FAMILY PARADIGMATIC PREFERENCES AND CHILDREARING PRACTICES OF
MOTHERS DIFFERING BY RACE AND EDUCATIONAL LEVEL
THERESA THWEATT RULIEN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
Theresa Thweatt Rulien
to my mother and father
This is as much yours as it is mine.
to my daughter, Caitlin
You did a lot of this with me and yes, we're
Unfortunately it is impossible to acknowledge every person who has contributed in some way to the completion of this dissertation, but I sincerely appreciate the countless acts of kindness, support, and empathy that allowed me to continue with this phase of my life. Specifically I want to thank Greg Dionne and Jeanne von Genk for their courage and willingness to tutor me through the basics of computer literacy. Without their patience I would have continued handwriting this study. I also want to graciously extend my appreciation and admiration to Mchael Lanier and Dr. Thom Borowy for their expert advice as well as their ongoing support and empathic understanding when I was sure I would never make it. The supportive guidance of my doctoral committee, Dr. Larry Loesch, Dr. Joe Wittmer, and Dr. David Mller, is gratefully acknowledged, and especially that of my chairperson, Dr. Ellen Amatea, who pleaded, pushed and never let me feel defeated.
Without family this would have been an impossible task. I thank my aunt, Mary Evelyn Albertson, for the use of her printer when I had none, and to my great aunt, Elizabeth "Jodie" Allen, for the neverending words and Bible verses of support and encouragement, and her "magical" way of knowing when I needed to receive a letter or phone call. And to Mr. McIver Brooks, who isn't really family, but should be, for never iv
hesitating to lovingly remind me to get to work. I thank my in-laws, Mr. and Mrs. Jack Rulien (Maureen), for babysitting and for understanding when I couldn't make it to a family function.
I especially wish to acknowledge my husband, Joey, and our three children, Caitlin, Dustin, and Dillin, for their patience at having me in school the entire time any of the four have known me. Their belief in me kept this a worthwhile goal. And to my parents, Dr. and Mrs. Robert Thweatt (Doris), I can not even begin to express my gratitude for their love, support, and faith in me. Without their parenting, urging, empathy, editing, advice, wisdom, and constant willingness to keep my children "just a few more hours" I would not have even come close to achieving this goal.
TABLE OF CONTENTS
ACKNOVrLEDGM ENTS ........................................................................................... -Av
LIST OF TABLES ........................................................................................................ viii
A B S T R A C T ................................................................................................................... ix
I INTRODUCTION ....................................................................................................... I
Scope of the Problem ................................................................. ............................. .... 4
Theoretical Framework ........... ............................................................................... 9
Need for the Study ...................................................................................................... 15
Purpose of the Study ................................................................................................... 16
Research Questions .................................................. ................................................. 17
Definition of Terms ..................................................................................................... 18
Organization of the Study ............................................................................................ 22
2 REVIEW OF THE LITERATURE ........................... ................................................. 23
Theories of Parenting and Parenting Education Programs ............................................ 23
Overview of Parenting Assessment Instrument Development ....................................... 28
Relating Parenting to Family Background Characteristics ............................................ 31
Relating Parenting to Parental Cognitive Level ............................................................ 40
Relating Parenting to the Nature of the Family Context ............................................... 46
C o n clu sio n ....................................................... .................................................... 7 5
3 METHODOLOGY ...................... .............................................................................. 78
Statement of the Purpose .............................. ............................................................. 78
H y p o th e se s .......................... ............................. ...................................................... 7 8
Design of the Study ...................................... ............................................................. 80
Description of the Population ...................................................................................... 82
Sampling Procedures... ........ .................................................................................... -83
S u bje c ts ................................ ..................................................................................... 8 4
Data Collection ........................................................................................................... 85
Instrumentation ........................................................................................................... 86
Data Analysis ............................................................................................................ 100
4 RESULTS ............................................................... ................................................ 102
Research Hypotheses.... .......................................................................................... 104
S u m m ary .................................................................................................... ............. 12 1
5 DISCUSSION .......................................................................................................... 122
Discussion of the Results .......................................................... ............................... 122
Limitations of the Study ................................................. ......................................... 132
Suggestions for Future Research .............................................................................. 133
Implications of the Study ................................................................. ........................ 134
Summary and Conclusions ........ ............................................................................... 135
A DEM OGRAPHIC INTERVIEW SCHEDULE ....................................................... 136
B PARENTING PREFERENCES INVENTORY ...................................................... 138
C L E T T E R ....................... ........................................................................................ 14 1
D TELEPHONE CONTACT .................................................................................... 142
E INFORM ED CON SENT .......... .................................. ......................................... 144
REFERENCES ............................................................................... .......... ............. 145
BIOGRAPHICAL SKETCH .................................... ................................................. 167
LIST OF TABLES
I Defining features, strengths, and limitations of family paradigms ................................ 12
2 P arenting instrum ents ......................................... ................................................ 32
3. Sample means and standard deviations of dependent variables by race and
edu catio nal level ...................................................................................................... 103
4. Analysis of variance to assess differences in family paradigmatic preference
due to m other's educational level .............................................................................. 105
5. Analysis of variance to assess differences in family paradigmatic preference
due to m other's race ....................... .................... ............................................ ..... 107
6. Regression analysis to assess the difference in level of verbal interaction
due to fam ily paradigm atic preference ...................................................................... 109
7. Regression analysis to assess the difference in level of behavioral interaction
due to fam ily paradigm atic preference .................................... ................................. In
8. Regression analysis to assess the frequency of democratic family-management
style due to family paradigmatic preference .............................................................. 113
9. Regression analysis to assess the frequency of laissez-faire family-management
style due to family paradigmatic preference .............................................................. 115
10. Regression analysis to assess the frequency of authoritarian fan-fily-management
style due to family paradigmatic preference ....... .......................................... .......... 117
11. Regression analysis to assess the frequency with which mothers use
person-centered communication due to family paradigmatic preference ................ 119
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
FAMILY PARADIGMATIC PREFERENCES AND CHILDREARING PRACTICES OF
MOTHERS DIFFERING BY RACE AND EDUCATIONAL LEVEL By
Theresa Thweatt Rulien
Chairperson: Dr. Ellen Amatea Major Department: Counselor Education
Family paradigmatic theory was utilized, in this study, as a framework for exploring and attempting to understand the differences in childrearing attitudes and practices of mothers differing in race and educational level. First, the author proposed to examine empirically the family paradigmatic preferences of women who were mothers of at least one elementary school aged child. Second, the author compared the childrearing practices and style of parent-child communication of mothers differing in family paradigmatic preference. Third, the author investigated the frequency with which mothers differing in educational level and race were characterized by differing family paradigmatic preferences.
The purpose of this descriptive comparative study was to determine whether women differing in terms of their family paradigmatic preference would differ in their parenting practices and parent-child communication. The sample consisted of 57 mothers representing four separate groups divided by educational level and race. Interviews were conducted with each of the mothers and consisted of the collection of demographic information as well as administration of four different instruments assessing (a) parent-child communication; (b) family-management style; (c) family paradigmatic preference; and (d) child-management values.
A series of chi-square analyses and a series of two-way analyses of variance were conducted to determine the frequency with which mothers differing by educational level and race chose each of the four family paradigms. A series of regression analyses were conducted to determine the effect of family paradigmatic preference on child-management values, family-management style, and child-centered communication.
No significant differences in family paradigmatic preference based on mothers' race or educational level were found. No significant differences were found in child-management values nor for levels of child-centered communication for mothers differing in family paradigmatic preference. A significant difference was found for mothers describing an open family paradigm and reporting a preference for a democratic
family-management style, however, no significant differences were found for the other family paradigms or family-management styles.
Implications and suggestions for future research were discussed in concluding this study.
Assisting parents in learning how to parent their children more effectively has been proposed as a viable solution to many of the social problems facing contemporary America (Alexander, Barton, Schiavo, & Parsons, 1976. Gordon & Davidson, 1981 Graziano, 1986; Klein, Alexander, & Parsons, 1977; Levant, 1983a; Levant, 1983b; Robin, Kent, OLeary, Foster, & Prinz, 1977). Whether it be in response to the problem of child neglect or abuse, to the rising tide of violence in schools, to the increase in juvenile crime and homicide, or to relatively normal "rebellion," improving parents' abilities to effectively rear their children has been viewed as an important step toward preventing social problems. But what style of childbearing and image of family life should be promoted in these parenting interventions?
Despite attempts to rethink "the family" as a social form and to dispel the myths and monolithic view of contemporary family life, relatively little research exists that focuses on the complexity and diversity of parenting beliefs and experiences or the pervasive impact of culture or class in shaping these beliefs. Instead most social scientists who have conceptualized and assessed childrearing/parenting practices and values have often conceptualized a continuum of "good" to"bad" (or functional to dysfunctional)
childrearing practices and values and have built assessment instruments that describe and evaluate parental behavior in terms of these types of continnua. Implicit in these researchers' (Dinkmeyer & McKay, 1976; Dreikurs & Soltz, 1964; Gordon, 1970, 1976; Graziano, 1983; Levant, 1983a; Levant, 1983b.- Moreland, Schwebel, Beck, & Wells, 1982; Popkin, 1983;- and Robin, 198 1) conceptualizations of parenting is the assumption that there is one optimal way to parent children and that deviations from this favored style represent less than adequate parenting. But could there be more than one way to effectively rear children? How might the nuances of culture, ethnicity, or class impact parents' beliefs about how families should be organized and childrearing practiced? Might there be different images or paradigms of family life preferred by parents of varying racial or educational backgrounds? Could there be a relationship between the ways parents report preferring to rear their children and their ideas about how their family life should be structured?
Although the impact of differing cultural, ethnic, and educational influences on the lives of families is beginning to be acknowledged, most researchers examining childrearing practices of parents differing by race and educational level have viewed any deviations from a proposed norm as a sign of deficit rather than as evidence of legitimately different styles of parenting shaped by different sociocultural traditions and contexts (Constantine, 1986, 1993; Fantini & Cardenas, 1980; Fine & Henry, 1989; Minuchin, 1974). Thus much
is known about how members of various sociocultural groups differ from the white, middle class standard of childrearing explicated in current childrearing theories, but relatively little is known about how these individuals effectively rear their children to be contributing members of society (Belsky, 1984; Bronfenbrenner, 1977; Goodnow, 1985; Nwachuku & Ivey, 1991).
How can the childrearing practices and values of parents living in distinctively different sociocultural contexts be explored? Might there be different images or paradigms of family life preferred by parents of varying racial or educational backgrounds? If it can be assumed that each sociocultural group offers to its members a sense of identity, direction, and explanation for the world around them and that such explanations may lead to the development of different processes for creating meaning in life, it may be found that members of different sociocultural groups hold different images or paradigms of family life. Could there be a relationship between the way parents report preferring to rear their children and their ideas about how their family life should be structured? In this study these questions were addressed by examining whether mothers' general beliefs about how families should be organized were associated with certain childrearing practices and interactional competencies, and determining whether such beliefs about family life occurred more frequently within mothers of certain racial or educational backgrounds. While it is certainly important to understand the preferences of all family members, the
author focused exclusively on the paradigmatic preferences of mothers as an initial step in studying the relationship between childrearing practices and family image.
Scope of the Problem
Childrearing practices have been questioned, debated, and obsessed upon by social scientists, educators and community leaders for over half a century (Carlson, 1985). As social institutions in the United States (e.g., day care and the schools) assume greater responsibility for childrearing, increasing attention is being paid to determining what might be the best way to rear children to be successful in society. Hand-in-hand with the increase in institutional involvement in childrearing has come the recognition that there are conflicting points of view as to how best to rear a child (Carlson, 1985; Dembo, Sweitzer, & Lauritizen, 1985; Gfellner, 1990). It is becoming increasingly evident that different subgroups of society, each operating with their own set of values, rules and expectations, may each have their own ideas about the "correct way" to rear a child (Ivey, 1988; Strom, Griswold, & Slaughter, 1981).
Is there only one correct method? This question has been raised only recently in
designing research and interventions on childrearing. Embracing the perspective that there was a single correct way to rear a child, researchers in the 1960s and 1970s often conceptualized parenting styles in terms of unidimensional constructs: love versus hostility or autonomy versus control (Baumrind, 1967, 1971, 1972; Becker, 1964; Kegan
& Moss, 1962; Maccoby & Martin, 1983; Schaefer, 1961; Sears, Maccoby & Levin, 1957). Baumrind (1967, 1971, 1972), for example, identified three basic styles of childrearing: authoritarian, permissive, and authoritative. According to Baumrind, authoritative parents produce competent, well-adjusted children with high self-esteem through clear, open communication. They also encourage independence while setting age-appropriate limits. The authoritarian style, based on dominating control, was found to be associated with poor social development and adjustment in children and adolescents (Lau, Cheung, Cheung, Lew, & Berndt, 1990). In this study, children perceived an increase in parental warmth and family harmony with a decrease in dominating control.
In the 1980s and 1990s with the growth of the family therapy movement, efforts
shifted to identifying how families function in rearing children. Attention has broadened to examine how a family, not just a parent, might structure themselves to conduct childrearing functions. Thus, the need for family communication, clarity of generational boundaries, and unity in parental decisionmaking was emphasized (Minuchin, 1974; Minuchin & Fishman, 1981; Satir, 1972). In addition, different models of healthy family functioning were proposed by family systems theorists (Beavers & Hampson, 1990; Olson, Portner & Bell, 1982; Stinnert & DeFrain, 1989). However, inherent in many of these family models was a continuum from healthy to pathological; there were right versus wrong ways of organizing as a family and rearing children.
In recent years, a second perspective for conceptualizing childrearing practices has emerged which focuses on parents' cognitive processes. Intrigued by questions such as "How do parents decide how they will rear their children?" and "What factors shape their personal 'theory of childrearing'?", researchers operating from this newer perspective seek to determine whether the very nature of adult cognitions shapes the ways parents respond to their children and the task of parenting. A key assumption underlying this research is that parents do not merely react to a particular child's actions in a situation, but that numerous previous interactions and knowledge about the child are cognitively structured and thus influence the manner in which the parent interprets and reacts to the situation (Dekovic & Gerris, 1992). Parents organize their actions depending on the goals they hope to achieve, supporting the idea that parental behavior is not merely reactive (Dix & Grusec, 1985; Goodnow, 1985; Miller, 1988; Sigel, 1985). For example, Dekovic and Gerris (1992) found, in their study of parental reasoning and childrearing behaviors, that parental reasoning complexity made an independent contribution to parental behavior beyond the contribution of social class or educational level.
This increased interest in studying childrearing from a cognitive perspective has resulted in the development of a variety of different cognitive theories which explain the cognitive development of adults (Ivey, 1991; Kohlberg, 1969; Loevinger, 1980). Cognitive developmental hierarchies have been constructed on the basis of both the
content and the structure of thinking regarding childrearing tasks. The associations among the structure of parental beliefs and parental communication characteristics (Applegate, Burleson, Burke, Delia & Kline, 1985), parental concepts of child development (Sameroff & Feil, 1985), parent-child relationships (Newberger, 1980), and parental behaviors (Dekovic & Gerris, 1992) have been examined. However, these models of parental reasoning complexity, with their emphases on developmental hierarchies and low to high levels of cognitive complexity, still have an inherent "right versus wrong" quality to their descriptions of parental thinking.
Recently, the need for understanding and appreciating the diversity in childrearing practices of individuals has resulted in a third approach to exploring parents' ideas about childrearing. This perspective focuses on understanding the world view of the parent. It underscores the normalcy of differing modes of childrearing which evolve as a result of a parent's social context, family organization, and cultural heritage. Based on the premise that there is more than one way to effectively rear children, this approach seeks to determine whether different childrearing practices are associated with particular normative assumptions, world views and frames of reference. This conceptual approach is illustrated by the thinking of Belenky, Clinchy, Goldberger and Tarule (1986):
Our basic assumptions about the nature of truth and reality and the origins of knowledge shape the way we see the world and ourselves as participants in it.
They affect our definitions of ourselves, the way we interact with others, our
public and private personae, our sense of control over life events, our views of
teaching and learning, and our conceptions of morality, (pg. 3).
Childrearing is, in itself, a cultural activity. Cultural and economic factors are at work in a child's life from the act of conception to the pregnancy and its associated customs to the birth ritual. How might the impact on parents' thinking of the cultural perspective be examined? Such a conceptualization of parenting would require that researchers go beyond examining mother-chid interaction and consider wider familial and cultural variables (Belsky, 1980; Hinde, 1979; LeVine, 1980; Patterson, 1982). Cultures and communities deliver many messages about parenting. Bronfenbrenner (1977) pointed out the importance of the broader community or macrosystem in setting the normative cultural standards about childrearing through advice from relatives and experts or through role modeling and witnessing interactions of other families. Goodnow (1985) suggested that culture is the primary source of information about the facts of childrearing including what children are like at different ages, what parenting techniques work, and what goals parents should value. Some researchers have extended this investigation to say that what parents think is important, guides or influences choices. Therefore, to define an effective parent requires some notion of the goals of parenting promoted in a specific cultural context (Gordon, 1980). Whether it be due to environmental risks or occupational and societal roles it has become increasingly evident that differing cultural perspectives have implications in the childrearing practices of parents.
Mother's race and educational level were selected as the independent variables for this study based on this review of the literature. A preliminary survey was conducted using social class as one of the independent variables, but it was determined that level of education provided a clearer picture of the mother's status level. Using Hollingshead's Four Factor Index of Social Status (1975) social class showed evidence of inconsistent categorizations due to the occupational level not necessarily correlating with the level of education.
Although this third perspective holds much promise for identifying the diverse ways in which parents might effectively rear their young, there has been only limited research exploring how a parent's view of family life is influenced by their cultural (i.e., racial) and educational status, and how this view of family life affects their childbearing practice.
A number of different typologies of family life have been proposed by family theorists to describe the diverse ways in which families organize themselves, structure their interactions, and give meaning to their fives together (Beavers, 1981; Olson, Sprinkle, & Russell, 1979). These typologies have been used to classify families in terms of a wide variety of characteristics (e.g., structural patterns, degrees of functionality or dysfunction, or patterns of communication). Constantine's (1986) theory of family
paradigms was chosen for this study because it is more inclusive than many existing family typologies, it includes all internal aspects of family life, including structural, psychological and philosophical contexts, and it provides an integrative framework for understanding the diverse ways in which families might organize themselves and function successfully.
Basing his paradigmatic model on Reiss' (198 1) notion of the family paradigm, Constantine (1993) described family paradigms as stable, implicit, overarching guides to family life which consist of the family's repertoire of understandings, shared assumptions, and traditions. He stated:
The paradigmatic framework encompasses three central ideas: (a) there is more than one way that families can organize themselves and function effectively; (b) families can be understood at more than one level of analysis; and (c) there are
relationships between how families construct their realities and how they function
and are organized (pg. 46).
Although the typology is conceptual, it is grounded in research and agrees with other empirically based models (Constantine, 1980, 1983). Its practical value also has been demonstrated in marriage and family therapy (Budd, 1990; Burr, Day & Bahr, 1989; Constantine, 1986, 1987; Constantine & Israel, 1985; Nugent & Constantine, 1988). The family paradigm framework distinguishes three levels of analysis by which families can be characterized: (a) paradigm (world view, model); (b) regime (organization, structure); and (c) process (behavior, interaction). Paradigms are reflected in process through regimes, the regulatory mechanisms that generate patterned collective behavior. The paradigm is the most abstract level and most difficult to assess; process is the most
concrete and most easily accessed. The regime is the means by which the paradigm, unobservable in itself, can be translated into observable behavior (Constantine, 1986).
According to Constantine, a paradigm is both a model and a world view, not only representing a family's values and ideals, but a construed reality of family and social relationships (Reiss, 1981). Thus, each family paradigm includes distinct approaches to the resolution of fundamental issues of family life such as the relative value given to stability versus change or the relative priority given individual versus the collective needs of family members. These variations in family values and family structure not only influence the relative value placed on childbearing in a family but also shape the distinctive ways in which chfldrearing is conducted by different family members. Different paradigms produce different perceptions and practices about childbearing.
Constantine (1986) described four distinctively different family paradigms: the closed, the random, the open, and the synchronous. The defining features of each paradigm are depicted in Table 1. The closed paradigm personifies the stable, secure family which relies on traditional authority and conformity to its norms to assure continuation of established family patterns. The closed family has a fixed, authoritarian, and hierarchical structure. Familial boundaries are closed and controlled, and roles are assigned and strongly emphasized. Truth is perceived as being absolute and permanent.
11 0 V
CZ C3 0 cz
0 Z cu
cl ca ca
cl 0 r 35
's .8 0
cu 0 C*
9 C) C:
Ln 02 a a -c 'e., A
t: cu 0 Z 0 g .6
e 2 ? 0
0 0 0 U ca
:0 z 0 z C)
60 - 2
E '6 0
The closed family is, in essence, the traditional American family. Family and family identity are primary with individual needs met through loyalty and service to the family.
The random paradigm is the antithesis of the closed. In this family paradigm the structure is changing and egalitarian. Roles are unassigned and undifferentiated; boundaries are open, uncontrolled and erratic. For the random family, change, freedom and individuality are of utmost importance. Truth is considered to be relative, transient, and personal. Members of such families favor change, novelty, and originality over stability and tradition. The individual is primary with the structure stressing creative individuality and egalitarian autonomy. The collective family needs are met through cooperative individual initiative and inventiveness.
The open paradigm presents a modem, communication-oriented, democratic family. Structure in this family is adaptive and heterarchical; roles are alternated and shared. Boundaries are defined, yet permeable and flexible. There is a balance between stability and change and between individuality and group needs. Process is valued most highly by this family. Truth is pragmatic; while some truths are absolute, many are relative. It attempts to achieve flexibility that integrates change with stability, and to incorporate the needs and interests of the individuals with those of the family as a unit. This attempt to synthesize the opposing values and approaches of the closed and random paradigms is achieved through a collaborative process of consensual negotiation, elevating
communication to a high art. Both stability and change are valued insofar as they enhance the family's ability to solve problems and adapt to challenges (Constantine, 1988).
The synchronous paradigm, the antithesis of the open paradigm, blueprints a quiet, harmonious family relying on tacit understanding and unstated rules rather than utilizing communication and negotiation. Roles are static and implicitly understood rather than assigned. Boundaries are rigidly defined, but not actively controlled. The synchronous family emphasizes neither change nor stability. Group concerns transcend individual needs. Truth is viewed as enduring and eternal. These families seek a perfection in which no one needs to be told what to do. Fairly uncommon in the United States (Budd, 1990, Constantine & Israel, 1985), the synchronous paradigm appears more frequently in Japan and other Asian countries and is a recognized form of organization manifest in Japanese management style (Constantine, 199 1; Oichi, 198 1).
Constantine's paradigms are based on family strengths and he, like Olson and his colleagues (1983), has made the assumption that any family paradigm is workable if affirmed and desired by family members. Constantine (1986, 1993) also has acknowledged that relatively few families are paradigmatically pure, that is, totally closed, completely random, absolutely open, or completely synchronous. However, families can be characterized as being more closely related to one paradigm than the other three, for example, more synchronous than open, random, or closed. Consequently, it was assumed
in this study that (a) the family's paradigm would influence the style of childrearing implemented; (b) there would be a variety of different acceptable patterns of childrearing demonstrated just as there are a variety of different acceptable family paradigms available; and (c) the types of interactional skills developed by parents would be related to the type of family paradigm valued.
Need for the Study
Because most research on childrearing practices and attitudes among individuals of varying ethnic and economic groups has focused on examining differences from one valued standard of childrearing practice, very little is known as to whether there are predictable differences among parents in different sociocultural contexts concerning how they think family life should be structured. Even less is known as to whether these differing paradigms of family life are related to specific childrearing practices and competencies.
Family paradigmatic theory posits that how one thinks is considered to be critical in the development of specific types of parent-child interaction. Thus, gaining an understanding of a parent's family paradigmatic preference seems extremely relevant to understanding parents' ideas about and practices of childrearing. Further exploration of parental differences in family paradigmatic preference is needed to increase understanding of the wide variety of responses parents have to the task of rearing children. Not only
would information regarding family life and childbearing preferences be useful in understanding why parents function the way they do, it would also be quite useful in designing more attractive and effective interventions for them.
Purpose of the Stu
In this study, family paradigmatic theory was utilized as a framework for exploring and attempting to understand the differences in childrearing attitudes and practices of mothers differing in race and educational level. According to family paradigmatic theory, a parent's beliefs and behaviors are shaped by a larger set of unifying beliefs about family life that guide action in implementing specific family roles and tasks. As a result, how parents think the family should operate will influence how they expect their child should function, how they think they should interact with their child, and what specific parent-child interactional competencies they develop.
Given the relative lack of knowledge about whether a distinctive sociocultural context may result in a particular family paradigm preference that may, in turn, result in the valuing of particular child behaviors and parenting attitudes and practices, and the potential of the family paradigmatic theory to identify a variety of different equally viable childbearing philosophies, the purpose of this study was three fold. First, the author proposed to examine empirically the family paradigmatic preferences of women who were mothers of at least one elementary school aged child. While paradigmatic choice is a
function of a mother's parenting experience, so that having more than one child could presumably affect her preference, this author focused on the effects of educational level and race as an initial step. Second, the author compared the childbearing practices and style of parent-child communication of mothers differing in family paradigmatic preference. Third, the author investigated the frequency with which mothers differing in educational level and race were characterized by differing family paradigmatic preferences.
The following research questions were posed in this study:
1. What are the differences in family paradigm preferences of mothers belonging to lower and higher educational levels?
2. What are the differences in family paradigm preferences of mothers belonging to Black Aftican American and White Anglo American racial groups?
3. After adjusting for educational level and race, what are the differences in child-management values demonstrated by mothers differing in family paradigm preferences?
4. After adjusting for educational level and race, what are the differences in family management styles demonstrated by mothers differing in family paradigm preferences?
5. After adjusting for educational level and race, what are the differences in levels of child-centered communication demonstrated by mothers differing in family paradigm preferences?
Definition of Terms
For the purpose of this study, key constructs and terms were defined as follows: Child-management value is defined as the sense of importance attributed by the mother to the two variables: level of behavioral interaction and involvement, and level of parent verbal interaction and encouragement of expressiveness. It was measured by the Parenting Preferences Inventory.
Level of parent behavioral interaction and involvement is defined as the degree of parental "hands-on" involvement in the day to day decision making and teaching of the child as measured by the Parenting Preferences Inventory.
Level of parent verbal interaction and encouragement of expressiveness is defined as the degree to which the mother encourages verbal interaction between herself and her child to elicit expression of thoughts and feelings as measured by the Parenting Preferences Inventory.
Race reflects self-identification and, following the guidelines of the U.S. Census Bureau, "does not denote any clear-cut scientific definition of biological stock." The categories of race used for purposes of census include both racial and national origin or
sociocultural groups as well as providing standards on ethnicity. For the purposes of the present study, categories were limited to either Black Afican American or White Anglo American.
Black Afican American is defined as a mother of Afican descent and at least a second generation citizen of the United States.
White Ang-lo American is defined as a Caucasian mother and at least a second generation citizen of the United States.
Family paradigm is the stable, implicit, overarching guide to family life that
consists of the family members' repertoire of understandings, shared assumptions, and traditions. It was measured by the Family Regime Assessment Scale (Imig & Phillips, 1992) and resulted in a classification of either closed, random, open, or synchronous family paradigm.
Closed family paradigm personifies the stable, secure family, relying on traditional authority and conformity to its norms to assure continuation of established family patterns. Roles are differentiated and boundaries controlled. It is one of the classifications established by the Family Regime Assessment Scale.
Random family paradigm encourages change, freedom, and individuality with roles unassigned and undifferentiated, and structure changing and egalitarian. It is one of the classificatons established by the Family Regime Assessment Scale.
Open family 12aradiam models a modem, communication-oriented, democratic
family with shared roles and defined, yet flexible boundaries. It is one of the classifications established by the Family Regime Assessment Scale.
Synchronous family paradigm blueprints a quiet, harmonious family relying on tacit understanding and unstated rules rather than utilizing communication and negotiation. Roles are static and boundaries rigidly defined, but not actively controlled. It is one of the classifications established by the Family Regime Assessment Scale.
Regim is the organization or structure by which the paradigm can be translated into observable behavior.
Child-centered communication refers to communication and strategies (both
regulative and comforting) used by the parent to give legitimacy to a child's feelings and encourage the child to reflect on and reason through the nature and consequences of his/her behavior. This variable was measured by the Person-Centered Communication Assessment (Applegate, et al, 1985).
Educational level refers to the number of years of formal education a mother has received, according to the categories used by Hollingshead (1975) in his index, and resulted in a classification of lower and higher level education. Lower level consisted of mothers with a high school education or less. 9gher level consisted of mothers having attended at least one year of college.
Famil-management style is defined as the specific style a mother employs with regards to orchestrating the day-to-day functioning of her family and will be categorized as democratic, laissez-faire, or authoritarian. It was measured by the democratic, laissez-faire, and authoritarian sub-scales of The Colorado Self-Report of Family Functioning (Bloom, 1985).
Democratic family stle denotes a family-management style encouraging negotiation and communication, and is considered comparable to the open family paradigm according to Constantine (1993). It was measured by the Democratic Family Style sub-scale of The Colorado Self-Report of Family Functioning (Bloom, 1985).
Laissez-faire family style denotes a family-management style that is externally focused regarding control and values flexibility, and is considered comparable to the random family paradigm according to Constantine (1993). It was measured by the Laissez-Faire Family Style sub-scale of The Colorado Self-Report of Family Functioning (Bloom, 1985).
Authoritarian family style denotes a family-management style that values
organization, rigid boundaries, and clear rules, and is considered comparable to the closed family paradigm according to Constantine (1993). It was measured by the Authoritarian Family Style sub-scale of The Colorado Self-Report Measure of Family Functioning (Bloom, 1985).
Organization of the Study
A five chapter model is presented in this research study. In Chapter 2, the author presents a review of the related literature. The methodology of the study is contained in Chapter 3, while a presentation of the results of the statistical analysis of data are included in Chapter 4. A discussion of the results, limitations of the study, and suggestions for future research are included in Chapter 5.
REVIEW OF THE LITERATURE
This chapter includes a brief review of the literature on parenting and childrearing with an emphasis on examining how parenting has been conceptualized and measured. Theories of childrearing or parenting are reviewed. Corresponding parent education programs are examined as well as the assessment instruments employed to assess parenting skills. Literature on contextual and racial differences in families and the implications of such differences for childrearing practice are reviewed, as well as research on family orientations.
The research on parenting has consistently depicted a "right versus wrong" mentality of childrearing. This representation is noted in the range of theory development supportive of one optimal and correct way for all parents. Parent education programs support this conceptualization as well. Even related assessment instruments are often designed in an evaluative manner due to the unidimensional conceptualization of parenting described in these theories.
Theories of Parenting and Parenting Education Programs
There are numerous theories of parenting and parent education. They coincide with and paralel the diverse theories of human interaction used in client therapy (Lamb and
Lamb, 1978). This review focuses on the three major theories used in parent education. It was noted that they support many aspects of the images of family life proposed by Constantine (1977, 1983, 1984a, 1984b, 1986, 1987, 1988). Adlerian
The basic assumption of this model is that people are social beings and want to be socialized. Behavior is seen as purposive and goal seeking. Cooperation, rather than permissiveness, is needed between parents and children, and it is believed that children's behavior is based on expectations. The goal of the Adlerian approach is to help parents understand their children--to know how they think and to comprehend the motives for their actions. The idea is to help parents improve the quality of the help they provide their children and to assist them in relating more effectively to their children. Techniques of parenting involve the use of natural and logical consequences as an alternative to powerful control of the child, a democratic approach to parenting, an emphasis on the goals of children's misbehavior and the use of the family council (Croake, 1983;- Lamb & Lamb, 1978). Adlerian programs include Children the Challenge (Dreikurs & Soltz, 1964); Systematic Training for Effective Parenting (STEP) (Dinkmeyer & McKay, 1976) and Active Parenting (Popkin, 1983).
The Adlerian approach, with its emphasis on family democracy and the importance of discussion by way of the family council, models the values of the open family paradigm.
The behavioral model is the most predominant model in use in the parent
education field (Lamb, 1986). The basic assumption within this model is that human behavior results from learning, much of which comes from interaction with the environment. Environmental change is therefore important. The goal is the assessment of problem areas with regards to excesses, deficits, and maintenance. This implies social value judgments. Techniques focus on increasing parental skills in influencing, controlling, and directing children's behavior and development. This is done through parents accepting responsibility for change in their children's behavior (Lamb & Lamb, 1978; Graziano, 1983; Levant, 1983a; Levant, 1983b). Instructions and procedures are clearly defined in concrete terms with instructors modeling concepts and parents role-playing. Communication skills are often emphasized in programs for families of adolescents (Moreland, Schwebel, Beck & Wells, 1982; Robin, 198 1).
The behavioral approach to parent education, with its use of techniques for
controlling and directing a child's behavior as well as the acceptance of responsibility by the parents for that behavior, models the closed family paradigm. Client-Centered / Humanistic
This approach, categorized as client-centered by Lamb and Lamb (1978) and humanistic by L'Abate (1981) and Medway (1989), has as its basic assumptions that
people are basically good and will do the right thing, needing only support to make their own choices. People are seen as responsible for their own decisions and are seen as capable of changing. The goal of this approach is an increase in the expression of feelings with a need for two-way communication. Few limits are placed on the child, due to the belief that when children are accepted by their parents, their self-esteem/concept and self-confidence will be high and they will be able to solve their own problems or appropriately seek assistance. Techniques focus on the acceptance and reflection of children's and parents' feelings to develop competent and well-adjusted children. The child's behavior is affected through the parent-child interaction (Tavormina, 1980). The Parent Effectiveness Training (PET) program (Gordon, 1970, 1976) is the most widely known of the client-centered approaches and shows support for its use as a preventive intervention (Cedar & Levant, 1990). The client-centered/humanistic approach to parent education, with its focus on the expression of feelings and the importance of communication, defines an open family paradigm.
While the three major theories in use in the parent education field appear to
support only two of the family paradigms proposed by Constantine, it should be noted that the closed and open paradigms are the only images of family life that define themselves by a value system capable of being taught by means of an educational program. Neither the random paradigm, with its values on laissez-faire parenting and anarchic control, nor the
synchronous paradigm, with its promotion of implicit and unstated expectations, and indirect and covert control, lend themselves to an educational or teaching model.
With the increasing support and need for the effectiveness of parent education, the consensus is that parent education should now be examined by techniques and their effects (Lambert, Shapiro & Bergin, 1986; Medway, 1989). It has also been suggested that research focus on: the limitations of parenting programs (Graziano, 1983; Lamb, 1986); the interaction of factors such as parent characteristics (Gordon & Davidson, 198 1; Martin, 1980; O'Dell, 1974; Tavormina, 1980); developmental stages (Dembo et al, 1985; Levant, 1983; O'Dell, 1974; Tavormina, 1980); styles of parenting (Fine & Henry, 1989; Harman & Brim, 1980; Medway, 1989); and the impact of parenting programs on family functioning (Gordon & Davidson, 198 1; Fine & Henry, 1989; Moreland, Schwebel, Beck, & Wells, 1982). But again, the attempts to actually organize parenting groups to fit the characteristics and needs of the parents have been limited. And yet, the parent's stage in the family life cycle is known to affect the parent's choice of techniques (Carter & McGoldrick, 1988) and that levels of parental flexibility and tolerance impact the parent's ability to recognize a need for change (Tavormina, 1980).The impact of the program, as well, was found to be dependent on parental beliefs, values, personal experiences and expectations (Dembo et al, 1980). In an attempt to determine the superiority of model type, Dembo, Sweitzer, and Lauritzen (1985) examined five comparison studies, but drew
no definite conclusions nor were they able to deter ine which type of program was best suited to a particular type of parent. Matching the type of program with the goals, needs and characteristics of parents might increase program efficiency (Fine & Henry, 1989; Tavormina, 1980). Therefore, a clearer understanding of the manner in which parents' beliefs about family life influence the acquisition of parenting skills would be helpful. An understanding of a parent's beliefs regarding family power, authority, roles, communication, and views on stability and adaptability might increase the positive outcome of parent education programs, by matching parents to groups that would fit their beliefs and images of family life.
Overview of Parenting Assessment Instrument Development
Most parenting assessment instruments have been developed to assess the type of knowledge or attitudes emphasized within a specific theoretical orientation. The optimal method of parenting is usually inferred.
The first parenthood instrument was developed in 1899 by Charles Sears. This was a survey to assess adult's attitudes toward punishment. Although a few instruments were developed in the intervening years, it was not until the 1930s that a wide array of instruments began to appear, designed to assess a wide range of constructs and variables related to parenthood. Sigmund Freud's theory of psychosexual stage development provided a major stimulus for understanding the role of parents in the development of
children's personalities. His ideas regarding these stages and the potential for fixations encouraged exploration in research into the role parents play in their children's development. His contributions did not actually spell out the impact of parents, but consisted in outlining some of the human dynamics which operate. Individuals such as Ernst Jones and Karen Homey began filling in Freud's outline (Holden, 1990).
Freud's work was influential in prompting physicians to become more attentive to parents and their childbearing practices. The impact of the family environment became an area of interest for study. This translated into research, by Allport, in efforts to study the construct of attitudes (Holden, 1990). This was the origin of the parental childbearing attitude questionnaires, as it was believed that the family atmosphere to which the child was exposed could be assessed by the parental attitudes. This model of parent-cMd interaction was simple, but inaccurate. Parental behavior was determined by assessing parental attitudes due to the (misconception) that parental behavior was simply a direct reflection of global attitudes of warmth, control, or punitiveness (Holden, 1990).
Some investigators, such as Koch, did not subscribe to such assumptions and
began recognizing the problems in the accuracy of the assessments (Holden, 1990). Other researchers recognized that children's perceptions were also important. A number of instruments assessing children's perceptions of their parents were available by the end of the 1930s (cited in Holden, 1990).
The bidirectionality and interplay of parent-child relations is now well
acknowledged (Bell & Harper, 1977 cited in Holden, 1990) as well as more complex conceptualizations including transactional and systems theories. Belsky (1984) has proposed that rather than viewing parenting as being guided by only one variable, the influence may be due to a complex interplay among the individual's developmental history, marital relations, work status, social network, and child characteristics in addition to the parent's attitudes and personality. Although these variables have been shown to impact a parent's childrearing ability as well as techniques (Dekovic & Gemrs, 1992; Dix, 1991; Simons, Whitbeck, Conger, & Chyi-In, 1991; Trickett, Aber, Carlson, & Cicchetti, 1991) it seems imperative to consider the overall interaction of these and other variables that define a family's image of themselves. According to Constantine (1986), a family's definition of themselves not only involves their demographic description, but also the value they place on the importance of member interaction, stability, autonomy, and control. If a family's self-perception or paradigm defines their sense of self then a portion of that image would include childrearing values. However, of the literature reviewed, there was no indication of parenting being studied from a family image perspective nor were any instruments located which assessed family paradigm in relation to parenting preferences.
Table 2 depicts the seven categories of parenting instruments used as the basis for reviewing the evolution of the conceptualization of parenting and its assessment by the Handbook of Family Measurement Techniques (Touliatos, Perlmutter & Strauss, 1990). A total of 204 parenthood instruments, developed from 1936 through 1974, were abstracted in Straus (1969) and in Straus and Brown (1978) as referenced in Touliatos and associates (1990).
Relating, Parenting to Family Background Characteristics
When examining a family and determining their organization, it appears helpful to examine family background characteristics (i.e. social status, parent educational level, marital status, and gender). This is the first perspective taken in studying parenting style. Based on family demographics it conceptualizes a "right vs. wrong" ideal of parenting. Social Class
Social class has been a popular means of studying differences in parenting styles. Prior to the I 980s, most efforts to explain why parents rear their children as they do focused on social class (Belsky, 1990). Researchers report that parents of differing social status demonstrate different value systems, levels of interaction, and childrearing techniques (Heffer & Kelley, 1987; Hess & Shipman, 1972; Schaefer & Edgerton, 1985; Segal in Sigel, 1985; Simons, Whitbeck, Conger, & Chyi-In, 1991; Strom, Griswold & Slaughter, 198 1; Trickett, Aber, Carlson, & Cicchetti, 199 1; Zegiob & Forehand, 1978).
Cd t. + a
0 Q 0
Cd E >
On cy Vj
Cd ca >
Cd rrA Cd cn m
0 ci ed L;' Q
Cd Cd 0
al 00 V Cd ICd 05 w
-Z:) cr cc &05
=S w cd :3
&n (40, M
CL a) CA
bo 0 Cl.
0 Cd 2
2 0 E
4 'd Cd
od Cd m 0 Cd
Many research findings indicated that parents from a lower socioeconomic class tend to have poorer parenting skills and higher drop out rates and outcome levels related to parent education programs than middle class parents (Clark & Baker, 1982; Dumas & Wahier, 1983; Furey & Basili, 1988; Johnson & Breckenridge, 1982; Trickett, et al, 1991; Wahler, 1980; Webster- Stratton, 1985). Due to this evidence, the appropriateness of existing parent education programs for lower social class parents based on white middle-class values has been questioned (Fine & Henry, 1989; Strom et al 1981).
Differences in parenting styles by social class noted in the literature depict
middle-class mothers as possessing an increased ability to positively communicate with their children as compared to lower social status mothers (Strom et al, 198 1), being more tolerant (Maccoby & Gibbs, 1964), less controlling and more permissive, (Maccoby & Gibbs, 1964; Strom et al, 1981), less directive (Sigel, 1982; Zegiob & Forehand, 1978) and less severe in punishment (Heifer & Kelley, 1987; Maccoby & Gibbs, 1964; Strom et al, 198 1). Middle-class mothers seem more likely to use reasoning, psychological methods of discipline, allow their children more freedom of choice and self-direction, show egalitarian parenting styles, express positive affect toward their children, and verbalize and support cognitive and academic growth (Gecas, 1979; Hess, 1970 both cited in Patterson, DeBaryshe & Ramsey, 1989). Lower-class parents are more likely to use physical discipline, be controlling of their child's behavior, exhibit authoritarian parenting
styles, and engage in less frequent verbal and cognitive stimulation. Further comparison regarding social status and child abuse was found in the literature. Simons, Whitbeck, Conger, and Chyi-In (1991) cited several studies including Bronfrenbrenner, (1977); Garbarino, (1976); Gil, (1970); Strauss and associates, (1980) which demonstrated a significant association between low social status and maltreatment. They suggested two means by which social class influences might correlate with parenting practices across generations: (a) replication of low social class and accompanying stressors and life-style (Burgess & Youngblade, 1988 referenced in Simons et al 1991) and (b) growing up in a lower social class family may influence a parent's childrearing approach and accompanying values regardless of the social status they achieved. Trickett, Aber, Carlson, and Cicchetti (1991) also found similar results reporting that the higher the socioeconomic status the lower the belief in authoritarian control techniques. They also reported that as stresses associated with poverty decrease, the expressed enjoyment of parenting increases. However, the expected relationship for each was not found for the abuse group studied. They also reported social class as a significant predictor of cognitive maturity. For example, middle-class parents are more concerned with motives and intentions of their children's acts whereas working class parents are more concerned with the overt consequences (Rubin, 1976).
Suggestions regarding working with low social class parents focus on
understanding concepts, verbal learning difficulties, training tactics, and participation, (Fine and Henry 1989; O'Dell 1974; and Wyckoff 1980. Matarazzo and Patterson (1986) found in their review of the literature that training effects were less well maintained in low social status groups. It was found that parents of lower social class and with lower education levels do better in parent education programs which minimized verbal learning and which emphasized direct training techiiques (Salzinger, Feldman & Portnoy, 1970 as cited in Matarazzo, & Patterson, 1986; O'Dell, 1974). Many researchers indicated that these same parents of low social class and educational levels are less likely to benefit from parent education programs than parents of middle and upper social class and education levels (Clark & Baker, 1982; Dumas & Wahler, 1983; O'Dell, O'Quin, Alford, O'Briant, Bradlyn & Giefenhain, 1982; Webster-Stratton & Hammond, 1990;, Furey & Basili, 1988) although there is some evidence to the contrary (Mira, 1970; Rogers, Forehand, Griest, Wells & McMahon, 198 1; Rose, 1974) Again, however, the emphasis is on the level of parenting skills development with an implicit judgment that these differences (ie. lower verbal skills, need for concrete directives and role modeling) are deficits and need to somehow be corrected. The nature of the family's organization is not taken into account (Constantine, 1986, 1993; Minuchin, 1974).
Research also focuses on the relationship between educational level and parenting style preferences. Again, a level of judgment is involved. Childrearing values, goals and behaviors are reported to be associated with parental educational level (Dekovic & Gemrs, 1992; Schaefer & Edgerton, 1985; Segal, 1985). Poorer child management skills (Patterson, Cobb & Ray, 1972) were determined to be related to lower levels of education as was the quality of the home environment (Marjoribanks, 1991).
Simons, Whitbeck, Conger, and Chyi-Ln (199 1) found that for both mothers and fathers the level of education was negatively related to the harsh parenting of adolescent males. Apparently parents of lower educational levels tend to utilize harsh discipline in raising boys. The relationship was not found to hold true for girls, with physical punishment apparently perceived as unnecessary or undesirable for parenting girls.
A positive correlation between parental reasoning complexity and educational level was reported by Dekovic and Gemrs (1992) which lends support to the findings of O'Dell and associates ( 1982) that less educated parents have poorer outcomes in parent education programs. As with socioeconomic level, it was determined that lower levels of education related to decreased understanding of complex concepts, poor attendance and decreased participation (Wyckoff, 1980; Mattarazzo & Patterson, 1986) as well as
difficulty with verbal learning expectations rather than direct training methods (Fine & Henry, 1989; O'Dell, 1974).
Gender of Parents
Research studies support the idea that parenting beliefs, values, and style differ depending on the gender of the parent (Cohen, Dibble & Grawe, 1977a, 1977b; Gilligan, 1982, Simons, Whitbeck, Conger & Melby, 1990; Simons, Whitbeck, Conger & Chyi-In, 1991; Goetting 1986 cited in Pittman, Wright & Lloyd, 1989). Mothers tend to be more child oriented than fathers. According to Chodorow (as cited in Gilligan 1982) girls experience themselves as like their mothers, thus fusing the experience of attachment with the process of identity formation. They, therefore, emerge from this period with a basis for empathy built into their primary definition of self, and with a stronger basis for experiencing another's needs or feelings as one's own. Because they are parented by a person of the same gender, girls come to experience themselves as less differentiated and more continuous with and related to the external world. Mothers demonstrate a love arising neither from separation nor from a feeling of being at one with the external world, but rather from a feeling of connection; a primary bond between other and self (Freud, 1930 cited in Gilligan, 1982).
Conversely, boys define themselves by separating their mothers from themselves, thus curtailing their primary love and empathic tie. They emerge more differentiated
(Chodorow, 1974 cited in Gilligan, 1982). Contrasting views were reported by Eisenberg and Lennon (1983), Kohlberg (1984) and Walker ( 1984) found in research edited by Gilligan, Ward and Taylor with Bardige (1988) in which they reported finding no evidence of sex differences in empathy or moral reasoning, but perhaps in moral behavior.
Reisinger (1982) found that fathers tend to be more consistent in their child
management style than mothers. Likewise, the major determinant of child behavior in mothers trying to gain situational compliance, is matemal strategy (Lytton, 1980; Maccoby & Martin, 1983; both cited in Dowdney & Pickles, 1991). Negative and inconsistent mothers are likely to engender negativity in their children.
"As long as the biological function of becoming a mother was thought to endow a woman with the requisite knowledge and ability, no professional training was considered necessary for her new responsibilities..." (Fitz-Simons, 1935). Positive characteristics of the mothers were consistently related to higher quality home environment (Luster & Dubow, 1990). With regards to parent education programs, highly educated women were more likely than women with little education to prepare for parenthood with books and courses (Simons, Whitbeck, Conger & Melby, 1990). Education level had no effect on the parenting styles of fathers, who were no more likely to acquire information on proper parenting if highly educated than if not. In the same study it was also determined that individualistic value commitments were associated with a woman's use of destructive
parenting practices, (coercion and hostility), while a man committed to such a value orientation could simply disengage from the parenting process. Compared to mothers, fathers have a great deal of cultural license regarding the manner in which they play the parenting role and so are more likely to engage in nurturing activities when they believe that such behaviors will make an important difference in the life of the child (Simons et al 1990). McBride (1990) determined in his review of the literature, that the degree of paternal involvement is related to the father's perceived sense of competence rather than his level of skill and knowledge. There was also some indication that patterns of fathering were more systematically related to patterns of marital interaction and satisfaction than was mothering (Belsky, Gilstrap & Rovine, 1984; Goldberg & Easterbrooks, 1984 both cited in Belsky 1990).
Research supports that marital relations influence the quality of parenting (Belsky, 1981, 1990; Dix, 1991). Segal (1985) found that obedience and sibling competition tended to occur more frequently in single-parent families. Members of these families also tended to suffer from social isolation, social status disadvantage and high levels of negative stress (Blechman, 1982).
Parents attempting to raise their children with little to no support from a partner, whether separated or together, reported having more difficulty than parents who were
supportive of each other (Belsky, 1981; 1990; Gordon & Davidson, 198 1; Marr & Kennedy, 1980; Simons et al, 1990). Leifer and Smith (1990) indicated that mothers who were able to break the abusive cycle were more likely to have established emotionally supportive relationships during childhood and/or with a current mate or to have participated in therapy.
With regards to parent education, Webster-Stratton (1985) reported evidence of lower success rates for single-parent families. Horton (cited in Coplin & Houts 1991) stated that if parents do not attend together, the non-attending parent will adopt the new strategies if the initial strategies are similar. However, if one parent opposes the change, new skills can be undermined (Forehand & McMahon, 1981).
The sample in the present study is grouped by race and educational level.
Although the literature noted a difference in middle-class parenting techniques and values, (Heffer & Kelley, 1987; Maccoby & Gibbs, 1964; Sigel, 1982; Strom et al, 1981) this investigator proposed that it is the family's image and definition of itself, encompassing race and educational level, but also involving issues of control, stability, interaction, and autonomy that actually influences the parenting preferences.
Relating Parenting to Parental Cognitive Level
The exclusive interest among developmental psychologists in the influence of overt
parenting practices and behaviors on child development has given way, in recent
years, to a broader interest in the experience of parenting. After nearly a half
century of research, developmental psychologists have discovered that parents,
like their developing offspring, are cognizing individuals with goals, plans,
motivations, intentions, and interests, (editor's notes, Smetana, 1994).
This is the second perspective, as defined in this paper, used as a means for studying childrearing.
Piaget's (1963) concept of the development of thought in children (originally
published in 1926) is the foundation of much of the research in developmental theory. His theory suggests that the thought processes of children move through distinct hierarchical stages of development. As thought progresses through the sensorimotor, pre-operational concrete operations, formal operations, and post-formal operations, it becomes increasingly complex, integrated, and differentiated. Each stage is necessary and can not be skipped because it builds on and incorporates the previous stage and readies the child for the next stage. Each stage is seen to be stable and predictable, and consistent across different contexts.
Areas of adult development, partly influenced by Piaget's work have followed. These include moral reasoning (Kohlber, 1969), ego development (Loevinger, 1980), educational development (Hunt, 1970), ethical/intellectual development (Perry, 1970), parental reasoning complexity (Newberger, 1980), cognitive complexity (Ivey, 1991), parental communication (Applegate et al, 1985) and parental conceptualization of development (Sameroff & Feil, 1985).
Consistent with Piaget's concepts, these developmental stages are arranged in
hierarchical order. Development in each stage is demonstrated by a gradual increase in the complexity of information processing and an increase in accessing capacity (Rozin, 1976 cited in Gelman, 1985).
Knowing the parents' level of reasoning might not enable us to predict a single
action of a parent, which is inherently difficult to predict without taking into
account the context in which the act takes place. But it might help account for the
variation in more global categories of parental behavior ( i.e. the parents' childrearing style), better than some specific parental beliefs (Sigel 1986).
Studying the parental reasoning complexity may be worthwhile for expanding our
knowledge about parental functioning, and in addition it can also enrich our
clinical understanding and serve as a basis for educational (preventive) or clinical
intervention for parents. In planning and evaluating efforts to improve pure
parenting skills or disturbed parent-child relationships, level of reasoning might be a useful criterion for observing whether or not a change took place in the parents'
deeper understanding of the parent-child relationship, (Selman, 1976 cited in Dekovic & Gemrs 1992, pg. 684).
Studies of parenting cognitions have been diverse. A summary by Goodnow and Collins ( 1990) listed a number of domains of parenting beliefs they characterized as dealing with directions of and conditions of development. Directions included parents' goals, starting points (expectations), and beliefs about child development. Conditions included ideas regarding the effects of internal and external events, the influence and responsibility of parents, and methods parents use to achieve goals. Researchers have primarily focused on the effects of these cognitions on parenting behavior and child outcomes (Sigel, McGillicuddy-DeLisi & Goodnow, 1992). Dekovic and Gemrs (1992)
pointed out the importance of studying the cognitive aspects of parental functioning as a critical step in understanding parental behavior, and Newberger (1980) suggested that understanding a parent's level of reasoning provides an idea of how the parent perceives the parent-child relationship and tells us which cognitive resources a parent has and utilizes to interpret and resolve their task as parents. The assumption is that parents do not merely react to the child's actions in the immediate situation, but that numerous previous interactions and knowledge about the child is cognitively structured and then influences the manner in which the parent interprets and handles the situation (Dekovic & Gerris, 1992). Parents organize their actions depending on the goals they hope to achieve, supporting the idea that parental behavior is intentional, not merely reactive. Dekovic and Gerris (1992) found, in their study of parental reasoning and childbearing behaviors, that parental reasoning complexity made an independent contribution to parental behavior. Conceptions about children, parenthood, and the parent-child relationship represent factors internal to the parent. Bandura (1977) reported that parents who do not believe they have the ability to parent successfully do not put their knowledge of parenting into action, become preoccupied with themselves, are emotionally aroused, and do not persist at the task of parenting. Belsky (1990) cited Cox, Owen, Lewis and Henderson (1989) who stated that parents who scored high on measures of ego development and ego strength behaved sensitively and responsively toward their infants. She also cited Frank,
Jacobson, Hole, Justkowski and Huyck (1986) who stated those parents developed feelings of confidence and control in their role as parents. Bugental (Bugental & Shennum, 1984; Bugental, 1991) focused on the amount of power parents believed they had in conflict situations with their children relative to the power they attributed to their children. Mothers low in perceived power experience negative affect, had problem-focused thoughts, and engaged in power-assertive or punitive childrearing. The attributions refer to the degree to which the parents believed the child's behavior was intentional and foreseen, as well as free from external control. Parents were assumed to assign blame and responsibility to their children in the case of misdeeds and to attribute behavior to dispositional characteristics rather than external situational constraints and forces. Dix and Grusec (1985) and Dix, Ruble and Zambarano (1989) demonstrated that these thoughts or attributions are associated with parental anger and punitive reactions in support of the idea that cognition is an important determinant of a large percentage of human emotions (Dix 1991).
Current research and theory conceptualizes emotions, cognitions, and behaviors as complimentary and interrelated, each altering and informing the other (Kiser, Piercy & Lipchik, 1993). Meichenbaumn ( 198 5) cited Piaget, "We must agree that at no level, at no stage, even in the adult, can we find a behavior or state which is purely cognitive without affect nor a purely affective state without a cognitive element involved. There is no such
thing as a purely cognitive state." The interplay of emotions and cognition gives rise to models of attachment hypothesizing that cognitive affective constructs are incorporated into the personality structure of the individual and are relatively stable overtime (Crowell & Feldman, 1991).
Parenting Style as Affected by Cootion
In studying preferences of disciplinary methods it was determined that the
significance of the goal is important. Are the parents' goals compliance or encouragement of a particular style of negotiating? The nature of the goal influences the extent to which parents prefer rationales over power assertion methods. An increased emphasis on conformity increased reports of using power assertion rather than reasoning (NEller, 1988). A sign of cognitive development is seen as the degree of differentiation or the extent to which people see differences among children rather than seeing them as all alike as well as the ability to see themselves and parenting as having shades of gray. Cooke (1991) offered further support in her study of the thinking that underlies parenting. She reported that parenting expertise in problem-solving situations with infants includes, (a) attention focus on cues relevant to the child's goals and needs in the problem-solving situation, (b) extensive specific knowledge of the child's behavioral characteristics and a strong foundation of domain knowledge related to child development and childbearing which is integrated into the specific knowledge of the child, (c) consciously considered
child-focused goals and subgoals, and plans for action which reflect thought about parental roles appropriate for the situational and response to cues from the child, and (d) action which provides opportunities for the child to be self-directive in the situation," (pg. 11). Main and Goldwyn, 1984 cited in Crowell and Feldman (1991) stated that a mother's attachment level is believed to be associated with her ability to read, understand, and respond to her child's behavior and needs. Sumlin (1979) as cited in Leggett (1982), however, reached the conclusion that attitudinal changes regarding parenting may not be equated with cognitive changes.
A number of studies confirmed the usefulness of coordinating learner
characteristics with educational approaches (Solomon, 1980 cited in Leggett 1982). Hunt (1970) cited matching by Internal Control (Rotter, 1966), Ego Development Scale (Loevinger, 1966) and Model Maturity Scale (Kohlberg, 1964). "If you wish to help students to learn you must know something about the underlying cognitive processes, and you also need to have some psychological insight into the individuals you want to help," (Howe, 1987, p. 145).
Relatina Parenting to the Nature of the Family Context
The third perspective for studying parenting, as defined for the purpose of this
paper, is a focus on the nature of the family. This includes the effects of race, contextual differences, and the orientation/organization of the family.
The decision to name this variable "race" has involved careful deliberation. The research contains an overlapping use of terms (i.e., culture, race, ethnicity, sociocultural) referring to the same categories. The U.S. Census Bureau uses the label "race", but acknowledges that this includes both racial and national origin or sociocultural groups. In an attempt to include all pertinent information, research involving any of the aforementioned terms was reviewed for relevancy.
The understanding of parenting requires that we go beyond mother-child
interaction and consider wider familial and cultural variables (Belsky, 1980 as cited in Dowdney & Pickles, 199 1). Cultures and communities deliver many messages about parenting. Bronfenbrenner (1977) pointed out the importance of the broader community or macrosystemn in the setting of the normative cultural standards about childrearing through advice from relatives and experts or through role modeling and witnessing interactions of other families. Goodnow (1985) suggested that one's culture is the primary source of information about the facts of childrearing including what children are like at different ages, what parenting techniques work, and what goals parents should value. Some researchers have extended this investigation as far as to say that what parents think, guides or influences those changes (Keller, Miranda and Gauda, 1984; Sameroff & Feil, 1985).
In Hoffman's (1986) study of cross-cultural differences in childbearing he cited numerous authors' differing answers to the variations. LeVine's 1974 research reported that the differences in childbearing patterns evolved in response to environmental risks threatening the child's survival and self-maintenance. Kohn (1969) stated that occupational roles affect an adult's attitudes and values and thus influence his/her role as a parent. Hoffman cited Barry, Bacon and Child (1967), Barry, Child and Bacon (1959), and Hoffman (1974, 1984, 1986) in reporting that parents rear their children so as to encourage the development of those qualities and attitudes needed for their expected roles, which differ from society to society. Therefore, to define an effective parent requires some notion of the goals of parenting (Gordon, 1980). LeVine stated that obedience is a necessary trait for surviving economically as an adult in a rural society, and Kohn reported that professional and managerial parents see initiative and independence as paying off, while blue-collar parents see obedience and sticking to the rules as most important. Their positions are supported by the Cross-National Value of Children study of 1973 involving eight countries. As well as the study by Harkness and Super (1992) supporting cross-cultural differences resulting from the fact that adult beliefs about the nature of children or about the world in general differ from group to group, and these beliefs affect parenting behavior. Gfellner (1990) also cited a 1986 Super and Harkness article stating that the psychology of caregivers refers to parents' beliefs and values or
ethnotheories that are regulated by the culture and which in turn regulate child development. Fine and Henry (1989) suggested that to be insensitive to the cultural implications which dictate the perceived need of the parents is possibly unethical.
The manner in which cultural groups approach communication, autonomy,
discipline, competition, and control influences parenting styles (Fantini & Cardenas, 1980. Ivey, 1986; McGoldrick, 1989; Sue, 198 1), as well as what new information they are both willing to and able to absorb. LeVine (1982) pointed out that the recognition that empathic listening in another culture is impossible without knowledge of the culture and the specific meanings and contexts through which feelings are expressed. Nwachuku and Ivey (1991) suggested the need for culture-specific training for counselors, and cited Minor's (1983) claim that counseling theory would be enriched if theorizing began from the point of view of the host culture. The client's behavior is looked at from the orientation of the insider, specifically persons from the client's culture. Nwachuku and Ivey (199 1) questioned the issue of "helper hierarchy" and suggested that our own counseling and psychotherapy theory is so culturally encapsulated that we have placed our ideas of helping on the client rather than consulting with the client to help him or her find his or her own culturally and individually appropriate solution.
The importance of being sensitive to the culturally different perspectives when attempting to assist parents is evidenced by returning to the previous mentioned
Cross-National Value of Children study. The needs children meet for parents vary. In the primarily rural countries studied, children were an economic utility. In the U. S. they were found to meet parental needs of primary ties and affection, wheras in the Asian culture they provide fun and stimulation. Super and Harkness (1981) stressed the importance of the "goodness of fit"; that the parents' and children's needs coincide. LeVine (1988) further denoted this view giving examples of parental strategies such as in an agrarian culture where "quantity" of children is important for the survival of the family. High fertility and the importance of infant care and nurturance for a child at risk are stressed with decreased attention paid to a child as it gets older. He contrasted this to an urban industrial society where "quality" of childrearing is the focus. Goals concern the child's acquisition of skills and the mental and social stimulation for a child with a future. Therefore, there is increased attention as a child grows older.
LeVine (1988) studied the effects of parental goals on parental behavior by stating several perspectives of the development of parental behavior. The phylogenetic perspective assumed innate sensitivity to infant signals for nurturance, but did not specify the variety of forms of infant care through which this sensitivity was implemented. The cultural perspective assumed parents were guided by culture-specific models of interpersonal relations, but did not indicate how these distant goals were integrated with the other aims of parents in a particular culture and with their perceptions of their child's
adaptive problems in the early years of life. He maintained that parental activity, while constrained by the human genome and directed by cultural values, must also be seen as adjusted, consciously and unconsciously, to those aspects of the environment that threaten or facilitate the attainment of parental goals. He concluded, by citing LeVine and White ( 198 7), and their view that each culture, drawing on its own symbolic traditions, supplies models for parental behavior that when implemented under local conditions become culture-specific styles of parental commitment.
Strom, Griswold and Slaughter (198 1) emphasized the importance of recognizing individual differences when choosing appropriate programs for parents. According to Gfellner's (1990) review of the literature, there are no clear norms of parenting behavior and this may contribute to the stress felt by parents in their parenting roles. A major challenge is the absence of reliable advice and valid modelling that can be implemented easily. Without a normative method of childbearing it is left to each parent to develop a parenting theory.
The American culture offers numerous scripts to women who become mothers. These offer opportunities for personal development. but also pose dilemmas or crises, when scripts are in conflict with one another or if there are no supportive social structures in the women's lives (Gilligan, Ward, & Taylor, 1988). Nucci and Weber (in press) cited by Nucci (1994) provided evidence that middle-class mothers act in ways indicative of a
conceptual differentiation between children's areas of personal choice and matters of moral and social regulation and prudence. Prudential issues (Nucci 1994) refer to children's actions that result in, or have the potential to result in, harm to the children. Children play an active role in the feedback through requests and resistance (Radke-Yarrow & Girnius Brown, 1981 cited in Nucci 1994). Depending on the cultural view of such resistance a child may or may not be seen as a problem. Super and Harkness (cited in Richman, Miller and Solomon, 1988) suggested that the notion of "difficult" temperament represents an interaction between certain characteristics of the infant and characteristics of the cultural context.
Nwachuku and Ivey (199 1) pointed to examples from previous microskill
multicultural studies by Ivey (1988) which consistently support cultural differences. For example, with an African American or Lebanese client, it may be more helpful to focus on relationships or on the family rather than use the typical person-centered approach through "I" statements. This realization would be imperative in designing a parenting program, for this clientele, to which they would be receptive.
Another example of the importance of fitting the program's style and information to the cultural priorities of the parents/families was found in a study on Mexican and Mexican American parents (Powell, Zambrana, & Silva-Palacios, 1990). It was found that these mothers prefer group and home visiting formats, that they need experienced role
models rather than educational reading material, they want participation by extended family members, rapport with staff members is extremely important, program content should emphasize the child and his/her future, and that all of these were more imperative when working with the Mexican immigrants than when serving the Mexican Americans.
A study of parent child development centers, conducted by Andrews, Blumenthal, Johnson, Kahn, Ferguson, Lasater, Malone and Wallace (1982) also recognized the role of culture in the evolution of parenting practices and emphasized letting parents make informed choices regarding children. Importantly, and in support of this paper's focus, was the perspective of these authors to avoid holding a single standard of parenting as absolute and correct.
This review of the literature found countless research articles which included even more numerous citations in support of a need to consider cultural orientation when studying parenting beliefs as well as techniques. Ethnicity is an important mediator of culture and a long neglected variable in family therapy (McGoldrick, Pearce, & Giordano, 1983). It contributes both paradigmatic preferences and nuances. Nucci's (1994) study on U.S. white middle-class mothers showed that research on parental authority has historically tended to frame the issue as a question of degree along a single dimension such as Baumrind's (197 1) division of parenting types into the categories permissive, authoritative, and authoritarian. Application of this typology to members of social class
and cultural groups that differ from the U.S. white middle-class have proven to be riddled with bias and misunderstanding. The cultural significance of parental behaviors labeled by the typology as authoritarian is not captured within the global, unidimensional framework of the typology. Baumnrind (1972) discovered that parental actions that fit within the authoritarian pattern within white families did not result in an authoritarian "syndrome" among Afican-American girls. Instead it fostered toughness and self-sufficiency and was perceived as "nurturant care-taking." Chao (1993) argued that depictions of parental behaviors as authoritarian, controlling, and restrictive are typically ethnocentric and misleading. Asian families have been found to obtain among the highest scores of unquestioning obedience to parents on the Baumrind measure. According to Chao, parental concern and love are equated in Asian cultures with firm control and governance. To label such tendencies as authoritarian is to misread behaviors that are based on Confucian conceptions of respect for elders. Likewise, Korean children's perceptions of parental warmth tend to increase as overall parental control increases (Rohner & Pettengill, 1985). This reflects a general cultural view of the individual as a fractional part of the family. Kim (1987) however felt that there is a gap between Korean parents' beliefs and their actual parenting, stating their ideology is democratic wheras in actuality it is authoritarian. He believes there are no clear rules, guidelines or expectations due to conflicting values concerning traditional versus modern parenting practices. In a six
country comparative study (Korean Gallup Research Center 1983 cited in Kim 1987) Korean children believed their parents to be too strict, and Korean teens reported the least family life satisfaction among adolescents of I11 countries. Kim described the Korean family by Minuchin's definition of enmeshed, most obvious in the parents' overinvolvement in their children's education, career, and marriages. Stating they have a greater degree of mutual care than of communication he believes there is potential for better relationships among family members with education in communication skills.
In their interactional assessment of white, black, and Mexican American families, Hampson, Beavers, and Hulgus (1990) also indicated the need to not pathologize differing family styles. They reported that their findings are consistent with and supportive of the hypothesis that any differences found between ethnicities are likely to be differences in style of structure or interaction rather than differences in competence or health. They cited Henggeler and Tavormina (1980) as depicting an alternative for interpreting ethnic family differences as not deficits or methodological shortcomings, but "socio-cultural differences in behavior, without any necessary implications of psychopathology."
Comparisons reported by Hampson and associates (1990) focused on cultural differences. Black families were noted as having consistent themes of strong kinship bonds and extended family ties (Hnes & Boyd-Franklin, 1982). Relatives expect and accept reliance on one another, so that various people interchange roles, jobs, and family
functions. There is a high degree of role flexibility and adaptability. This boundary flexibility may look more like chaos to the outside observer accustomed to more predictable patterns of a nuclear household. The church is a major source of socialization and prestige, hence, ministers are more often sought in times of need than are mental health professionals (9nes & Boyd-Franklin, 1982). An egalitarian power structure has evolved earlier than in many Anglo families and there is also a higher rate of "father-absent" homes. Members tend to internalize feelings (Pinderhughes, 1982) and their research supported that Black families were rated as significantly less able to express or state feelings and thoughts directly and clearly, less apt to respond openly and receptively to efforts of family members to interrupt or make personal statements (permeability), and to show a more consistent, less varied range of feelings in their various interactional exchanges. Black families were also less encouraging of dependency needs in their children. The Mexican American family's kinship network was described as extremely important with affiliation and cooperation stressed, while overt confrontation and competition discouraged. The focus on relationships is viewed as more important than task or role performance (Eshleman, 1985 cited in Hampson et al 1990). A high degree of cohesion and hierarchical organization is normal. Interaction is characterized by generational interdependence and involvement, and internal control (Falicov 1982). Respect is significant and autonomy is less important than dignity. The Mexican American
families were rated as showing more dominant power exchanges and were rated as congruent with their family concept by an outside rater. They showed a greater range of overtly expressed feelings and a more congruent sense of family concept than White or Black families, as well as more attuned to and encouraging of dependency needs in their children and more likely to describe themselves as close-knit and discouraging of aggressive, defiant, or disruptive behavior.
Nucci (1994) reported, in looking at Brazilian mothers, that across social classes and geographical regions, these mothers expressed beliefs that children require areas of choice for personal growth. The manner in which these beliefs were expressed varied as a function of the mother's underlying assumptions about the nature of their children's needs and capacities. Middle-class mothers (modem Southern region) held views essentially like those of mothers of U.S. middle-class; that children are to be treated as individuals from infancy and given opportunities to exercise choices to enhance their individual talents and personalities. Their reasons were aimed at supporting their children's emerging autonomy, agency, and personal competence. The more traditional lower-class mothers viewed the limited cognitive capacity of infants and young children as an indication that they were not yet to be considered individuals with a choice or opinions of their own. They gave prudential (risks to the child) and pragmatic reasons for their responses.
There is evidence that ethnicity and social status are two separate variables (Laosa, 1981) although it is often difficult to distinguish to which a specific difference is due (Strom et a] 1981). Sameroff and Feil (1985) determined that parental reasoning complexity was generally lower in lower social status groups in all cultures, however, cultural beliefs in traditional, categorical concepts can prevent or eliminate the need for more complex thought.
The failure of Moss and Jones (1977) as cited in Grusec, Hastings, and
Mammone (1994) to find differences between middle-class and lower-class mothers in their childrearing attributions suggests that these particular cognitions might be tied to something other than messages from cultural and socioeconomic groups. General belief systems having to do with developmental timetables, methods of change, and values may be affected by culture, but there is also a great deal of individual variation within groups for these systems. Goodnow (1985) argued that parents are not passive recipients of their culture's messages, but filter them. Parents seek out each other to discuss childrearing values, and the networks they form build their belief system and reinforce those that fit their community's expectations. If there are certain cultural values, individuals may selectively attend to the messages matching their own beliefs or may choose to decrease contact with contradictory opinions, rather than conform to those opinions.
The Laboratory of Comparative Human Cognition (1982, in press) as cited in Simmons (1994) defines context as interactions between individuals, and between individuals and objects, that are delimited by a unique arrangement of goals, behaviors, expectations, demands, and rules constructed by the participants. Culture influences the presence and arrangements of context. Childhood, therefore, is not a given and consistent phenomena, but exists in an historical and societal context, one that is in great part defined by the needs of adults enmeshed in the conflict (Aptekar, 1990). Behaviors giving rise to judgments of intelligence, behaviors that have traditionally been treated as general mental abilities are, instead, best thought of as performances specific to particular contexts. Unlike a general ability framework, the cultural practice approach (Simmons, 1994) points to evidence that culture-specific knowledge and activities constitute contexts that organize the development and deployment of a repertoire of task-specific cognitive skills. The generality of the skills depends on the extent to which the contexts have common features. Instead of emphasizing the representation of situations internally, Simmon's cultural practice perspective maintains that schemes for guiding behavior do not exist solely in the individual, but are constructed in the interaction between individuals or between the individual and the task in a particular context. These schemata are construed as internal representations of situations and knowledge sets. Processing is difficult if the schemata is inappropriately matched with the input. The importance of matching a family's contextual
representation with the type of education offered is obvious; so they can learn (accretion), revise (tune), and replace (restructure). These functions are seen as an ongoing part of social interactions that are an important source of feedback related to monitoring, checking and evaluation. Grusec, Hastings and Mammone (1994) also supported that parenting beliefs are affected by the particular context in which parents find themselves as well as the specific experiences they have had with their own children, and Rigazio-DiGilio (1993) stated that according to Developmental Counseling and Therapy (DCT) and Systemic Cognitive Developmental Therapy (SCDT), individual development and family development occur within a social and cultural environment. DCT and SCDT practitioners underscore the importance of helping clients expand their understandings and behavior at their current orientation. The emphasis is first to build a strong foundation before moving on. After that they suggested that assisting clients to access other orientations offers opportunities for them to co-construct a broader range of perspectives and behaviors and to develop a more flexible approach to the world.
Belsky (1984) mused that it is of interest to learn that while great effort has been expended studying the characteristics and consequences of parenting, much less attention has been devoted to studying why parents parent the way they do and investigating the effect of the child on parenting behavior. He continues with a report on the available research, concluding that influence on parental functioning is based on three general
sources: (a) the parents' ontogenic origins and personal psychological resources, (b) the child's characteristics of individuality, and (c) contextual sources of stress and support.
Stressors impinging on the family such as unemployment, family violence, marital discord, and divorce are associated with both delinquency and child adjustment problems in general, according to research cited by Patterson, DeBaryshe, and Ramsey ( 1989). Although stressors may have direct and independent effects on child behavior, we assume that the major impact of stress on child adjustment is mediated by family management practices. If the stressors disrupt parenting practices, then the child is placed at risk for adjustment problems (Patterson et al, 1989). For example, financial strain, while demonstrating no direct influence upon parenting, did show an indirect effect upon the parenting practices of mothers, suggesting that financial strain leads to marital problems and that marital difficulties reduce the amount of warmth and nurturance the mother shows. The strain increases the probability that a child will be perceived as difficult, thus increasing the probability of parental rejection and hostility. Rosenblum and Paully (1984) agreed that mother-infant interaction under conditions of economic insecurity, maternal employment, and environmental instability may elicit conflicts between parental and infant needs. They cited several authors who reported that such conflicting demands on a mother, which strain her coping capacity, also undermine her ability to mediate her infant's successful interaction with its surroundings, in turn decreasing the infant's own emerging
sense of competence and mastery and may increase its vulnerability. For both mother and infant, one key factor influencing coping capacity and the sense of self-efficacy is the predictability of events and the relationship of their own actions to outcomes. A setting that does not permit anticipation and strategic response to environmental events may result in negative consequences, with evidence that chaotic, unstable environments alter mother-infant dyadic interaction and can be the source of developmental pathology (Pavenstadt 1965; Turnbull, 1973 cited in Rosenblum & Paully 1984). The most critical predictors of the home environment are the matemal characteristics; age, education, ethnicity, and initial self-esteem and locus of control (Menaghan & Parcel, 1991).
Support is also seen as relevant in predicting parenting difficulty. Mothers with fewer social supports and more family stressors expressed less nurturant and more restrictive attitudes regarding their children, therefore behaving in more hostile and restrictive ways with them, while social support from both relatives and friends, particularly when children are young, has been found to be important to parental satisfaction, parental confidence, and positive parenting behavior (Pittman, Wright & Lloyd 1989; Belsky, 1984). Social support functions in three general ways by providing
(a) emotional support, (b) instrumental assistance, and (c) social expectations (Belsky 1984),
Belsky (1984) noted that the work on child abuse highlights three distinct sources of stress and support: (a) the marital relationship, (b) social networks and (c) employment, which can either promote or undermine parental competence. According to research reviewed by Belsky (1984), marital relations do not influence parenting directly, but impact it indirectly by affecting the general psychological well-being of parents and therefore the skills they exercise in the parenting role. In the same manner he reported that the benefit accrued from social networking is the enhancement of the parent's emotional support. Belsky stresses the importance between matching the support desired and the support received. He cited references to unemployment and an increase in child maltreatment, but beyond those stated the majority of work is on maternal employment, suggesting that the mother's employment status influences the quality and quantity of her and her spouse's parenting behavior. Kohn's (1963) work demonstrated that working class men whose jobs required compliance to authority tended to stress obedience and conformity in their children and favor physical punishment whereas middle-class fathers whose jobs required self-direction and independence tended to value the same in their children. Marjoribanks (1991) determined that family context defined by parent's aspirations for their children is a moderate to strong predictor of young adult's social-status attainment. The implication of this observation is that families that more intensively constrain may have particular difficulty permitting change, such as new
interactions (Hauser, Powers, Noam, Jacobson, Weiss & Follansbee, 1984). Pittman, Wright and Lloyd (1989) in their review of the literature, however, determined that the research findings about the effects of background characteristics such as income, employment and number of children on parental attitudes were inconsistent. Baumrind (1993) also found this, but questioned the reliability of the childbearing assessments.
Radke (1969) reported a need for the education of parents in the
authority-discipline area of home relations. Parents still failed to recognize that children were entitled to respectful treatment as demonstrated in children's descriptions of punishments, in the substantial proportion of parents described as showing anger in discipline, and in the overwhelming proportion of discipline procedures depending on sheer power of the adult or undermining the child's power. There was no stimulation for growth of self-dependence in the child. Because they fail to recognize that the child also strives for power, these techniques frustrate and the probable outcome is aggression and further attempts to gain power (Radke 1969; Strom, Barros & Strom, 1990).
A child's behavior standards correspond closely to the standards required by the parents, and a child emulates by his own behavior with other children the behavior of the parent. Radke (1969) gives the following example. If there is an autocratic, restrictive and severe disciplinary atmosphere at home then the child is viewed as more unpopular with other children. They are seen as inconsiderate, emotionally unstable, frequently
fighting and quarreling, uninhibited and daring, less rivalrous, and more insensitive to criticism and blame. Simons, Whitbeck, Conger and Chyi-In (1991) offered that rural families may engage in more physical discipline than do urban families, but the impact of aggressive parenting on a child's beliefs and personality development are likely to be the same regardless of setting. The picture that is beginning to emerge is that the meaning of deprivation is a deprivation of meaning; a cognitive environment in which behavior is controlled by status rules rather than by attention to the individual characteristics of a specific situation and one in which behavior is not mediated by verbal cues or by teaching that relates events to one another and the present to the future. This environment produces a child who relates to authority rather than to rationale, who although often compliant, is not reflective in his behavior, and for whom the consequences of an act are largely considered in terms of immediate punishment or reward rather than future effects and long-range goals (Hess & Shipman, 1965). Ofientation/Orizanization of the Fgmfly
Family images/constructs do not exist apart from the individuals, neither do they consist merely of the sum of the individual images. The family's distinct image of itself emerges from the interaction of its members' individual images of reality and their behavior. This is formed over time with crucial features often linked to certain critical events in the family history (Constantine, 1986).
Constantine (1986) suggested that families have distinct, stable styles of
functioning based on fundamentally different paradigms. He defined paradigm as a model of both the actual and the ideal; a way of seeing and knowing; a world view. In this way, a family's paradigm is their organization. It is an image and a point of reference for checking the family's realization of its own concept of itself Furthermore, it puts boundaries around a family's understanding, shaping what they can see and do and the manner in which they view their environment. For this reason, it seemed imperative that attitude assessment of target populations be a necessary first step in the design and implementation of an intervention program (Meichenbaum, 1985; Strom, Barros & Strom, 1990). Although the use of specific techniques of discipline may vary, a study by Roberts, Block and Block (1984) indicated that parents have fundamental, pervasive, and enduring childbearing orientations that color their use of specific discipline techniques. These authors also cited other researchers for further support of the viewpoint that parents have coherent and long-lasting underlying childbearing philosophies. The overall picture that emerges from the data is one of considerable continuity in the general attitudes, values and goals of the parents. Likewise the family paradigm appears to be a central and stable feature of a family, Family paradigms therefore seems a very natural way to study parenting beliefs and styles.
Paradigms are distinguished on the basis of their goal-directed behavior differing in how they are oriented to the duality of continuity and change. The viability of all living systems is determined by the interplay of stability and adaptability. Taxonomy is the art and science of classification, with a taxon simply being one of the units of classification, not something real or objectively true about the things being classified. Thus, the abstract ideal of any particular family paradigm can be represented by a point where the dimensions correspond to various properties by which the paradigms can be distinguished. The coordinates of the different taxons correspond to how much of each property is present in a given paradigm. Because the taxons are dialectically interrelated and maximally distinct, there is some characteristic property for each family paradigm that will be maximally present in a pure or ideal example of the type and absent in the pure or ideal forms of the other types. Each taxon then can be characterized by a unique measure which corresponds mathematically to a vector, an arrow pointing in the directions of paradigmatic purity. This vector not only defines, but distinguishes it from alternative types and therefore incorporates and models interrelationships among types (Constantine, 1993). Each family paradigm can therefore be identified by a representative vector representing a similarity to the type with some degree of departure from the others. The four taxonomic vectors, in their pure form, can be identified as hierarchy: corresponding to the closed paradigm, with a degree of reliance on a hierarchy of authority to regulate
process and determine solutions; divergence: corresponding to the random paradigm, with a degree of reliance on completely different and independent action by individuals as acceptable forms of group solution and activity; reflexivity: corresponding to the open paradigm, with a degree of reliance on exploration and examination of family's own structure and process for developing solutions and making decisions; and alignment: corresponding to the synchronous family, with a degree of reliance on preexisting, tacit agreement with a shared set of values, goals, and ideas to regulate process and define solutions (Constantine, 1993).
An example of theory that is primarily dimensional while also including typological features is the Olson Circumplex Model (Olson, Russel & Sprenkle, 1983; Olson, Sprinkle & Russell, 1979). Dimensions are components argued to be basic aspects of the variation in families and their process. The paradigmatic framework provides conceptual connections between the dimensions in Olson's model, that of cohesion and adaptability, and the family types representative of the family paradigms. For example, the closed family paradigm corresponds with a low level of adaptability and a high level of cohesion creating structural connectedness, which may become rigid enmeshment during times of stress.
In comparison, the absence of dimensional features has been noted as a limitation of the paradigmatic framework (Broderick, 1986). With the addition of a fourth family
paradigm, the framework was transformed into the conventional two-by-two arrangement. The basic dimensions in which family paradigms differ can now be determined. One of the aspects of tetrahedral geometry is that the degrees of hierarchy, divergence, reflexivity, and alignment for any family add up to a fixed value. The total can be distributed in any way among the vectors, but the scores add up to the same total. This property supports the assumptions, held about family paradigms, that family style results from tradeoffs among competing typal goals (Kantor & Lehr, 1975). Family paradigms represent the ways in which families assign priorities among competing values and objectives (Kantor & Lehr, 1975; Kluckhohn, 1958). To determine the basic dimensions in which family paradigms differ it is possible to construct a set of dimensions that is directly related to the underlying taxonomy of paradigms. The midpoint of each edge is joined with the midpoint of the opposite edge creating three orthogonal axes which differentiate one pair of paradigms from an opposite pair. The three axes created represent: "engagement," varying from totally enmeshed to completely disengaged; "variability," with extremes from chaos to rigidity; and "convergence," from dichotomized/highly skewed to highly integrated synergy.
The paradigm is the template for the patterns we see (Constantine, 1986).
Paradigms (models, images, worldviews, references, standards) are related in process (behavior, interaction) through regimes (organization, structures, regulating mechanisms).
If the family's regime is successful in translating its paradigm into process the family succeeds at being itself However, not all regimes are compatible with all paradigms. It is possible for a family to be guided by one concept of family, yet be unable to put this into practice either because its organization as a family is incompatible with its paradigm or because its organization is ineffective at generating the desired behavior. A family's greatest strength is also its greatest flaw when the family remains true to its paradigm. A mismatch between paradigm and regime creates stress. Different members may hold competing paradigms and subsystems may operate under different rules. For example, family images are often shaped more by some members' personal images than by others. These are the "architects" of the family system. Because collective family images emerge in an historical process, the parents are most likely to be the architects (Satir, 1972), however, this may not always be the case. In a study reviewed by Tolan, Cromwell, and Brasswell (1986), it was found that in families of delinquents, the delinquent child was often more influential than one or both parents on the tenor and direction of family interactions. No matter what the family's paradigm, delinquency is a sign of family stress. Johnson (cited in Tolan et al 1986) also explained that delinquency serves as a family homeostatic device that signals a failing family system.
Family paradigms are distinguished by finer gradations and variations and
combinations. Because of the fundamental distinctions the paradigm establishes the range
of themes upon which variations can be built. Kantor and Lehr (1975) identified three conceptually distinct paradigms through participant-observation research, with the premise that there is no right answer or single best formula for family living. Families guided by different paradigms will use different approaches to problem solving and be especially good at different things as well as be prone to different kinds of difficulties. The closed paradigm encourages stability, security, and belonging. These families are "continuity-oriented" and homeostatic, operating so as to correct deviations from established patterns. Continuity and uniformity are the priority so the family as a group is seen as more important than the individual. Their motto: "stability through tradition and loyalty." The random paradigm promotes novelty, creativity, and individuality. These families are "discontinuity-oriented." There is not an absence of pattern, but it is a pattern of continual change. The individual is paramount and the source of the family's variety. Their motto: "variety through innovations and individuality." The open paradigm supports adaptability, efficacy, and participation. These families are "consequence-oriented" with resolution through negotiation and collaboration. This paradigm stresses adaptiveness to the needs of both the individual and the system, flexibility and balance. There is an adaptive mix of continuity and discontinuity. They are dedicated to communication. Their motto: "adaptability through negotiation and collaboration." The original, research-based model was extended, based on formal,
systems theoretical arguments regarding the relationship of individual to collective action and its regulation in systems, in a series of papers by Constantine ( 1977, 1983, 1984a). The fourth paradigm, the synchronous paradigm, encourages harmony, tranquility, and mutual identification. These families are "coincidence-oriented" meaning they simply expect there to be a consensus of goals and world views among members. There is no conflict between the individual and the group as neither comes first. There is a non-intellectual sense of unity and harmonious agreement. Their motto: "harmony through perfection and identification."
Including the synchronous paradigm with Kantor and Lehr's types accounts for all four extremes identified in the Olson Circumplex Model and results in a reasonable correspondence (Constantine, 1986) with the four family paradigms derived by Reiss (1971, 1981).
Families differ in the setting of priorities and deciding between competing goals. Their paradigm offers them the guidelines on which to build and by which they view the world. A family operating within a closed paradigm sees the random and open families as chaotic, and views the synchronous as utopian. A random family sees the closed and open as authoritarian. An open family sees the closed and random as alternatives to be used or mixed as needed (Constantine, 1986). Therefore, therapy (or parent training) that
recognizes only a single ideal for family living may try to push families into becoming what they are not, rather than becoming better at what they are.
All paradigms are considered equal in their capacity to serve as effective models for successful family functioning. Each vector represents purity rather than extremeness so that all points within the tetrahedron represent workable family configurations. Most other models define a single ideal of family life (Beavers, 198 1; Olson et al, 1983). The extremes on the dimensions of cohesion and adaptability in the Olson Circumplex Model can be shown to describe extreme patterns associated with each of the four paradigms. These extremes would correspond to the most probable direction of faure for a family and is referred to as an "error of substantiation," a family becoming an exaggerated version of its own paradigm (Kantor & Lehr, 1975). The degree of a family's enablement was observed to not be dependent on a family's style (Kantor & Lehr, 1975). They argued that it must be assessed independently based on the family's ability to function as a system. Any form of family can therefore succeed and any can fail. However, what is functional for one family may not be for another. The paradigmatic framework distinguishes a family's organization or regime from its behavior or process. In the most straightforward situation, a family's regime is successful in translating its paradigm into process-, the family succeeds at being itself Not all regimes, however, are compatible with all paradigms and it is possible for a family to be guided by one concept of family, yet be unable to put it into
practice, either because its organization is incompatible with its paradigm or because its organization is ineffective at generating the desired behavior. The greater the disparity between a family's actual process/behavior and the ideal balance for its specific paradigm, the more dysfunctional that family is likely to be. Each variation of family style has its own vulnerabilities and strengths, however the more the variation from its particular mean, the less likely the family will remain functional. The probability of dysfunction is higher when there is greater disparity between paradigm and process. Families have a probable direction of failure when trying to resolve problems because they are most likely to draw on the methods and resources of their own paradigm. By doing this they intensify their paradigmatic commitment so that, for example, a closed system would move toward even greater rigidity and enmeshment, while an open paradigm would become chaotic through their increased enmeshment. A synchronous system would move toward greater rigidity, but with increased disengagement, while a random paradigm would become increasingly chaotic and disengaged (Constantine, 1983). Problems may also occur if a family gives up on its commitment or is unable to establish an effective organization to its style and slips in the opposite direction. Again, in a closed paradigm an ineffective authority structure might result in a decline into greater chaos or disengagement. Either extreme represents movement along the taxonomic vector away from the dominant paradigm. This perspective can help the clinician (or parent educator) fully respect the family in its chosen
style of organization and operation. The key to identifying a family's dysfunctional excesses or inadequacies is understanding what is congruent with that family's unique mix of paradigms. It is not the amount of disorder, differentiation, closeness, that is functional or dysfunctional except as it exceeds a family's own tolerance and capacities. The direction of incongruence may also suggest whether the family is caught in a spiral of exaggeration or inadequacies. This also helps us to understand what happens when external factors introduce a certain level of imbalance into a family. Again, the issue is not how extreme the factor is, but how extreme relative to each family's preferred and actual organization. Any degree of cohesion or adaptability can be functional if validated by the family (Olson et al, 1983).
It seemed evident by reviewing this body of knowledge, that it is imperative to
research how differing family paradigms relate to a family's parenting beliefs and practices. This line of thought supports the idea of no "one right way" of parenting, yet also acknowledges that problems can ariseAithin any family type. Simons, Whitbeck, Conger and Melby (1990) distinguished between "constructive" and "destructive" parenting practices. Constructive parenting practices offer a style characterized by warmth, inductive reasoning, clear communication, and appropriate monitoring and tend to promote a child's cognitive functioning, social skills, moral development, and
psychological adjustment. Destructive practices involve hostility, rejection, and coercion and are associated with delinquency, psychopathology, academic failure, and substance abuse. The use of these two terms in the description of parenting styles would appear to blend well with the family paradigmatic view by not incorporating a single right definition of parenting, but still acknowledging some fairly agreed upon desired outcomes for our children. The constructive parenting characteristics are in support of Garbarino, Sebes and Schellenbach (1984) low-risk families characterized by a flexibly connected family system, a disavowal of coercion, and a more supportive, less punishing style of parenting. The destructive parenting practices support their definition of high-risk families characterized by chronic internalized developmental problems, positive values and attitudes concerning coercion, and a chaotically enmeshed interpersonal system.
It is of specific interest to this researcher to acknowledge the limitations cited in the literature as to the lack of attention given to the specific and unique characteristics of parents as learners. In a review of the parenting literature conducted by Lamb (1986), not a single article focused on the need to match parental characteristics to the parent education model utilized. Rather, it was found that parenting programs are often organized based on the characteristics of the children of involved parents: by developmental stages (Croake, 1983; Forehand & Long, 1988; Tavormina, 1980); types of problems (Gordon & Davidson, 198 1 Levant, 1983 Robin et al, 1977); or severity of
problems (Alexander et al, 1976; Gordon & Davidson, 198 1; Graziano, 1986; Klein et al, 1977). Numerous other researchers have cited the need for a thorough examination of the interaction between the characteristics of the parent and the parent education experience (Fine & Henry, 1989; Gordon & Davidson, 198 1; Graziano, 1983; Lamb, 1986; Levant, 1986a; Levant 1986b; Medway, 1989; Moreland et a], 1982; Tavormina, 1980).
Statement of the Purpose
The purpose of this study was to determine whether women differing in terms of their family paradigmatic preference would differ in their parenting practices and parent-child communication. Women's family paradigm preferences were classified into one of Constantine's family paradigms. Differences among the paradigm groups were then examined in terms of (a) parental discipline practices; and (b) style of parent-child communication. In addition, differences in race and educational level among participants in different paradigms were explored.
In this chapter, the research hypotheses, design of the study, population, sampling procedures, subjects, and data collection are described. The instrumentation, data analysis, and methodological limitations are also discussed.
The following research hypotheses were examined in this study:
Hypothesis One- Mothers differing in educational level demonstrate no differences in family paradigmatic preferences as measured by The Family Regime Assessment Scale.
Hypothesis Two: Mothers differing in race demonstrate no differences in family paradigmatic preferences as measured by The Family Regime Assessment Scale.
Hypothesis Three: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in child-management values for level of verbal interaction and expressiveness as measured by the Parenting Preferences Inventory.
Hypothesis Four: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in child-management values for level of behavioral interaction and involvement as measured by the Parenting Preferences Inventory.
Hypothesis Five: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in family-management style as measured by the Democratic sub-scale of the Colorado Self-Report Measure of Family Functioning.
Hypothesis Six: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in family-management style as measured by the Laissez-Faire sub-scale of the Colorado Self-Report Measure of Family Functioning.
Hypothesis Seven: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in family-management style as measured by the Authoritarian sub-scale of the Colorado Self-Report Measure of Family Functioning.
Hypothesis Eight: After controlling for educational level and race, mothers differing in family paradigmatic preference demonstrate no differences in levels of child-centered communication as measured by the Person-Centered Communication Assessment.
Design of the Study
This investigation was a descriptive comparative study in which the relationship between a parent's family paradigm (or world view) and their parenting was examined. The independent variables included the following parental characteristics- educational level, race, and family paradigm preference. The dependent variables included: parenting style and child-centered parent-child communication. Each of these variables is described below:
Educational level. For the purposes of this study, educational level was divided into two levels: lower and higher. The educational level of each mother was based on
information gathered in the demographic interview schedule and then determined according to years of formal education.
Race. The race of the mothers participating in this study were defined by their
self-identification. This study focused on two groups: black Affican American, at least a second generation in the United States and white Anglo American, at least a second generation in the United States. The Census Bureau reports race as self-identified and denoting no clear-cut scientific definition of "biological stock." Due to this manner of determining racial categories, White, for example, is noted to include other entries such as Canadian, German, Italian, Lebanese, Near Easterner, Arab, and Polish. For this reason the limitation of "at least second generation in the United States" was placed on the sample to attempt to control for other cultural background influences.
Family paradigm preference. The family paradigm preference of the mothers
involved in this study was categorized into one of Constantine's (1986) four paradigms: closed, random, open, or synchronous. This represents their image or view of the way in which they feel their families should operate. Dependent Variables
Parenting Mle. The parenting style is those means of discipline, including behavior management and verbal negotiation, employed by the mothers participating in this study.
Child-centered parent-child communication. This type of communication
represents a belief on the part of the mother that the expression of feelings and thoughts by both her and her child is an important aspect of the parent-child relationship.
Description of the Population
The population consisted of mothers of elementary school children, ages five years to eight years, who attend public school in Duval County. According to the U.S. Bureau of the Census (1990) the total population of Duval County was 672,971. TheStateof Florida (1994) projected a 1995 population for Duval County of between 698,300 and 741,500. Of the total 1990 population, 163,902 were reported to be black citizens and 489,604 were reported to be white citizens. This is approximately 24% black and 76% white.
According to the annual "20th Day Count" of the Duval County Public School System, there are approximately 123,470 students enrolled in grades pre-kindergarten through l2th grade, attending 153 schools. There are 102 elementary schools (pre-k through 5th grade) serving 65,409 students, 31,956 of which are in the target population of kindergarten, first, and second grade. The overall racial mix of the school system is 40% black and 60% white, although individual schools vary from I% black and 99% white to 99% black and 1% white. The three schools chosen for this study approximate the racial mix of the county of 24% black and 76% white.
The administrators of three different public elementary schools in Duval County were contacted and asked to (a) provide a classroom roll, including parents'names and telephone numbers for the kindergarten, first, and second grade classes to the researcher; or (b) send a letter to parents written by the investigator which briefly explained the purpose of the study and requested parents to call if willing to participate (Appendix Q. All parents responding from the groups receiving one of the letters were contacted to participate in the study. Of the parent names provided through classroom fists, attempts were made to equally fill each of the four groups by locating areas of residence through matching names with the telephone book. Mothers were interviewed until the minimum criteria of 12 subjects per group was achieved. The researcher contacted mothers by phone and explained the purpose of the study, the procedure, and the time commitment (Appendix D). Upon determining that the mother met criteria for participation in the study she was invited to participate. Criteria included (a) having at least one child between the ages of five years and eight years, living in the home, who was the mother's natural (biological) child; (b) be either white Anglo American of at least a second generation or black Affican American of at least a second generation; and (c) having two adults, acting as parental figures, living in the home together for at least three years.
The sample consisted of 57 mothers representing four different groups: (a) lower educational level (high school graduate or less) white Anglo American, second generation;,
(b) higher educational level (at least one year of college) white Anglo American, second generation; (c) lower educational level (high school graduate or less) black Afican American, second generation;, and (d) higher educational level (at least one year of college) black Afican American, second generation.
Demographic information regarding the mother's race, marital status, educational level, and occupation along with source of additional income and that person's educational level and occupation was gathered. Twenty-seven Afican-American mothers were interviewed of which 15 were higher educational level and 12 were lower educational level. Thirty caucasian mothers were interviewed of which 15 were higher educational level and 15 were lower educational level. All mothers were involved in significant relationships, but classified themselves differently. Of the African-American mothers 20 were married, 3 were single, 3 were separated, and I was divorced. Of the caucasian mothers 28 were married, 1 was single, and I was divorced. Occupational levels were determined according to criteria of the Four Factor Index of Social Status by Hollingshead (1975). In the Afican-American sample I mother was not employed, I was employed at a menial service or unskilled worker level, 12 were employed at a semiskilled, skilled or
clerical level, 10 were employed at a semiprofessional or minor professional level, and 2 were employed at a lesser or major professional level. In the caucasian sample 5 mothers were not employed, 3 were employed at a menial service or unskilled worker level, 14 were employed at a semiskilled, skilled or clerical level, and 8 were employed at a semiprofessional or minor professional level.
Upon giving her agreement to participate, an appointment was made with each mother to meet with the researcher at a mutually agreeable site for an interview, lasting approximately one hour. The researcher again explained the purpose of the study and addressed any questions the mother may have had at that time. The informed consent was signed and she was provided with a copy (Appendix E).
The interview began with the collection of demographic information, (Appendix A). Following this, the Person Centered Communication Assessment was administered with responses audio-taped. Three sub-scales of the Colorado Self-Report Measure of Family Functioning followed along with the Family Regime Assessment Scale. Finally, the Parenting Preferences Inventory (Appendix B) was administered. The assessment process took approximately one hour and was conducted with the interviewer present to answer any questions which may have arisen.
Demographic information was collected verbally and included mother's race, marital status, educational level, occupation, number and ages of children living in the household, as well as marital status, educational level, and occupation of the other parental figure in the home. Four assessment tools were used in this study: (a) the Person-Centered Communication Assessment; (b) the Democratic, Laissez-Faire, and Authoritarian sub-scales of the Colorado Self-Report Measure of Family Functioning; (c) the Family Regime Assessment Scale; and (d) the Parenting Preferences Inventory. The Person-Centered Communication Assessment
Measurement of the participants' styles of parent-child communication was accomplished through the use of the Person-Centered Communication Assessment developed by Applegate, Burke, Burleson, Delia, and Kline (1985). This instrument is based on a constructivist's approach to communication, which emphasizes the interrelations among stable individual differences in social-cognitive and verbal strategies. The development of communicative abilities is seen as grounded in the development of a system of social cognitive schemes, the most basic of which are interpersonal constructs. These constructs are defined as bipolar cognitive structures developed and used to interpret, anticipate, and evaluate the thoughts and behaviors of others.
Subjects are asked to respond to seven hypothetical situations which describe typical parent-child communication exchanges. Five of the situations describe circumstances requiring the mother to regulate or discipline her child's behavior. Two of the situations present circumstances requiring the mother to provide comfort or emotional support to her child. Mothers are asked to state explicitly what they would say in each situation. Probe questions or role-playing (with the interviewer playing the role of the child) are used as needed to encourage mothers to respond fully and to provide actual messages rather than abstract discussion.
Individual differences in the person-centered quality of the mother's
communication is assessed by the use of two parallel six-level hierarchical coding systems. The specific aspect of person-centered communication focused on is the extent to which mother's regulative and comforting strategies encourage the child to reflect upon and reason through relevant behaviors, feelings, and circumstances. Regulative strategies are scored by the degree to which the mother encourages the child to modify his/her behavior as a function of reflecting on and reasoning through the nature and consequences of his/her inappropriate behavior. Comforting strategies are scored by the degree to which the mother's response grants legitimacy to the child's feelings and encourages him/her to reflect upon and seek an understanding of his/her feelings as related to the circumstances. The seven individual situation scores are scored within the appropriate hierarchical coding
system, and are summed across regulative and comforting strategy to form a single index of person-centered parental communication.
Interrater reliability coefficients (by intraclass correlation) were .85 for regulative strategies and .90 for comforting strategies. Cross-situational consistency for the seven situations, as assessed by Cronbach's coefficient alpha was .90 for the five regulative situations and .80 for the two comforting situations. Construct validity was determined by a comparison of three indices of interpersonal construct system development and the assessments of person-centered parental communication.
The indices of construct differentiation and construct system integration were assessed using a modified form of Crockett's (1965) Role Category Questionnaire. The third index, construct system structure, was assessed using a modified form of Kelly's (1955) Role Construct Repertory Test and coded for level of construct abstractness. In general, the indices of interpersonal construct system development were moderately associated with the assessments of person-centered parental communication. All three of the construct system indices were significantly and positively associated with the degree of person-centeredness exhibited in regulative strategies with construct abstractness being the strongest predictor (r =.49). Abstractness was the only significant construct system predictor for the degree of person-centeredness exhibited in comforting strategies (r =.59). Results from the multiple regression analysis were consistent with previous findings
indicating that construct abstractness is the most important aspect of construct system development contributing to person-centered communication. The three indices of person-centered parental communication were positively intercorrelated at highly significant levels for both regulative and comforting strategies, indicating that a considerable degree of coherence exists in the quality of communicative strategies used across communicative functions, and justifying summing across them to form a single index of person-centered parental communication. There were also significant associations between the two indices of socioeconomic status and both the construct system and communicative behavior indices. Even when the influence of socioeconomic status was partialed out, the construct system indices generally remained significant predictors of person-centered communication. Finally, the results of the path analysis, for assessment of the extent to which the impact of social class on person-centered communication is mediated by the interpersonal construct system, support the notions that
(a) the level of construct system development is a direct determinant of person-centeredness in parental communication patterns, and (b) the effect of social-class on parental communication patterns is mediated through the interpersonal construct system.