Part I
Understanding family structures: Theory, assessment and methodology
1 Concept and psychometric properties of the FAST
Thomas M. Gehring and Daniel Marti
Introduction
The consideration of the family is a crucial addition to the well-recognized notions of individual and environmental risk factors and resources of human development. Undoubtedly, there is still a great need for further development of comprehensive concepts for family diagnosis as it applies to health promotion and to the prevention and therapy of biopsychosocial problems. Suitable family assessment calls for a particular set of methodological techniques that allows the integration of individual, interpersonal and contextual parameters into a systemic perspective based on our current knowledge of the organization of family structures.
Any test reduces the complexity of human systems to a few parameters, a simplification required by working models. Moreover, the structure of a family system cannot be fully derived from the reports of its individual members, just as it is not possible to determine the characteristics of the family members solely by analysing their interactions. According to Jameson and Alexander (1994), family measures should consider normative aspects and, at the same time, be appropriate for each case in its particular circumstances. Comprehensive family evaluation, therefore, requires standardized test procedures that allow a quantitative and qualitative analysis of interpersonal constructs as well as of the interactional patterns of parents and children in various settings (L'Abate, 1994; Wilkinson, 1998; Werner-Wilson et al., 1999).
This chapter describes the fundamental concept of the Family System Test FAST (Gehring, 1998, 2000), a clinically derived figure placement technique for the assessment of family cohesion and hierarchy structures in individual and joint settings. First, important aspects of family psychology are summarized, with emphasis being placed on well-adjusted and problematic relational structures. Second, psychometric properties of the FAST and its use to evaluate family functioning in research and clinical practice are described.
Describing family structures
Systemic approaches provide a comprehensive framework for describing the relational structure of families on different levels such as dyads, triads or the family as a unit (Ackerman, 1985; Minuchin, 1985; Kaslow, 1996). Families are characterized by features such as intimacy, generational differentiation and continuity. They are organized systems that are conditioned by their environment, which, in turn, receives inputs from these systems. These interdependent changes always occur in relation to time. Therefore, the interpretation of respective processes should include the past and the present, as well as the future.
Cohesion and hierarchy
It is well known that cohesion and hierarchy are two fundamental dimensions that describe the interpersonal structure of family systems. Cohesion is generally defined as emotional bonding or attachment between family members (Bowen, 1960; Stierlin, 1974). The structure of cohesion includes the regulation of closeness and distance between family members and their respect of personal privacy. In reference to family systems, cohesion is defined as the extent to which the family members are organized as a coherent whole. Hierarchy covers several theoretical assumptions and cannot be attached to a single definition (Kranichfeld, 1987). It can be referred to authority, decision-making power or the amount of influence exercised by one family member over another. The concept of hierarchy has also been used to study the flexibility of roles and rules within the family (Olson, 1986).
Family boundaries
One of the major issues for the description of family structures is the concept of boundaries, and this is pertinent to the understanding of the family as an organized system. The construct of boundaries is used to describe relations between families and their social environment, as well as relations between various subsystems within the family (e.g. individuals, dyads or triads). Family boundaries can be defined by the rules that determine who belongs to a given system or subsystem (Minuchin, 1974). External family boundaries manifest themselves by the fact that family members behave differently towards each other than they do towards people outside their community of life. Internal family boundaries are marked by differences in the behaviour among members of distinct subsystems. Studies of dyads and triads, including parent, parentāchild and sibling relations, demonstrate that interactions among family members and the way relations are perceived differ according to type and size of the subsystem (Rabinowitz and Eldan, 1984). The construct of generational boundaries refers to the fact that patterns of cohesion and hierarchy are influenced by whether a subsystem consists of members of the same or different generations (Beavers, 1985; Wood, 1985).
Family stress and interpersonal structures
Research, including various methodological approaches, has shown that well-adjusted and troubled families display different relational structures (Lebow and Gurman, 1995). Using self-report or observation methods, it has been demonstrated that members of non-clinical families are emotionally close to one another and that the relationships between the generations are neither egalitarian nor very hierarchical. The structure of such family relations is flexible in response to situational and developmental demands (Minuchin, 1985). Furthermore, depending on their age or specific roles, family members perceive their relations differently and are able to express their views coherently, as well as to work towards a consensus based on common goals (Oliveri and Reiss, 1982; Feldman and Gehring, 1988).
In contrast, troubled families often display centrifugal patterns. Their ties are disengaged, with few reciprocal relationships and an atmosphere of remoteness, which may lead to the expulsion of a family member (Minuchin and Fishman, 1981; Beavers and Voeller, 1983). In addition it has been reported that their structures are likely to be either rigid or chaotic (Green et at., 1985; Selvini-Palazzoli, 1986). Such patterns are reflected in the difficulties that family members have in communicating their needs adequately and this, in turn, limits healthy individual and family development. Research that determined types of family structures on the basis of both cohesion and hierarchy dimensions showed that while cohesion is generally low in distressed families, the dimension of hierarchy tends to be either high or low (Friedman et al., 1987; Preli and Protinsky, 1988; Anderson and Gavazzi, 1990; Green et al., 1991). Therefore, on the basis of studies comparing distressed and non-distressed samples, it can be concluded that there is evidence for a linear relation between cohesion and family functioning, while no such relation exists with hierarchy.
Structural family theory predicts that in well-functioning families the rules governing interactions between parents differ from those in parentāchild subsystems; a fact that is reflected by clear cross-generational boundaries. In other words, fatherāmother relationships generally display a higher degree of cohesion than parentāoffspring relationships. There are also clear generational boundaries regarding hierarchy, in that parents have a relatively large voice in decision-making because of their experience and material resources. Distressed families, in contrast, display unclear generational boundaries (Haley, 1973; Wood and Talmon, 1983). This becomes manifest in cross-generational coalitions where cohesion between a parent and child is stronger than between the parents. Moreover, there are hierarchy reversals, where the power or influence of a child exceeds that of the parents.
Spatial representations of family relationships
The first attempts at spatial representation of interpersonal structures were sociograms that quantified and visualized emotional bonding and hierarchical structures in social groups (Chapin, 1953). Spatial representations of family relations in clinical practice were introduced on the basis of āhuman sculpturesā (Satir, 1967; Simon, 1972; Duhl et al., 1974). These techniques, however, have the potential to trigger strong emotional reactions in patients with serious psychological disorders and can therefore be used only by experienced therapists. Further limitations lie in their inability to quantify relational patterns or to allow independent studies of individual perceptions of family relations. Figure placement techniques involve figures that respondents arrange on a board to depict dyadic, triadic or larger subgroupings in order to operationalize their perceptions of emotional bonding. These techniques represent an advance because they can be used with parents and children in both individual and interactive settings, and therefore provide a standardized evaluation of the family membersā interpersonal constructs as well as of their interaction.
The use of figure placement techniques in clinical practice and research has many advantages. These techniques are very versatile and fulfil the basic requirements of comprehensive family evaluation (Rigazio-DiGilio, 1993). For example, they can be applied with preschool children because they do not require reading or writing skills. As representations of family constructs are language independent, these tools are also well suited for cross-cultural research. Furthermore, they are simple to administer and the representations are easy to analyse both quantitatively (e.g. types of family structure) and qualitatively (e.g. follow-up interview). The family can be portrayed under various conditions (e.g. in a conflict situation), and family members unavailable for the study (e.g. grandparents) or significant persons outside the family (e.g. teachers) can also be included in the portrayals. An individual representation of the family takes only 5 minutes, or between 10 and 30 minutes when carried out in a group setting. This economic method elicits the family membersā perceptions of structural patterns within the family, including all of its subsystems. When used as an interactive task, figure placement techniques provide information derived from standardized observation about how family members negotiate when attempting to reach a consensus.
Using figure placement techniques, scoring of cohesion is based on assessment of distances between figures. The construct of hierarchy is operationalized in different ways. For example, it has been measured by comparing the extent to which individual representations of cohesion agreed with the representation produced by the same family members in a group setting (Russell, 1980) or, in the case of figure schemata, by interpreting a horizontal line-up to indicate egalitarian relations and a vertical line-up to represent extremely hierarchical relations (Madanes et al., 1980). Numerous research projects investigating cohesion in the family as a unit and also as distinct dyads yielded results that were convergent with structural theory. It has been shown that distances between figures vary as a function of ...