Future Directions for FAST and Familiy Evaluation
Marianne Debry, Peter K. Smith and Thomas M. Gehring
 

 
  Family models represent attempts to establish links between theory, research and clinical practice. Such models usually employ cohesion, hierarchy and the flexibility of both these dimensions as key parameters in desribing the structure of family relationships. However, there are at present no generally recognised models which can provide a differentiated and empirically sound evaluation of the functionality of family systems in various contexts and at different developmental stages.

Systematic description of the structure of family relations is an important step in the creation and evaluation of health-related interventions strategies. It is noteworthy, that the structure of a family systems cannot be fully derived from the characteristics of its members just as it is not possible to describe in any detail the characteristics of family members solely by analysing their interactions. A suitable family evaluation preferably requires standardised test procedures which allow an analysis of the relationship between family members' individual and interpersonal constructs as well as their interations.

Because systematic analyses of interations are time-consuming and involve a great deal of committment of staff, observation methods are not standard in clinical practice. For economial reasons, questionnaires which provide subjective information from a family insider perspective are frequently used in clinical practice and research. However, these self-report methods in general focus either solely on the family as a whole or on individual dyads.

The use of spatial representations in clinical and research settings has many conceptual advantages. The first attempts at spatial representation of interpersonal structures were sociograms. Representation of family cohesion and hierarchy in clinical practice were introduced on the basis of "human sculptures", an approach which resemble psychodrama. Figure placement techniques such as the FAST clearly represent an advancement in family sculpture because they allow a higher degree of standardisation. This method can be used to elicit interpersonal patterns within whole family, as well as its subsystems in various situations either by one or several family members individually, as a family task or in any combination thereof. However, despite its great versatility, good psychometric properties, construct validity and a three-decade history, figure placement techniques have only recently gained recognition as promising clinical and resarch tools. The following sections provide some reflections about possible future directions for FAST and family evaluation.

Family evaluation: much remains to be done
Back to top of page
 

There is a delay in the field of family evaluation with regard to individual assessment. Snyder, Cavell, Heffer & Mangrum (1995) notice the broad range of assessment techniques addressed to individuals and dyads, in contrast to fewer measures at the extended family or community levels.

L'Abate & Snyder (chap. 3) emphasise the inadequacies of family assessment techniques. A large number of them are not theory-driven. Some present poor psychometric properties: standardisation and sampling, internal consistency and temporal stability, discriminant and construct validity do not meet the expected standards. Nevertheless, there are some well-known exceptions: for example, the tests of Olson, Beavers, Epstein or Moos (Lebow, chap. 4).

Almost all the existing measures are verbal. Most of them are self-report questionnaires, with all their inherent drawbacks: not suitable for young children and for people with limited language skills. They are more appropriate for research than for clinical practice. On the other hand, some measures are highly qualitative and emotional (e.g., sculptures): applying them needs experienced therapists (Gehring & Marti, chap. 1) and they are not compatible with the usual requirements of research.

L'Abate & Snyder (chap. 3) deplore that family evaluation is not yet in the mainstream of clinical practice. Family therapists are not prone to use standardised measures. In a survey of 598 marital and family therapists, Boughner, Hayes, Bubenzer & West (1994) found that only 33% of them regularly used assessment instruments. In family therapy, only 13% of the respondents reported using at least one instrument. All the measures were questionnaires. Many of them evaluated personality instead of dyads or systems. A large number of respondents (70%) held standardised assessment to be not at all or not very important. It is obvious that in the field of family therapy, practitioners fail to plan and to evaluate the treatment with the help of standardised assessments.

The status of research in family therapy still remains marginal (Gehring, Page & Marti, chap. 15). Research findings still have a modest impact upon clinical practice. But it must be admitted that few of them are clinically meaningful and provide relevant information for the therapeutic process (Pinsof & Wynne, 2000; Shields & Wynne, 1994).

Hence, there is a gap between family theory and evaluation, between assessment and therapy, and between researchers and clinicians. L'Abate & Snyder (chap. 3) and Gehring et al. (chap. 15) plead for a more comprehensive approach of family, integrating research and clinical practice. They advocate a more scientific approach of therapy, including systematic family assessment and appraisal of the treatment outcomes.

Family psychology can be defined as a science as well as a profession (Pinsof, 1992). Shields & Wynne (1994) claim that family therapy is at a turning point in its development. The discipline risks becoming marginalised in the field of health care if research on the process and efficacy of family therapy is not strengthened. This book is an attempt to institute a dialogue and a cooperation between researchers and therapists around a common technique, the FAST. It is designed to begin to address these deficiencies.

What family evaluation should be
Back to top of page
 

Family appraisal is necessarily multidimensional and highly complex. Snyder et al. (1995) suggest a multifaceted, multilevel approach.The following criteria have to be met:

  • Theory-driven and empirically grounded: constructs need to have theoretical relevance and techniques must give guarantees of excellent psychometric properties.
  • Multiple levels: the interactions of individuals, dyads, triads, nuclear family are worth being evaluated and articulated with extended and community systems.
  • Multiple dimensions: relevant dimensions like intimacy, power, communication, problem-solving, support, adaptability are to be appraised (Seywert, 1990).
  • Multiple methods: assessment strategies addressed to a wide range of ways of expression. The aim is to propose to family members many media to represent and to express themselves about their family: words, drawings, postures, figurines, so that everyone can find his own language.
  • Multiple contexts: the family must be observed across several settings.
  • Assessment leads to a product but is also an interactive process which learns as much as the result. Assessment techniques should allow comparisons between family members, between stages of life cycles, between different types of family configurations, across cultural backgrounds. It is important that they are as much culture-free as possible and free from stereotypes. They must be open to a large diversity of cases and settings.
The FAST, a breakthrough in family evaluation
Back to top of page
 

The FAST is theory driven: it is clearly in line with the theory of structural therapy (Minuchin) (Debry, chap. 2) and empirically grounded (Gehring & Marti, chap. 1). It assesses different dimensions like cohesion, hierarchy and flexibility, which have been proved to be relevant in family studies (Fontaine, 1985). These dimensions can be determined simultaneously but are independent (Gehring & Marti, chap. 1; Debry, chap. 2). The FAST also addresses multiple levels of the system: the whole nuclear system, parents, siblings and every dyad.

The FAST is a task: people are asked to do something, alone or together. It is also a figure placement test requiring visual, motor and verbal activities. It looks like a game and is easily understandable.

It is also a semiprojective technique. Participants are asked to represent family relations with a symbolism shared by them and the psychologist: distance on the square board for emotional proximity, relative height for hierarchy. Figures are identified as men and women. Coloured figures can be used in the way the respondents want.

Three representations are commonly used. The first (typical family representation) awakes the defenses and it is the representation where usually boundaries are the most clear. The second (ideal representation) is characterised by an increase of cohesion and a decrease or a status quo of hierarchy. The plots for both these representations can be influenced by desirability. In this respect the third (conflict representation) is perhaps the most interesting because it leads the participant to spread out the figures and show more extreme positions in the family (Gehring & Debry, 1995). We can thus compare the family representation in three different settings and appraise the flexibility of the representations through different contexts.

As a product the FAST provides quantitative and qualitative results. Cohesion and hierarchy scores can be calculated for the family as a unit as well as for its subsystems either based on an arithmetical or on a categorical procedure (Gehring & Marti, chap. 1). A relational structure is determined according to the degree of cohesion and hierarchy. The disturbances of boundaries are qualified: excessive intergenerational proximity by coalitions and deficient parental authority by hierarchy reversals.

As a process, the therapist can observe the interactions between the members of the family when they are making the joint representation. This representation provides a lot of information on the relational patterns that can or not be in accordance with the FAST plot. For Rigazio-DiGilio (chap. 14) the application of the FAST represents both an assessment tool and a therapeutic intervention.

The FAST, a technique in full development
Back to top of page
 

Since the publication of the FAST, clinicians and researchers have contributed to broaden its scope. The contributors of this book demonstrate how versatile and flexible it can be.

Some of them have created new instructions. In chapter 7 (Berdondini & Genta) strangers have been added: two hypothetical neighbours, one of which is a drug addict. Hunter & von Balmoos (chap. 9) asked for a representation of the mother's family of origin and a representation showing a triad. Smith, Myron-Wilson & Sutton (chap. 8) asked the child, after he or she had completed the FAST, to imagine how they thought their parents would represent the family. The parent was invited to imagine how their child had represented their family.

New scorings are proposed. Some spatial configurations are characterised [no separation of figures, on a continuous line, one or two corner(s)] by Berdondini & Genta (chap. 7); symbiosis/amalgamation for enmeshed figures by Meyer (chap. 6).

The research reported in this book cover many areas of psychology:

 
  • Developmental psychology: attachment (Morley-Williams & Cowie, chap. 5), theory of mind ability during middle childhood (Smith et al., chap. 8), preadolescence (Debry, chap. 2), puberty (Kim & Wongyannava, chap. 10), transition to adulthood (Mayseless & Scharf, chap. 11).
  • Family psychology: single parent families (Hunter & von Balmoos, chap. 9).
  • School psychology: bullying at school (Berdondini & Genta, chap. 7).
  • Work psychology: supervision and team development (Steinebach, chap. 16).
  • Cross-cultural psychology: comparison between Japanese and Swiss families (Ikeda & Hatta, chap. 12); three-generation Chinese families (Shu & Smith, chap. 13).
  • Family therapy: interactive assessment and treatment in family psychotherapy (Rigazio-DiGilio, chap. 14), parental interventions in child psychiatry (Gehring et al., chap.15).
  A promising development of the FAST lies in its application to other systems than the family. For instance, Steinebach (chap. 16) used it to investigate the therapeutic relationship. Individual and family outcomes can be measured, predicted and evaluated. The FAST can also portray the structure and dynamics of professional teams (Steinebach, chap. 16), with perhaps a revised scoring procedure and interpretation frame. The relation between behavioural patterns in teams and in family of origin is worth being studied by this means.

The samples used in the research presented in this book are more often clinical ones (7 samples with 20 to 30 participants, mean = 23) than large samples (4 large samples: from 102 to 265 participants, mean = 177). One reason for this could be that the FAST application is time consuming, being given on an individual basis to a participant or members of a family; although, less time consuming than many clinical measures, so large samples are feasible.

Of the twelve empirical studies reported in this book, five include young children under 7 years; five are with children between 8 and 12 years; three are with adolescents, and seven include adults. Children or adolescents alone do the FAST in five studies; children and mothers, in three studies; both parents in two studies; and grandparents in one study. One study included representations of team structures by health professionals.

Among the three types of representation it is the typical one that is the most required from the participants, and the conflict one the least. The typical representation is used in all the reported researches, the ideal representation in five of the twelve studies and the conflict representation, in only two of them. However a variant procedure (see new instructions, above) is used in three studies.

This overview demonstrates that the FAST can be used in many contexts of research, with different generations and cultural backgrounds. The standardised test instructions and scoring procedures can be modified according to specific needs and types of relational systems (various family constellations, therapeutic relationships, professional teams, groups).

Some limitations
Back to top of page
 

As Gehring & Marti (chap. 1) wrote, any test reduces the complexity of human systems to a few parameters, a simplification required by working models. The working models reflect the zeitgeist of the period. In the case of the FAST, its concepts originate with structural therapy, and some limitations are ascribable to this framework. For instance, it captures the family map at a given moment and does not shed light on the family history. The FAST should not be the only measure in a comprehensive family evaluation, it should be complemented by other, and by counseling with family members.

Nevertheless, the family history could easily be integrated with modification of the test instructions. For example, it is possible to reconstruct important stages of the family development by asking the family members about the relationships when some important life events occured (birth of one of the children, death of a grandparent). Future aspects of the family development can also be anticipated. For example, one can ask how the family looks like when the children will leave home (in five years).

The concept of hierarchy is not as clear as the concept cohesion. It encompasses many aspects such as: who makes the decision? who sets the limits? who has influence? The real meaning it has for the respondent is not always mastered and necessitates further research (Kahn & Meyer, 1999).

The material itself (square board, figures) presents limitations. Some positions are impossible because of the number of figures (Shu & Smith, chap. 13). There are more freedom, more position opportunities for the figures placed first than for those placed last. But geometric constraints force respondents to make decisions about who is closer to whom. Respondents very often do not want to differentiate between various degrees of closeness (for social desirability reasons), when they are not pushed to do it systematically.

Some limitations
Back to top of page
 

As Gehring & Marti (chap. 1) wrote, any test reduces the complexity of human systems to a few parameters, a simplification required by working models. The working models reflect the zeitgeist of the period. In the case of the FAST, its concepts originate with structural therapy, and some limitations are ascribable to this framework. For instance, it captures the family map at a given moment and does not shed light on the family history. The FAST should not be the only measure in a comprehensive family evaluation, it should be complemented by other, and by counseling with family members.

Nevertheless, the family history could easily be integrated with modification of the test instructions. For example, it is possible to reconstruct important stages of the family development by asking the family members about the relationships when some important life events occured (birth of one of the children, death of a grandparent). Future aspects of the family development can also be anticipated. For example, one can ask how the family looks like when the children will leave home (in five years).

The concept of hierarchy is not as clear as the concept cohesion. It encompasses many aspects such as: who makes the decision? who sets the limits? who has influence? The real meaning it has for the respondent is not always mastered and necessitates further research (Kahn & Meyer, 1999).

The material itself (square board, figures) presents limitations. Some positions are impossible because of the number of figures (Shu & Smith, chap. 13). There are more freedom, more position opportunities for the figures placed first than for those placed last. But geometric constraints force respondents to make decisions about who is closer to whom. Respondents very often do not want to differentiate between various degrees of closeness (for social desirability reasons), when they are not pushed to do it systematically.

Convergence and divergence
Back to top of page
 

Are family representations similar when realised by members from the same family? By respondents whose ages or cultures are different? The contributions to this book make some comparisons possible.

Different members of the same family
Back to top of page
 

Family representations tend to vary according to the members. Smith et al. (chap. 8) found 73% of the eligible measures to be different between an English child's representation of the family and his/her mother's representation. Chinese children represent more balanced relational structures and more coalitions than their parents and grandparents (Shu & Smith, chap. 13). In comparison with their parents, Israeli adolescents represent patterns with less cohesion (Mayseless & Scharf, chap. 11). English adolescent daughters reported less cohesion and more maternal power than did their mothers (Kim & Wongyannava, chap. 10).

Other studies report similar findings for children and parents. In single parent families, mothers and daughters both represent unbalanced relational structures with low hierarchy (Hunter & von Balmoos, chap. 9). In the same way, at the onset of the therapy, unbalanced structures and unclear generational boundaries were represented by parents, individually and jointly (Gehring et al., chap. 15).

Gehring & Marti report that there is no correlation between FAST representations of members from the same family. Nevertheless 2/3 of the fathers portrayed the same pattern as the family jointly (chap. 1). These findings suggest that family members perceive their relationships differently and that it is necessary to gather family plots made by each of them.

There is no unique answer to the question of similarity between family constructs of family members. Results of Western studies suggest that in families with severe psychological problems, all members, generally, do agree that their relationships are conflictual and thus unbalanced and rigid (no changes between typical and conflict representations). However the category unbalanced includes different patterns of cohesion and hierarchy.

In nonclinical families, the members have distinct views of their relationships, but they generally agree that the structure is balanced. Nevetherless, mothers have a less positive view than fathers who represent the family most often as balanced. The representations tend to be more balanced if performed jointly by the family as a group than individually by their members (except fathers). This can be explained as an effect of desirability.

Participants of different ages
Back to top of page
 

Completing the FAST seems easy and suitable for children. Hunter & von Balmoos (chap. 9) stress that the FAST looks, to children, like a play activity. Studies with boys and girls in middle childhood show that they understand the instructions. Nonclinical samples differentiate cohesion and hierarchy well. They represent clear generational boundaries (Debry, chap. 2). The boundary with the external environment seems permeable (Berdondini & Genta, chap. 7).
Nevertheless, the task does require some ability of decentration. One has to represent the relations from one's own viewpoint but also to take account of the relations between others. Some degree of social cognition must be attained (Smith et al., chap. 8). Morley-Williams & Cowie tried to use the FAST with preschool children (chap. 5). They concluded that the FAST lacked meaning and relevance for many of them. Children below six years may have some difficulties in understanding the instructions. The concept of hierarchy is often interpreted in terms of age or height.

With this proviso concerning preschool children below six years, the studies in this book and elsewhere demonstrate that the FAST is appropriate for different generations. The best example is Shu & Smith's research applying the FAST to respondents from three generations in China (chap. 13). It is relatively seldom that a technique can be used with so a wide a range of ages (and with illiterate persons some of the older participants in this study could not read). This generalisability over age gave Smith et al. (chap. 8) the idea of trying out the FAST as a theory of mind measure. They explored whether children and their mothers could have an insight in to what the otherís family representations might be.

Participants from different cultures
Back to top of page
 

Results from different continents are at our disposal and make comparisons possible between family structures in the United States and in Europe (Gehring & Marti, chap. 1) and between Western and Eastern cultures (Ikeda & Hatta, chap. 12; Shu & Smith, chap. 13).

There are some common trends. Cohesion is higher in the ideal representation and lower in the conflict representation (Gehring & Marti, chap. 1). Conflict representations generate more enmeshed boundaries (especially coalitions) than typical and ideal ones. The mother seems generally closer to the children than does the father (Mayseless & Scharf, chap. 11; Ikeda & Hatta, chap. 12) and is more frequently involved in coalitions (Debry, chap. 2).

However, some findings look quite different according to the culture. Eastern families seem to be more cohesive, with less clear generational boundaries. In Japan, the relationship between mothers and children is very close and egalitarian. Fathers are given more power than mothers. They are less integrated in the family. In three-generation Chinese families, hierarchy tends ëto run along lines of generation and gender (Shu & Smith, chap. 13). Paternal grandfathers received the most powerful status. Chinese parents and children seem closer than do spouses.

Japanese and Chinese cultures were grounded in Confucian philosophy. Respect and obedience to the eldest are required. Priority to the eldest and economic reasons lead to a pattern of patriarchal three-generation families. In China the One-child Policy and the three-generation family can lead sometimes to the 4-2-1-syndrome, a family where six adults (four grandparents and two parents) care for one child. Although these countries have different economic and political systems, although they underwent many changes in family life, they share this common historical background.

The issue of relational structure
Back to top of page
 
The FAST is underlined by two assumptions in line with the structural theory:
  • Parents are supposed to be closer to each other than to their children
  • Parents are supposed to have more power than their chidren.
If this is not the case, boundaries might be seen as problematic and families could be dysfunctional. Clinical families display unclear generations boundaries (cross-generational coalitions and hierarchy reversals). Balanced relational structures with moderate cohesion and moderate or strong hierarchy are expected to be optimal. Very high or very low cohesion or hierarchy lead to unbalanced structures, suggesting that these families might be dysfunctional.

One can wonder if balanced structures necessarily mean healthy family functioning. Conversely, particular boundaries (coalitions, hierarchy reversals) and unbalanced structures are more associated with pathology? The results presented in this book suggest that they are frequently found in nonclinical samples.
The high prevalence of unbalanced structures in nonclinical samples is associated with variables such as a period of development (puberty): 57% of unbalanced structures versus 9% of balanced ones were identified in a sample of nonclinical preadolescents (Debry, chap. 2).
Type of representation: in comparison with the typical FAST, the conflict FAST generates significantly more unbalanced relational structures, coalitions and hierarchy reversals. Unbalanced structures can be more reactional than structural.
Family configuration: girls from divorced families showed more unbalanced family structures than semi-orphans; single mothers portrayed their families as unbalanced: high cohesion, low hierarchy, hierarchy inversions (Hunter & von Balmoos, chap. 9).
cultural background: typical structures of Japanese adolescents are more unbalanced than those of Swiss ones. Low cohesion is due to the father's isolation, and low hierarchy due to weak maternal power (Ikeda & Hatta, chap. 12). In three generation Chinese families, Shu and Smith found only 16% of balanced structures (chap. 13). In this case high cohesion and low hierarchy (present in 78% of the plots) are associated. Cross- generational coalitions are present in 37% of the plots. In comparison with US children who portray high parental cohesion, Chinese children see more inter-generational closeness. In Israel, none of the family members depicted the parental dyad as closer than the parent-son dyads. The bond of adolescent with mother is perceived as stronger than with father, at least from the mother's and adolescent's perspective (Mayseless & Scharf, chap. 11).
These findings suggest that extreme scores in cohesion and hierarchy can be associated in some contexts with wellbeing. High cohesion and low hierarchy are especially frequent in family representations of Western and Eastern nonclinical samples. Cross-generational coalitions and hierarchy reversals can be understood as close and egalitarian relations, frequently between mother and child, and are not necessarily to be considered as dysfunctional patterns.

The link between relational structures and normality for clinical and nonclinical families in Eastern and Western cultures is very complex. The literature shows that families with psychological problems are more likely to show unclear generational boundaries. However, it remains to study what it does mean, when dysfunctional families show clear boundaries? Or, when healthy family members show unbalanced patterns? Furthermore, we need to find out wether the FAST shows also clinical discriminant validity for Asian families. How do stressed and nonstressed families differ in these cultures?

Future prospects
Back to top of page
 

During recent decades, great changes have occured in family configurations, in sexual roles, and in rearing practices. Family psychology and therapy have to take this paradigmatic revolution into account. Family evaluation proved to be an important field for research and for clinical practice, a field which deserves to be developed on a larger extent. Some challenges are to be taken up.

The challenge of relational diagnosis
Back to top of page
 

For a long time clinicians saw the origin of psychological disorders exclusively within the individual, in his or her genes, mind or psyche. Diagnosis and treatment were focused on the individual. Since the 1960s, the systemic movement has brought the idea that systems can be dysfunctional and produce symptoms both in persons and in their relationships. The analysis of the interpersonal and contextual factors lead to a relational diagnosis complementary to the appraisal of an individual's personality (Lebow, chap. 4).

Individual disorders are not correlated with specific relational dysfunctions. We have to come to terms with the lack of this kind of linear causality. Gehring and Marti, for instance, found that the type of children's disorders is unrelated to the portrayal of family structures through the FAST representations of parents, patients and their siblings (chap. 1). As Denton (1995: 35) puts it:
The ultimate problem in such reconciliation between relational and individual disorders is that they describe different and unique levels of organisation of nature. There exist interactions between these levels, but the levels also function independently to each other. We should continue to try to understand these connective links but appreciate at the same time that it is unlikely that we will find simple, linear linkages between what goes on within and between people.

The challenge of interactive assessment and therapy
Back to top of page
 

Historically, the therapist was placed in the position of expert. Following the medical model, he or she labelled the disorders and led a passive patient towards healing. Current therapy models are predicated on more collaborative methods. Figure placement techniques like the FAST can be an assessment tool which can be interpreted by both clinicians and clients. These types of interactive assessments, geared toward social constructivism, become the vehicle for sharing power in the therapy sessions by stimulating collaboration, reflexivity, and a multiplicity of ideas and possibilities (Rigazio-DiGilio, chap. 14).

Gehring, and Rigazio-DiGilio, have proposed a model for participative-discursive intervention, seen as a solution-focused dialogue. The aim is to empower individuals and systems. Family members (parents) define the problem, explain their family constructs and their attempted changes. The clinical procedure includes three steps: planning, problem-solving and evaluation. A transparent conceptualisation of the clinical procedure is likely to succeed in turning parents into partners for empirically-derived semi-standardised treatmentí (Gehring et al., chap. 15).

The challenge of family's diversity
Back to top of page
 

The meta-frameworks perspective (Breunlin et al., 1992) emphasises the importance of culture, gender and development in family psychology. There are so many ways to live in a family, depending on the context of time and space. Family measures should cope with this paradox: to consider normative aspects and to address the singularity of each family system (Gehring & Marti, chap. 1).

The worst thing to wish the FAST for the future is for it to become a test which indicates what a normal family is, a kind of universal model of sane family relationships. It is much better suited for describing family relationships in their detail and in their variety, for helping families to put their intimate experience of family life into words. For this reason, relational structures and specific boundaries might be seen more as dimensions than discrete categories.

Being more descriptive than prescriptive, the FAST can fit the extreme diversity of current family configurations. This does not preclude statistical analyses on larger samples, describing this diversity. Further research is needed on, for example: nuclear and extended families; families in transition (single parent families, divorced, blended families); adoption; trans-generational patterns; and cross-social cross-cultural comparisons. Future research using longitudinal designs can provide more insight into the meaning of structural patterns and their transformation across time.

The concepts of structure and boundary must remain independent of pathology, as recommended by Minuchin. Unbalanced structures or disturbed boundaries are not necessarily synonymous of disorders, even if dysfunctional families are more likely to show them.

Family issues are to be considered more in relative terms than in absolute ones. The point at which family patterns become pathogenic depends heavily on context (cultural background, family style, community support system), on timing (when they occur in the family cycle), and how long transitional pattern last (Wood & Talmon, 1983: 356).

This suggests the relevance of contextual and qualitative information, the interest of longitudinal studies capturing family development. For family clinicians and for researchers alike, the way towards the future should necessarily be paved with relativism and complexity.

References
Back to top of page
 

Boughner, S.R., Hayes, S.F., Bubenzer, D.L. & West, J.D. (1994). Use of standardized assessment instruments by marital and family therapists: a survey, Journal of Marital and Family Therapy 20: 69-75.

Breunlin, D., Schwartz, R. & Mackune-Karrer, B. (1992). Meta-frameworks: transcending the models of family therapy. San Francisco: Jossey-Bass.

Denton, W. H. (1995) Problems encountered in reconciling individual and relational diagnoses, in F. W. Kaslow Handbook of relational diagnosis and dysfunctional family patterns. New York: Wiley, pp. 35-45.

Fontaine, P. (1985). Familles saines 1. Esquisse conceptuelle générale, Thérapie Familiale 6: 267-282.

Gehring, T.M. & Debry M. (1995). L'Evaluation du système familial: Le FAST, ATM: Braine-le-Château.

Kahn, J.S. & Meier, S.T. (1999). Level of measurement and the Family System Test: The relationship between participants' measurement constructs and interpretation's score, Constructivism in the Human Sciences 4: 103-115.

Pinsof, W.M. (1992). Toward a scientific paradigm for family psychology: the integrative process systems perspective, Journal of Family Psychology 3-4: 432-447.

Pinsof, W.M. & Wynne, L.C. (2000). Toward progress research: Closing the gap between family therapy practice and resaerch, Journal of Marital and Family Therapy 26: 1-8.

Seywert, F. (1990). L'Evaluation systèmique de la famille, Paris: PUF.

Shields, C.J. & Wynne, L.C. (1994). The marginalisation of family therapy: A historical and continuing problem, Journal of Marital and Family Therapy 20: 117-138.

Snyder, D.K., Cavell, T.A., Heffer, R.W. & Mangrum, L.F. (1995). Marital and family assessment: A multifaceted, multilevel approach, in R.H. Mikesell, D.D. Lusterman & S.H. McDaniel (eds), Integrating family therapy: Handbook of family psychology and systems theory. Washington, DC: American Psychological Association.

Wood, B. & Talmon, M. (1983). Family boundaries in transition: A search for alternatives, Family Process 22: 347-357.

Home