.m. 5%» in: W?
a gerfimzw
:5... L

M...

- y. I. i
. ‘7']

:1: mm”.
..

Ecémq

 

This is to certify that the
thesis entitled

THE RELATIONSHIP OF FAMILY-RELATED
DEVELOPMENTAL ASSETS TO POSITIVE VALUES AND
SOCIAL COMPETENCIES

presented by

BARBARA D. HILLAKER

has been accepted towards fulfillment
of the requirements for the

MS. degree in Family and Child Ecologx

 

 

 

Major Professor's'SIgnamre' '

/9~/11/Q.1

Date

0., ,
a

MSU is an Allinnetive ActionIEquel Opportunity Institution

 

 

LIBRARY

Michigan State

University

 

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

15%“, 792$”

U7

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65-p.15

 

THE RELATIONSHIP OF F AMILY-RELATED DEVELOPMENTAL
ASSETS TO POSITIVE VALUES AND SOCIAL COMPETENCIES

By

Barbara D. Hillaker

A THESIS

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

MASTER OF SCIENCE
Department of Family and Child Ecology

2004

ABSTRACT

THE RELATIONSHIP OF FAMILY-RELATED DEVELOPMENTAL ASSETS TO
POSITIVE VALUES AND SOCIAL COMPETENCIES

By

Barbara D. Hillaker

This study examined the relationships of three parenting processes, independently
and jointly, to positive values and social competencies in middle school students. The
processes, or “family-related developmental assets” studied were: maintaining standards
(demandingness), positive family relationships (responsiveness), and positive family
communication.

The study utilized secondary data from 10,623 Michigan middle school students
who took the Search Institute’s Profiles of Student Life: Attitudes and Behavior survey in
1998/ 1999. Stepwise regression procedures found that while all three parenting
dimensions were positively associated with positive values and social competencies,
positive family communication had the strongest predictive power. Positive family
communication was the family variable most lacking among students.

Demographic variables were also entered into the analysis. Males scored
significantly lower than females on positive values and social competencies, but not on
family variables. Grade level, family structure, and mother’s educational level showed

some small but statistically significant relationships with outcome variables.

 

_—r a..- WEN—”m“, , .

Cepyright by
BARBARA D. HILLAKER
2004

 

 

 

 

 

 

 

"P
Pm I:

 

 

 

ACKNOWLEDGEMENTS

I want to give special thanks to all who have helped make this thesis possible: to
my committee members, Dr Joanne Keith, Dr. Francisco Villarruel, and Dr. June Youatt.
My major professor, Dr. Keith, initiated this project and was very supportive throughout
the process with her warmth, humor, direction, and encouragement. Dr. Villarruell’s
input sharpened and shaped the final draft, and Dr. Youatt’s assistance led to my research
focus on parenting adolescents. I also want to thank Dr. Dawn Contreras and Dr. Bruce
Haas, whom I consider defacto committee members. I am greatly indebted to the
generosity of Dr. Haas who taught me how to use SPSS and spent many hours both
teaching and directing the statistical operations involved in this study. Dr. Contreras took
me under her wing, introduced me to programming for the parents of adolescents, and
never failed to brighten my day and inspire me.

I also want to thank my family. My heartfelt gratitude to my husband and
parenting partner, Glen, who believed I could and should write this thesis before I did. I
couldn’t have done this without his support. I want to acknowledge my children, Peter,
Earl, Anna, and John who have taught me that parenting adolescents can provide
immeasurable joy. It is to them and their generation that I dedicate this thesis. I also
thank my parents who listened, discussed ideas, and encouraged me—they are the ones
who first taught me that parenting is a noble enterprise worthy of dedication and

commitment.

iv

 

 

IlEl

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................... vii
LIST OF FIGURES ................................................................................................. ix
CHAPTER 1
INTRODUCTION .................................................................................................... 1
Rationale for the Study .................................................................................. 1
Theoretical Framework ............................................................. 8
Application of Theoretical Framework to This Study ........................... 12
CHAPTER 2
LITERATURE REVIEW .................................................................. 13
Early Adolescence ................................................................... l3
Puberty ............................................................................................... 13
Primary Developmental Tasks of Adolescence ................................. 15
Identity Formation ................................................................. 16
Social Changes ....................................................................... 16
The “Asset Framework” ................................................................................ 17
“Positive Values” Assets .................................................................... 19
“Social Competencies” Assets ............................................. 23
Family—Related Assets ....................................................................... 26
Positive Family Communication ............................................ 27
Parenting Styles ..................................................................... 3O
Authoritative Parenting Style ..................................... 31
Research Supporting Directional Impact ............................... 38
Demographic Variables ................................................................................. 40
Grade Level ........................................................................................ 40
Gender ................................................................................................ 41
Mother’s Education ............................................................................ 41
Family Structure ................................................................................. 42
Measurement Issues ....................................................................................... 42
CHAPTER 3
METHODOLOGY .......................................................................... 44
Research Hypotheses ..................................................................................... 44
Research Questions ........................................................................................ 45
Participants ........................................................................... 46
Measures .............................................................................. 46
Demographic Variables ............................................................. 47
Independent Variables ............................................................... 48
Dependent Variables ................................................................. 51
Procedures ............................................................................ 54

 

 

CHAPTER 4
RESULTS ..................................................................................... 55
Research Question 1: What are the Levels of Each Family Variable
in Michigan Middle School Students? ....................... 55
Research Question 2: What are the Levels of Positive Values and
Social Competencies among Michigan

Middle School Students? ........................................... 57
Research Question 3: What are the Relationships between each Family
Variable and Positive Values? ................................... 58
Research Question 4: What are the Relationships between each Family
Variable and Social Competencies? ........................... 65
Research Question 5: What are the Relationships Among the Family
Variables? .................................................................. 71

Research Question 6: What are the Relationships among Demographic
Variables, Family Variables, and Positive Values

and Social Competencies? ......................................... 73
Grade .................................................................................................. 80
Gender ................................................................................................ 82
Two-Parent Home .............................................................................. 84
Educational Level of the Mother ....................................................... 85
CHAPTER 5
CONCLUSIONS, DISCUSSION, AND IMPLICATIONS .................................. 86
Review of Major Findings ............................................................................. 86
Findings About Family Variables ...................................................... 86
Findings From Demographic Variables ............................................. 88
Implications .................................................................................................... 90
Positive Values, Social Competencies and
Ecological Considerations ................................................................. 90
Program Development ....................................................................... 91
Program Evaluation ........................................................................... 95
Limitations and Implications for Future Research ......................................... 96
APPENDIX ............................................................................................................... 102
University Consent ......................................................................................... 103
REFERENCES ......................................................................................................... 104

vi

 

 

 

 

 

 

 

4.3

Table
Posm

4.413;
4-44.51'
4.4ch

Table
Social

4.51.3)

LIST OF TABLES

TABLE Page
3.1 Number of Students in Each Grade ............................................................... 46
3.2 Survey Items Used for Family Variables ....................................................... 50
3.3 Survey Items Used for Social Competencies ................................................. 52
3.4 Survey Items Used for Positive Values ......................................................... 53
4.1 Descriptive Statistics for Family Variables ................................................... 56
4.2 Descriptive Statistics for Dependent Variables ............................................. 58
4.3 Correlations between Family Variables and Outcome Variables .................. 61

Table 4.4 (a-c): Results of SPSS Stepwise Regression for
Positive Values--Only Family Variables Entered.

4.4(a) Regression Model Summary for Positive Values .......................................... 63
4.4(b) ANOVA Results for Positive Values ............................................................. 64
4.4(c) Coefficients in Regression Model for Positive Values .................................. 65

Table 4.5 (a-c): Results of SPSS Stepwise Regression for
Social Competencies--Only Family Variables Entered

4.5(a) Social Competencies Model Summary .......................................................... 69
4.5(b) ANOVA Results from Regression Model for Social Competencies ............. 70
4.5(c) Coefficients from the Regression Model for Social Competencies ............... 71
4.6 Correlations Among the F amin-Related Variables ....................................... 72

4.7 Demographic Variables Correlations with Family Variables, and
Outcome Variables ......................................................................................... 74

vii

Table 4.8 (a-c): Results of SPSS Stepwise Regression for Positive Values—-Family and
Demographic Variables Entered

4.8(a) Statistically Significant Variables Entered in Order of Relative Strength ..... 7S
4.8(b) Regression Model Summary for Positive Values .......................................... 75
4.8(c) ANOVA Results for Positive Values ............................................................. 76

Table 4.9 (a-c) Results of SPSS Stepwise Regression for Social Competencies--Family
and Demographic Variables Entered

4.9(a) Statistically Significant Variables Entered in Order of Relative Strength ..... 77
4.9(b) Regression Model Summary for Social Competencies ................................. 78

49(0) ANOVA Results for Social Competencies .................................................... 79

viii

1.1

1.2

1.3

1.4

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

4.10

4.11

4.12

4.13

4.14

LIST OF FIGURES

Theoretical Model Proposed for this Study (for Positive Values) ................. 7
Theoretical Model Proposed for this Study (for Social Competencies) ........ 7
Bronfenbrenner’s Theory of Human Development ....................................... 8
Youth Embedded in Interacting Systems ....................................................... 11

Percentages of Students Reporting High and Low Levels of
Family Variables ........................................................................................... 57

The Relationship Between Positive Family Communication and
Positive Values ............................................................................................... 59

The Relationship Between Positive Family Relationships and

Positive Values .............................................................................................. 59
The Relationship Between Maintaining Standards and Positive Values ....... 60
The Variance in Positive Values Explained by Family Variables ................. 62

The Relationship Between Positive Family Communication and
Social Competencies ...................................................................................... 67

The Relationship Between Positive Family Relationships and
Social Competencies ...................................................................................... 67

The Relationship Between Maintaining Standards and
Social Competencies ..................................................................................... 68

The Variance in Social Competencies Explained by Family Variables ........ 68

Bar Graph Showing Mean Social Competencies by Grade Level ................. 81
Bar Graph Showing Mean Positive Values by Grade Level .......................... 81
Bar Graph Showing Mean Communication by Grade Level ........................ 82
Bar Graph Showing Mean Positive Values by Gender .................................. 83
Bar Graph Showing Mean Social Competencies by Gender ......................... 84

ix

CHAPTER 1

INTRODUCTION

Most parents want to raise “good kids”---children who are responsible, caring,
honest, and hard-working, and who grow into adults with these same qualities. Society,
naturally has a vested interest in these outcomes as well, so that competent, healthy adults
with positive values and social competencies, become the next leaders to better society
and effectively nurture the next generation. However, it is ofien the children and
adolescents with risk factors who become the focus of research, with the main endeavor
that of preventing or fixing serious problems, such as drug abuse or teen pregnancy.
Brown, D’Emidio-Caston, and Benard (2001) state that a "cross-discipline, integrated
look at nearly 40 years of research tells us that identifying and nurturing an individual's
capacities rather than focusing on his or her deficits creates a capable, productive, and
compassionate person" (p. viii).

Rationale for the Study

In the last decade or so, a new stream of research has begun to counter the
problem-focused perspective outlined above. Such research can be found at the Search
Institute in Minnesota which has taken a more positive, holistic approach towards youth
development, studying what internal and external (family, school and community)
variables comprise the “developmental assets” or “building blocks” required for all
children to grow into healthy, responsible, caring citizens (Keith, Huber, Griffin, &
Villarruel, 2002; Scales, 1999). These assets collectively are associated with thriving

behaviors, and the research base from which the assets are identified demonstrates a wide

range of positive outcomes associated with most of the assets (Leffert et al., 1998; Scales,
Benson, Leffert, & Blyth, 2000).

However, most of the initial studies examining the effects of developmental
assets, as identified in Search Institute’s 40 asset framework, have looked at the
relationships between the number of developmental assets adolescents possess and the
various risk behaviors they display, rather than examining the various associations with
positive outcomes (Keith & Perkins, 1995; Leffert, et al., 1998). The large data set,
generated by the surveys developed by Search Institute researchers, also presents an
opportunity to look at processes that might be associated with desired outcomes, such as
positive values and social competencies.

The Search Institute’s asset framework was informed by a body of previous
research performed on adolescence (Scales, Leffert, & Lerner, 1999). The assets of
family boundaries and family support were identified and defined as representing family
processes associated with numerous positive outcomes in youth (Leffert, et al., 1998;
Scales, et al., 1999). These two family processes correspond to what Smetana (1995, p.
299) calls the “two orthogonal dimensions of demandingness and responsiveness” widely
used in the typology of parenting styles developed by Baumrind (Baumrind, 1975, 1991).

Further, in the synthesis of scientific research supporting developmental assets,
family support and positive family communication have often been dealt with as a unit
(Scales, et al., 1999). However, although a substantial body of literature simultaneously
looks at the parenting processes conceptually linked to family boundaries and family
support, this same body of research does not routinely identify and utilize positive family

communication as a core variable (Mattanah, 2001; Roberts & Steinberg, 1999;

Shucksmith & et al., 1995; Steinberg, 2001). While, positive family communication is
addressed in some of the parent-education programming, along with limit-setting and
relationship-building skills (Rick Kosterrnan, Hawkins, Spoth, Haggerty, & Zhu, 1997;
Kumpfer & Tait, 2000), in the research literature, it is often examined alone (Jaccard,
Dittus, & Gordon, 2000; Lytle, Bimbaum, Boutelle, & Murray, 1999; Miller-Day, 2002).
This study capitalized on the opportunity to examine these three parenting processes
together: family boundaries, family support, and positive family communication.

The importance of families to developmental outcomes for youth and children has
been well documented (Kumpfer & Kafiarian, 2000; Scales, et al., 1999). The processes
of maintaining standards (demandingness) and developing positive relationships
(responsiveness) have particularly strong research backing. Thus the primary
independent variables for this study became the three family processes of maintaining
standards, developing positive family relationships, and having positive family
communication. These processes may be affected by demographic variables, however,
and this had to be accounted for in the study. Therefore, the demographic variables
relating to the adolescent student were gender and grade level, while the demographic
variables accounted for that represented the external influences were family structure and
the educational level of the mother.

The study, itself, examined the relationships between positive parenting and
positive values and social competencies in middle school students. The relationships
between parenting styles and many developmental outcomes in children and adolescence
have been studied extensively. However, studies examining the relationship between

family dynamics and the outcomes of positive values and, to a lesser extent, social

competencies are relatively scarce (Scales, et al., 1999). The social competency assets
have a stronger research base than the positive values assets (Scales, et al., 1999).
Parenting styles have been related to various measures of social competence (Leffert, et
al., 1998). Fewer studies evaluating social competence outcomes include positive family
communication, the demandingness/standards dimension of parenting and the
responsiveness or positive family relationships dimension.

This study, therefore, replicated and expanded on previous studies of the
relationship between family variables and social competencies in youth. Positive values
and social competencies, including good decision-making and interpersonal skills, are
characteristics expected of adults as competent, productive citizens. Thus, positive
values and social competencies were the dependent variables selected for this study.

Although Search data have been widely disseminated as summary data and the
results have been used to inform policy and programming decisions, the data have been
under-utilized as a research tool in order to better understand the relationship among the
various assets (Leffert, et al., 1998). Search researchers themselves state: “Some assets,
for example, may well function as precursors of other assets. Support may increase the
internalization of positive values. . .Research is needed, of course, to deepen the
understanding of interactions among the assets, their developmental sources, and the role
they may have as mediators or moderators of behavior" (Leffert, et al., 1998, p. 227).
This study began to address those gaps in the research.

Adolescence is a period of transition to adulthood. It is a critical period in which
life-altering choices are often made. As a child progresses through adolescence, there are

increasing opportunities for high risk behaviors and choices with potentially serious

consequences. There is also potential for constructive development. Researchers and
practitioners, dedicated to the prevention of drug abuse, have developed and tested
intervention programming aimed at the parents of early adolescents. These proven
programs are designed to teach and enhance parenting practices and processes that have
been shown to be effective in delaying or preventing drug and alcohol use (Hogue &
Liddle, 1999; Rick Kosterman, etal., 1997; Kumpfer & Tait, 2000) and are usually
intentionally targeted at the early years of adolescence, before serious, intractable
problems have developed (Hamburg, 1997). The parenting processes taught in these
programs are aimed at reducing high risk behaviors in adolescents and are the same skills
and processes associated with positive outcomes in youth, namely warmth and
relationship-building processes (responsiveness), better discipline skills (demandingness),
and better communication skills (Hogue & Liddle, 1999; Kumpfer & Tait, 2000).
Relatively little research attention has been given to how family assets are related
to the positive values and social competencies expected of “good kids.” Given the
ultimate goal of raising youth to be competent, caring adults with positive values and the
important role families and parents have to play, this study examined the relationships of
family-related assets to positive values and social competencies using available Michigan
data from the Search Institute’s Profiles of Student Life: Attitudes and Behavior survey.
Given the critical nature of the early adolescent years as a period where
intervening in parenting processes can still effectively and positively alter outcomes for
youth, this study analyzed only data fi'om youth in middle school. It examined the
relationships among demographics, family-related assets, and positive values and social

competencies in middle school youth. The goal of this study was to understand better the

influence of family variables on positive values and social competencies and to assess the
relative strengths and weaknesses of family-related assets as reported by middle school
youth.

See Figures 1.1 and 1.2 on the following pages for conceptual models of

the study variables.

 

Figure 1.1

Theoretical Model Proposed for This Study

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

p
g Maintaining
I T Standards
T
l \
v
Demographics E Positive
—Gender P Family _
-Grade A Relationships Positive
Mother’s Education R “F Values
-2 parent home __> 5
I V Positive
N Family
6 Communication /
Figure 1.2
Theoretical Model Proposed for This Study
p
g Maintaining
| T Standards
T
r \A
V
Demographics E Positive
-Gender P Family S i I
-Grade A Relationships 03 a
-Mother's Education R —> Competencies
-2 parent home _’ 5
1" V Positive
N Family
G Communication

 

 

 

 

 

 

Theoretical Framework
Previous theories and approaches developed over the years have gradually been

incorporated into a larger, more encompassing framework that acknowledges interacting
systems from the intra-personal to the global level. This is the systems approach of Urie
Bronfenbrenner’s “ecological theory of human development” which has provided the
theoretical framework for much of the research on adolescence over the past decades and
is also the model for this study (Bronfenbrenner, 1979; Gegas & Sefl’, 1990; Steinberg &
Morris, 2001). Figure 1.3 below is a model of Bronfenbrenner’s theory (adapted from
Santrock, 1996, p.51).

Figure 1.3. Bronfenbrenn er’s Ecological Theory of Human Development

    
 

Macrosystem

 

Mesosystems: the interactions
among microsystems

 

 

Chronosystem: Patterning of environmental
events and transitions over the life course;
socio-historical conditions

 

 

 

 

Bronfenbrenner’s ecological theory sees each individual as embedded in multiple,
interconnected systems (Bronfenbrenner, 1979, 1986). These systems can be visualized
as nested systems, a smaller system within a larger system and both within even larger
systems. The most basic system is the individual. The unique physical, mental, and
psychological processes within an individual influence both that person’s development
and the environment in which s/he lives. This is especially important to keep in mind
when studying early adolescents because puberty and maturation are creating large
changes within the adolescent. According to ecological theory, changes within one
system are bound to affect other systems, and adolescence, in particular, is a time of
ongoing transition and change between individuals and their contexts (Bronfenbrenner,
1979; Lerner & Galambos, 1998).

Bronfenbrenner labels the most immediate systems in which the individual resides
as microsystems. A family is a primary microsystem, but schools, sport teams, clubs, and
churches are also microsystems. These are environments where individuals interact with
one another on a regular, face-to-face basis, forming, what Bronfenbrenner calls, the
“proximal level of the environment” (Muuss, 1996). These environments have physical
contexts, but also include interacting and changing patterns of relationships, roles, and
symbols.

Individuals participate in more than one microsystem, however. When
microsystems interact they form a mesosystem. A mesosystem is “a set of interrelations
between two or more settings in which the developing person becomes an active
participant” (Bronfenbrenner, 197 9, p. 209). Parents may attend parent-teacher

conferences or participate in fundraisers for sports teams or bands. Microsystems can

 

support and reinforce each other or provide conflicting or counteracting influences. For
instance, a family and a church may reinforce a certain set of values, while a peer group
may or may not support those values. Increased supportive linkages between various
microsystems create a synergistic effect promoting positive youth development.

According to Bronfenbrenner, another system affecting development is the
exosystem. This system is one in which an individual does not directly participate, but it
none-the-less, affects him or her. For instance, a parent’s place of employment affects
how much money is available for a family and perhaps the level of stress within a family.
These may significantly affect a child’s life even though the child does not directly
participate in the microsystem of the parent’s place of employment. A school board is
another exosystem that impacts the lives of adolescents.

The macrosystem is the most distal system in Bronfenbrenner’s theory.
Macrosystems are broad societal systems consisting of the patterns characteristic of a
given culture or subculture, including belief systems, resources, hazards, lifestyles,
opportunities, and patterns of social interchange that are embedded in such systems.
Increasingly, macrosystern variables are taken into account among researchers who may
compare adolescents and families from various countries or cultures.

Bronfenbrenner also speaks of the chronosystem (Bronfenbrenner, 1986). This
refers not only to the developmental changes over time within an individual, but also
changes over time within the environment. This allows an examination of the interplay
between change (and continuity) within an environment and within the individual.

The study of adolescence is especially suited to an ecological perspective.

Biological and physical changes are systemic changes within an individual. A transition

10

from elementary to middle school is a change in a microsystem environment. The
macrosystem, American society today, has norms and expectations for adolescence,
communicated in large part through the media, that are different fiom those of earlier
generations, other cultures, and different from the norms and expectations for young
children. The period of early adolescence fits what Bronfenbrenner refers to as an
“ecological transition,” where major changes are taking place in one or more systems,
including role and environmental changes (Bronfenbrenner, 1979). The development of
the survey used in this research project, the Search Institute’s Profiles of Student Life:
Attitudes and Behavior (the PSL—AB), was clearly rooted in ecological theory (Leffert, et
al., 1998). It inquires about the multiple contexts that adolescents find themselves in and
measures whether variables such as “boundaries” are present in family, school,
neighborhood, and community contexts (Scales, Benson, Leffert & Blyth, 2000). See

Figure 1.4 below:

Figure 1.4 Youth Embedded in Interacting Systems.

 

      
 

 

Community Q = Same Youth

/

Other theories may explain certain processes in ways that are compatible with

 

 

 

ecological theory. For instance “social control theory” explains deviant behavior as

11

resulting from the lack of positive connections to social institutions such as family,
school, or organized religion and the lack of personal or internal controls (Dukes & Stein,
2001). Family interactional theory suggests that a close parent-child mutual attachment
ultimately leads to more conventional behavior in the adolescent and less association with
deviant peers (Brook, Whiteman, & Finch, 2000). Both of these theories broadly propose
that interactions in one system impact other systems, which is an ecological principle
(although they do not spell out the reciprocal, interacting influence). Bronfenbrenner’s
theory is not a developmental stage theory, but earlier work of developmental theorists
such as Piaget and Erikson can be incorporated into ecological theory as some of the
influences impacting development.
Application of Theoretical Framework to This Study

Research coming out of an ecological theoretical framework can never address all
the relevant variables in one analysis. This particular study focused on the family system,
while acknowledging that other systems impact youth and interact with the family system
(see Figure 1.4 on previous page). With that in mind, this report examined the
relationships between demographics, family variables, and positive values and social
competencies in middle school youth. The family variables reflected the processes
involved in maintaining standards, positive family relationships, and positive family
communication and the relationships among the family variables were examined. The
goal was to understand better the influence of family variables on positive values and
social competencies and to assess the relative strengths and weaknesses of family—related

assets as reported by 6’'1 through 8th grade students.

12

CHAPTER 2

LITERATURE REVIEW

Early Adolescence
Puberty

There is no question early adolescence, typically defined as ages 10-14, is a
significant time in children’s development, with complex and inter-related changes taking
place in the physical, social, contextual, and behavioral domains (Hamburg, 1997; Lerner
& Galambos, 1998; Petersen, Leffert, & Graham, 1995). Except for the first three years
of life, no period during a child’s life brings as much physical change. With the onset of
puberty, a child’s body changes into that of a man or woman. Hormonal changes bring on
the development of secondary sex characteristics, sex drive, and mature reproductive
capabilities (Petersen, et al., 1995). Along with these come marked changes in physical
appearance and size. The appearance of breasts in girls and the deepening of the voice
and the emergence of facial hair for boys are biological changes that are apparent to all,
not just to the adolescent.

This process requires transitional adjustment on the part of individuals, parents,
and peers (Steinberg & Morris, 2001). Generally, boys tend to view the growth in height
and strength and more “masculine” characteristics positively, while girls may view
pubertal changes more negatively (Petersen, et al., 1995). Complicating the social scene
is the fact that these changes do not occur at the same time or at the same rate for all
children. For example, girls may start the process of puberty as early as 8 years old and

as late as 13. Boys typically start a few years later than girls, sometime between 9 ‘/2 and

13

14 '/2 years old (Petersen, et al., 1995). An adolescent’s adjustment can be affected by
whether the onset of puberty for the individual is out of sync with that of his or her peers
(Compas, Hinden, & Gerhardt, 1995). The duration or tempo of puberty also varies
considerably, ranging from 1 '/2 years to six years, with puberty generally lasting about 4
years. The question remains, however, if this physical transformation normally brings
with it turmoil and distress (for the child and/or the family system). Previously, the
“storm and stress” among adolescents was taken as a given, especially among those of
psychoanalytic orientation. However, research has shown that nonclinical families are
not universally distressed (Lerner & Galambos, 1998; Montemayor, 1986). Indeed, most
of the research fiom the 1980’s onward has suggested that, for most youth, adolescence is
not a “particularly turbulent” time (Gegas & Seff, 1990, p. 942) and less than 20% show
signs of serious problems that warrant adult intervention (Irvin, 1996).

“If most teenagers pass through adolescence relatively problem free, then why
does such a negative stereotype of that age exist? (Irvin, 1996, p. 223)”. In other words,
why does the “storm and stress” myth persist? First, there is a large enough minority of
students and families who experience significant distress for most people to be personally
aware of such situations. Secondly, the risks of serious and potentially permanently
damaging problems do increase during adolescence. Drug abuse, sexually transmitted
diseases, and pregnancy are practically non-existent risks in early and mid-childhood, but
are increasingly likely as a child goes through adolescence. By 17 years old about one-
quarter (Hamburg, 1997) to about one-half of all youth have engaged in risky or harmful
behaviors (Dryfoos, 1990; Keith, et al., 2002; Lerner & Galambos, 1998). Thirdly, while

most adolescents do not exhibit serious problems, “Even well-adjusted, intelligent, and

14

reasonable adolescents do, on occasion, exhibit truly obnoxious behavior. . .. (T)hey are
not like this all of the time, but probably all adolescents behave this way some of the
time” (Newman, 1985, p.636 ). Puberty generally brings some increase in parent-child
conflict and negativity (Steinberg & Morris, 2001). Montemayor’s (1983) article titled
“Parents and Adolescents in Conflict: All of the Families Some of the Time and Some
Families Most of the Time” expresses it succinctly. While adolescence is not a
universally turbulent time for adolescents, most parents do experience increased stress,
and the majority of parents find adolescence to be the most difficult stage of parenting
(Gegas & Seff, 1990; Montemayor, 1986)). The increasing autonomy of the adolescent
often causes parents to feel a “loss of control” and have increased fears for the safety of
their child. Conflict, mostly over relatively minor issues, is a problem in some families.
About 25% of parents in one study complained about such problems as defiance,
arguments, tasks not finished, and conflict with siblings. Parent-child conflict increases
between childhood and early adolescence, then tapers off after about age 15
(Montemayor, 1986).
Primary Developmental Tasks of Adolescence

Given the transitional nature of the early adolescent period and the adjustments
both adolescents and their parents must go through, an understanding of the primary
developmental tasks of adolescence is in order to set the context for this study. The
changes and behaviors of adolescence are not simply, or even primarily, a result of
increased hormone levels (Lerner & Galambos, 1998). In the cognitive domain (as
originally theorized by Piaget) the younger adolescent is developing the capacity to use

formal logic abstract reasoning (Muuss, 1996; Petersen, et al., 1995). While decision-

15

making capabilities are increasing, capabilities in one domain do not necessarily hold
across all domains. Adolescents may not truly understand the risks involved in such
behaviors as sexual activity and drug, tobacco, or alcohol use, often thinking “it’ll never
happen to me” (Scales, 1997). Understanding adolescent decision-making must take into
account cognitive development, the knowledge the youth possesses, social and
psychological factors, and cultural and societal influences (Gordon, 1996).

Identity formation. In the psychological domain, adolescence is a period where,
according to Erikson’s theory, the primary developmental task is identity formation
(Muuss, 1996; Sartor & Youniss, 2002). This includes not only dealing with such
questions as “who am I?” and “what do I want to become”, but also forming a personal
value system or philosophy of life. The beginning of identity formation is developed in
social contexts with interaction and feedback from peers. Thus, adolescence is a period
when the peer group becomes increasingly important.

Some research shows that attachment to the same-sex parent decreases slowly
during adolescence (Buist, Dekovic, Meeus, & Aken, 2002). However, parental
knowledge of their adolescent’s daily activities and emotional support are associated with
higher identity achievement (Sartor & Youniss, 2002). Research also supports the notion
that healthy identity formation involves a restructuring of the parent-child relationship,
rather than a process of breaking ties and attachment to parents (Brook, et al., 2000;
Liddle, Rowe, & Diamond, 2000; Sartor & Youniss, 2002). Subsequent research has
shown identity formation continues into young adulthood.

Social changes. On the social level, there are changes in social contexts for most

early adolescents, such as the move from an elementary school to a middle school.

16

Increasing autonomy and individuation changes family relationships, but for the majority
of adolescents, it does not require detachment from the parents (Steinberg, 1990; van
Wel, ter Bogt, & Raaijmakers, 2002). Peer relationships and time spent with peers
becomes increasingly more important. The influence of peers, both positive and
negative, is arguably more influential than that of parents, particularly during later
adolescence, although parenting style and parental values influence the choice of peers
(Petersen, et al., 1995; Pilgrim, Luo, & Urberg, 1999).

The "Asset F ramework"

Optimal outcomes for youth include both avoiding the many risky or dangerous
behaviors that compromise their health and well-being as well as developing the positive
competencies and values needed for successful adult outcomes (Garnbone, Klem, &
Connell, 2002). In regard to avoiding participation in potentially harmful activities,
research (and programming) has focused on identifying risk factors for drug abuse, teen
pregnancy, delinquency, smoking, drinking, dropping out of school, etc. Partly in
reaction to the research narrowly focused on youth “at risk” and problem-centered
application and policy (Brown, D'Emidio-Caston, & Benard, 2001), some researchers
have taken the approach of targeting the factors that “build best lives” for all youth
(Keith, et al., 2002).

Researchers associated with the Search Institute have developed a framework that
is used by communities to find positive ways to reduce risk and promote positive youth
development for all youth. Its purpose is “to provide a language for core elements of
positive human development” that would unite and mobilize communities behind a

shared vision (Leffert, et al., 1998, p. 21 1). Since the early 1990’s, researchers at the

17

Search Institute first identified 30, then 40 developmental assets—internal factors and
external supports fi'om communities, schools, families, and other adults that comprise the
key ingredients all youth need to build constructive lives (Benson, 1997; Keith, et al.,
2002; Keith & Perkins, 1995; Leffert, et al., 1998). The assets relevant to this analysis
will be discussed in more detail later.

The premise of the Search asset framework is that it is not just one factor that is
all-important. Rather, the larger the total number of “assets” a child has, the less likely he
or she is to engage in risky behavior (Benson, 1993; Dukes & Stein, 2001; Keith &
Perkins, 1995; Keith, Huber, Griffin, & Villarruel, 2003). Additionally, the more assets
an adolescent has, the more likely he or she is to “thrive” (Scales, et al., 2000). Thriving
was defined in this case as exhibiting leadership, having school success, valuing
diversity, maintaining physical health, helping others, delaying gratification, and
overcoming adversity. Over and above the variance accounted for by the demographic
variables, 10% to 43% of the variance across ethnic groups was explained by the
developmental assets (Scales, et al., 2000). With the exception of “overcoming
adversity,” the developmental assets had moderate predicting power. The assets, as
conceptualized and identified by the Search Institute (“Search”), fall into two broad
categories, internal and external. Within the external category are the assets of support,
empowerment, boundaries and expectations, and constructive use of time. The internal
assets are commitment to learning, positive values, social competencies, and positive
identity. Search developed an instrument for assessing asset levels in youth from 6th
through 12th grades, called the Search Institute Profile. The earlier version, which

assessed 3O assets, has been given to over a 100,000 students. Across the United States,

18

99,462 students took the PSL--AB in the 1996-1997 school year alone (Leffert, et al.,
1998). In Michigan, 20,872 students in 35 counties took the survey during the 1998-1999
academic year (Keith, et al., 2002). The results from Michigan follow the national trend
and clearly show that students with more assets engaged in fewer risk behaviors.
Building Best Lives (Keith, et al., 2002) shows the relative strengths and weaknesses of
various assets among these Michigan youth.

While every developmental asset adds a “building block” (Keith, et. al., 2003) to
the creation of a healthy, productive, young person, some of the assets are clearly more
important than others, some have been researched more than others (Garnbone, Klem,. &
Connell, 2002, Scales, 1999), and some, if missing, are more easily developed than
others. Search research identifies a broad range of factors that collectively correlate with
positive outcomes for youth. These assets are supported by empirical research to varying
degrees.

The assets of support (including family support), boundaries and expectations,
constructive use of time, and commitment to learning are asset categories supported by an
extensive body of research (Leffert, et al., 1998). Leffert and colleagues admit that a
“smaller base of empirical studies supports the empowerment and positive values
categories,” while the categories of social competencies and positive identity have a
moderate level of empirical support (Leffert, et al., 1998, p. 212). A review of the
research related to the particular assets addressed in this study follows.

“Positive Values” Assets
One of the strengths of the asset framework is that it assesses both internal and

external assets. The internal assets include “positive values” and “social competence.”

l9

The more limited research base supporting the positive values asset category is weakened
by lack of clarity concerning what is meant by values. The Webster Illustrated
Contemporary Dictionary defines value as “something regarded as desirable, worthy, or
right, as a belief or ideal.” Values serve as “guiding principles. . .used to select and justify
actions...” (Knafo & Schwartz, 2003, p.595). In this sense of the word, Search included
6 “positive values” as belonging to the 40 assets. These are caring, equality and social
justice, integrity, honesty, responsibility, and restraint. Scales admits that this is not an
exhaustive list of the potentially important values (Scales, et al., 1999). It does, however,
reasonably reflect widely accepted values.

The PSL—AB measures values by, asking the student “how important is each of
the following to you in your life?” Thus, these questions are not measures of behavior or
living up to one’s ideals and beliefs. An example of a discrepancy between values
espoused by youth and youthfirl behavior is the fact that 85% of the undergraduate
students in one study agreed that “basically I am an honest person” although 65%
admitted to at least sometimes lying to their parents about their whereabouts while in
high school (Knox, Zusman, & McGinty, 2001). Other items on the PSL: AB do measure

behaviors that may reflect underlying values (e. g. “During an average week, how many

 

hours do you spend . .. helping other people without getting paid. . .”). However, these are
not the items used to measure the positive values asset.

One study that measured a construct similar to the Search Institute’s positive
values and social competencies assets and its association with parenting style is a study
by Gunnoe, Hetherington, and Reiss (1999). This study utilized data from 486 mostly

Caucasian families with adolescents ages 10 through 18, who participated in the Non-

20

shared Environment (NSE) Study. Qualities, such as honesty, perseverance at hard tasks,
empathy, trustworthiness, self-control, and obedience to parents, teachers, and police,
were assessed by parental report and adolescent self-report. Parenting style was
measured by trained observers. That study looked for the relationships of parental
religiosity and parenting style to outcomes reflecting adolescent “social responsibility”
and used stepwise regression analysis. The aspects of the study relating parenting style to
measures of adolescent values and social competencies are informative to this study
(Gunnoe, Hetherington, & Reiss, 1999).

Another study, less directly comparable to research using Search data, defined
“prosocial values” as belief-based assessment of adolescents’ perception of self--- an
investment in socialized values, such as a sense of belonging, security, and a sense of
accomplishment (Ludwig & Pittman, 1999). Their research indicates that these prosocial
values are associated with fewer behaviors such as delinquency, risky sex, and drug use.
Ludwig and Pittman studied prosocial values and self-efficacy among adolescents,
looking for interaction effects (Ludwig & Pittman, 1999). They found that more
prosocial values, greater self-mastery, and a feeling of trustworthiness were associated
with less delinquent behavior, drug use, and risky sex. However, in contrast to the
general benefit of prosocial values, increased feelings of personal power were associated
with more delinquency, risky sex, and drug use.

In their summary of research related to developmental assets, Scales and Leffert
list studies that support the following positive direct or indirect associations with positive
values (Scales, et al., 1999):

0 Higher levels of prosocial behavior

21

0 Better means-end problem solving skills

0 Better formal reasoning skills

0 Higher conflict resolution skills

0 Greater overall well-bein g

0 Higher self-esteem

o More hopefulness

0 Greater belief in male responsibility to prevent pregnancy

0 Less inclination to have sexual intercourse, and greater use of condoms or

other contraception
0 Less affiliation with deviant fiiends less likelihood of being solicited to
sell crack cocaine

0 Greater competence among Afiican American 9th grade males

0 Greater competence among 9th grade females

0 Higher grades and math and reading scores

0 Higher perceived scholastic competence, less worry about school

Various factors influence the likelihood that adolescents adopt their parents’

values. First they must accurately perceive what those values are. Studies show that
adolescents’ accuracy in perceiving parental values are associated positively with value
agreement between the parents, and with parental warmth and responsiveness, but
negatively with indifferent or autocratic parenting and with conflict regarding values.
(Knafo & Schwartz, 2003). Parent-child value similarity is also associated with parental
warmth and responsiveness. Open, reciprocal communication, promotes the adopting of

parental norms (Kelly, Comello, & Hunn, 2002). As previously mentioned, adolescent

22

social responsibility, measured as reportedly practicing values such as honesty, taking
responsibility, and empathy, have been associated with parental religiosity and
authoritative parenting (Gunnoe, et al., 1999).

Having positive values can be seen as one aspect of the broader category of moral
development. While research based on Kohlberg’s theory of moral development provides
strong theoretical and research support that contends increases in moral reasoning are
developmental (Muuss, 1996), there is nonetheless consistent research support indicating
the influence of parents on the values of their children (Scales, et al., 1999). White
(2000), for example, found support for the hypothesis that greater perceived
connectedness or cohesion in the family system by adolescents would be associated with
attributing greater influence to the family as a source of moral authority. Boyes and
Allen (1993) found 10th and 12th grade youth who perceive their parents as
“authoritative” were more likely to use post-conventional moral reasoning, with the effect
greatest with the older youth. However, research findings are not entirely consistent
(Pratt, Arnold, Pratt, & Diessner, 1999). These studies do not directly correspond to the
asset fiarnewor ’3 “positive values” which asks students the importance of specific
values, rather than measuring the stage or complexity of moral thinking or the perceived
source of moral authority.

“Social Competencies " Assets

Search categorizes five of the assets as social competencies. These are planning
and decision making, interpersonal competence, cultural competence, resistance skills,
and peaceful conflict resolution. Scales admits social competence is “a difficult concept

to define, but it generally refers to adaptive functioning” in social situations that obtains

23

beneficial outcomes (Scales, et al., 1999, p. 173). Research literature supporting the social

9, 6‘ ,9 “

competencies assets uses the terms “social skills, social problem solving skills, social

orientation” (the adolescents beliefs and values regarding their own use of social problem

,9 66

solving skills), “social adjustment, social performance,” and “social competence”
(Cavell, 1990; Kuperrninc & Allen, 2001; Matlack, McGreevy, Rouse, Flatter, & Marcus,
1994; Scales, et al., 2000; Schoenrock, Bell, & Sun, 1999). The social competency assets
of resistance skills and conflict resolution skills are often addressed separately in the
research literature, and are particularly prominent in research focusing on prevention of
alcohol and drug use and violence. Decision making is also researched as it contributes to
adolescent risk-taking (Gordon, 1996; Rolison & Scherrnan, 2002).

Interpersonal competence is probably what most people would identify as the core
dimension of social competence. The PS L--AB survey asks three questions to measure
interpersonal competence. These involve asking the students how “people who know you
well” would rate them on “caring about other people’s feelings,” “feeling really sad when
one of my fiiends is unhappy,” and “being good at making and keeping friends.” The
first two items measure aspects of empathy. Family cohesion, parental support, self-
esteem, and communication were significantly related to empathy in adolescents (Henry,
Sager, & Plunkett, 1996). “Having friends” generally contributes to positive
development, but Hartup adds the importance of clarifying the identity of the fiiends and
the quality of the friendships in assessing the developmental significance of having
fiiends (Hartup, 1995).

Interpersonal competencies are skills and attitudes that begin in early childhood,

as children learn to share, to say please and thank-you, and to ask, rather than hit or

24

snatch, to get what they want. Robert Selman’s stage theory of interpersonal
understanding or social cognition (Selman & Schultz, 1990), maintains that children go
through stages in which they gradually learn to distinguish the self from others and begin
to take another’s viewpoint while simultaneously maintaining their own. According to
Selman, early adolescents, ages 10 to 15, are in the mutual perspective-taking stage.
They are beginning to be able to step back and see social situations and interactions from
a third-party perspective---a development which coincides with an increasing ability to
process abstract thought. They are also beginning to reflect on their own inner processes.
Interpersonal competencies also affect the social dynamics of the adolescent’s

world as well as promote positive future outcomes. Scales and Leffert report the
following findings associated in the literature with interpersonal competence and cultural
competence (Scales, et al., 1999). (Scales reports that these constructs are usually dealt
with together in the literature.)

0 Protection against adversity

- Improved adjustment

0 Positive self-esteem

o Perceived self-confidence

0 Improved peer competence

0 Peer acceptance

0 Improved ability to form friendships

0 Increased problem solving ability

0 Decreased loneliness

0 Decreased depression

25

o Decreased problem behaviors such as substance abuse
0 Lower risk of negative consequences from early sexual activity among African-

American youth

However, greater social competencies do not necessarily, or automatically, lead to
positive outcomes. As previously mentioned, Ludwig and Pittman (1999) found that
greater feelings of personal power (in contrast to self-mastery), were associated with
more risky behaviors.

Other researchers use measures that correspond somewhat to Search’s delineation
of the internal assets. Steinberg, Elmen, and Mounts measure “psychosocial maturity”
which is defined as having three sub-dimensions of autonomy: work orientation, self-
reliance, and identity. Identity includes self-esteem, internalization of values and clarity
of self-concept (Steinberg, Elmen, & Mounts, 1989). Many researchers measure aspects
of commitment to learning, self-esteem, locus of control, and identity development. A
high level of parental responsiveness, in combination with either a high level of parental
strictness or an absence of strictness, was associated with increased social competence
(Lambom, Mounts, Steinberg, & Dombusch, 1991).

F amily-Related Assets

In the Search Institute framework, family support and positive family

communication are two separate constructs, or assets. In the 40-asset framework, “family

9, ‘6

support, positive family communication,” and “parent involvement in schooling,” are
designated as part of the external asset category of support. The asset of “family

boundaries” is considered an external asset in the “boundaries and expectations”

category. Family boundaries is defined as “family has clear rules and consequences and

26

monitors the young person’s whereabouts.” Gambone and associates (2002) take all the
Search constructs related to family and lump them into one concept called family support.
Much of the research literature uses the term “authoritative” to describe the “supportive
parents” who are “emotionally close with their children, communicate openly with them,
engage them in democratic discussions about family rules and decisions, and provide
clear but sometimes negotiable boundaries and norms” (Scales, et al., 1999).

Interestingly, the findings from Michigan students taking the PSL--AB show
overall (6th through 12th graders) 68% of the students report having family support—
defined as “family life provides high levels of love and support,” but only 28% report
having positive family communication (Keith, et al., 2002). The percentage of 6th grade
youth reporting positive family communication is 45%. For 12‘h graders, the percentage
is only 17% (Keith, et al., 2002). Much of the research literature on the assets of family
boundaries and family support is contained in the literature on authoritative parenting,
which will be discussed in greater detail later.
Positive Family Communication

This asset is measured by the PSL-—AB with the following three items:

0 I have lots of good conversations with my parents: [Strongly agree, agree, not
sure, disagree, or strongly disagree?]

0 If you had an important concern about drugs, alcohol, sex, or some other serious
issue, would you talk to my parent(s) about it? [yes, probably, I ’m not sure,
probably not, or no]

0 In an average week, how many times do all of the people in your family who live

with you eat dinner together? [ eight possible responses: none through seven]

27

Other items, conceivably related to “positive communication,” assessed whether parents
asked about school or homework, and agreement with the statement that parents “often
tell me they love me.” Those items, however, are used to measure other constructs.

Parents report high levels of communication with their middle school students,
although the research does not always identify the quality of the conversations. In a
phone survey of 309 families of 7th and 8th graders, 80% of the parents reported talking
about school work with their child almost every day (Lytle, et al., 1999). Chores around
the home and getting along with other family members were the other topics most
frequently discussed on a daily basis. Socio—economic status (SES) predicted differences
in the amount of communication and the topics discussed, with lower SES families giving
risk-reduction messages more frequently. For low SES families, the fourth most
fi'equently discussed t0pic was avoiding alcohol and drugs, but for high and mid-SES
families the fourth-ranked issue was safety. Not surprisingly, the topic least likely to be
talked about on a daily basis for all SES groups was sexual activity. Overall, increased
conversation was associated with lower SES. Significantly greater proportions of lower
SES parents talked to their children about each issue on a daily basis as compared to mid
and high-SES parents.

Much of the literature on parent adolescent communication focuses on parents
talking to their child about sex, drugs, contraception, tobacco use or other risky behaviors
(Jaccard, et al., 2000; Miller-Day, 2002). In a study of 67 teens ages 11-17, less than half
spoke with a parent about alcohol, drugs, or tobacco (Miller-Day, 2002). However, in a
large study using secondary data, 74% of adolescents had at least one conversation about

the dangers of alcohol, although only 12-15% had a conversation with a parent about

28

various drugs or alcohol in the previous year (Kelly, et al., 2002). In a study of 751
African-American youth, ages 14-17, regarding parent-teen communication about sexual
issues, the mothers perceived more communication with their teen than the teens
themselves reported (Jaccard, et al., 2000). While it is unclear how much parents, in
general, do communicate with their children about these issues, the research supports that
communication of parental values does influence teens. For example, Jaccard, Dittus, and
Gordon (2000) report that “the extent to which mothers communicate with their teens
about sex is important in predicting adolescent sexual behavior” (p. 195). Perceived
parental sanctions concerning drug use was associated with lower drug involvement
(Kelly, et al., 2002).

Many different factors are associated with communication between an adolescent
and his or her parents. Not surprisingly, a positive overall relationship between parents
and teens is predictive of greater communication about sexual issues, including birth
control (Jaccard, et al., 2000). Family cohesion (emotional bonding and family
togetherness) is linearly related to positive family communication, defined as
expressiveness, clarity, and problem solving (Perosa & Perosa, 2001). However, when it
comes to talking about sexual issues, parents often have specific concerns or an
uneasiness about talking about sex that inhibits conversations, even when the overall
relationship is good. Additionally, communication alone was not suflicient for single
mothers’ accurate awareness of their adolescents’ stressors. Monitoring promoted more
accurate awareness of the adolescents’ stressors in the mothers and mediated the
relationship between mother’s awareness and better adolescent adjustment (Hartos &

Power, 2000)

29

Parenting Styles

The influence of parents and parenting style on children and adolescents
has been studied extensively (Holmbeck, Paikoff, & Brooks-Gunn, 1995; Larnborn, et al.,
1991). Much of the research examining the parenting dimensions corresponding to the
assets of family boundaries and family support is contained in the research on parenting
styles. Much of this research has tested, refined, or expanded upon the seminal research
of Baumrind (Baumrind, 1975, 1991). She developed terminology for three primary
parenting styles: authoritarian, authoritative, and permissive. Authoritarian parenting is a
style of parenting that is high on control and strictness, but low on warmth and
acceptance. Permissive parenting is a style of parenting that is low on strictness and
control, but high on warmth and acceptance. Authoritative parenting, the style of
parenting most researchers and child development experts advocate, is one that utilizes
behavioral control and standards or strictness coupled with high levels of warmth and
acceptance. Baumrind used the terms, nurturance, control, and autonomy to define the
primary dimensions or qualities of her rich portrait of the authoritative parenting concept
(Baumrind, 1975). In addition, she identified a “Rejecting-Neglecting” parenting style
(neither demanding nor responsive) and a traditional parenting style (father more
demanding, mother more responsive) (Baumrind, 1991). Demandingness and
responsiveness correspond roughly to the Search constructs of family boundaries and
family support. An in depth examination of the authoritative parenting style is warranted
as it encapsulates two of the three main independent variables in this study: maintaining

standards and positive family relationships

30

 

 

 

Authoritative Parenting Style. Research has generally conceptualized
authoritative parenting as mothers and fathers who are high on the dimensions of both
demandingness/control and responsiveness. Jackson, Henriksen, and Foshee (1998), who
developed a measure for assessing authoritative parenting, state:

The demanding behaviors characteristic of authoritative parenting include

setting and enforcing clear standards of behavior, actively monitoring and

supervising a child’s activities, maintaining structure and regimen in a

child’s daily life, and making maturity demands consistent with the

developmental phase of the child. The responsive behaviors characteristic

of authoritative parenting include being affectionate and accepting,

providing comfort and support, being involved in children’s academic and

social development, and recognizing children’s achievements (pg. 319).
Sometimes the “autonomy” dimension is also included (Gray & Steinberg, 1999;
Mattanah, 2001; Steinberg, et al., 1989). Although this typology is widely used, it
doesn’t capture the richness or breadth of the parent-child relationship, or spell out the
processes or mechanisms that may impact the development of children and adolescents
(Holmbeck, et al., 1995).

Researchers have used a myriad of terms for these two major components of
demandingness/control and responsiveness in the authoritative parenting style. The first
component of demandingness/control is the parenting process variously referred to in the
literature as control, demandingness, strictness/supervision, limit-setting, or boundaries
(Gray & Steinberg, 1999; Kurdek & Fine, 1994; Mattanah, 2001; Paulson & Sputa, 1996;
Scales, etal., 1999; Shucksmith & et al., 1995).

More recent research has sometimes distinguished between behavioral control and

psychological control (Galambos, Barker, & Almeida, 2003; Pettit, Laird, Dodge, Bates,

& Criss, 2001; Smetana & Daddis, 2002). Behavioral control is used to refer to making

31

 

 

 

. W

and enforcing rules and monitoring a child’s whereabouts or activities. This form of
control respects the child’s autonomy and is the beneficial form of control that is part of
the authoritative parenting construct (Smetana & Daddis, 2002). Psychological control
negatively affects the child’s development and refers to control by parental intrusiveness,
guilt induction, and love withdrawal (Smetana & Daddis, 2002). Psychological control is
control by manipulation, love withdrawal, and guilt-induction and has as antecedents
harsh parenting early in childhood and a mother’s judgment of child externalizing
behavior problems (Pettit, et al., 2001). An earlier proactive parenting style (aiming to
prevent problems) was predictive of later monitoring behaviors. Pettit et al (2001) found
that psychological control, but not monitoring, uniquely predicted adolescent anxiety and
depression. Baumrind found that, overall, children of parents who exercised control had
fewer behavior problems, but the type of control used was differentially associated with
other child characteristics. “Children of assertive and rationally controlling parents,
unlike those of restrictively controlling parents, tended to be more competent and
communal (Baumrind, 1991, p. 148).” Psychological control does not predict monitoring
and vice versa (Smetana & Daddis, 2002). The PSL—AB measures monitoring and
behavioral control, but not psychological control; therefore the detrimental effects of
psychological control should not be expected to be associated with the family assets
measured by the PSL—AB.

The second component is variously labeled as responsiveness, acceptance,
nurturance, warmth, and/or involvement (Gray & Steinberg, 1999; Kurdek & Fine,
1994; Mattanah, 2001; Paulson & Sputa, 1996; Shucksmith & etal., 1995). This factor

is associated with socially competent children, whether used together with

32

demandingness in an authoritative parenting style, or by non-demanding parents
(Jackson, Henriksen, and Foshee, 1998; Baumrind, 1991). Some researchers refer to
this dimension as “family support” (Galambos, 2003; Scales, 1999).

Baumrind’s description of authoritative parenting is parenting that is high on both
demandingness and responsiveness. An interaction effect between demandingness/control
and support/responsiveness is assumed by many scholars in the field, but that assumption
has often been untested (Galambos, et al., 2003; Smetana & Daddis, 2002). Much of the
current research is built on a two-factor model of parenting that posits
demandingness/control and responsiveness/acceptance as both positively related to
adjustment (Kurdek & Fine, 1994). Kurdek and Fine tested that notion against the
possibility that demandingness and responsiveness interact synergistically to promote
adjustment. Another complex relationship they tested for was a curvilinear relationship
between adjustment and acceptance and adjustment and control. Their study found
acceptance and control both positively related to adjustment. For psychosocial
competence and self-regulation, no complex relations were found for one sample, and
only a curvilinear effect for family control found in the other sample. The form of
control assessed in their study was limited to monitoring-type behaviors. Another study
found a curvilinear effect for monitoring with respect to grade point average, but a linear
relationship for other aspects of psychosocial development and mental health (Roberts &
Steinberg, 1999).

The effectiveness of the authoritative parenting style and its perception by
adolescents is affected by contextual variables, such as culture and the internal beliefs of

both parents and children. In regard to the control dimension of authoritative parenting, it

33

matters not just how much control is exerted by parents, but what type, over what
domains, and how it is legitimized in the eyes of the parent and the adolescent
(Baumrind, 1991; Smetana, 1995; Smetana & Daddis, 2002). Both parents and
adolescents can differ over what areas or domains of life are legitimately areas over
which parents should exercise control. Authoritative parents tend to view issues of
morality and conventionality as domains in which they are legitimately obliged to make
rules. They also tend to exercise more control over “multi-faceted and fiiendship issues”
than other types of parents. However, they make different justifications for the control
depending on the type of issue involved and clearly articulate their reasoning. Rational
control is not perceived by adolescents to be restrictive (Baumrind, 1991). Permissive
parents allow more domains to be under the adolescent’s personal control (Smetana,
1995)

The authoritative parenting style occurs in a wide variety of familial, ethnic,
cultural, and economic contexts (Lambom, et al., 1991). Researchers have examined the
relationships between authoritative parenting and numerous contextual, cultural, and
demographic variables. F amily income, parent education, family type, and ethnicity have
sometimes been associated with authoritative parenting (Gunnoe, et al., 1999), while the
importance of religion in daily lives and decision making processes, has been found to be
associated as well (Gunnoe, et al., 1999). .

Authoritative parenting does tend to differ by family structure—an intact two
parent home or a single parent home. Two-parent families are more likely than single
parent families or step families to have authoritative parenting styles (Gunnoe, et al.,

1999). Hartos and Power (2000) summarize the literature stating that communication and

34

monitoring are associated with fewer behavioral problems in youth from both two-parent
and single parent homes. Single parents are more likely to have a “permissive” parenting
style than parents in intact two-parent homes.

Researchers have tested the relationship of authoritative parenting to a broad
range of outcomes or characteristics for children and adolescents. These include
delinquency and violence, psychological distress, tobacco use, drug and alcohol use
(Dombusch, 1987; Fletcher & Jefferies, 1999; Jackson, Henriksen, & Foshee, 1998;
Jones, Forehand, & Beach, 2000; Klein, Forehand, & Armistead, 1997; Lambom, 1990;
Laurence, D, & S, 1989; Leech, Day, Richardson, & Goldschmidt, ; Park & Bauer, 2002;
Pilgrim, et al., 1999; Shucksmith & et al., 1995; Slicker, 1996; Steinberg, 1990; Steinberg
& Elmen, 1986a; Steinberg, et al., 1989). Children with authoritative parents have been
shown to score higher on “measures of self-esteem, self-control, adjustment at school,
peer acceptance, and conflict resolution ability” (Jackson, et al., 1998). They are less
likely to engage in delinquent or risky behavior such as substance abuse
(Adalbjarnardottir & Hafsteinsson, 2001; Jackson, et al., 1998), and more likely to
succeed in school , a fact that is well documented in the literature (Dombusch, 1987;
Gray & Steinberg, 1999; Laurence, et al., 1989; Mattanah, 2001; Steinberg & Elmen,
1986b; Steinberg, et al., 1989).

The previously mentioned NSE study of 486 mostly Canadian youth found
authoritative parenting associated with greater adolescent social responsibility defined as
“prosocial attributes such as honesty, perseverance at hard tasks, empathy,
trustworthiness, self-control, and obedience to parents, teachers, and police (Gunnoe, et

al., 1999, p. 210). This study found parental religiosity predicted authoritative parenting

35

and also had a direct effect on adolescent social responsibility, not mediated by
authoritative parenting.

A study of 4,100 high school students examined four sets of outcomes by
parenting style typology (Lambom et al.,1991). The sample was diverse in terms of
ethnicity, type of community, family structure, and socioeconomic status. Authoritative
parenting was associated with adolescents scoring highest on measures of psychosocial
development and academic achievement and lowest on measures of internalized distress
and problem behavior. Authoritarian parenting, which is high on demandingness, but
low on responsiveness, was associated with less school misconduct and less drug use and
a more positive orientation toward school than were parenting styles characterized by
high levels of responsiveness and low levels of demandingness. High levels of
responsiveness without high levels of demandingness—the study labeled it an indulgent
parenting style—was associated with high levels of self-confidence and social
competence but more problem behavior and lower grade point averages and less school
orientation. Overall, an authoritative parenting style was associated with more positive
results than any of the other parenting styles, with equal or better scores on all outcome
measures. A neglectful parenting style, low on both demandingness and responsiveness,
was associated with poorest adolescent outcomes.

The relationship between authoritative parenting and moral reasoning or
development has been studied in a relatively limited number of studies. Pratt et a1. (Pratt,
et al., 1999) studied a small sample of Canadian youth and found authoritative parenting
to be associated concurrently with higher moral reasoning, supporting a previous study

by Boyes and Allen (1993). However, an authoritative parenting style did not predict a

36

higher moral reasoning score two years later (at age 16). Students with authoritative
parents were shown to be more responsive to parents’ views (Mackey, Arnold, & Pratt,
2001). In general, a more supportive and democratic family style and the use of Socratic
discussion techniques have been associated with higher levels of adolescent moral
reasoning (Pratt, et al., 1999). These family dynamics are compatible with an
authoritative parenting style, but—except for supportiveness---not usually measured or
delineated as part of authoritative parenting.

The effects of authoritative parenting have been tested in a wide variety of
contexts with diverse populations. Studies in England (Shucksmith & et al., 1995)
replicate the findings of American studies. Some studies have shown differential effects
of various parenting styles on developmental outcomes, depending on ethnicity (Chao,
2001; Park & Bauer, 2002). However, studies show benefits are associated with an
authoritative parenting style consistently across many family structures, socio-economic
categories and ethnicities (Steinberg, 1990). Authoritative parenting was associated with
less involvement in substance use for Chinese, European American, and African
American adolescents (Pilgrim, etal., 1999). Both authoritative and authoritarian
parenting styles have been associated with positive outcomes among Chinese-Americans
(Chao, 2001). The strength of various parenting processes have been shown to differ by
ethnicity. White adolescents report higher levels of parental responsiveness, while
Afiican American adolescents report higher levels of parental control (Freeman &

Newland, 2002)

37

Research Supporting Directional Impact

To zero in on the questions of “what matters most?” and “how much does it
matter?” a recent meta-analysis by Gambone, Klem, and Connell (2002) is most helpful.
Their extensive literature review and meta-analysis of re-configured longitudinal data sets
provides solid answers. They confirm that the importance of “supportive adult
relationships” is documented by the largest research base. The authors state :

“the dimensions of support from parents that matter are: they offer help

when needed, discuss school and future plans with their child, check up on

homework, know what the child is doing with his/her time, know his/her

fi'iends, discipline consistently, and are emotionally supportive. When

children have these supports they get better grades, are more engaged in

school, have higher test scores, better attendance, participate in more

extra-curricular activities, and are less likely to drop out.. .. are more likely

to have adaptive coping mechanisms and less likely to engage in risky

behavior.” (p. 29 — 30)

Questions on the PSL--AB directly measure all of the dimensions listed above
except whether the parent discusses future plans with their child. Search, however,
assigns the dimensions to various asset categories: boundaries and expectations, family
support, positive communication, parent involvement in schooling.

Longitudinal research provides support for a causal link between aspects of
family-functioning (similar to those measured by the Search Institutes assets) and
positive youth outcomes. The PSL—-AB surveys and others which provide only cross-
sectional data, cannot provide this support (Dukes & Stein, 2001; Gambone, et al., 2002;
Shucksmith & et al., 1995). Controlled longitudinal studies of successful prevention-
oriented parent-education programs in early adolescence also provide indirect support for

the contention that assets, such as parental monitoring (boundaries) and positive

communication, reduce the likelihood of risk behaviors, such as substance use/abuse in

38

teens. A random-assignment, longitudinal, and controlled study of the Iowa Extension’s
Strengthening Families Program: 10-14 showed fewer conduct problems among youth,
increased parental effectiveness in setting limits, increased parental affection, and lower
rates of alcohol, tobacco, and marijuana use among participating youth even years after
completing the program. In fact, the differences between control and participating youth
increased over time. A major component of the Strengthening Families Program: 10-14 is
teaching parents skills in setting boundaries and positive communication. The Iowa State
University Extension website summarizes this NIMH, 1992-1997 study at

httm//www.extension.iastate.edu/sfp/sfpeval.html.

 

Preparing for the Drug-Free Years is another program which teaches parenting
skills and has also been rigorously tested (Rick Kosterrnan, Hawkins, Haggerty, Spoth,
& Redmond, 2001; Rick Kosterrnan, et al., 1997). Pre-and post intervention videos were
made of family interactions. Statistically significant improvements in mother’s proactive
communication in general family interactions were found, as well as statistically
significant improvements in proactive communication in problem-solving for both
parents, and a decrease in negativity (Kosterman, et al., 1997). A four-year study utilizing
random assignment showed both the Iowa Strengthening Families Program and Preparing
for the Drug Free Years delayed the onset of initiation of alcohol use by adolescents over
a minimal contact randomly assigned control group (R. L. Spoth, Guyll, & Day, 2002).
Other rigorous studies of this and other programs that intervene preventatively in families

of young adolescents demonstrate the power of family assets on healthy youth outcomes.

39

Demographic Variables

Previous studies, to be described subsequently, have sometimes found
demographic variables, such as those identified in this study, to be significantly
associated with either parenting styles and processes or with adolescent outcomes, such
as positive values or social competencies. Thus, in this study examining the relationships
of parenting processes to adolescent outcomes, accounting or controlling for the effects
of demographic variables is warranted. The selection of demographic variables was
restricted by the information requested by the PSL—AB surveys. The demographic
variables in this study are the grade level and gender of the adolescent, the educational
level of the mother, and family structure (two-parent home or other).
Grade Level

The study examined the data from middle school students, so grade level was an
important demographic variable to include. The results of Michigan survey data on
which this study was based was also analyzed by grade (Keith, et al.,2002). Most of the
asset scores for external assets decline from 6th through 12th grades, with the exception of
safety, other adult relationships, high expectations, and youth programs. For internal
assets, generally the older youth scored higher. Most of the exceptions are among the
asset items used in this study, namely caring, restraint, equality and social justice, and
resistance skills. homework, achievement motivation, and cultural competence are also
assets with lower scores for older students. For some of the positive values and social
competencies, the “assets appear to decrease from 6th grade level, but show some increase

later”(Keith, et al., 2002, p. 29).

40

Gender

Some studies found gender differences in assets related to parenting, with girls
reporting more parental support and higher levels of monitoring (Sartor & Youniss,
2002). Girls have been found to perceive parental values slightly better than males
(Knafo & Schwartz, 2003). In high school students, sons of recently divorced mothers
reported less parental control than did daughters (Freeman & Newland, 2002). Analysis
of data from the Nonshared Environment Study (NSE) repeatedly found girls more
prosocial than boys (Gunnoe, et al., 1999). Michigan female students in the 6th through
12th grades taking the PSL--AB scored at least 8% higher on each of the positive values
and social competencies items used in this study (Keith, et al.,2002). This is a reporting
of descriptive differences, not statistical differences. They also scored 6% higher on
family boundaries than male students. Other analyses of PSL—-AB data have looked for
sex (gender) by asset level interactions (Leffert, et al., 1998).

Mother ’s Education.

The NSE study also found mother’s education, income, and family type to be
significant control variables in stepwise regression procedures predicting authoritative
parenting (Gunnoe, et al, 1999). SES was associated with differences in family
communication, as previously mentioned (Lytle, et al., 1999). Lambom et al.(l99 1)
found an interaction between parental education and parenting style in predicting internal
distress. The educational level of the mother may be associated with various parenting
processes or outcomes as a function of the education or as a function of other factors

associated with SES.

41

Family Structure

Again, in the NSE study, family structure was found to be a significant predictor
of authoritative parenting for mothers, but not for fathers. The family structure options in
that study were non-divorced families and step families (including blended families) who
were together for at least five years. “National survey studies of family structure and
children’s outcomes consistently find that children raised in two-parent homes do better
than children raised in single parent homes on measures of educational achievement and
adjustment”(Jaffe, Moffitt, Capsi, & Taylor, 2003, p. 109).

Measurement Issues

Research on parenting styles has employed a variety of assessment measures.
Two common methods are parents’ self-assessment and the child’s assessment of the
parent. Both observer-rated parenting with combinations acceptance, autonomy-
allowing,, and appropriate discipline and adolescent-reported perceptions of such are
associated with similar prosocial outcomes (Johnson, Shulman, & Collins, 1991; Paulson
& Sputa, 1996). However, the reports of parents and adolescents do not necessarily
match. High school students rated their parents lower on all parenting dimensions than
the parents themselves did (Paulson & Sputa, 1996). Likewise, researchers using the
Community Asset Development for Youth survey with youth, their parents, and other
adults in the community found that 24% to 35 % more parents reported teaching or
supporting an asset than students reported having the asset (Keith, et al., 2002). For
example, 88% of parents report supporting family bonding and communication, while
63% of the youth report having it. For setting limits for youth the percentages are 41%

and 86% respectively. Hartos and Power (2000) also found mothers reported more

42

monitoring and better communication than their adolescents did. Thus, it is emphasized
here that the measurements of parenting processes or family assets, as shown by the
responses on the PSL—AB survey, are indicators of the student’s perceptions and might
well be lower than parental assessments of those processes.

Another issue is that some research measures do not differentiate between the
father’s parenting style and the mother’s parenting style. While they are often
moderately correlated, one study found 21% of 5th, 8th, and 1 1th graders reported
“incongruent” parenting styles, with one parent authoritative and the other
authoritarian/rejecting (Johnson, et al., 1991). Paulson & Sputa report mothers scored
higher than fathers on both demandingness and responsiveness, as reported by both the
teens and their parents (Paulson & Sputa, 1996). The PSL—AB does not differentiate
between a mother’s parenting and a father’s parenting, asking instead about “parents” or
“family.”

As previously mentioned, the PSL--AB, from which the data for this thesis were
utilized, has been used with over 100,000 students across the United States. However, the
data collected is from predominantly White communities. The Michigan data set is 90%
White (Keith, et al., 2002). There is some evidence in the literature that the Search assets
might not be reliable constructs for inner-city, racially diverse populations (Price, Dake,
& Kucharewskil, 2002). An analysis of data from 288 primarily African-American inner-
city youth found that the construct validity of the Search categories was not supported,
with weak internal consistency and stability reliabilities. These students took the survey a

second time after a two week delay (Price, et al., 2002).

43

CHAPTER 3

METHODOLOGY

A review of the literature has not identified research analyzing Search data for
relationships found between parent-related assets and any of the internal assets.
Although Search data have been widely disseminated as summary data and the results
have been used to inform policy and programming decisions, the data have been under-
used as a research tool to understand better the relationship between assets (Leffert, et al.,
1998). This study utilized secondary data from 6‘h through 8th grade Michigan students
who took the Search Institute Profiles of Student Life: Attitudes and Behaviors survey in
the 1998-1999 school year. Using that data base, this study began to address those gaps in
the research, focusing on the relationships of family variables to positive values and
social competencies in middle school youth and looking at the relative strength and
weakness of those assets. Based on the review of the literature, the following hypothesis
and research questions were proposed for this study.

Research Hypotheses

2 H1.1: Higher levels of maintaining standards will be positively associated with

higher levels of positive values and social competencies.
:> H1.2: Higher levels of positive family relationships will be positively associated

with higher levels of positive values and social competencies.
2 H1.3: Higher levels of positive family communication will be positively

associated with higher levels of positive values and social competencies.

 

 

Hl.4: Positive parenting, the combination of all three parenting related variables

will be positively associated with positive values and social competencies.
Research Questions

What are the levels of each family variable among Michigan middle school

students?

What are the levels of positive values and social competencies among Michigan

middle school students?

How are family-related assets related to positive values and social competencies

among middle school youth?

What are the relationships among the family related variables?

What are the relationships among demographic variables, family variables, and

the positive values?

What are the relationships among the demographic variables, family variables and

social competencies?

For each question the family variables are:

I Maintaining standards
I Positive family relationships
I Positive family communication

The demographic variables to be accounted for are:

I Grade

I Gender

I The educational level of the mother (which can be considered a proxy for
SES)

45

I Family structure
Participants

The 10,623 participants in this study are a subset of the 20,872 Michigan students
who took the Profiles of Student Life: Attitudes and Behavior survey in the 1998/1999
school year. The participants were predominately White, middle class students from two-
parent homes in 35 Michigan counties. The participants were mostly from suburban
areas, and small to mid-size cities, many in northern Michigan (Keith, etal., 2002).

The respondents who were in the 6m, 7m and 8‘h grades, the typical middle school
grades, were the subjects studied here. The frequency by grade is as follows:
Table 3.1

Number of Students in Each Grade

 

 

Grade Frequency
6th 1 ,5 89

7th 4,062

8th 4,972

Total 10,623

Measures
The initial data base for this study was the responses of Michigan 6‘”, 7‘“, and 8th
graders on the 156 item PSL—AB survey, with only 37 items from the survey used. Of
the items selected, 12 were selected as measures of the family variables. Thirteen items

were selected for the positive values measure. Seven items were selected for the social

46

competence measure and five of the items, reflecting demographic data such as grade
level, gender, and family structure, were used.
Demographic Variables

To reiterate, family structure, grade of the student, gender, and educational
attainment of the student’s mother were the demographic variables accounted for in this
study. Ethnicity was not used as a demographic variable as the data set is 90% White,
with only 2% American Indian, 1% Asian/Pacific Islander, 1% Black/Afiican American,
2% Hispanic origin, and 4 % multiracial. Grade was treated as a continuous variable and
gender was scored as a dichotomous variable.

The Search item assessing “family structure” gives four choices: a two-parent
home, a one-parent home (mother only), a one-parent home (father only), and living
sometimes with one’s mother, sometimes with the father. A substantial majority, 74.2%,
reported living with two parents. Of the others, 12.4% reported living with their mother
only, 2.6% with their father only, and 10.7% part-time with each parent. For the purpose
of this study, family structure was measured as a binary variable: two-parent home, yes or
no? Living with one’s mother or father only, or part-time with each was recorded as “not
a two-parent home.” The students were asked to interpret “parents” and “mother” and
“father” as referring to the adults “most responsible for raising you” whether they were
foster parents, step-parents, or other relatives or guardians. The survey questions about
parenting and family dynamics did not distinguish between mothers and fathers.

One survey item asked about the mother’s education. This item can be thought of
as a proxy for SES as well, since no direct reference to income level or SES was included

in the survey. The survey responses were recoded into the following categories: l= did

47

not complete high school, 2 = completed high school, 3 = some college, 4 = completed
college, 5 = graduate or professional school after college. Very few mothers (7.6%) did
not complete high school. The majority (64.7%) had at least some college, while 27.7%
had only a high school education. Over 30% of the mothers (30.2% to be exact)
completed college and 15.9% had graduate or professional school after college. This
indicates a predominately “middle class” population.

Independent Variables

Three family dynamic variables were decided upon, based on review of the
research literature, the Search construct categories, and the available items in the PSL-
AB. Questionnaire items having face validity as representing the demandingness
dimension of authoritative parenting were tested for reliability and used as the
maintaining standards variable. Items were selected if they added to the reliability. The
question, asking how many hours on an average school day the student spent at home
without an adult present, was not used because the responses did not follow the pattern of
a five-point scale.

Items for the positive family relationships and the communication variables were
also first selected by face validity then tested for reliability. A few items that potentially
could relate to more than one construct were tested for reliability with both sets of items
determining the best match. For instance, “My parents often tell me they love me” was
tested for reliability as part of the positive communication variable and the positive
family relationships variable. The reliability score for the positive family relationships
variable increased from .8257 to .8597 with the addition of that item, so it was retained as

a positive family relationships item. The “positive family relationships” variable was

48

designed to correspond to the dimension referred to most often in the literature as
“responsiveness” or “acceptance.” It included items Search identified as “support.” The
term support was not used here as the label because Gambone and colleagues (2002)
include monitoring and consistent discipline as dimensions of “support.” Although
positive verbalizations may possibly be a part of the expression of the affection,
acceptance, comfort, and support that are a core elements of responsiveness, typical
definitions of responsiveness, such as that of Jackson et a1. (1998), do not mention
communication. The 40-assets framework also treats positive family communication as a
variable separate from support/ positive family relationships. Therefore, the third family
construct measured here was “positive family communication,” which was dealt with as a
separate variable. The questionnaire items selected for each variable and the scale
reliabilities for each family variable are listed in Table 3.2 on the next page.

The family variables were used individually as independent variables. In
addition, as researchers have debated synergistic, additive, or independent effects for
parental demandingness and parental responsiveness, a variable was created to reflect the
combined strength of maintaining standards, positive family relationships, and positive
family communication. All 12 family-related items were combined for a “positive
parenting” variable, with the scores for each item added and then averaged. “Positive

parenting” was used separately then as the fourth independent variable.

49

Table 3.2

Survey Items Used for Family Variables

 

 

 

 

 

 

 

Item Survey Item Alpha if
number item
deleted

Items used for “Maintaining Standards”: reliability = .7072

51 In my family there are clear rules about what I can and cannot do. .6190
(Strongly agree, agree, not sure, disagree, strongly disagree.)

43 If I break one of my parents’ rules, I usually get punished. . (Strongly .6608
agree, agree, not sure, disagree, strongly disagree.)

122 How much of the time do your parents ask you where you are going and .6873
with whom you will be? Wever, Seldom, Some of the time, Most of the
time, All of the time.)

30 My parents push me to be the best I can be. (Strongly agree, agree, not .6532
sure, disagree, strongly disagree.)

23 How often does one of your parents ask you about homework? .6720
(Very often, often, sometimes, seldom, never.)
Items used for Positive Family Communication: Reliability = .7409

135 I have lots of good conversations with my parents. (Strongly agree, .5624
agree, not sure, disagree, strongly disagree.)

121 If you had important concerns about drugs, alcohol, sex, or some other .6910
serious issue, would you talk to your parents about it? (Yes, Probably,
I 'm not sure, Probably not, No)

22 How often does one of your parents talk to you about what you are doing .7145
in school? (Very often, often. sometimes, seldom, never.)
Items used for Positive Family Relationships: Reliability = .8597

49 In my family, I feel useful and important (Strongly agree, agree, not .8247
sure, disagree, strongly disagree)

40 I get along well with my parents. (Strongly agree, agree, not sure, .8269
disagree, strongly disagree)

44 My parents give me help and support when I need it. (Strongly agree. .8084
agree, not sure, disagree, strongly disagree)

48 My parents often tell me they love me. (Strongly agree. agree, not sure. .8257

disagree, strongly disagree)

 

50

Dependent Variables

The two dependent variables are positive values and social competencies. The
positive values Search includes are caring, [promoting] equality and justice, integrity,
honesty, responsibility, and restraint. For each value, there are only one, two, or three
questionnaire items, so reliability measures are not meaningful for each individual
positive value. However, the reliability for all the items together is .8853. All
questionnaire items added reliability to the construct except the two restraint items
(which were reverse-coded to keep a positive scale). Without the restraint items, the
reliability alpha = .8877. Interestingly, reliabilities increased further when each positive
value item (including restraint) was tested with the items Search used to measure social
competencies. However, it was decided that “positive values” and “social competencies”
were more meaningful as separate constructs.

The items selected for the social competencies variable were the items Search
used to measure the “interpersonal competencies,” planning and decision making,” and
“resistance skills.” Each of these items added reliability to the construct. Again, each
asset was measured by only one, two, or three questionnaire items. The conflict
resolution item was not included in this study because it was constructed such that it
could not be converted to a likert-type scale; there were arguably three acceptable or
“right” choices and two “wrong” choices. The reliability alpha for the social competence
items selected was .7780. These reliability results justified using positive values and
social competencies as the dependent variable constructs. The questionnaire items used
for the social competencies variable and the positive values variable are listed in tables

3.3 and 3.4 respectively on the following pages.

51

All items were coded so that the desired or ideal response was given the highest

response value (5) and the least desired response was given the lowest value (1). To

obtain this, all of the questionnaire items used in the independent variables and the two

restraint items used in the positive values dependent variable were reverse coded.

Table 3.3
Survey [tents Used for Social Competencies
Reliability = .7780

 

 

 

Item Survey Item Alpha
number (People who know me would say ...this is "not at all like me ” to “very If rtem
. ., deleted
much like me )
Statements Related to Interpersonal Competence:
69 Caring about other people’s feelings .7360
75 Feeling really sad when one of my friends is unhappy .7652
76 Being good at making and keeping friends .7661
Statements Related to Decision Making:
70 Thinking through the possible good and bad results of different choices .7293
before I make decisions
79 Being good at planning ahead .7588
Statements Related to Resistance Skills:
68 Knowing how to say “no” when someone wants me to do things I know .7426
are wrong or dangerous
74 Staying away from people who might get me in trouble .7514

 

52

Table 3.4

Survey Items Used for Positive Values
Reliability = .8853

 

 

 

 

 

 

 

 

Item Survey Item Alpha if
number (not important, somewhat important, not sure, item
quite important, or extremely important) deleted

Statements Related to Caring:

6 Helping other people .8757

8 Helping to make the world a better place in which to live .8730

13 Giving time or money to make life better for other people .8730
Statements Related to “Equality and Social Justice”

7 Helping to reduce hunger and poverty in the world .8752

10 Helping to make sure that all people are treated fairly .8742

12 Speaking up for equality (everyone should have the same rights and .8774
opportunities)
Statements Related to Integrity:

14 Doing what I believe is right even if my friends make fun of me .8755

15 Standing up for what I believe, even when it’s unpopular to do so .8796
Statements Related to Honesty and Responsibility

l6 Telling the truth, even when it’s not easy .8734

17 Accepting responsibility for my actions when I make a mistake or get in .8763
trouble

l8 Doing my best even when I have to do a job I don’t like .8791
Statements Related to Restraint (Strongly agree, agree. not sure,
disagree, strongly disagree.)

37r It is against my values to drink alcohol while I am a teenager .8837
45r It is against my values to have sex while I am a teenager .8845

 

53

Procedures

The PSL—-AB was given in 35 self-selected school districts during the 1998/1999
school year. Surveys were administered in a classroom setting according to standardized
instructions. Generally, school districts administered the survey to all 6th through 12th
grade students in attendance on the day the survey was administered (Leffert, et al.,
1998). Students were assured their surveys would remain anonymous. After completing
the surveys, the students put them in an envelope that was sealed and mailed back to
Search Institute for processing.

SPSS 11.5 was used to perform the statistical operations administered in this
study and multiple regression analyses were used to answer the proposed questions. Step-
wise multiple regression procedures were first performed using the three independent
variables of maintaining standards, positive family relations, and positive family
communication. The step-wise regression was repeated including the demographic
variables and the independent variables. Additionally, separate correlations were
performed between each independent variable, including the combined positive parenting
score, and each dependent variable. Frequency data was also tabulated for each variable,
creating profiles of the strength and weaknesses of each family-related variable and each

outcome variables for various demographic groupings.

54

CHAPTER 4

RESULTS

This study examined the relationships of the parenting processes to positive
values and social competencies in middle school youth. The parenting processes were
represented by the family variables: maintaining standards, positive family relationships,
positive family communication, and positive parenting (which is the combination of the
aforementioned family variables). Demographic variables were also accounted for. The
results of this study clearly demonstrated that higher levels of positive parenting were
associated with higher levels of positive values and social competencies as reported by
middle school age youth. All the family/parenting variables individually and collectively
were associated positively with both positive values and social competencies. This
chapter will analyze the relationships between each variable, compare their relative
importance, and answer each research question in turn.

Setting the context, the first two questions ask for descriptive results for each

variable.

Research Question 1:
What are the Levels of Each Family Variable in
Michigan Middle School Students?
The results for all variables were skewed towards high scores. The mean level of
positive family communication is 3.56—about halfway between “not sure” or
“sometimes” and “often” or “agree.” The mean level of maintaining standards and

positive family relationships is 4.02 and 4.08 respectively—4.0 being a conventional cut-

off point for “having” an asset measured by a five point Likert scale. See Table 4.1 below

55

for the descriptive statistics results for the family variables. A large majority, 68.4%, of
all middle school students reported “having” positive family relationships, with scores of
4.0 or above. Most middle school students, 63.1%, also reported maintaining standards
scores of 4.0 or above. Less than half (43.9%) of all middle school students reported
positive family communication scores of 4.0 or above. Conversely, while less than 10%
of the students reported low scores (below 3.0) for maintaining standards and positive
family relationships, about 23 % of the students reported low levels of positive family
communication.

Table 4.1

Descriptive Statistics for Family Variables

 

 

Positive Family Positive Family Maintaining
Communication Relationships Standards
N Valid 10168 10298 10168
Missing 355 225 355
Mean 3.5564 4.0828 4.0218
Median 3.6667 4.2500 4.2000
Mode 4.33 5.00 4.20

 

Figure 4.1 on the next page, shows the percentages of students reporting having
the family-related assets at average scores of 4.0 or above compared to the percentages
reporting scores below 3.0. For clarity and comparison purposes, the figure does not
show “neutral” scores, those ranging from 3.0 to less than 4.0, therefore not showing the

students whose responses averaged “not sure” or “sometimes.”

56

Figure 4.1 Percentages of Students Reporting High and Low Levels of Family Variables

 

 

52% Positive Family Communication
I Positive Family Relationships

(Percentages Maintaining Standards
rounded)

Research Question 2:
What are the Levels of Positive Values and Social Competencies
Among Michigan Middle School Students?

About half of the students (51.9%) reported having positive values, with scores at
or above 4.0. The mean score for positive values was 3.85. Many of the students (3,146
to be exact), did not answer one or more of the positive values survey items, so the
number of cases used in the analysis was 7,477. It is not clear how the missing data
affected the results. About a third of the students (32.7%) reported mean scores for social

competencies at or above 4.0. The mean score for social competencies was 3.49. The

number of cases analyzed for social competencies was 10,014. See Table 4.2 following:

57

Table 4.2

Descriptive Statistics for Dependent Variables

 

Social Competencies Positive Values
N Valid 10014 7477
Missing 509 3046
Mean 3.4905 3.8545
Median 3.5714 4.0000
Mode 3 .5 7 4. 15

 

Research Question 3:
What are the Relationships Between each
Family Variable and Positive Values?

As predicted in the hypotheses, higher levels of each family variable were
associated with higher levels of positive values. Simple correlations yielded statistically
significant positive associations between each family variable and positive values. The
positive linear relationship between each family variable and positive values can be seen
graphically in Figures 4.2, 4.3, and 4.4. It should be noted that there is no evidence of a
curvilinear relationship for any of the variables—-as tested for by some researchers,
particularly concerning measures of “control” similar to the maintaining standards

measure(Gray & Steinberg, 1999; Kurdek & Fine, 1994). This study does not show that

any level of the positive parenting variables is “too much.”

58

Figure 4.2

The Relationship Between Positive Family Communication and Positive Values

 

Dot/Lines show Means

5.00-

Positive Values
(.0
8.

2.00"

1.00" I I I l I
1.00 2.00 3.00 4.00 5.00

Positive Family Communication

Figure 4.3

The Relationship Between Positive Family Relationships and Positive Values

 

Dot/Lines show Means

5.00-

Positive Values
00 .e
3 8.

'8.

100' I I I I I
1.00 2.00 3.00 4.00 5.00

Positive Family Relationships

59

Figure 4.4

The Relationship Between Maintaining Standards and Positive Values

 

Dot/Lines show Means
5.00-
4.00-
8
2
a
>
o 300-
.2
.2:
°' 2.00-
1.00' I I I I l
1.00 2.00 3.00 4.00 5.00
Maintaining Standards

Positive family relationships had a Pearson correlation coefficient of r = .420 with
positive values, significant at <.000 level (See Table 4.3 on next page). Maintaining
standards was correlated with positive values with a Pearson correlation coefficient of r =
.419, significant at p < .000. Thus, Maintaining standards and positive family
relationships each account for about 17.6% of the variance in positive values. Positive
family communication and positive values had a Pearson’s correlation of .459, significant
at the .000 level, two-tailed. Therefore, positive family communication had the strongest
association with positive values, accounting for about 21% of the variance in positive
values. Positive parenting (the average of all the parenting items) had a correlation
coefficient of .504, significant at the .000 level. In totality, all the parenting variables

combined account for 25% of the variance in positive'values.

60

Table 4.3

Correlations between Family Variables and Outcome Variables
All Pearson Correlations, significant at the .000 level (two-tailed).

 

 

 

Positive Values Social Competencies
Positive Family .459 .469
Communication
Positive Family .420 .436 _
Relationships
Maintaining .419 .389
Standards I
Positive Parenting .504 .507
Total _ ..

 

These relationships can also be seen in the results of the stepwise regression
procedures done on the same data. See Figure 4.5 and Table 4.5 (a-c) on the next pages
for results of the regression procedures. In the regression model, the strength of the
communication variable is born out. All the relationships were significant at the .000
level, (see Table 4.4 (b). Positive family communication is the strongest predictor—see
the betas in Table 4.4(c) for the relative weighting of the variables. Most of the
relationship of maintaining standards and positive family relationships to positive values
is accounted for in the regression model by their overlap with positive family
communication. After the variance that overlaps with positive family communication is

accounted for, then maintaining standards accounts for an additional 3.6% of the

variance. (See the differences in the R square (r2) values in Table 4.4 (a)). Most of the

association of positive family relationships to positive values is absorbed in the

61

regression model by its overlap with the other family variables, but it adds an additional
1.2% to the variance in positive values that is accounted for by the regression model.
Figure 4.5

The Variance in Positive Values Explained by Family Variables:
The Results of the Stepwise Regression Model

 

All Three Variables Entered: Total r2 =.256
(Total Shaded Area)

 

 

 

 

 
  

. . Maintaining
Posrtrve Values Standards

 

 

 

 

Positive Family Communication:
r2=.208

 

 

 

62

Tables 4.4 (a-c)

Results of SPSS Stepwise Regression for Positive Values: Only Family Variables Entered

 

 

 

 

Table 4.4 (a)
Regression Model Summary for Positive Values
N = 6948
Model R R square Adjusted R Std. Error of
Square the Estimate
1 .456“ .208 .208 .67002
2 .494b .244 .244 .65451
3 .506c .256 .256 .64935
a. Predictors: (Constant), Positive Family Communication
b. Predictors: (Constant), Positive Family Communication, Maintaining Standards
c. Predictors: (Constant), Positive Family Communication, Maintaining Standards

Positive Family Relationships

63

 

Table 4.4(b)

ANOVA Results for Positive Values

 

 

Model Sum of (If Mean F Sig.
Squares Square
1 Regression 819.385 1 819.385 1825.2 .0003
Residual 3 1 18.71 6947 .449
Total 393 8. 10 6948
2 Regression 962.504 2 481.252 1123.4 .000b
Residual 2975.59 6946 .428
Total 393 8. 10 6948
3 Regression 1009.69 3 336.563 798.191 .000c
Residual 2928.41 6945 .422
Total 3938.10 6948

 

Predictors: (Constant), Positive Family Communication

Predictors: (Constant), Positive Family Communication, Maintaining Standards

Predictors: (Constant), Positive Family Communication, Maintaining Standards

64

 

Table 4.4(c)

Coeflicients in Regression Model for Positive Values: Only Family Variables Entered

 

Standardized

 

 

Model Unstandardized C ffi . t t Sig.
Coefficients oe cren S
B Std. Beta
Error

(Constant) 2.560 .032 81 . 185 .000
Positive Family .355 .008 .456 42.722 .000
Communication
(Constant) 1 .897 .048 39.847 .000
Positive Family .254 .010 .326 25.795 .000
Communication

.000
Maintaining .254 .014 .231 18.278
Standards

.000
(Constant) 1.71 1 .050 33.930
Positive Family .192 .01 l .246 16.869 .000
Communication
Maintaining .213 .014 .193 14.857 .000
Standards
Positive Family .140 .013 .152 10.578 .000
Relationships

Question 4:

What are the Relationships Between Each
Family Variable and Social Competencies?

Each family-related variable is also positively associated with social
competencies, as predicted in the hypotheses. The graphs in Figures 4.6 through 4.8 show
this positive association. Again, there is no evidence of a curvilinear relationship for any
of the independent variable. Table 4.3 on page 61 lists the correlation coefficients.

Positive family communication is associated with social competencies with a correlation

65

coefficient (Pearson) of .469 significant at the .000 level. Positive family relationships
had the next highest correlation coefficient, r = .436, significant at the .000 level. For
maintaining standards, r = .389, significant at the .000 level. Again, communication was
the strongest variable, accounting for almost 22 % of the variance in social competencies.
The correlation coefficient for the positive parenting variable is .507, significant at the
.000 level, accounting for 25.7% of the variance. Stepwise regression confirmed positive
family communication as the strongest variable, with positive family communication
first, then positive family relationships and maintaining standards. (See Table 4.5 (a-c)).
All were statistically significant at the .000 level (Table 4.5(b)). For social competencies,
the relative strengths of positive family relationships and maintaining standards was
reversed from that of the regression model for positive values. Positive family
relationships explained another 3% of variance after that accounted for by positive family

communication. Maintaining standards accounted for an additional 1.3% of the variance.

66

 

Figure 4.6

The Relationship Between Positive Family Communication
and Social Competencies

Dot/Lines show Means

.e
8.

Social Competencies
N on
a 8.

1.00.. I I I |
1.00 2.00 3.00 4.00 5.00

Positive Family Communication

Figure 4.7
The Relationship between Positive Family Relationships and Social Competencies

Dot/Lines show Means

5.00'

Social Competencies
w .e
8. 8.

3.3

1,003 I I I I
1.00 2.00 3.00 4.00 5.00

Positive Family Relationships

67

Figure 4.8

The Relationship between Maintaining Standards and Social Competencies

5.00 ‘

4.00 '

3.00 '

2.00 '

Social Competencies

1.00 ‘I I I I I
1.00 2.00 3.00 4.00 5.00

Maintaining Standards

Figure 4.9

The Variance in Social Competencies Explained by Family Variables:

The Results of the Stepwise Regression Model

 

(Total Shaded Area)

 

All Three Variables Entered: Total r2 =.26l

 

 

 
   
  

Social Competencies

 

68

 

 

Positive Family
Relationships

 

 

 

Positive Family
Communication:

r2=.218

 

 

 

Table 4.5(a-c).

Results of SPSS Stepwise Regression for Social Competencies
Only Family Variables Entered

 

Table 4.5(a)

Social Competencies Model Summary

 

 

N=9320
Model R R Square Adsjc‘l’jg‘: R militia: the
1 .467(a) .218 .218 .71845
2 .498(b) .248 .248 .70472
3 .511(c) .261 .261 .69849

 

 

Predictors: (Constant), Positive Family Communication

b Predictors: (Constant), Positive Family Communication, Positive Family
Relationships

c Predictors: (Constant), Positive Family Communication, Positive Family

Relationships, Maintaining Standards

69

Table 4.5(b)

ANO VA Results from Regression Model for Social Competencies

 

 

Model Sum of Squares df Mean F Sig.
Square

1 Regression 1343.374 1 1343.374 2602.546 .0003
Residual 4810.252 9319 .516
Total 6153.626 9320

2 Regression 1525.987 2 762.994 1536.329 .000b
Residual 4627.639 93 18 .497
Total 6153.626 9320

3 Regression 1608.010 3 536.003 1098.628 .000c
Residual 4545.617 93 l 7 .488
Total 6153.626 9320

 

Predictors: (Constant), Positive Family Communication
b Predictors: (Constant), Positive Family Communication, Positive Family

Relationships
c Predictors: (Constant), Positive Family Communication, Positive Family

Relationships, Maintaining Standards

70

 

Table 4.5(c)

Coeflicientsfiom the Regression Model for Social Competencies

 

 

 

Model Unstandardized Standardized t Sig.
Coefficients Coefficrents
B Std. Beta
Enor

(Constant) 2. 1 15 .028 75.043 .000
Positive Family .389 .008 .467 51.015 .000
Communication
(Constant) 1 .639 .037 44. 149 .000
Positive Family .261 .010 .314 26.084 .000
Communication

.000
Maintaining .228 .012 .231 19.176
Standards
(Constant) l .283 .046 27.927 .000
Positive Family .217 .01 l .260 20.602 .000
Communication
Maintaining .187 .012 .190 15.389 .000
Standards
Positive Family .169 .013 .144 12.966 .000
Relationships

 

Question 5:
What are the Relationships Among the Family-Related Variables?

Simple correlations also show that the family variables are positively associated
with each other. Parents and families with one of the family-related assets tend to have
the other family-related assets also (See Table 4.6, next page). The association between
positive family relationships and positive family communication is the strongest variable,
indicating that families with positive relationships tend to have more positive family

communication. However, the fact that far fewer families have positive family

71

 

communication (44%) than have positive family relationships (68%), indicates that the
association between positive family relationships and positive family communication is
not a perfect correlation. The stepwise regression models for both positive values and
social competencies with the three family variables entered showed the overlapping
relationships between the family variables. Positive family communication was the
strongest family variable and hid most of the variance predicted by maintaining standards
and positive family relationships. However, the models still showed additional predictive
value for each family variable. (See Figures 4.5 and 4.9). Similarly, the correlation table
on page 61 (Table 4.3) shows that the correlations between the positive parenting total
and the outcome measures are greater than that between the each individual family
variable and the outcome measures, supporting hypothesis # 4.

Table 4.6

Correlations among the F amily-Related Variables
All Pearson ’3 Correlations, significant at the. 000 level.(Z-tailed)

 

 

Positive Positive Family Maintaining
Family Relationships Standards
Communication
Positive Family .663 .561
Communication
Positive Family .663 .533
Relationships
Maintaining .561 .533
Standards

 

72

Research Question 6:
What are the Relationships among Demographic Variables, Family Variables, and
Positive Values and Social Competencies?

A brief review shows that the demographic variables included were grade, gender,
the educational level of the mother, and family structure (presence or absence of a two-
parent home. The model proposed in Figures 1.1 and 1.2 theorized an indirect effect for
these demographic variables on positive values and social competencies, namely that
these demographic variables would influence parenting style, which would, in turn, affect
the dependent variables. The results obtained only partially support this model, resulting
in a more complex picture. Due to the large sample size, almost all the correlations
between one variable and another were statistically significant. (See Table 4.7 on the
next page.) However, some of the correlations were too small to be meaningful. A
stepwise regression with all the demographic and family variables entered was done with
positive values as the dependent variable (Table 4.8 (a-c)) and again with social
competencies as the dependent variable (Table 4.9 (a-c)). Adding the demographic

variables accounted for an additional 5.3 % of the variance in positive values, increasing

the total r2 from .256 to .309 (see Table 4.8(b)). Adding the demographic variables to

the regression model for social competencies added another 6.1% to the total variance

explained, increasing the total r2 from .261 to .322 (see Table 4.9(b)). In the regression
model for positive values, 5 variables were statistically significant at the .000 level (Table
4.8 (c). For social competencies, seven variables were statistically significant at the .000
level (Table 4.9(c).) The paragraphs following the output tables will take a closer look at

the results pertaining to each demographic variable.

73

 

I

Table 4.7

Demographic Variables Correlations with Family Variables, and Outcome Variables

 

Positive Family
Communication

Positive Family
Relationships

Maintaining
Standards

Positive Values

Social
Competencies

 

Grade Gender Mother’s Two Parent
Education Home
-.l31** .046** .117** .100"
-.121** -.018 .140** .104**
—.083** .012 .119** .120**
-.145** .249** .082** .080**
-.094** .239** .145** .095**

 

Pearson’s Correlations

”significant at the .000 level

Table 4.8 (a-c)

Results of SPSS Stepwise Regression for Positive Values Family and Demographic

Variables Entered

 

74

 

Table 4.8(a)

Statistically Significant Variables Entered in Order of Relative Strength
N = 6013

 

 

 

Model Variable Entered
1 Positive Family Communication
2 Gender
3 Maintaining Standards
4 Positive Family Relationships
Grade
Table 4.8(b)

Regression Model Summary For Positive Values

 

Adjusted R Std. Error of the

 

Model R R Square Square Estimate
1 .453(a) .205 .205 .66884
2 .507(b) .257 .257 .64640
3 .540© .292 .292 .63122
4 .554(d) .307 .306 .62472
5 .557(e) .310 .309 .62329

 

Predictors: (Constant), Positive Family Communication
Predictors: (Constant), Positive Family Communication, Gender

0"”

c Predictors: (Constant), Positive Family Communication, Gender, Maintaining

Standards

(1 Predictors: (Constant), Positive Family Communication, Gender, Maintaining

Standards, Positive Family Relationships

e Predictors: (Constant), Positive Family Communication, Gender, Maintaining

Standards, Positive Family Relationships, Grade

75

 

Table 4.8©

ANOVA Results for Positive Values

 

 

 

 

 

Model Sum of Squares df Mean F Sig.
Square

1 Regression 692.784 1 692.784 1548.631 .000“
Residual 2689.483 6012 .447
Total 3382.267 6013

2 Regression 870.688 2 435.344 1041.915 .000b
Residual 2511.579 6011 .418
Total 3382.267 6013

3 Regression 987.674 3 329.225 826.295 .000°
Residual 2394.593 6010 .398
Total 3382.267 6013

4 Regression 1037.078 4 259.270 664.318 .000d
Residual 2345. 1 89 6009 .390
Total 3382.267 6013

5 Regression 1048.222 5 209.644 539.640 000°
Residual 2334.045 6008 .388
Total 33 82.267 6013

Table 4. Predictors: (Constant), Positive Family Communication

b Predictors: (Constant), Positive Family Communication, Gender
c Predictors: (Constant), Positive Family Communication, Gender, Maintaining
Standards
(1 Predictors: (Constant), Positive Family Communication, Gender, Maintaining
Standards, Positive Family Relationships
Table 4. Predictors: (Constant), Positive Family Communication, Gender,
Maintaining Standards, Positive Family Relationships, Grade

76

 

Table 4.9 (a-c)

Results of SPSS Stepwise Regression for Social Competencies
Family and Demographic Variables Entered

 

Table 4.9(a)

Statistically Significant Variables Entered in Order of Relative Strength
N =8072

 

 

 

Model Variable Entered
1 Positive Family Communication
2 Gender
3 Positive Family Relationships
4 Maintaining Standards
5 Mother’s Education
6 Two-parent Home
7 Grade
77

Fae.

 

Table 4.9(b)

Regression Model Summary For Social Competencies

 

Adjusted R Std. Error of

 

 

Model R R Square Square the Estimate

1 .468(a) .219 .219 .71001

2 .519(b) .269 .269 .68692 _

3 .551(c) .303 .303 .67081 [A
4 .562(d) .315 .315 .66508

5 .567(e) .321 .321 .66240

6 .567 (f) .322 .321 .66206

7 .568( g) .322 .322 .66180

 

Predictors: (Constant), Positive Family Communication
b Predictors: (Constant), Positive Family Communication, Gender
c Predictors: (Constant), Positive Family Communication, Gender,
Positive Family Relationships
d Predictors: (Constant), Positive Family Communication, Gender,
Positive Family Relationships, Maintaining Standards
e Predictors: (Constant), Positive Family Communication, Gender,
Positive Family Relationships, Maintaining Standards, Mother's Education
f Predictors: (Constant), Positive Family Communication, Gender,
Positive Family Relationships, Maintaining Standards, Mother's Education,
Two-parent Home
g Predictors: (Constant), Positive Family Communication, Gender,
Positive Family Relationships, Maintaining Standards, Mother's Education,

Two-parent Home, Grade

78

 

Table 4.9 (0)

ANOVA Results for Social Competencies

 

 

 

Model Sum of (If Mean F Sig.
Squares Square
1 Regression 1143.998 1 1143.998 2269.299 .000a
Residual 4068.748 807 1 .504
Total 52 12.746 8072
2 Regression 1404.822 2 702.411 1488.595 .000b
Residual 3807.924 8070 .472
Total 5212.746 8072
3 Regression 1581.834 3 527.278 1171.774 .000c
Residual 3630.912 8069 .450
Total 5212.746 8072
4 Regression 1644.047 4 411.012 929.202 .000“I
Residual 3568.699 8068 .442
Total 5212.746 8072
5 Regression 1673.122 5 334.624 762.628 .0006
Residual 3539.624 8067 .439
Total 52 12.746 8072
6 Regression 1677.215 6 279.536 637.736 .000f
Residual 3535.530 8066 .438
Total 5212.746 8072
7 Regression 1680.468 7 240.067 548.128 .000g
Residual 3532.278 8065 .438
Total 5212.746 8072

a. Predictors: (Constant), Positive Family

b Predictors: (Constant), Positive Family Communication

c Predictors: (Constant), Positive Family Communication, Positive Family

79

 

 

d Predictors: (Constant), Positive Family Communication, Positive Family
Relationships, Maintaining Standards
e. Predictors: (Constant), Positive Family Communication, Positive Family
Relationships, Maintaining Standards, Mother’s Education
f. Predictors: (Constant), Positive Family Communication, Positive Family
Relationships, Maintaining Standards, Mother’s Education, Two-Parent Home
g. Predictors: (Constant), Positive Family Communication, Positive Family
Relationships, Maintaining Standards, Mother’s Education, Two-Parent Home
Grade
Social competencies and positive values were negatively associated with grade
level. Youth in higher grades reported lower levels of social competencies and positive
values (See Figures 4.10and 4.11). They also reported lower levels of each parenting
variable. The correlations between grade and the family variables were statistically
significant, although very small---the highest being with Communication ( -.131) (See
Figure 4.11.) For both social competencies and positive values the association with grade
level was statistically significant and was retained in the stepwise regression model. See
the preceding tables, Table 4.7, Table 4.8 (a) and Table 4.9 (a). Grade came in fifth--
after communication, gender, positive family relationships, and maintaining standards for

positive values. For social competencies the results were similar, again grade level came

in 1ast--seventh place.

80

 

 

Figure 4.10

Bar Graph Showing Mean Social Competencies by Grade Level

 

Bars show Means

 

 

 

Social Competencies

 

Figure 4.1 1

Bar Graph Showing Mean Positive Values by Grade Level

 

Bars show Means

5.00 -

4.00-

Positive Values

 

 

 

81

 

Figure 4.12
Bar Graph Showing Mean Communication by Grade Level

Bars show Means
5.00-

 

Positive Communication

   

 

Grade

Gender

Gender was among the strongest independent variables, with males scoring
significantly lower on both positive values and social competencies (see Figures 4.13 and
4.14). This is indicated by the Pearson’s correlations of .249 and .239, respectively—
both significant at the .000 level. Gender was also statistically significant in the stepwise
regression models. In the regression model for positive values, gender explained an
additional 5.2% of the variance after communication was accounted for. In the regression
model for social competencies, gender explained an additional 5%. Gender had no
statistically significant association with maintaining structure or positive family
relationships. The association of gender with positive family communication did reach
the level of statistical significance, but the correlation was so small (.046) that it is
essentially meaningless, a product of the large sample size (Males had a mean

communication score of 3.5 1. For females it was 3.60). Therefore, as measured by this

82

study, the effect of gender on positive values and social competencies is a direct effect,
rather than one mediated through differing parenting styles for male adolescents and
female adolescents.

Figure 4.13
Bar Graph Showing Mean Positive Values by Gender

Bars show Means

5.00 -

 

Positive Values

   

0.00-
1=MALE 2=FEMALE

Gender

83

 

Figure 4.14

Bar Graph Showing Mean Social Competencies by Gender

Bars show Means

5"
o
q:

.e
3

3.00-

2.00-

Social Competencies

1 .OO‘

 

0.00-

1=MALE 2=FEMALE
Gender

T wo-Parent Home

Family structure was coded binomially as the presence or absence of living in a
two-parent home. The question asked did not differentiate between living with a parent
and a step-parent and living with the student’s biological or adoptive parents. Living
sometimes with one’s mother and sometimes with one’s father was coded as not living in
a two-parent (intact) home. Of the middle school students surveyed, 74.2% lived in a
two-parent home. Living in a two-parent home had small positive associations with
family/parenting variables and social competencies and positive values (See Table 4.7).
The correlations were statistically significant at the .000 level, but again too small to
contribute much to the overall picture. The strongest association was with maintaining
standards, r =. 120, mildly supporting previously reported findings that single parent

homes monitor less or are less effective at maintaining standards. The associations of a

84

 

 

two-parent home with the parenting variables were stronger than the associations with the

outcome variables, supporting the hypothesized model of demographic influence

mediated by differences in parenting. For social competencies, however, the two-parent

home variable was found statistically significant in the stepwise regression model,

indicating it explained a tiny fraction of additional variance over and above that

explained by family/parenting variables, and therefore, was not retained in the regression re

for positive values.

 

Educational Level of the Mother

All correlations for the educational level of the mother were again statistically
significant, but of relatively minor importance (See Table 4.7). Families with more
educated mothers were somewhat more likely to maintain standards, have positive family
relationships, and have positive communication. Children of more highly educated
mothers also scored slightly higher on social competencies and positive values. The
regression model for social competencies retained educational level of the mother after
the parenting variables and gender, adding an additional .006 to the R square value(See
Table 4.9). In the regression models for positive values the educational level of the
mother was not statistically significant. The strongest (but still weak r = .145)
association of any parent demographic variable with another variable was the association
of mother’s educational level with social competencies. The educational level was not
recorded as an interval or continuous variable, so the results are tainted by that fact. The
survey responses were recoded into the following categories: l= did not complete high
school, 2 = completed high school, 3 = some college, 4 = completed college, 5 = graduate

or professional school after college.

85

I I
I “Ii-:21.

CHAPTER 5

CONCLUSIONS, DISCUSSION, AND IMPLICATIONS

Conventional wisdom and scientific research has generally advocated a two-
dimensional approach to parenting: love and limits, warmth and firmness. This study
confirmed the importance of those two dimensions: maintaining or enforcing standards ’7“

and warm, positive relationships. Both of these dimensions were associated positively

 

with positive values and social competencies in the middle school youth surveyed.
However, this study added an important factor to the composite of positive parenting that
is often left out. It found frequent, positive, and open communication between
adolescents and their parents to be the aspect of the parent-adolescent relationship most
strongly associated with positive values and social competencies. It is also the most
likely—out of the three parenting variables assessed in this study—to be weak or
missing. This presents direction and opportunities for improving the outcomes for
children. This chapter will review the major findings of this study and examine the
implications for both future research and practical programming for families and
adolescents.
Review of Major Findings

Findings About Family Variables

All the hypotheses proposed in this study were statistically supported by the
results of this study. Higher levels of positive family communication, positive family
relationships, and maintaining standards were each associated with higher levels of

positive values and higher levels of social competencies, as reported by Michigan middle

86

 

school students taking the Search Institute’s PSL—-AB survey. It was also hypothesized
that “students scoring higher on all three parenting variables will score higher on positive
values and social competencies.” This hypothesis was also supported. The correlations
between the positive parenting total and positive values and social competencies were
positive and were higher than the correlations between the individual parenting variables
and the outcome variables. Thus, some additive effect is present. Having a parenting style I'-
that includes all three dimensions of parenting is associated with somewhat higher levels
of positive values and social competencies than any single parenting dimension by itself.

'/

This adds to the considerable body of previous research demonstrating the positive

 

effects of an authoritative parenting style. It also contributes to the research associating
increased levels of developmental assets with positive developmental outcomes for
youth.

Of the variables examined, positive family communication had the strongest
relationship with both positive values and social competencies, although all the
parenting/family variables were associated positively with the outcome variables.
Positive family communication, maintaining standards, and positive family relationships
also had a strong tendency to co-occur. Not surprisingly, families that had positive
relationships with each other, where parents provide help and support when needed and
often express love verbally, were also the families most likely to have positive family
communication. Still, although over two-thirds of the students reported having positive
relationships in their families, only about 44% reported having positive communication
overall. Looking at the specific communication items measured, about half (52.1%) of

the youth respondents in this survey reported that they would probably talk to their

87

parents about serious issues such as drugs, alcohol, or sex. Further, 60% agreed that they
had lots of good conversations with their parents and 62% reported that their parents
“often” or “very often” talk about what the child did in school. Clearly, there is
considerable room for improvement on this dimension of parenting and family life.

In summary, the four most important findings concerning the parenting/family
variables are as follows:

1. Positive family communication, positive family relationships, and maintaining

standards were each found to have a positive relationship with positive values and

social competencies.

 

2. Positive family communication was found to be strongest predictor of positive
values and social competencies in this investigation.
3. The combination of all three parenting variables was found to strengthen the
relationship with positive values and social competencies.
4. Positive family communication was found to be the weakest family variable for
most middle school students surveyed.
Findings From Demographic Variables
Gender was one of the strongest predictors in the regression models for both
positive values and social competencies, second only to positive family communication.
Males reported significantly lower levels of positive values and social competencies than
females. The correlations between family variables and gender were not statistically
significant (positive family relationships and maintaining standards) or so low as to have
no real world significance ( r = .046 for gender and positive family communication). In

other words, these young males did not report differences from females in

88

 

parenting/family dynamics. This demonstrates a direct effect of gender on positive
values and social competencies rather than an effect mediated by parenting style.

Grade level was associated with the family variables, and the positive values and
social competencies variables, with small, but statistically significant negative
correlations. Grade level was significant in the regression models for both positive values
and social competencies. This contradicts a common hope or expectation of parents and
educators that positive values and the various social competencies would improve with
age. As this study involved only student-reported perceptions, it may be that the decrease
is more perceptual than actual. Possibly younger children have a slightly more idealistic
perception of themselves and their families than do older children. During the middle
school years, with so much biological and social adjustment, it also may be that there is a
slight overall increase in general negativity of attitudes and behavior. While most of the
current literature discounts the notion of drastic upheavals and storm and stress across the
board, this interpretation is not incompatible with prevailing research and knowledge
about early adolescence.

The following summarizes the major findings regarding the demographic
variables:

1. Gender was the strongest predictor of positive values and social competencies

after positive family communication.

2. The mother’s educational level and having a two-parent home had very small,

but statistically significant, association with social competencies.

3. Grade level was slightly negatively associated with positive values, social

competencies, and the family variables.

89

 

 

 

 

Implications

Positive Values, Social Competencies and Ecological Considerations

The development of positive values, morals, character, and ethics does not occur
in a vacuum. Parents and families do not exist in a vacuum. Nor does the association
with positive parenting account for all, or even most, of the variance in adolescents’
values and social competencies. Ecological theory emphasizes context and the interplay r
between systems. Today’s youth grow up in a society that conveys mixed messages
about such values as integrity, honesty, restraint, and responsibility. Adults do not

always provide good role models in restraint, good decision-making, or quality

 

fiiendships.
An effort to promote positive values and social competencies among youth must
address these factors as well. “Moral development, like all human development,
proceeds through social experience” (Pace, 2003, p. 258). These social experiences
ideally include consistent messages, positive relationships, and opportunities for both
discussing and practicing the positive values promoted. This study has confirmed the
importance of certain social experiences within the family for the development of
positive values and social competencies. However, these opportunities and caring
relationships with adults who are genuine role models can and should occur in a variety
of settings. According to Pace, “connections to the sacred and the cultivation of
spirituality are also important aspects to moral and character education” (2003, p. 259).
Churches and faith-based institutions, extra-curricular youth organizations, schools,
communities, and the media all have a role to play. They can directly support and

enhance the deveIOpment of character and competencies in youth by offering specific

90

2'31

character education, opportunities for service to others, constructive, character-building
activities, and positive messages about values through a variety of media.

They can also provide support, encouragement, advice, and training for parents,
thus changing the norm for parents of adolescents to “go it alone” (YMCA & Search-
Institute, 2002). Parents and others involved in building assets among youth should be
viewed as partners with a common goal (Simpson & Roehlkepartain, 2003). This study
confirms that the advice and training for parents should include supporting parents in
maintaining standards, developing positive family relationships, and promoting positive
family communication, all of which are associated with positive values and social
competencies in youth.

Program Development

The most important and useful implication derived from this study is that there is
considerable room for improvement in the dimension of parenting and family life found
to be most strongly associated with positive values and social competencies, namely,
positive family communication. It is a dimension of family dynamics that can be
strengthened through programs designed to teach families parenting skills. Programs like
this have been developed for the primary purpose of preventing drug abuse. Two well-
researched and proven programs are Preparing for the Drug Free Years and the
Strengthening Families: 10-14 program put on by Iowa State University Extension (Rick
Kosterman, et al., 2001; Rick Kosterman, et al., 1997; Kumpfer & Tait, 2000). These
programs have not only been documented in controlled studies to reduce or prevent drug
and alcohol use, but have also been documented to improve skills, such as setting

appropriate limits, increasing proactive communication, strengthening bonds and

91

 

improving relationships between parents and their children (Rick Kosterman, et al., 1997;
R. Spoth, Redmond, Haggerty, & Ward, 1995). These 5 to 7-week programs involved
parents and their children and were intended as a universal preventative intervention for
the general population. Other versions of the Strengthening Families Program were
designed for specific ethnic audiences. While these parent-training programs were not
designed specifically to improve positive values and social competencies, this study
suggests that programs that improve communication between youth and their parents and
help parents maintain or develop appropriate standards and improve relationships may,
indeed, produce those positive benefits as well as reduce risk behaviors.

A large obstacle preventing parenting programs from having a broad societal
impact is that few parents choose to attend them voluntarily (McCurdy & Daro, 2001).
Most parents expect to “go it alone” (YMCA & Search-Institute, 2002). Recruitment
issues, barriers to attendance, and parental attitudes and concerns are areas beginning to
be studied by researchers. A preliminary study from Search Institute and the YMCA of
the USA reported five key findings (2002):

1. Most parents do not turn to friends, family, or community resources for

support in parenting.
2. Parents with good relationships with their partners are more likely to feel
successful as parents.
3. Most parents interviewed do feel successful as parents most of the time.
4. Job demands, sibling rivalry, over-scheduling, and financial stresses are

among the challenges that make parenting harder.

92

 

 

5. The things parents thought would help them include talking to other parents,
being affirmed for their parenting, advice from trusted professionals, and
having other trusted adults spend time with their children.

Additional studies found that time-related factors were the reason most often cited

by parents for not attending preventative programs for parents of young adolescents (R.
Spoth & Redmond, 2000; R. Spoth, Redmond, Hockaday, & Shin, 1996). Additionally, F
certain segments of the parent population have “relatively low levels of motivation” for
attending such programs (R. Spoth, et al., 1996). Many parents perceive their children as

“already doing fine” and do not believe their child is at risk, so attending a parent-training

 

program does not seem worthwhile (R. Spoth, etal., 1996). An exploratory study of the
concerns of parents of middle school students found that parents were not worried about
issues like sex and drug use—at least in the middle to upper-middle class community
studied (Lewis, 2002).

Findings from this study, particularly if confirmed and expanded by further
research, could be used to provide increased motivation for parents to attend preventative
programming. Parents who are not concerned about potential drug use by their children
may be motivated to improve the levels of positive values and social competencies in
their children—especially if they realize there is a tendency toward decreases in these
areas in the middle school years. Parents could be informed that, even in families where
children report positive relationships, there may be a lack in the positive family
communication that is most strongly associated with positive values and social

competencies.

93

 

Parents surveyed for the YMCA/Search Institute study expressed opinions that
affirmation as parents, talking to other parents, and receiving advice from trusted
professionals would be helpful to them. These are things that can be provided by parent-
training programming. The results of this study can also be used to provide the
affirmation parents desire, as well as motivation. Affirmation does come from students
who report that their parents are generally doing well in the important dimensions of
positive relationships and maintaining standards in their families. The motivation is that
improving these areas and especially improving communication may make a significant
difference in the development of positive values and social competencies. 1f parents are
not worried, and therefore motivated by, concerns of high-risk behaviors, perhaps a more
positive motivation might seem compelling. It makes sense to promote and develop
programs that assist parents proactively during early adolescence, before problems that
are more serious develop, and before communication deteriorates further.

This study also points to another area of program design that warrants attention.
Programs, organizations, mentors, and models that promote positive values and social
competencies are important during the middle school years. While this research
highlighted the role of families in relation to positive values and social competencies,
ecological theory suggests families benefit from the support of the larger community.
The reported decreases in positive values and social competencies during the middle
school years can be addressed on many fronts. Findings from the data set used in this
study show that time spent at home and involvement in religious communities drops

during the middle school years (Keith, et al., 2002). This presents a challenge: keeping

94

 

TT-il

youth involved in activities and environments that actively promote positive values and
social competencies.

In regard to gender, this study raised questions rather than providing answers.
Further research is needed. The middle school males surveyed reported significantly
lower levels of positive values and social competencies. The differences in mean scores
for positive values and social competencies by gender persist in late high school, with
mean scores of 3.27 and 3.20 for males and 3.68 and 3.67 for females. The middle
school males did not report differences in the family and parenting dimensions studied,
however. Males in middle school and high school also report lower levels of
involvement in religious communities, achievement motivation, school engagement,
service to others, and are less likely to have relationships with adults other than their
parents (Keith, et al., 2002). How might males be encouraged to participate in
constructive activities? Are our institutions, such as schools and places of worship,
designed to accommodate and engage young males? How might virtues, character, social
competencies, and spirituality be modeled and presented so that they appeal to young
males? Where are the masculine role models for strength of character and caring? What
role does the media play? What roles do fathers play? It is clear there is a need for
answers and the will to implement them.

Program Evaluation

Programs for youth, if they are evaluated for effectiveness at all, hope and expect
to see improvement in measured outcomes. This study, along with the larger study from
which the data was derived, found decreases in many assets as grade levels increased.

Whatever the explanation for the decreases with age, if this pattern holds up after further

95

F..-

 

I

research, it is important knowledge for program developers, program evaluators, parents,
and all those working with early adolescents. If, in general, self-reported levels of
developmental assets tend to decrease in the middle school years, then programming for
adolescents that appears to have had no positive effect, may have indeed prevented an
expected decrease in developmental assets. It will be important for evaluators and others
to understand whether these self-reported decreases are functions of perceptual changes
or reflections of actual decreases in assets substantiated by other objective measures.
Further research that involves some objective measure of developmental assets by outside
observers can illuminate these issues.
Limitations and Implications for Future Research

This study utilized secondary data generated from the Search Profiles of Student
Life: Attitudes and Behavior survey. This imposed several limitations. First, the only
questions available were those designed to suit the original researchers’ purposes. The
PSL—-AB provided a broad overview of adolescent life, but generally often had only one,
two, or three items to measure each construct. The greater clarity, depth, and precision
that may have resulted from additional questions related to the variables in this study
were not available. For instance, students were asked about the likelihood of punishment
if they broke one of their parents’ rules. They were not asked what type of punishment
would be imposed, or whether it seemed reasonable and fair, or harsh and excessive. The
research literature would indicate that those are important distinctions.

The survey questions about parenting did not distinguish between mother-child
dynamics and father-child dynamics—another instance where additional depth or

precision may have been illuminating.

96

 

Future research about the nature, extent, quality, content, and dynamics involving
family communication in the adolescent years is clearly warranted. What types of
communication are most associated with positive values and social competencies? What
can families do or not do to promote that type of communication? What factors make it
most likely that youth will talk to their parents about serious issues? How much does the
quantity of daily “what did you do in school today” type of communication matter?
These are questions for additional research.

Additional research is also needed to address the perceptual versus the behavioral

components of the variables studied. How well would the perceptions of middle school

 

students regarding their own values and competencies match the perceptions of adults
who know them or with objective measures of behaviors? In other words, are the
parenting measures studied associated with adolescents who actually behave more
honestly, responsibly, and with more caring and integrity? Also, previous research
indicates that adolescents’ perceptions of parenting practices do not necessarily match the
perceptions of the parents themselves. Do parents perceive more positive communication
in the home than do their children? Do parents accurately assess whether their children
would be likely to come to them with serious issues or concerns? Additional research is
also needed to provide an observational or third-party measure to assess the decrease in
assets with grade level found in this study. Extending this research past the middle
school age group is also an important next step, to assess whether the decrease in assets,
found in this study to be associated with increasing grade level, continues into high

school.

97

 

Another limitation posed by the use of the PSL--AB survey is that all the survey
items do not use the same scale for the response options. All the “positive values”
questions except the two measuring restraint had not important, somewhat important, not
sure, quite important, and extremely important as the response options. For the other
dependent variable, social competence, the response options stated, “People who know
me would say that this is. . .not at all like me, a little like me, somewhat like me, quite like
me, [or] very much like me.” Most of the independent variable items used a Likert scale
of strongly agree, agree, not sure, disagree, and strongly disagree. Some of the items
however had a response scale of very often, often, sometimes, seldom, and never.
Differences between variables may be affected by differences in the possible responses
offered.

Additionally, this study is limited to adolescent-reported measures. Data obtained
by direct observation of the researcher or the reports of others are not available. Multiple
sources of data would have been ideal. Shared source variance may produce
overestimates of the relationship between family variables and the internal assets of
adolescents (Jackson, et al., 1998). For example, an extremely optimistic or pessimistic
child could over- or under-report both family related assets and internal assets. However,
many researchers have concluded that both objective and subjective assessments of
parenting practices provide important information to the researcher (Lambom, et al.,
1991; Muuss, 1996; Shucksmith & et al., 1995). The adolescents’ perception of
parenting practices may be more predictive of outcomes than the parent’s actual behavior
(Paulson & Sputa, 1996). The use of the subjective point of view is also consistent with

ecological theory (Muuss, 1996).

98

 

The conclusions and results of this study can only be generalized to similar
groups of adolescents. Because the Michigan data from the PSL--AB is primarily from
small and mid-sized cities and suburban areas, the conclusions cannot be applied to large-
city, urban populations. The respondents in this study were also 90% White/non-
Hispanic. A broad base of previous research on the effects of parenting style finds an
authoritative parenting style generally effective across many cultures and ethnicities, but
also finds differences among communities. Additional research among diverse
communities would ascertain whether these findings apply to other populations of youth
and families. The responses cannot be assumed to be representative of communities with
a different cultural or ethnic composition.

Research should also be connected to community development efforts,
cooperating with and sharing information with those involved in policy-making,
programming, and intervention. Findings must be made available to parents and others
motivated to promote positive youth development. The Search Institute’s PSL—AB
survey was designed with these youth development goals in mind. Research should
continue to inform program development and include studies of intervention
effectiveness. Recruitment barriers and programming specific to particular populations
are areas for research focus (Lerner & Galambos, 1998).

Another limitation to the analysis presented here is the uneven distribution of
missing data. Thousands of cases—roughly a third of the data set—were not able to be
used in statistics using the positive values variable. This was not the ease for statistics

using the social competencies variable. How this affected the outcomes is not known.

99

 

 

Comparisons between the positive values variables and the social competencies variables
should be made cautiously for this reason.

This study was also limited by its cross-sectional nature. Although previous
research on parenting, including longitudinal studies, lends credence to the notion of
causality, this study could not make that claim. It is possible that parents with children,
who, for some other reason, are higher in positive values and social competencies, are
more likely than other parents to have positive communication, positive relationships,
and maintain standards. Certainly, a common sense view would maintain that it would be
easier to have positive relationships with children who are socially competent.

It is also possible that some genetic disposition toward constructive social
dynamics and values (shared by the parents and their children) influences both the
parenting styles of the parents and positive values and social competencies of their
adolescent children. In fact, it seems likely that these possibilities do account for some
measure of the observed relationships. An ecological perspective expects bi-directional
influence. Reality is generally more complex than any one study can explain. Even the
most recent and most extensive studies of the relationships between genetic influences,
family influences, and non-shared environmental influences, do not fully resolve or
explain these issues.

A large, longitudinal study of families with adolescents that included sibling and
twin comparisons found genes and genetic change important factors, but also found more
family or shared environment influence than the authors expected (Reiss, Neiderhiser,
Hetherington, & Plomin, 2000). The best research portrays a complex, multi-facetted

picture of adolescent development, with much of the detail still out of focus and unclear.

100

 

Yet, the stakes are high. A generation of socially competent youth with strong and
positive values is needed to shape the society of the future. While research continues,
working together and putting into practice the best knowledge we have about raising

competent, caring, moral, young people, is essential.

 

lOl

 

APPENDIX

102

 

MICHIGAN STATE

UNIVERSITY
May19,2003

 

TO: JoanneKErI'H
203mm

RE: R8 0 01-307 CATEGORY: 1-4 EXEMPT

APPROVAL DATE: March 4, 2003
EXPIRATION DATE: February 4. 2004

TITLE: BUILDING WON BEST LIVES: A STATEWIDE AGGREGATE
ThelhiverslyCommltteeonResearchlnvolvthmanSlflects' (UCRIHSMevIewoflrlsprojectB
mmwlmmmmmmmmwwmammmpeembm
adequadypmtededmdmemodstoobtainflamedcmsuuareappmprhle. Thereforethe
UCRIHSAPPROVEDTHSPROECTSREVISION.

REVISION REQUESTED: May 14. 2003

REVISIONAPPROVALDATE: Math. 2003

Changelnthestudylnvestlgators. AdditionallnvestigatorswiillncludeBruceHaas
andBaIbaraHiilaker.Changeslnthedataanaiysls.

 

REVISIONS: UCRIHSmustmvlewanydraroeshprocedmeskwolvhgtammufledsmrbrb
hltiationofthedrange. lithislsdoneatthefinedrenewalmleaseuselhegeenrenewdform.
Tommwmdwmmmhmnwmmmbh
UCRlHSChaI, requesthgrevlsedqaprovalandmflreproject‘smandfle. Includeh
yourequestadescrbtionoflhechangel'd ”www.mma
advertisementslhatareappliceble. .

PROBLEMSICHANGES: Shoudeitherotmelolowhgarlseduhgthecouseoflhewknofly
UCRIHS promptly: 1) problems (ur'rexpected side effects, comptalnts, etc.) Inolvhg human
Wamdmrgeshmreseadrmalamflmfionkmmmkbm
mmmmmuwmmwmw
flmmbeoifutherasslstanoe, pleaseoonlactuset(517)355-21800rvlaemal:
UCRlHSerisuHedu

Sincerely,

 

. Franciso A. VILLARUEL
°°- 107 Human Ecology
usu

Wh-m
M“

103

 

REFERENCES

 

Adalbjamardottir, S., & Hafsteinsson, L. G. (2001). Adolescents' perceived parenting
styles and their substance use: Concurrent and longitudinal analyses. Journal of
Research on Adolescence, 11(4), 401-423.

Baumrind, D. (1975). Early socialization and the discipline controversy. Morristown,
N.J.: Silver Burdett Company. a...

Baumrind, D. (1991). Effective parenting during the early adolescent transition. In P. A.
Cowan & E. M. Hetherington (Eds.), Family transitions (pp. 1 1 1-163). Hillsdale,
New Jersey: Lawrence Erlbaum Associates.

 

Benson, P. L. (1997). All kids are our kids: What communities must do to raise caring .i‘w
and responsible children and adolescents (1st ed.). San Francisco: J ossey-Bass.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature
and design. Cambridge, Massachusetts: Harvard University Press.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development:
Research perspectives. Developmental Psychology, 22(6), 723-742.

Brook, J. S., Whiteman, M., & Finch, S. (2000). Longitudinally foretelling drug use in
the late twenties: Adolescent personality and social-environmental antecedents.
The Journal of Genetic Psychology, 161(1), 37-51.

Brown, J. H., D'Emidio-Caston, M., & Benard, B. (2001). Resilience education.
Thousand Oaks, CA: Sage Publications.

Buist, K. L., Dekovic, M., Meeus, W., & Aken, M. A. G. v. (2002). Developmental
patterns in adolescent attachment to mother, father, and sibling. Journal of Youth
and Adolescence, 31(3), 167-176.

Cavell, T. A. (1990). Social adjustment,social performance, and social skills: A tri-
component model of social competence. Journal of Clinical Child Psychology,

19(2), 111.
104 j g

Chao, R. K. (2001). Extending research on the consequences of parenting style for
chinese americans and european americans. Child Development, 72(6), 1832.

Compas, B. E., Hinden, B. R., & Gerhardt, C. A. (1995). Adolescent development:
Pathways and processes of risk and resilience. Annual Review of Psychology, 46,
265.

Dombusch, S. M. (1987). The relation of parenting style to adolescent school
performance. Child Development, 58(5), 1244-1257.

Dryfoos, J. G. (1990). Adolescents at risk: Prevalence and prevention. New York:
Oxford University Press.

Dukes, R. L., & Stein, J. A. (2001). Effects of assets and deficits on the social control of
at-risk behavior among youth. Youth and Society, 32(3), 337-359.

Fletcher, A. C., & Jefferies, B. C. (1999). Parental mediators of associations between

perceived authoritative parenting and early adolescent substance use. The Journal
of Early Adolescence, 19(4), 465.

Freeman, H. S., & Newland, L. A. (2002). Family transitions during the adolescent
transition: Implications for parenting. Adolescence, 37(147), 457-475.

Galambos, N. L., Barker, E. T., & Almeida, D. M. (2003). Parents do matter:
Trajectories of change in externalizing and internalizing problems in early
adolescence. Child Development, 74(2), 578.

Gambone, M. A., Klem, A. M., & Connell, J. P. (2002). Finding out what matters for
youth: Testing key links in a community action framework for youth development.
Philadelphia: Youth Development Strategies, Inc., and Institute for Research and
Reform in Education.

Gegas, V., & Seff, M. A. (1990). Families and adolescents: A review of the 1980's.
Journal of Marriage and the Family, 52, 941-958.

Gordon, C. P. (1996). Adolescent decision making: A broadly based theory and its
application to the prevention of early pregnancy. Adolescence, 31 (123), 561.

105

 

 

Gray, M. R., & Steinberg, L. (1999). Unpacking authoritative parenting: Reassessing a
multidimensional construct. Journal of Marriage and the Family, 61(3), 574-587.

Gunnoe, M. L., Hetherington, E. M., & Reiss, D. (1999). Parental religiosity, parenting
style, and adolescent social responsibility. Journal of Early Adolescence, 19(2),
199-225.

Hamburg, D. A. (1997). Toward a strategy for healthy adolescent development. The
American Journal of Psychiatry, 154(6), 6.

Hartos, J. L., & Power, T. G. (2000). Relations among single mothers' awareness of their
adolescents' stressors, maternal monitoring, mother-adolescent communication,
and adolescent adjustment. Journal of Adolescent Research, 15(5), 546-563.

Hartup, W. (1995). The company they keep: Friendships and their developmental
significance. In R. M. Lerner & D. F. Perkins (Eds), Social interactions in

adolescence and promoting positive social contributions of youth (pp. 1-13). New
York: Garland Publishing.

Henry, C. S., Sager, D. W., & Plunkett, S. W. (1996). Adolescents' perception of family
system characteristics, parent-adolescent dyadic behavior, adolescent qualities,
and adolescent empathy. Family Relations, 45(3), 283.

Hogue, A., & Liddle, H. A. (1999). Family-based prevention intervention: An approach
to preventing substance use and anti-social behavior. American Journal of
Orthopsychiatry, 69(3), 278-290.

Holmbeck, G. N., Paikoff, R. L., & Brooks-Gunn, J. (1995). Parenting adolescents. In M.
H. Bornstein (Ed.), Handbook of parenting (Vol. 1, pp. 91-111): Lawrence
Erlbaum, Associates.

Irvin, J. L. (1996). Developmental tasks of early adolescence: How adult awareness can
reduce at-risk behavior. The Clearing House, 69, 222-225.

J accard, 1., Dittus, P. J ., & Gordon, V. V. (2000). Parent-teen communication about pre-
marital sex. Journal of Adolescent Research, 15(2), 187-208.

106

 

 

Jackson, C., Henriksen, L., & Foshee, V. A. (1998). The authoritative parenting index:
Predicting health risk behaviors among children and adolescents. Health
Education and Behavior, 25(3), 319.

J affe, S. R., Moffitt, T. E., Capsi, A., & Taylor, A. (2003). Life with (or without) father:

The benefits of living with two biological parents depend on the father's antisocial

behavior. Child Development, 74(1), 109-126.

Johnson, B. M., Shulman, S., & Collins, W. A. (1991). Systemic patterns of parenting as
reported by adolescents: Developmental differences and implications for
psychosocial outcomes. Journal of Adolescent Research, 6(2), 235-252.

Jones, D. J ., Forehand, R., & Beach, S. R. H. (2000). Maternal and paternal parenting
during adolescence: Forecasting early adult psychosocial adjustment.
Adolescence, 35(139), 513-530.

Keith, J. G., Huber, M. Q., Griffin, A., & Villarruel, F. A. (2002). Building best lives:
Profiles of 24, 000 michigan youth from 2 asset approaches.

Keith, J. G., & Perkins, D. F. (1995). 13,000 adolescents speak: Michigan State
University.

Kelly, K. J., Comello, M. L. G., & Hunn, L. C. P. (2002). Parent-child communication,
perceived sanctions against drug use, and youth drug involvement. Adolescence,
37(148), 775-787.

Klein, K., Forehand, R., & Armistead, L. (1997). Delinquency during the transition to
early adulthood: Family and parenting predictors from early adolescence.
Adolescence, 32, 61-80.

Knafo, A., & Schwartz, S. H. (2003). Parenting and adolescents' accuracy in perceiving
parental values. Child Development, 74(2), 595.

Knox, D., Zusman, M. E., & McGinty, K. (2001). Deception of parents during
adolescence. Adolescence, 36, 61 1-614.

107

 

 

 

Kosterman, R., Hawkins, J. D., Haggerty, K. P., Spoth, R., & Redmond, C. (2001).
Preparing for the drug free years: Session-specific effects of a universal parent-
training intervention with rural families. Journal of Drug Education, 31(1), 47-68.

Kosterman, R., Hawkins, J. D., Spoth, R. L., Haggerty, K., & Zhu, K. (1997). Effects of a
preventive parent-training intervention on observed family interactions: Proximal

outcomes from preparing for the drug free years. Journal of Community
Psychology, 25(4), 337-352.

Kumpfer, K. L., & Kaftarian, S. J. (2000). Bridging the gap between family-focused
research and substance abuse prevention practice: Preface. The Journal of
Primary Prevention, 21(2).

Kumpfer, K. L., & Tait, C., M. (2000). Family skill straining for parents and children:
Office of Juvenile Justice and Delinquency Prevention Bulletin, US Dept of
Justice.

Kuperminc, G. P., & Allen, J. P. (2001). Social orientation: Problem behavior and
motivations toward interpersonal problem solving among high risk adolescents.
Journal of Youth and Adolescence, 30(5), 597.

Kurdek, L. A., & Fine, M. A. (1994). Family acceptance and family control as predictors
of adjustment in young adolescentszLinear, curvilinear, or interactive effects?
Child Development, 65, 1137-1146.

Lambom, S. D. (1990). Patterns of competence and adjustment among adolescents from
authoritative, authoritarian, indulgent, and neglectful families. Madison,
Wisconsin: National Center on Effective Secondary Schools.

Lambom, S. D., Mounts, N., Steinberg, L., & Dombusch, S. M. (1991). Patterns of
competence and adjustment among adolescents from authoritative, authoritarian,
indulgent, and neglectful families. Child Development, 62(5), 1049-1065.

Laurence, S., D, E. J ., & S, M. N. (1989). Authoritative parenting, psychosocial maturity,
and academic success among adolescents. Child Development, 60(6), 1424.

Leech, S. L., Day, N. L., Richardson, G. A., & Goldschmidt, L. Predictors of self-
reported delinquent behavior in a sample of young adolescents. The Journal of
Early Adolescence, 23(1), 78.

108

 

 

Leffert, N., Benson, P. L., Scales, Sharma, A. R., Drake, D. R., & Blyth, D. A. (1998).
Developmental assets: Measurement and prediction of risk behaviors among
adolescents. Applied Developmental Science, 2(4), 209-230.

Lerner, R. M., & Galambos, N. L. (1998). Adolescent development: Challenges and
opportunities for research, programs, and policies. Annual Review of Psychology,
49, 413.

Lewis, H. M. (2002). Worries among parents of early adolescents transitioning to or
currently enrolled in middle school: A descriptive study. Michigan State
University, East Lansing, MI.

 

Liddle, H. A., Rowe, C., & Diamond, G. M. (2000). Toward a developmental family
therapy: The clinical utility of research on adolescence. Journal of Marriage and
Family Therapy, 26(4), 485-499. _f-.-:—--

 

Ludwig, K. B., & Pittman, J. F. (1999). Adolescent prosocial values and self-efficacy in

relation to delinquency, risky sexual behavior, and drug use. Youth and Society,
30(4), 461-482.

Lytle, L. A., Bimbaum, A., Boutelle, K., & Murray, D. M. (1999). Health and risk
communication from parent to teen: The "parental energy index". Health
Education, 99(5), 207-214.

Mackey, K., Arnold, M. L., & Pratt, M. W. (2001). Adolescents' stories of decision
making in more and less authoritative families: Representing the voices of parents
in narrative. Journal of Adolescent Research, 16(3), 243-268.

Matlack, M. E., McGreevy, M. J., Rouse, R. E., Flatter, C., & Marcus, R. F. (1994).
Family correlates of social skill deficits in incarcerated and nonincarcerated
adolescents. Adolescence, 29(113), 1 l7.

Mattanah, J. F. (2001). Parental psychological autonomy and children's academic
competence and behavioral adjustment in late childhood: More than just limit-
setting and warmth. Merrill-Palmer Quarterly, 47(3), 355-376.

McCurdy, K., & Daro, D. (2001 ). Parent involvement in family support programs: An
integrated theory. Family Relations, 50(2), 1 13-121.

109

 

Miller-Day, M. A. (2002). Parent-adolescent communication about alcohol, tobacco, and
other drug use. Journal of Adolescent Research, 1 7(6), 604-616.

Montemayor, R. (1983). Parents and adolescents in conflict: All of the families some of
the time and some of the families most of the time. Journal of Early Adolescence,
3(1-2), 83-103.

Montemayor, R. (1986). Family variation in parent-adolescent storm and stress. Journal
of Adolescent Research, 1(1), 15-31.

Muuss, R. E. (1996). Theories of adolescence (6 ed.): McGraw-Hill.

Newman, j. (1985). Adolescents: Why they can be so obnoxious. Adolescence, 20(7 9),
636-646.

Pace, K. L. (2003). The character of moral communities: A community youth
development approach to enhancing character development. In F. A. Villarruel,
D. F. Perkins, L. M. Borden & J. G. Keith (Eds.), Community youth development:
Programs, policies, and practices (pp. 248-272). Thousand Oaks: Sage
Publications.

Park, H.-S., & Bauer, S. (2002). Parenting practices, ethnicity, socioeconomic status and
academic achievement in adolescents. School Psychology International, 23(4),
386.

Paulson, S. E., & Sputa, C. L. (1996). Patterns of parenting during adolescence:
Perceptions of adolescents and parents. Adolescence, 31, 369-381.

Perosa, L. M., & Perosa, S. L. (2001). Adolescent perceptions of cohesion, adaptability,
and communication: Revisiting the circurnplex model. Family Journal, 9(4), 407-
419.

Petersen, A. C., Leffert, N., & Graham, B. L. (1995). Adolescent development and the
emergence of sexuality. Suicide & Life - Threatening Behavior, 25, 4.

110

 

 

 

 

Pettit, G. S., Laird, R. D., Dodge, K. A., Bates, J. E., & Criss, M. M. (2001). Antecedents
and behavior-problem outcomes of parental monitoring and psychological control
in early adolescence. Child Development, 72(2), 583-598.

Pilgrim, C., Luo, Q., & Urberg, K. A. (1999). Influence of peers, parents, and individual
characteristics on adolescent drug use in two cultures. Merrill-Palmer Quarterly,
45(1), 85-107.

Pratt, M. W., Arnold, M. L., Pratt, A. T., & Diessner, R. (1999). Predicting adolescent
moral reasoning from family climate: A longitudinal
study. The Journal of Early Adolescence, 19(2).

Price, J. H., Dake, J. A., & Kucharewskil, R. (2002). Assessing assets in raciallly diverse,
inner-city youths: Psychometric properties of the search institute asset
questionnaire. Family and Community Health, 25(3), 1-9.

Reiss, D., Neiderhiser, J. M., Hetherington, E. M., & Plomin, R. (2000). The relationship
code: Decifering genetic and social influences on adolescent development (Vol.
1). Cambridge, Massachusetts: Harvard University Press.

Roberts, M., & Steinberg, L. (1999). Unpacking authoritative parenting: Reassessing a
multidimensional construct. Journal of Marriage and the Family, 61 (3), 574.

Rolison, M. R., & Scherrnan, A. (2002). Factors influencing adolescents' decisions to
engage in risk-taking behavior. Adolescence, 37(147), 585.

Santrock, J. W. (1996). Child development (7th ed.). Madison, Wisconsin: Brown and
Benchmark.

Sartor, C. E., & Youniss, J. (2002). The relationship between positive parental
involvement and identity achievement during adolescence. Adolescence, 3 7(146),
221.

Scales, P. C. (1997). The role of family support programs in building developmental
assets among young adolescents: A national survey of services and staff training
needs. Child Welfare, 76(5), 611.

111

 

 

Scales, P. C. (1999). Reducing risk and building developmental assets: Essential actions
for promoting adolescent health. Journal of School Health, 69(3), 113-1 19.

Scales, P. C., Benson, P. L., Leffert, N., & Blyth, D. A. (2000). Contribution of
developmental assets to the prediction of thriving among adolescents. Applied
Developmental Science, 4(1), 27-46.

Scales, P.C., Leffert, N., & Lerner, R. M. (1999). Developmental assets : A synthesis of
the scientific research on adolescent development. Minneapolis: Search Institute.

Schoenrock, C. J ., Bell, N. J., & Sun, S.-W. (1999). Family correlates of adolescent self-
monitoring and social competence. The Journal of Psychology, 133(4), 377-393‘.

Selman, R. L., & Schultz, L. H. (1990). Making a fi'iend in youth: Developmental theory
and pair therapy. Chicago: The University of Chicago Press.

 

Shucksmith, J ., & et a1. (1995). Models of parenting: Implications for adolescent well-
being within different types of family contexts. Journal of Adolescence, 18(3),
253-270.

Simpson, A. R., & Roehlkepartain, J. L. (2003). Asset building in parenting practices and
family life. In R. M. Lerner & P. L. Benson (Eds.), Developmental assets and
asset-building communities: Implicationsfro research, policy, and practice. NY:
Kluwer Academic/Plenum Publishers.

Slicker, E. K. (1996). Parenting style and family environment as they relate to academic
achievement and problem behaviors in older adolescents. US; Tennessee.

Smetana, J. G. (1995). Parenting styles and conceptions of parental authority during
adolescence. Child Development, 66(2), 299-316.

Smetana, J. G., & Daddis, C. (2002). Domain-specific antecedents of parental
psychological control and monitoring: The role of parenting beliefs and practices.
Child Development, 73(2), 563-580.

Spoth, R., & Redmond, C. (2000). Research on family engagement in preventive
interventions: Toward improved use of scientific findings in primary prevention
practice. The Journal of Primary Prevention, 21 (2).

112

 

Psi

Spoth, R., Redmond, C., Haggerty, K., & Ward, T. (1995). A controlled parenting skills
outcome study examining individual difference and attendance effects. Journal of
Marriage and the Family, 57(2), 449.

Spoth, R., Redmond, C., Hockaday, C., & Shin, C. Y. (1996). Barriers to participation in
family skills preventive interventions and their evaluations: A replication and
extension. Family Relations, 45(3), 247.

Spoth, R. L., Guyll, M., & Day, S. X. (2002). Universal family-focused interventions in
alcohol-use disorder prevention: Cost-effectiveness and cost-benefit analyses of
two interventions. Journal of Studies on Alcohol, 63(2), 219.

Steinberg, L. (1990). Authoritative parenting and adolescent adjustment across varied
ecological niches. Madison, Wisconsin: Corp Author: National Center on
Effective Secondary Schools.

 

Steinberg, L. (2001). We know some things: Parent-adolescent relationships in retrospect
and prospect. Journal of Research on Adolescence, 11(1), 1-19.

Steinberg, L., & Elmen, J. D. (19863). Adolescent responsibility, parent-child relations,
and school performance. U.S.; Wisconsin: Corp Author National Center on
Effective Secondary Schools Madison W. I.

Steinberg, L., & Elmen, J. D. (1986b). Authoritative parenting promotes adolescent
school achievement and attendance. U.S.; Wisconsin: Corp Author National
Center on Effective Secondary Schools Madison W. I.

Steinberg, L., Elmen, J. D., & Mounts, N. (1989). Authoritative parenting, psychosocial
maturity, and academic success among adolescents. Child Development, 60(6),
1424-1436.

 

Steinberg, L., & Morris, A. S. (2001). Adolescent development. Annual Review of
Psychology, 52, 83.

van We], F., ter Bogt, T., & Raaijmakers, Q. (2002). Changes in the parental bond and
the well-being of adolescents and young adults. Adolescence, 37(146), 317.

113

 

YMCA, & Search-Institute. (2002, November 2002). Building strong families: A
preliminary study from ymca of the USA and search institute on what parents
need to succeed. Retrieved 3/12/2003, 2003, fiom http://www.search-
instituteorg/families/

 

 

114 F7

 

t

I» -

<r—

 

 

 

 

’1!

   

1ljiriljlijlljlllily1