions:
A comparison between normal and developmentally delayed groups. In E. J. Mash, L. A.
Hamerlynck, & L. C. Handy (Eds), I_3_e1iavior modificationa and families. New York:
Brunner/Mazel.

Thomas, A., Chess, 8., & Birch, H. G. (1968). Temperaaentand behavior
disorders in children. New York: New York University Press.

Thomas, A., Chess, 8., Birch, H. G., Hertzig, M. E., & Korn, S. (1963).
Behavioral individuality irfiarly childhood. New York: New York University Press.

Travis, G. (1976). Chronic Illness in Children. Stanford, CA: Stanford University
Press, 1-9, 42-49, 164-197, 233-268, 343-366, 367-402.

Ventry, I. M. Effects of conductive hearing loss: fact or fiction. Journal of Speech
and Hearing Disorders, 1980, Q, 143-156.

Vernon-Feagans, L. V., Manlove, E. E. & Volling, B. L. (1996). Otitis media and
the social behavior of day-care-attending children. Child Development. 1996, 61, 1528-
1539.

Vygotsky, L. (1962). Thought and Language. New York: Wiley.

Wahler, R. G., Hughey, J. B., & Gordon, J. S. (1981). Chronic patterns of
mother-child coercion: Some differences between insular and noninsular families. Analysis
and intervention in development disabilities. 1(2), 145-156.

Wakschlag, L. S. & Hans, S. L. (1997). Relation of Early Parenting to the
Development of Disruptive Behavior Disorders in High Risk Youth. Paper presented at

 

149
the Biennial Meeting of the Society for Research in Child Development, Washington, DC.

Wallace, 1. F., Gravel, J. S., Schwartz, R. G. & Ruben, R. J. (1996). Otitis Media,
Communication Style of Primary Caregivers, and Language Skills of 2 Year Olds: A
Preliminary Report. Developmental and Behavioral Pediatrics, 17 (1), 27-35.

Webster-Stratton, C. (1985). Comparisons of behavior transactions between
conduct-disordered children and their mothers in the clinic and at home. Journal of
Abnormal Child Psychology, Q0), 169-184.

Webster-Stratton, C., & Eyberg, S. M. (1982). Child temperament: Relationship
with child behavior problems and parent-child interactions. Journal of Cliniaal Child

Psychology, 11(2), 123-129.

Wulbert, M., Inglis, S., Kriegsmann, E., & Mills, B. Language delay and
associated mother-child interactions. Developmental Psychology, 1975, 11,61-70.

Wylie, H. L., F ranchack, P., & McWilliams, B. J. (1965). Characteristics of
children with speech disorders seen in a child guidance center. Perceptual and Motor
Skills, 22,1101-1107.

Yarrow, M. R., Campbell, J. D., and Burton, R. V. (1968). Child-rearing: An
inquiry into research a_nd methods. San Francisco: Jossey-Bass.

Zinkus, P. W. & Gottlieb, M. I. (1980) Patterns of perceptual and academic
deficits related to early chronic otitis media. Pediatrics, 66(2), 246-252.

 

"I11111111111111111111113