U.S. Department of Health and Human Services

Identifying Successful Families: An Overview of Constructs and Selected Measures

Maria Krysan, Kristin A. Moore, Ph.D., and Nicholas Zill, Ph.D.

Child Trends, Inc.

May 10, 1990

PDF Version

---

This report was prepared under contract #HHS-100-89-0041 between HHS's Office of Social Services Policy (now the Office of Disability, Aging and Long-Term Care Policy) and Child Trends, Inc. For additional information about this subject, you can visit the ASPE home page at http://aspe.hhs.gov. The Project Officer was Gerald Silverman.

---

TABLE OF CONTENTS

INTRODUCTION

DEFINING A SUCCESSFUL FAMILY

CONSTRUCTS

MEASURES

Identifying Strong or Successful Families

Measurement Strategies

Measures of Communication

Measures of Encouragement, Appreciation and Commitment

Measures of Religious Orientation

Measures of Adaptability

Measures of Social Connectedness

Measures of Clear Roles

Measures of Time Together

METHODOLOGICAL ISSUES

ISSUES RELEVANT TO APPLICATION OF FINDINGS

REFERENCES

APPENDIX A. THE CHARACTERISTICS OF STRONG FAMILIES

APPENDIX B. PROPOSED MODEL OF FAMILY STRENGTHS

INTRODUCTION

Researchers, policy makers, and the media have focused considerable attention on the problems of families and how some families are failing to meet the needs of their members. Much less attention has been paid to strong, healthy families, and the characteristics that make them successful. Understanding the "secrets" of strong families would be useful to clinicians and counselors as they plan intervention programs designed to help families. The extent to which the characteristics of healthy families are skills that can be taught has implications for policy development and intervention programs. Information on family strengths can also be useful to those seeking to improve family life.

The study of family strengths has been pursued by researchers from a variety of disciplines, including psychiatry, sociology, psychology, and family/marriage counseling. This paper will present an overview of this growing literature, focusing on four topics:

• the definition of strong families;
• the characteristics of strong families as identified by various researchers;
• the operationalization of these characteristics; and
• the methodological, policy, and intervention issues that the research raises.

A review of the literature on successful families suggests two general strategies that have been used to uncover the ingredients that make for family success. First, there is the study of family strengths per se. This is a body of research that looks at strong families in order to identify what makes them strong. Stinnett and DeFrain's work at the Family Strengths Research Project represents such a perspective. They focus on strengths. Alternatively, there are researchers who study healthy or normal families and the characteristics that differentiate them from their dysfunctional counterparts. These researchers often begin by examining problem families (such as those of mental patients) and contrasting them with healthy families. The work of Olson, McCubbin, Beavers, and Lewis illustrate this approach.

Research focusing on topics other than successful families is often relevant to the study of strong families. For example, research on family communication, time use in families, and conflict resolution among families and couples can help shed light on the processes within families that make them strong or that undermine their ability to be successful. This research can provide empirical support for hypotheses about the importance to family functioning of a given family characteristic. In addition, the constructs and measures developed in these more narrow and intense studies can potentially be adopted for use in broad assessments of the multiple factors that determine family success. This overview will focus, however, primarily on studies specifically on successful families research.

DEFINING A SUCCESSFUL FAMILY

Any attempt to understand the "secrets" of strong families must begin with an understanding of the definition of a successful family. A family can be identified as successful by the characteristics of the individual members, by the characteristics of the family interaction, or by the extent to which it fulfills certain functions considered to be the responsibility of the family.

One criterion for a successful family is that it is able to reproduce itself; i.e. to raise children who go on to establish stable and harmonious families themselves. This is, of course, a broad and somewhat circular definition that is difficult to apply at a given point because of the time involved in waiting for children to establish families of their own. A number of additional definitions are available for consideration.

Some researchers have approached the question of defining a successful family by developing models and perspectives about the functions of the family. The functions that have been used have tended to be psychological ones, rather than societal or economic functions of the family. For example, Lewis and colleagues, approaching the study of the family from a clinical perspective, use Parson's two cardinal tasks of a family to guide their research: "A family ought to raise children who become autonomous, and it should provide sufficient emotional support for stabilizing the parents' personalities and continuing their emotional maturation. To the extent a family accomplishes these tasks, it can be considered competent; to the extent it fails at one or both tasks, it can be considered less competent or dysfunctional" (Lewis and Looney, 1983, p. 4). Thus, a family is successful to the extent that it provides an environment appropriate for the development of child and parent alike.

Stinnett, coming from a family strengths research perspective, proposes that a successful or strong family "creates a sense of positive family identity, promotes satisfying and fulfilling interaction among members, encourages the development of family group and individual members, and is able to deal with stress" (Stinnett, 1979). This definition used by Stinnett differs from Lewis' in that it includes an element of satisfaction with family life in addition to positive individual outcomes. This difference has implications for operationalizing the definition, which will be discussed in a later section.

David Olson and colleagues propose that families should be able to: 1) "cope with stress and problems in an efficient and effective way;" 2) "have and use coping resources both from within and from outside the family;" and 3) "have the ability to end up being more cohesive, more flexible and more satisfied as a result of effectively overcoming stress and problems" (Olson, 1986, p. 104). Their focus is on a family's ability to adjust, in the face of change or crisis, with an emphasis on the changes across the family life cycle. Their definition of a strong family is contingent on family interaction rather than the characteristics of the individual.

CONSTRUCTS

Based on various assumptions about what a strong family does, researchers have developed lists of particular characteristics that are common among successful families. The traits identified by some of these researchers are found in Appendix A. These represent a combination of structural and behavioral attributes of the family. Much of the research that went into identifying these characteristics was based on models of family functioning that point to specific areas of family dynamics that are critical to successful family functioning.

Despite some disparities, a perusal of Appendix A reveals a number of similarities across researchers. Thus, in spite of differences in discipline and perspective, there seems to be a consensus about the basic dimensions of a strong, healthy family. In an unpublished review of the literature, Judson Swihart identifies those characteristics of strong families that are most frequently mentioned by researchers.

These constructs, which are often interrelated and complex, will form the framework for the remainder of the discussion on the characteristics of strong families. These constructs are:

• communication
• encouragement of individuals
• commitment to family
• religious orientation
• social connectedness
• ability to adapt
• expressing appreciation
• clear roles
• time together

In a later section, each of these constructs will be identified, defined and described briefly as it exists in strong, healthy families. In addition, a sample of the instruments used to measure it will be presented. First, we briefly describe several approaches to operationalizing what it means to be a strong or successful family.

MEASURES

Identifying Strong or Successful Families

In empirical research, criteria must be selected to use in identifying a strong family so a set of characteristics common to them can be assembled. Despite general agreement on the characteristics of strong families, the criteria that are used in a research study to differentiate between a functional and a dysfunctional family will depend on the perspective of the researcher and what s/he considers to be the functions of a family, and how s/he defines success. For example, is the education of children a family function or not? Is it more important to children to grow up to be happy or economically self-sufficient, or are both essential? Researchers have operationalized the concept of a strong family in a number of ways.

Some researchers identify strong families as those that are nonpathological or nonclinical (for example, Olson, Lewis, Beavers, etc.). That is, they study clinical and nonclinical families and then identify dynamics or characteristics which distinguish between the two. Thus, a strong family is one whose members are not seeking professional help, are not alcoholics, are not schizophrenics, etc. These researchers have developed models of family functioning based on different assumptions of what a family is supposed to do. They use these models to identify patterns of interaction that distinguish between well-functioning and poorly functioning families, based on the differences between clinical and nonclinical families. They describe the characteristics in terms of the way these families deal with certain issues (power, conflict, affective issues, etc.)

Other researchers, those studying family strengths per se, use a different method for distinguishing between strong families and weak families. In their research, Stinnett and DeFrain have selected strong families based on three assumptions: 1) "they would have a high degree of marital happiness;" 2) "they would have satisfying parent-child relationships;" and 3) "family members would do a good job of meeting each other's needs" (Stinnett and DeFrain, 1985, p. 9).

They identify strong families in their research in two ways. In an initial study, subjects were located through a nomination process. Most of the respondents for a second data collection project were volunteers responding to a newspaper ad requesting participants for a research project on strong families. Participants in both of these studies were screened using a self-report family and marital satisfaction scale. Those who scored very high were included in the study. Thus, for the most part, successful families have been operationalized as those people who are identified by themselves or others to be successful. The characteristics of these families are then studied. The nomination process has also been used by Lawrence Gary, in his study of the strengths of stable, black families (Gary et al., 1983).

A result of this type of research is the identification of a set of characteristics that families identify as important to family functioning, traits that make them a strong family. In their study of American Indian, black, Chicano, white, and Hmong families, Abbott and Meredith find a number of similarities across all these groups as to what parents consider to be traits of healthy families (Abbott & Meredith, 1985).

Measurement Strategies

Researchers have devised a number of means to measure characteristics of family life. The methods range from large scale surveys to observation techniques. The assessments vary in scope from measures of total family functioning to the measurement of component constructs. Prior to reviewing each individual construct, the issue of "total family functioning" must be addressed.

Much research on family strengths and successful families has been in the assessment of total family functioning. In these inventories, the entire set of characteristics that are identified by the researcher to be important to the functioning of a successful family is built into one scale which measures "total family functioning." These instruments are often used as screening mechanisms for clinicians or as an identifier of a strong family for research purposes. That is, families who score high on this overall assessment are considered strong, successful families.

One such paper-and-pencil measure is Olson's Family Adaptability and Cohesion Evaluation Scales (FACES). This scale was designed specifically to test hypotheses derived from his Circumplex Model. The two main dimensions measured are cohesion and adaptability, which are operationalized in terms of measures of component constructs such as commitment, encouragement of individual members, time together, and social connectedness (Olson, 1982).

Moos' Family Environment Scale, another "total family functioning measure" consists of three subscales focusing on different dimensions of the family: relationship, personal growth, and system maintenance. Again, these dimensions are operationalized in terms of cohesion, expressiveness, conflict, control, organization, etc., and are assessed by having study participants fill out forms that describe their family (Moos and Moos, 1986).

The Beavers Timberlawn Family Evaluation Scale (BTFES) is an example of an observation technique for assessing total family functioning. A trained clinician observes a video tape of family interaction and assesses the family across several dimensions to achieve an overall family functioning measure. This measurement technique has been used by Lewis in studies of well-functioning working class blacks, as well as married couples making the transition to parenthood (Lewis and Looney, 1983; Lewis, 1989).

In the following sections, we will discuss component constructs and describe how they manifest themselves in successful families. Where possible, specific examples of instruments used to measure them will be identified. Of course these measures often are subscales of total family functioning instruments, such as those described above.

Measures of Communication

The presence of effective communication patterns is one of the most frequently mentioned characteristics of strong families (Swihart, 1988). Researchers characterize the communication patterns of strong families as clear, open, and frequent. Family members talk to each other often, and when they do, they are honest and open with each other (For example, Stinnett & DeFrain, 1985; Lewis, 1979; Epstein, 1983; Olson, 1986). Indeed, the Circumplex Model, developed by Olson and colleagues, identifies communication as the facilitating dimension of a functional family.

Many people have studied communication patterns in the family and have used a variety of methods, including both observational and self-report survey instruments. Grotevant's work on the contribution of the family to identity formation in adolescents provides an example of observation work which links family communication patterns with positive individual outcomes. This study involved the observation of 121 families with a high school senior as they perform a decision-making task. In his work, Grotevant finds that individuals from families characterized by open and clear communication score higher on measures of self-esteem (Grotevant, 1983).

Barnes and Olson have developed a self-report instrument that focuses on the nature of communication during the often difficult period of adolescence. This inventory collects perceptions from three family members: both parents and the adolescent. In this instrument, respondents are questioned about "the amount of openness, the extent of problems or barriers to family communication and the degree to which people are selective in their discussion with other family members" (Olson et al., 1982, p. 34).

Swihart points out that communication skills are a particularly fruitful area in terms of intervention. Skill in communication and teaching families to communicate better has been a successful intervention technique and is the focus of many programs (Swihart, 1988).

Measures of Encouragement, Appreciation and Commitment

The next three characteristics are closely linked, and so will first be addressed as a whole, and then individually. The encouragement of individual members, expressing appreciation, and commitment to family are all affective aspects of family life that are closely related to a dimension identified by several researchers as cohesion. Olson defines cohesion as the feeling of closeness and attachment of family members to each other. The instrument he uses to measure this construct (FACES) consists of subscales that include the above three characteristics. Thus, Olson defines cohesion as a sum of, among other things, commitment, appreciation and support, and the encouragement of the individual within the context of the family (Olson et al., 1982).

Numerous researchers have identified the construct of cohesion as important, though sometimes with slightly different definitions. Furthermore, there have been a number of techniques developed to measure the construct. In addition to verbal self-report techniques, non-verbal techniques have been used to measure cohesion. For example, Feldman, in her study of changing perceptions of family power and cohesion across adolescence, had adolescents represent the amount of closeness between family members by placing figures on a board. Closeness was defined by how far apart members were placed in relation to each other. She used this as an indication of cohesion (Feldman and Gehring, 1988).

Although the summary concept of cohesion has been measured by a variety of techniques, measures of the component characteristics of encouragement of individual members, commitment to family, and expression of appreciation are much less common.

The encouragement of individual members encompasses a wide range of affective dimensions related to support, recognition and respect. Swihart reviews the basic concept as it manifests itself in strong families: "Strong families appreciated the uniqueness of each family member while cultivating a sense of belonging to the family. Individuals enjoyed the family framework which provided structure but did not confine" (Swihart, 1988, p. 3). Thus, the development of the autonomy of its members, within a supportive environment, is an important function of the family, and a characteristic of strong families.

Grotevant's observational study of family characteristics and adolescent development is relevant to this dimension. This study describes the form that this takes in families with successful identity formation of the children. He notes: "...these studies converge on the conclusion that both connectedness (as indicated by support, cohesiveness, and acceptance) and individuality (as indicated by disagreements) in family interaction are related to identity formation in late adolescence" (Grotevant, 1983, p. 231). Grotevant identifies the nature of the family interaction and provides evidence for the support of the importance of the encouragement of the individual as a characteristic of a successful family.

Stinnett describes commitment to the family as follows: "Commitment ... goes in two directions. Each family member is valued; each is supported and sustained. At the same time they are committed to the family as a unit. They have a sense of being a team; they have a family identity and unity. When outside pressures (work, for example) threaten to remove family from its top priority, members of strong families take action and make sacrifices if necessary to preserve family well-being" (Stinnett, 1986, p. 48).

"Delivering a high level of positive reinforcement to family members, day in and day out, doing things that are positive from the other person's perspective, just for their sake, not merely as a strategy for "buying their love," etc.", is Schumm's description of appreciation as an important characteristic of strong families (Schumm, 1986, p. 122). Measures of commitment to the family and expressing appreciation are most often found as subscales of family functioning scales, such as Moos and Moos' Family Environment Scale, or Stinnett and DeFrain's Family Strengths Inventory. Few instruments offer assessments of these constructs on their own.

Measures of Religious Orientation

Religious orientation is identified by many researchers as being an important component of strong families. Most researchers point out that religious orientation/spiritual wellness is not necessarily contingent on membership in any particular denomination, or on frequency of church attendance. Stinnett offers a sample of the various forms religious or spiritual wellness can take: "For many, the yearnings of their spiritual nature are expressed by membership in an organized religious body such as a church, synagogue, or temple. For others spirituality manifests a concern for others, involvement in worthy causes, or adherence to a moral code" (Stinnett, 1986, p. 48). Thus, this construct has not always been defined in terms of measures of frequency of church attendance or adherence to a specific belief. Rather, the emphasis is that strong families are guided by an underlying moral or value system shared by all members. Subscales in a number of the self-report inventories designed to assess total family functioning measure this construct, in particular, Moos and Moos' Family Environment Scale and Olson's Enriching and Nurturing Relationship Issues, Communication, and Happiness Scale (ENRICH). Stinnett and DeFrain, in their Family Strengths Inventory, also assess this dimension. In the Family Environment Scale, respondents are asked to respond true or false to such questions as: "Family members attend church, synagogue, or Sunday school fairly often," and "We don't believe in heaven and hell" (Moos and Moos, 1986).

Measures of Adaptability

A family's ability to adapt to stressful and potentially damaging events, as well as to predictable life-cycle changes, has been identified as an important characteristic of strong families. Researchers note that ability to adapt is contingent upon a number of other characteristics common to strong families such as effective communication, affective involvement, external resources, etc. (for example, Stinnett and DeFrain, Lewis, Olson, McCubbin). This section focuses on family characteristics not yet discussed that have been linked to successful adaptation.

Like cohesion, adaptability is a major dimension identified by Olson's Circumplex Model. Indeed, much of his research focuses on the ability or inability of different families to overcome stress and crises. This concept can be broken down into several individual constructs. Among these are two of the nine characteristics of strong families on which this paper is focusing. First, a family's social connectedness affects its ability to seek external assistance in the face of problems. In addition, the clear definition of roles (in particular those related to the power structure) within the family is an important factor in its ability to adapt to changes. These will be addressed following a broader discussion of adaptability.

Adaptability, as a central dimension of the Circumplex Model, has been defined as "the ability of a marital or family system to change its power structure, role relationships, and relationship rules in response to situational and developmental stress" (Olson et al., 1989, p. 12). It is hypothesized that well-functioning families have modes of adaptation that are structured or flexible rather than rigid or chaotic. Olson qualifies this hypothesis in two important ways, which apply to both the adaptability and cohesion dimensions. First of all, in measuring these constructs, family members are asked to rate their families on both perceived and ideal situations. Thus, a level of satisfaction with current family dynamics is obtained. It is hypothesized that if all family members are satisfied with the family as it is, even though it may be considered extreme on either cohesion or adaptability, then the family will function effectively (Olson et al., 1989).

Second, Olson proposes that families at different stages in the life cycle (childless, with infants, with elementary school children, with adolescents, etc.) require different levels of cohesion and adaptability. That is, a level of cohesion that is functional in a family with an infant may not be once the child is an adolescent (Olson et al., 1989).

Olson uses the FACES self-report scale to measure the level of adaptability in the family. This is made up of questions developed to assess the family power, negotiation style, role relationships and relationship rules within the family.

Measures of Social Connectedness

Successful families are not isolated; they are connected to the wider society. One effect of such a connection is the subsequent availability of external resources, identified by researchers as important to effective coping by families. An intervention study by Cowan and Cowan offers support for the importance of social support in overcoming a stressful period in family life. In their research, they identified couples who were childless but planning to have their first child soon. This time period, the birth of a first child, is often identified as a particularly stressful one for families. Intervention consisted of group meetings at which couples discussed problems and issues pertinent to the arrival of a first child, and the husband-wife relationship. The authors provide evidence, based on comparisons to control groups receiving no intervention, that this form of social support was helpful in the partner's transition to parenthood (Cowan and Cowan, 1987).

One way to measure social connectedness is to ask families to identify those external resources available to them. For example, McCubbin's self-report F-COPES (Family Coping Strategies) Scale assesses the extent to which a family turns to friends, family, neighbors, and community resources when they are facing a crisis (Olson et al., 1982). In their study of stable black families, Gary and colleagues assessed this dimension by asking questions about respondents' participation in organizations in the community, as well as the presence of relatives in the area, and neighbors who were considered close friends (Gary et al., 1983).

Measures of Clear Roles

Many researchers identify clear role definition as an important characteristic of family functioning, and as essential for a family's ability to adapt to changing situations. With a clear, yet flexible structure in place, family members are aware of their responsibilities in and to the family. Consequently, in the face of crisis and problems, members know their roles. The McMaster Family Assessment Device (Epstein, et al, 1983), a self-report instrument, has a subscale that measures roles in the family. The authors note the nature of this aspect of family functioning: "[it] focuses on whether the family has established patterns of behavior for handling a set of family functions ... In addition, assessment of the roles dimension includes consideration of whether tasks are clearly and equitably assigned to family members and whether tasks are carried out responsibly by family members" (Epstein, et al, 1983). An eight-item subscale provides a measure of this construct.

Related to the concept of role definition is the issue of power. Who decides what and how it is decided are indicators of the role structure in the family. Within healthy families, there is a clear recognition that the parents are in charge. At the same time, parents are open to their children's input -- the parents are rarely seen as authoritarian (Lewis, 1979, pp. 87-88).

Power and control have been measured by a number of techniques. In her study on adolescent's perception of power and cohesion in the family, Feldman uses a projective instrument to assess the power structure of the family. Subjects symbolically represent their family's power structure by elevating on blocks figures that represent individuals in the family. The relative height of the figures is taken to indicate the perceived ability of that member to influence other members (Feldman and Gehring, 1988).

In the Family Assessment Device described above, the nature of behavior control is assessed in terms of the "way a family expresses and maintains standards for the behavior of its members. Behavior in situations of different sorts is assessed as are different patterns of control" (Epstein, et al, 1983). This nine-item self-report subscale asks questions about family rules and expectations. A number of other self-report instruments have been designed with subscales that tap the power dynamics in the family, including Moos and Moos' Family Environment Scale.

Measures of Time Together

A final ingredient in family strength is spending time together. Researchers note that this includes quality as well as quantity of time. Self-report instruments assessing family functioning address this topic in terms of the quality of time spent together, and the extent to which families enjoy spending time together. Questions are asked about whether "family members like to spend their free time with each other" (Olson's FACES) and about "spending time together and doing things with each other" (Stinnett and DeFrain, Family Strengths Inventory). Detailed methodological work on this construct has been conducted by researchers interested in time use per se, which could be adapted for use in the study of successful families.

METHODOLOGICAL ISSUES

The following methodological limitations and issues must be considered in evaluating the study of family strengths and successful families:

The use of small and non-random samples.

Samples used in research on family strengths studies are often small and seldom representative of the national population or even of a known subpopulation. Studies on strong families by clinicians are also characterized by small, homogenous samples. For example, Beavers and Lewis studied 33 families, the majority of which were white, Protestant, middle- to upper-class, biologically intact families with an oldest child in adolescence (Beavers, 1977). Research using self-report surveys has a tendency to employ larger samples, but these are rarely representative. The majority of the 1,146 families who participated in the study by Olson and colleagues, for example, were Caucasian and Lutheran (Olson et al., 1982). Stinnett and DeFrain have analyzed data from more diverse samples, but participant families were mostly self-selected. Few studies have focused on black, Hispanic, or low-income families. The extent to which one can generalize the findings of family strength studies to the population as a whole is limited by the nature of the samples.

The applicability of findings-based on white, middle-class families to other groups.

A review of the family strengths literature reveals that much of the work in this field has focused on white, and/or middle-class families. This has obvious implications for the ability to generalize the findings to the population as a whole, as well as to subgroups such as minorities and low-income families. Studies of minority families that are relevant to successful family research range from those which focus on minority families and attempt to identify their particular strengths, to those in which minority families and/or low income families are compared with white and/or middle-class families across a variety of dimensions. In the latter case, researchers often use models, constructs, and instruments that were initially developed using white middle-class samples to study successful minority and low-income families.

Building on an initial study of 33 intact white, middle-class families in which Beavers and Lewis identified a number of characteristics of successful families, Lewis and Looney conducted a study that compared this sample with a sample of 18 working-class black families. Although they hypothesized that the differences in socioeconomic status would translate into differences in family characteristics, they found that there were few differences between the competent families in both samples. For the most part, the same traits that characterized the most competent white, middle-class families were characteristics of well-functioning, working-class black families (Lewis and Looney, 1983).

Similarly, Abbott and Meredith studied family strengths across five groups. Their sample consisted of 210 white parents, 105 Chicano parents, 103 black parents, 80 Hmong parents, and 57 American Indian parents. The two objectives of their study were to: 1) identify those characteristics considered critical to effective family functioning; and 2) assess the level of family strengths across the ethnic groups. They made use of family strengths inventories developed by Stinnett and DeFrain, Olson, Curran, and others. They found that parents across all five ethnic groups generally agreed on the traits of healthy families. However, they also found that the white and Hmong families achieved the highest scores on the assessment instruments, followed by blacks, Chicanos, and American Indians, respectively.

The researchers acknowledged that a weakness of the study is that the assessment instruments were created and standardized using white, middle-class samples, thus possibly affecting the validity of the scores for the minority families. In an attempt to compensate for this problem, the researchers conducted in-depth interviews with a subgroup of the sample. These interviews revealed that subjects from all five groups viewed the items on the scales as fair indicators of family strengths (Abbott & Meredith, 1985).

Lawrence Gary and his colleagues conducted an exploratory and descriptive study of 50 black families identified by community groups to be particularly strong and stable (Gary et al., 1983). They sought to identify the critical factors and conditions that contribute to strong black family life, as well as their coping strategies. They found that the families in their sample possessed many of the same characteristics identified previously by other researchers in the field, such as Hill, Scanzoni, Royce and Turner, and Stinnett. (See appendix A for a list of the characteristics identified by these researchers.)

Other researchers have explored differences between white families and black or Hispanic families across specific dimensions of family life. For example, Vega studied cohesion and adaptability in Hispanic and Anglo families using Olson's FACES instrument. In this study, he found that there were no important differences between the low-income Mexican American and middle-income Anglo parents in their perception of levels of family cohesion and adaptability (Vega et al., 1986).

In summary, an important methodological weakness of successful families and family strengths research has been the homogeneity of samples. While the studies described above have used more diverse samples, they have used instruments and measures that were created and standardized, for the most part, on white, middle-class families. The development of measures sensitive to various subpopulations seems an important step if successful family research is to become more relevant to the population as a whole.

Use of measures based on a single family members' perceptions.

In a discussion of research issues related to the study of family interaction, Christensen and Arrington provide a number of methodological comments relevant to family strengths research. One is that a family is composed of individuals interacting. The study of families, they argue, should therefore use techniques that assess interaction dynamics rather than individual characteristics. In much family interaction research, however, the unit of observation (the parent) is often different from the object of study (the family). That is, frequently, studies of families are based on perceptions of one parent. Some researchers have taken care to develop and assess measures capable of being completed by multiple family members (Olson's FACES, for example). The difficulty then becomes designing appropriate methods for combining the scores of a number of family members to arrive at a "family score".

Findings reported by several researchers that members of the same family often score quite differently on family assessment measures further supports the importance of obtaining measures from multiple family members (Christensen and Arrington, 1987). For example, Schumm, in his study of family satisfaction among Anglos and Mexican Americans, found that using just the scores from the mothers results in different conclusions than if the scores of the fathers and adolescents are considered as well (Schumm et al., 1988).

In addition to only assessing one family member's perspective, Christensen and Arrington point out that often times, the actual items on a self-report instrument designed to assess the family focus on the individual rather than on the interaction in the family. For example, a survey question might find out how often one member argued with another, but assess nothing about the circumstances that led to the argument, or how it was resolved. The latter are important descriptors in terms of understanding the nature of family interaction (Christensen and Arrington, 1987).

Even assessment techniques that observe family interaction are not immune to the problem of individual versus family. Christensen and Arrington note: "Observational analyses focus on rates of behavior for each individual present ... the data tell us nothing about who criticized whom, what each person did to elicit criticism, or what each person did in reaction to the criticism" (Christensen and Arrington, 1987, p. 262). They note as an exception the techniques of Gottman and Patterson, who examine, based on observation of family interactions, the "likelihood that a particular behavior by one family member will lead to or follow a behavior by another" (Christensen and Arrington, 1987, p. 262).

The effects of family life-cycle stages.

Family strengths research often, though not always, fails to recognize the impact of various life-cycle stages on the characteristics of strong families. Research by Olson and colleagues, for example, which does explore the effects of life cycle stage, has suggested that the dynamics most conducive to well-functioning families can vary depending on the structure of the household. Power structures and communication patterns that are effective for families with infants and toddlers may differ widely from families with adolescents and young adults. What is functional at one stage may be dysfunctional in another (Olson et al., 1989). Researchers studying family strengths must bear this in mind as they design studies and interpret findings.

Intercorrelations among family strengths characteristics.

The nine characteristics identified by researchers as important to successful family functioning are often highly correlated. For example, the ability to adapt is frequently associated with effective communication skills and connectedness to society, or clear role definitions may be impossible to achieve in families lacking effective communication skills. Schumm proposes a model that addresses the relationships among various characteristics of strong families.

Schumm argues that, "to date it appears that most work associated with the topic of family strengths has focused upon the identification and specification of concepts rather than attempts to interrelate concepts. For the family strengths literature to move in the direction of genuine theory construction as opposed to purely conceptual work, it is critical that a beginning be made in terms of interrelating concepts in the family strengths literature" (Schumm, 1985, p. 5). In light of this, Schumm proposes an initial model of how family strengths fit together. He identifies a number of ways in which some of the six traits identified by Stinnett are interrelated. For example, he links appreciation and commitment to spending time together. Schumm's model is diagrammed in Appendix B.

Schumm's model of family strengths has a number of implications for intervention. It suggests that when determining intervention techniques, the interrelatedness of family characteristics must be considered. As Schumm points out, focusing on one family dynamic may not be helpful until other interaction characteristics are addressed. It may be that a family's communication problems actually reflect an underlying lack of time together or lack of expression of appreciation among family members. Thus, a family's participation in a workshop to improve communication skills may not be helpful because the main problem lies elsewhere, and until that area is resolved, movement along the dimension of communication will be difficult (Schumm, 1985).

ISSUES RELEVANT TO APPLICATION OF FINDINGS

In addition to methodological issues, there are two broad questions that need to be answered about family strengths research before we can tell how much this body of research can contribute to the improvement of family functioning. They are:

Are family strengths causative, or merely symptomatic?

If the family strengths constructs were measured in a representative sample of families, along with other demographic, socioeconomic, and psychological characteristics of family members, would the constructs be predictive of whether the family would still be together and still be functioning harmoniously five or ten years in the future? How much predictive power would the constructs add, over and above that provided by measures of current marital satisfaction, marital conflict, education, income, and other family social assets? As far as we can tell, this question has never been empirically tested.

A related issue is whether traditional theoretical perspectives do not provide a more parsimonious and powerful understanding of family dynamics. None of the work uncovered to date compares the predictive power of successful family models with the power of models generated from other perspectives. It does not appear that the complementarity of models has yet been considered.

How transferable are "family strengths" to families that are not functioning well?

Even if it is demonstrated that the family strength characteristics are strong and unique predictors of successful functioning, there is the important question of whether these characteristics can be taught to those in families that are not functioning well. It may be that the family strengths depend on individual characteristics of family members such as intelligence, flexibility, emotional stability, etc., that have a substantial genetic component or depend on early upbringing. Teaching strengths to individuals who are unintelligent, rigid, hostile, or unstable might well prove to be a daunting undertaking. On the other hand, carefully constructed programs that introduce new strategies of living to families with problems might result in improved childrearing and family interaction at relatively low cost.

Obviously, the transferability issue is a significant one if the family strengths research is to have important policy applications. Again, as far as we can tell, this question has been little tested or even addressed in the existing literature. In addition, there are questions of transferability across class and ethnic lines, and applicability to single-parent and other non-traditional families, which could be profitably addressed in future studies.

REFERENCES

Abbott, Douglas A., & Meredith, William H. 1985. "Minority families: Strengths of four ethnic groups." Paper presented at the Annual Meeting of the National Council on Family Relations, Dallas, TX.

Beavers, W. Robert. 1977. Psychotherapy and growth: A family systems perspective. New York: Brunner/Mazel.

Bowman, Ted W. 1983. "Promoting family wellness: Implications and issues." In D. Mace (ed.) Prevention in family service approaches to family wellness, pp. 39-48. Newbury Park, CA: Sage Publications.

Christensen, Andrew, & Arrington, Angela. 1987. "Research issues and strategies." In T. Jacob (ed.) Family interaction and psychopathology. New York: Plenum Press.

Cowan, Carolyn P., & Cowan, Philip A. 1987. "A preventative intervention for couples becoming parents." In C. F. Zachariah Boukydis (ed.) Research on support for parents and infants in the postnatal period. Norwood, NJ: Ablex Publishing Co.

Epstein, Nathan B., Lawrence M. Baldwin, & Duane S. Bishop. 1983. "The McMaster Family Assessment Device." Journal of Marital and Family Therapy 9(2): 171-180.

Feldman, Shirley S., & Gehring, Thomas M. 1988. "Changing perceptions of family cohesion and power across adolescence." Child Development 59: 1034-1045.

Gary, Lawrence E., Lula A. Beatty, Greta L. Berry, et al. 1983. Stable black families: Final report. Washington, DC: Mental Health Research & Development Center, Institute for Urban Affairs & Research, Howard University.

Grotevant, Harold D. 1983. "The contribution of the family to the facilitation of identity formation in early adolescence." Journal of Early Adolescence 3(3): 225-237.

Hill, Robert B. 1971. The strength of black families. New York: Independent Publishers Group.

Jacob, Theodore (ed.). 1987. Family interaction and psychopathology. New York: Plenum Press.

Lewis, Jerry M. 1989. The birth of a family. New York: Brunner/Mazel.

Lewis, Jerry M. 1979. How's your family? A guide to identifying your family's strengths and weaknesses. New York: Brunner/Mazel.

Lewis, Jerry M., & John G. Looney. 1983. The long struggle:. Well-functioning working class black families. New York: Brunner/Mazel.

Moos, Rudolf H., & Moos, Bernice S. 1986. Family environment scale manual (Second Edition). Palo Alto, CA: Consulting Psychologists Press.

Olson, David H. 1986. Prepared statement before the House Select Committee on Children, Youth, and Families. In: The diversity and strength of American families. Washington, DC: US Government Printing Office.

Olson, David H., Hamilton I. McCubbin, Howard Barnes, et al. 1983. Families: What makes them work. Beverly Hills, CA: Sage Publications.

Olson, David H., Hamilton I. McCubbin, Howard Barnes, et al. 1982. Family inventories: Inventories used in a national survey of families across the family life cycle. St. Paul, MN: University of Minnesota.

Olson, David H., Candyce S. Russell, & Douglas H. Sprenkle (eds.). 1989. Circumplex model: Svstemic assessment and treatment of families. New York: Haworth Press.

Schumm, Walter R. 1986. Prepared statement before the House Select Committee on Children, Youth, and Families. In: The diversity and strength of American families. Washington, DC: US Government Printing Office.

Schumm, Walter R. 1985. "Beyond relationship characteristics of strong families: Constructing a model of family strengths." Family Perspective 19(l): 1-9.

Schumm, Walter R., McCollum, E. E., Bugaighis, M. A., Jurich, A. P., Bollman, S. R., & Reithz, J. 1988. “Differences between Anglo and Mexican American family members on satisfaction with family life." Hispanic Journal of Behavioral Sciences 10(l): 39-53.

Skinner, Harvey A., Steinhauser, Paul D., & Santa-Barbara, Jack. 1983. "The family assessment measure." Canadian Journal of Community Mental Health 2: 91-105.

Stinnett, Nick. 1986. Prepared statement before the House Select Committee on Children, Youth, and Families. In: The diversity and strength of American families. Washington, DC: US Government Printing Office.

Stinnett, Nick, Chesser, Barbara, & DeFrain, John (eds.). 1979. Building family strengths. Blueprints for action. Lincoln, NE: University of Nebraska Press.

Stinnett, Nick, & DeFrain, John. 1985. Secrets of strong families. Boston: Little, Brown & Co.

Swihart, Judson. 1988. "Characteristics of strong families." Unpublished paper, International Family Center, Logos Research Institute.

U.S. House of Representatives Select Committee on Children, Youth, and Families, 99th Congress, 2nd Session. February 25, 1986. The diversity and strength of American families. Washington, DC: US Government Printing Office.

Vega, William A., Thomas Patterson, James Sallis, et al. 1986. "Cohesion and adaptability in Mexican-American and Anglo families." Journal of Marriage and the Family 48(Nov): 857-967.

Walsh, Froma. Normal family processes. 1982. New York: Guilford Press.

APPENDIX A. The Characteristics of Strong Families

W. Robert Beavers (Psychotherapy and Growth: A Family Systems Perspective, 1977)

1. Connectedness with other social systems, and open to other viewpoints, lifestyles and perceptions. A respect for differences and awareness of individual boundaries. Intimacy is attained via skillful communication.
2. Solid parental coalitions and clear role definition without rigid stereotyping.
3. Complementary rather than symmetrical power roles.
4. An encouragement of autonomy. The family is comfortable with differences of opinion. There is an absence of invasiveness. The family has a degree of flexibility and adaptability.
5. The belief that human behavior is limited and finite and that human behavior is the result of a number of variables, not one clear-cut cause.
6. Family members are involved with each other. Conflict may exist between members, but not unresolvable conflict.
7. Effective negotiation and task performance.

8. Transcendent values.

A. Billingsley, 1968. (In Gary, Lawrence, Stable Black Families: A Final Report., 1983) (Black family strengths.)

1. A set of values -- with an accompanying pattern of behavior which is consistent with those values -- and a certain degree of independence from the control of the forces affecting the lives of family members.
2. Strong religious convictions and behaviors.
3. Educational achievement or educational aspirations of one or more family members.
4. Economic security and possession of property.
5. Strong family ties.

6. Community centered activities (i.e., associational ties, role models, and child advocacy).

Dolores Curran. (As cited in Hearing before the Select Committee on Children, Youth, and Families, House of Representatives, 1986)

1. Communicates and listens.
2. Affirms and supports one another.
3. Teaches respect for others.
4. Develops a sense of trust.
5. Has a sense of play and humor.
6. Exhibits a sense of shared responsibility.
7. Teaches a sense of right and wrong.
8. Has a strong sense of family in which rituals and traditions abound.
9. Has a balance of interaction among members.
10. Has a shared religious core.
11. Respects the privacy of one another.
12. Values service to others.
13. Fosters family table time and conversation.
14. Shares leisure time.

15. Admits to and seeks help with problems.

Nathan Epstein, and others. Based on the McMaster Model of Family Functioning. (in "The McMaster Family Assessment Device," Epstein, Baldwin, and Bishop, 1983)

1. An ability to solve problems in a way which maintains effective family functioning.
2. Communication that is clear and direct.
3. Established patterns of behavior, clear and equitable assignment of family tasks, and responsible completion of assigned tasks.
4. Affective responsiveness.
5. Family members are interested in each other and value each other's activities and concerns. Intermediate levels of involvement with each other.

6. Behavioral control of family members.

Robert B. Hill. (The Strengths of Black Families, 1971) (Black family strengths)

1. Strong kinship bonds.
2. Strong work orientation.
3. Adaptability of family roles.
4. Strong achievement orientation.

5. Strong religious orientation.

Jerry Lewis (How's Your Family? A Guide to Identifying Your Family's Strengths and Weaknesses, 1979)

1. Parental relationship characterized by shared power, deep levels of intimacy, intimate communication, strong bonds, shared as well as individual activities, good sexual relations, openness with feelings, and high levels of empathy.
2. Power in the family is firmly in the parent's hands, but includes input from the children. There are clear differences in the roles between parent and child.
3. Close-knit families who share a great deal with each other and feel strongly connected, but not at the expense of individuality.
4. Communication among family members is open, clear and spontaneous.
5. Early identification of problems, a lack of blaming individuals for the problem, and a negotiation and compromise approach to problem solving.
6. Expressive of feelings, and empathic responses, with a general family mood that is characterized by warmth, humor, and concern for each other.
7. The ability to accept and deal with change and loss.
8. A basic belief that people are mostly good, rather than mostly evil. A realization that humans make mistakes, and an understanding of the complexity of human motivation.

9. An encouragement of both intimacy and autonomy, as well as involvement of the family in the world around them.

David H. Olson. (in Prepared Statement before the Select Committee on Children, Youth and Families, U.S. House of Representatives, 1986)

1. Parents have a strong and happy marriage.
2. Members feel close to each other but also allow each other privacy and freedom to act independently.
3. Family members are flexible, creative as a group, and able to solve problems together.
4. Family members are able to listen and share both negative and positive feelings with each other.

5. Family members are able to cope effectively with stress.

H.A. Otto, 1975 (Otto, in Bowman, T.W., 1983)

1. The ability to provide for the physical, emotional, and spiritual needs of a family,
2. The ability to "give and take" in the area of child-rearing practices and discipline.
3. The ability to communicate effectively.
4. The ability to provide support, security, and encouragement.
5. The ability to initiate and maintain growth-producing relationships and experiences within and without the family.
6. The capacity to maintain and create constructive and responsible community relationships in the neighborhood, town, school, and so on.
7. The ability to grow with and through children.
8. An ability for self-help, and the ability to accept help when appropriate.
9. An ability to perform family functions and roles flexibly.
10. Mutual respect for the individuality of family members.
11. The ability to use a crisis or a seemingly injurious experience as a means of growth.

12. A concern for family unity, loyalty, and intra-family cooperation.

Royce and Turner, 1980. (In Gary, Lawrence, Stable Black Families: A Final Report., 1983) (Black family strengths.)

1. Validated those identified by Robert Hill.
2. Teaching children to respect themselves.
3. Teaching children how to be happy.
4. Stressing cooperation within the family.

5. Disciplining the children.

Virginia Satir. (Satir, in Bowman, T.W., 1983)

1. Promotion of positive self-worth.
2. Open communication system.
3. Clarity as to family rules and expectations

4. Link to the wider society -- commitment beyond the family.

Scanzoni, 1971. (In Gary, Lawrence, Stable Black Families: A Final Report. 1983) (Black family strengths.)

1. Economic and social advantages as well as employment opportunities provided by experiences in the city.
2. Social status obtained from religious involvement, parental presence in the home, and the number of children.
3. The ability to provide economic, social, and educational resources for children in the family.
4. Role models for children in the community (i.e., school teacher, minister, etc.)
5. The quality of the husband-wife relationship.

6. The level of interaction between parent and child.

Walter R. Schumm. (In collaboration with Judson Swihart, in Prepared Statement before the Select Committee on Children, Youth and Families, U.S. House of Representatives, 1986.)

1. Time. Time spent together in a variety of supportive, enjoyable activities as opposed to being so tied up with work and children as to have no pleasant times with family members.
2. Positiveness. Otherwise identified as appreciation in some models. Delivering a high level of positive reinforcement to family members, day in and day out, doing things that are positive from the other person's perspective, just for their sake, not merely as a strategy for "buying" their love, etc.
3. Commitment. Being committed to building a good marriage and family, not merely staying together in terms of not getting divorced and being willing to adapt positively to change.
4. Person esteem. Valuing oneself and other family members as worthy of lifelong commitments and one's support and appreciation.
5. Openness. Being open to one's own needs and wants and willing to share them openly, while also being open to hearing and truly understanding the needs and wants of other family members. This is more than just having "good communication skills" but includes the desire to know oneself and be known and to know others.

6. Value system. A value system that supports the other five areas, often provided within the context of an intrinsic religious faith (not merely a lip service to a set of doctrines) or absorbed as a child by having been raised in such an environment.

Harvey Skinner. ("The Family Assessment Measure," 1983)

1. Differentiation and performance of roles.
2. Communication that is clear, direct, and sufficient and relates information essential to task accomplishment and ongoing role definition.
3. Expression of affect.
4. Degree and quality of family members, interest in one another.

5. Control of family members by processes that are characterized as constructive, consistent, and responsible.

Nick Stinnett and John DeFrain. (Secrets of Strong Families, 1985)

1. Commitment.
2. Time together.
3. Appreciation.
4. Coping with crisis.
5. Spiritual wellness.

6. Communication.

Judson Swihart. (In an unpublished review of the literature, identifies the following as the characteristics of strong families most frequently mentioned by researchers.)

1. Communication
2. Encouragement of individual family members.
3. Commitment to family.
4. Religious orientation.
5. Social connectedness.
6. Ability to adapt.
7. Expressing appreciation.
8. Clear roles.

9. Time together.

APPENDIX B. Proposed Model of Family Strengths