Journal of Family Therapy

JOFT 22.
2 - Main Text 14/4/0 10:59 am Page 144
 The Association for Family Therapy 2000. Published by Blackwell Publishers, 108 Cowley
Road, Oxford, OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
Journal of Family Therapy (2000) 22: 144–167
0163–4445
Circumplex Model of Marital and Family Systems
David H. Olsona
The Circumplex Model focuses on the three central dimensions of marital

and family systems: cohesion, flexibility and communication. The major
hypothesis of the Circumplex Model is that balanced couple and family
systems tend to be more functional compared to unbalanced systems. In
over 250 studies using the Family Adaptability and Cohesion Scales
(FACES), a linear self-report measure, strong support has been found for
this hypothesis. In several studies using the Clinical Rating Scale (CRS), a
curvilinear observational measure, the hypothesis was also supported.
These two assessment tools, the FACES and the CRS, are designed for
research, clinical assessment and treatment planning with couples and
families.
Overview
The Circumplex Model of Marital and Family Systems was devel-
oped in an attempt to bridge the gap that typically exists between
research, theory and practice (Olson et al., 1989). The Circumplex
Model is particularly useful for ‘relational diagnosis’ because it is
system-focused and integrates three dimensions that have repeat-
edly been considered highly relevant in a variety of family theory
models and family therapy approaches (see Table 1). The model,
and the assessment instruments which have been developed from it,
is specifically designed for clinical assessment, treatment planning
and research on outcome effectiveness of marital and family ther-
apy (Olson, 1993, 1996).
Family cohesion, flexibility and communication are the three
dimensions in the Circumplex Model. These three dimensions
emerged from a conceptual clustering of over fifty concepts devel-
oped to describe marital and family dynamics. Although some of
these concepts have been used for decades (for instance, power and
a Family Social Science, University of Minnesota, 290 McNeal Hall, St Paul,

Minnesota. Mailing Address: Life Innovations, PO Box 190, Minneapolis,
Minnesota 55440, USA.
 2000 The Association for Family Therapy and Systemic Practice

JOFT 22.2 - Main Text 14/4/0 10:59 am Page 145
14676427, 2000, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-6427.00144 by Tunisia Hinari NPL, Wiley Online Library on [22/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Circumplex Model 145
roles), many of the concepts have been developed by family ther-
apists observing problem families from a general systems perspec-
tive.
A variety of other family models has focused independently on
variables related to the cohesion, flexibility and communication
dimensions. Table 1 summarizes the ideas of a number of family
theorists who have worked on describing marital and family systems.
Most of these models have been developed in the past twenty-five
years by individuals who utilize a family systems perspective. The
value and importance of these three dimensions is underlined by
the fact that these theorists have independently concluded that
these dimensions were critical for understanding and treating mari-
tal and family systems.
Marital and family cohesion (togetherness)

Family cohesion is defined as the emotional bonding that family members
have towards one another. Within the Circumplex Model, some of the
specific concepts or variables that can be used to diagnose and
measure the family cohesion dimensions are emotional bonding,
boundaries, coalitions, time, space, friends, decision-making, inter-
ests and recreation. The focus of the construct of cohesion is how
family systems balance the separateness of their members versus
togetherness.
There are four levels of cohesion ranging from disengaged (very
low) to separated (low to moderate) to connected (moderate to high)
to enmeshed (very high) (See Figure 1). It is hypothesized that the
central or balanced levels of cohesion (separated and connected)
make for optimal family functioning. The extremes or unbalanced
levels (disengaged or enmeshed) are generally seen as problematic
for relationships over the long term.
In the model’s balanced area of cohesion (separated and
connected), individuals are able to experience and balance these
two extremes and are also able to be both independent from and
connected to their families. Many couples and families that go for
therapy often fall into one of the extremes or unbalanced areas.
When cohesion levels are very high (enmeshed systems), there is
too much consensus within the family and too little independence.
At the other extreme (disengaged systems), family members do
their own thing, with limited attachment or commitment to their
family.

JOFT 22.2
- Main Text
TABLE 1 Theoretical models using cohesion, flexibility and communication
14/4/0 10:59 am
Cohesion Flexibility Communication
Beavers and Hampton (1990) Stylistic dimension Adaptability Affect

Benjamin (1977) Affiliation Interdependence
Epstein et al. (1983) Affective Behaviour control Communication
involvement Problem-solving Affective responsiveness
Page 146
French and Guidera (1974) Capacity to
change power
Gottman (1994) Validation Contrasting
Kantor and Lehr (1975) Affect Power
Leary (1975) Affection Dominance
Hostility Submission
Leff and Vaughn (1985) Distance Problem-solving
Parsons and Bales (1955) Expressive role Instrumental role
Reiss (1981) Co-ordination Closure
Balanced couple and family systems (separated and connected
types) tend to be more functional across the life-cycle. More specif-
ically, a separated relationship has some emotional separateness, but it
is not as extreme as the disengaged system. While time apart is more
important, there is some time together, some joint decision-making
and marital support. Activities and interests are generally separate
but a few are shared. A connected relationship has emotional closeness
and loyalty to the relationship. Time together is more important
than time alone. There is an emphasis on togetherness. There are
separate friends, but also friends shared by the couple. Shared inter-
ests are common with some separate activities.
Unbalanced levels of cohesion are at the extremes either very low
(disengaged) or very high (enmeshed). A disengaged relationship
often has extreme emotional separateness. There is little involve-
ment among family members and there is a great deal of personal
separateness and independence. Individuals often do their own
thing, separate time, space and interests predominate, and
members are unable to turn to one another for support and prob-
lem-solving. In an enmeshed relationship there is an extreme amount
of emotional closeness, and loyalty is demanded. Individuals are
very dependent on each other and reactive to one another. There
is a lack of personal separateness and little private space is permit-
ted. The energy of the individuals is mainly focused inside the
family and there are few outside individual friends or interests.
Based on the Circumplex Model, very high levels of cohesion
(enmeshed) and very low levels of cohesion (disengaged) tend to
be problematic for individuals and relationship development in the
long run. On the other hand, relationships having moderate levels
of cohesion (separated and connected) are able to balance being
alone versus being together in a more functional way. Although
there is no absolute best level for any relationship, many will have
problems if they function at either extreme of the model (disen-
gaged and enmeshed) for too long.
Marital and family flexibility

Family flexibility is the amount of change in its leadership, role relation-
ships and relationship rules. The specific concepts include leadership
(control, discipline), negotiation styles, role relationships and rela-
tionship rules. The focus of flexibility is on how systems balance
stability versus change.

148 David H. Olson
Figure 1 Circumplex Model: Couple and family map
The four levels of flexibility range from rigid (very low) to struc-
tured (low to moderate) to flexible (moderate to high) to chaotic (very
high) (See Figure 1). As with cohesion, it is hypothesized that
central or balanced levels of flexibility (structured and flexible) are
more conducive to good marital and family functioning, with the
extremes (rigid and chaotic) being the most problematic for fam-
ilies as they move through their life-cycle.
Basically, flexibility focuses on the change in a family’s leadership,
roles and rules. Much of the early application of systems theory to
families emphasized the rigidity of the family and its tendency to
maintain the status quo. Until the work of recent theorists, the

importance of potential for change was minimized. Couples and
families need both stability and change, and the ability to change
when appropriate distinguishes functional couples and families
from dysfunctional ones.
Balanced couple and family systems (structured and flexible types)
tend to be more functional over time. A structured relationship tends to
have a somewhat democratic leadership with some negotiations
including the children. Roles are stable with some sharing of roles.
There are few rule changes with rules firmly enforced. A flexible rela-
tionship is characterized by egalitarian leadership and a democratic
approach to decision-making. Negotiations are open and actively
include the children. Roles are shared and there is fluid change when
necessary. Rules can be changed and are age-appropriate.
Unbalanced marriages and families tend to be either rigid or
chaotic. A rigid relationship is where one individual is in charge and
is highly controlling. There tend to be limited negotiations with
most decisions imposed by the leader. The roles are strictly defined
and the rules do not change. A chaotic relationship is characterized
by erratic or limited leadership. Decisions are impulsive and not
well thought out. Roles are unclear and often shift from individual
to individual.
Based on the Circumplex Model, very high levels of flexibility
(chaotic) and very low levels of flexibility (rigid) tend to be prob-
lematic for individuals and relationship development in the long
run. On the other hand, relationships having moderate levels of
flexibility (structured and flexible) are able to balance change and
stability in a more functional way. Although there is no absolute best
level for any relationship, many relationships tend to have problems
if they always function at either extreme of the model (rigid and
chaotic) for an extended period of time.
Marital and family communication

Communication is the third dimension in the Circumplex Model and
is considered a facilitating dimension. Communication is considered
critical for facilitating movement on the other two dimensions.
Because it is a facilitating dimension, communication is not graphi-
cally included in the model along with cohesion and flexibility.
Couple and family communication is measured by focusing on
the family as a group with regard to their listening skills, speaking
skills, self-disclosure, clarity, continuity tracking, and respect and

150 David H. Olson
regard. In terms of listening skills, the focus is on empathy and atten-
tive listening. Speaking skills include speaking for oneself and not
speaking for others. Self-disclosure relates to sharing feelings about
self and the relationship. Tracking is staying on topic, and respect
and regard relate to the affective aspects of the communication and
problem-solving skills in couples and families, and it has been found
that balanced systems tend to have very good communication,
whereas unbalanced systems tend to have poor communication.
Three-dimensional (3-D) Circumplex Model and linear scoring

Both self-report instruments and rating scales were developed to
assess the status of couples and families on the dimensions of the
Cirucumplex Model and to test hypotheses derived from the model.
These assessment instruments are described later in the paper. An
interesting development arose from studies in which self-report
assessment instruments (the FACES II and FACES III) were used to
evaluate the cohesion and adaptability dimensions of the
Circumplex Model. These studies demonstrated that in FACES II
and III cohesion and flexibility are linear rather than curvilinear
dimensions (Olson, 1991). Because FACES assesses cohesion and
flexibility in a linear manner, it was necessary to develop a concep-
tual version of the Circumplex Model to capture the balanced
versus unbalanced aspects of cohesion and flexibility.
The result was a three-dimensional version of the Circumplex
Model (Olson, 1991) (Figure 2) which enables the findings from all
the FACES studies to be better integrated into the model. High
scores on FACES cohesion and flexibility scales are indicative of
balanced systems, while low scores on FACES reflect unbalanced
systems.
One advantage of this 3-D Model is that it is conceptually and
empirically more related to other family models such as the Beavers
System Model (Beavers and Hampson, 1990) and the McMaster
Family Model (Epstein et al., 1993). This is also demonstrated
statistically in that FACES III is highly correlated in a positive
manner to the Self-report Family Inventory (SFI) (Beavers and
Hampson, 1990), the Family Assessment Measure (FAM III)
(Skinner, 1983) and the McMaster Family Assessment Device (FAD)
(Epstein, et al., 1993).
Another advantage of the three-dimensional design is that it
incorporates first- and second-order change more effectively into

Figure 2 Three-dimensional family Circumplex Model
the Circumplex Model (Olson, 1991), an idea suggested by Lee

(1988).
First-order change is curvilinear in that too much or too little change
is problematic and it relates directly to the flexibility dimension.

152 David H. Olson
More specifically, either too much change (i.e. chaotic system) or
too little change (i.e. rigid system) is related to a less functional
pattern in families. In contrast, the two balanced types of flexibility
are called structured and flexible because they represent more
balanced levels of change.
Second-order change is change from one system type to another
system type. It is ‘change of the system itself’ and can only be assessed
over time. Under stress, patterns of change become more apparent.
Second-order change can occur in times of normative stress, such as
the birth of a child, or non-normative change, such as when a parent
is injured in a car accident. Second-order change is linear with higher
change in the balanced systems and the lowest level of change in
unbalanced systems. In times of stress, balanced systems will tend to
change to another system type to adapt, while unbalanced systems
tend to stay stuck in their extreme pattern, which can often create
more stress. Second-order change in this model is thereby similar to
Beavers’ concept of competence (Beavers and Hampson, 1990).
The three-dimensional model also has the advantage of demon-
strating more clearly the dynamic similarity within the balanced and
unbalanced types. This model more clearly illustrates that the four
balanced types are more similar to each other dynamically in terms of
second-order change than they are to any of the unbalanced types.
Conversely, the four unbalanced types are similar to each other
dynamically in that they are all low in second-order change. This clari-
fies the dynamic similarities within balanced or unbalanced types that
are often lost when looking at the Circumplex Model when it is laid
out in the two-dimensional (four levels x four levels) design.
Hypotheses derived from the Circumplex Model

The most basic hypothesis derived from the Circumplex Model is
that balanced types of couples and families will generally function more
adequately than unbalanced types.
Balanced types of couples and families on cohesion allow their
members to experience being both independent from and
connected to their family. On flexibility, balance means maintaining
some level of stability in a system with openness to some change
when necessary. Extreme behaviours on these two dimensions
might be appropriate for certain stages of the life-cycle or when a
family is under stress, but it can be problematic when families are
stuck at the extremes.

The Circumplex Model is sensitive to ethnic and cultural diver-
sity as reflected in the following hypothesis: If a family’s expecta-
tions support more extreme patterns, families will then operate in
a functional manner as long as all the family members like the
family that way.
Unbalanced types of couple and family systems are not neces-
sarily dysfunctional, especially if a family belongs to a particular
ethnic group (i.e. Hispanic, Southeast Asian) or religious group
(i.e. Amish, Mormon) that has normative expectations which
support extreme behaviours on these dimensions. Ethnicity is a
central trait of families and needs to be seriously considered in
assessing family dynamics. What might appear to be an ‘enmeshed’
family of colour to a white outsider may be functional for some
ethnic groups. This hypothesis necessitates measuring family satis-
faction for each family member. A Family Satisfaction Scale has
been developed for this purpose that is based on dimensions from
the Circumplex Model.
An important hypothesis linking communication and the
Circumplex Model states: Balanced types of couples and families
will have more positive communication compared to unbalanced
systems. In general, positive communication skills are seen as help-
ing couple and family systems facilitate and maintain a balance on
the two dimensions. Conversely, poor communication impedes
movement in the unbalanced systems and increases the chances
that these systems will remain extreme.
Changes in couple and family systems over time in response to stress

The Circumplex Model allows one to integrate systems theory and
family development theory, a proposal made more than two
decades ago by Reuben Hill (1970). Building on the family devel-
opmental approach, it is hypothesized that the stage of the family
life-cycle and composition of the family will have considerable
impact on the type of family system (Carter and McGoldrick, 1988).
A hypothesis related to change is: Couple and family systems will
change to adapt to developmental needs and situational stress.
The Circumplex Model is dynamic in that it assumes that
changes can and do occur in couple and family types over time.
Families can move in any direction that the situation, stage of the
family life-cycle or socialization of family members may require.
The model can be used to illustrate developmental change of a

154 David H. Olson
couple as they progress from dating to marriage, to pregnancy,
childbirth and child-rearing, raising and launching adolescents,
and moving into life as a couple again.
When one family member’s desires change, the family system
must somehow deal with that request. For example, increasing
numbers of wives want to develop more autonomy from their
husbands (cohesion dimension) and also desire more power and
equality in their relationships (flexibility dimension). If their
husbands are unwilling to understand and change in accordance
with these expectations, the marriages will probably experience
increasing levels of stress and dissatisfaction. Another common
example of changing expectations occurs when a child reaches
adolescence. Adolescents often want more freedom, independence
and power in the family system. These pressures to change the
family system by one member can facilitate change in the family,
despite the resistance of the family to change.
An example of how the Circumplex Model can be used in both
understanding and graphing the changes in a family system over
time is a family where the husband, Peter, aged 53, had a heart
attack. His wife, Martha, was a homemaker and they had three
teenagers living at home, one of whom was attending college.
The changes in this family system are illustrated in Figure 3.
Before the heart attack the family was flexibly separated (point A)
which was generally appropriate for that stage of the family life-
cycle. Once the heart attack occurred, however, the family quickly
shifted to becoming more chaotically enmeshed (point B). Very high
levels of closeness, characterized by enmeshment, occurred because
the illness brought the family closer together emotionally. It also
created chaos among family members because they needed to
dramatically shift many of their daily routines.
From about the third to sixth week, the family became rigidly
enmeshed (point C). In an attempt to stabilize the chaos by reorgan-
izing some of the routines in their family system, the family became
rigid. Six months later, the family was functioning as a structurally
connected family (point D). Some of the rigidity and extreme cohe-
sion decreased, but it remained a close family with a more struc-
tured system because of Peter’s disability.
In summary, because of Peter’s heart attack, this family’s system
changed several times over the course of the next six months as it
adapted to this crisis. It started as a balanced system (flexibly separ-
ated), moved to two unbalanced types (chaotically enmeshed and

Figure 3 Family change before and after husband’s heart attack
rigidly enmeshed), before ending up once again as a balanced

system (structurally connected).
It is expected that family systems will change in response to a
crisis. As hypothesized in the Circumplex Model, it is the balanced
families that would have the resources and skills to shift their system
in an appropriate way to cope more effectively with a crisis. In
contrast, it is hypothesized that unbalanced families will not have

156 David H. Olson
the resources that are needed to change, and therefore will have
more difficulty adapting to a crisis. Balanced families are higher in
second-order change because they are able to alter their family
system to adapt to family crises.
Research validating the Circumplex Model
Balance versus unbalanced families

A central hypothesis derived from the model is that balanced
couples and families will function more adequately than unbal-
anced couples and families. More than 250 studies using FACES
(Olson, 1996) have supported the major hypothesis that balanced
systems are more functional than unbalanced systems. These stud-
ies have generally compared families with a variety of emotional
problems and symptoms compared to non-clinical families. Almost
all of these studies have been done using the FACES self-report
scales, which measure cohesion and flexibility in a linear way. These
findings should be interpreted using the three-dimensional (3-D)
Circumplex Model with high scores on cohesion and flexibility
representing balanced systems.
Strong support for the curvilinear nature of the Circumplex Model
has been found using the Clinical Rating Scale (Olson, 1990, 1994).
The findings using the Clinical Rating Scale have also provided
support for the major hypothesis that balanced families are more func-
tional than unbalanced families (Thomas and Olson, 1993).
A pioneering study by Clarke (1984) focused on families with
schizophrenics, neurotics, families who had completed therapy
sometime in their past and a no-therapy control group (See Figure
4). He used FACES II, a self-report scale that assesses family cohesion
and flexibility. He found a very high level of unbalanced families in
the neurotic and schizophrenic groups compared to the no-therapy
group. Conversely, he found a significantly higher level of balanced
families in the no-therapy group compared to the other groups.
Figure 4 illustrates the flexibility between these groups. While the
percentage of unbalanced family types decreased dramatically from
the symptomatic to no-therapy groups (neurotic, 64%; schizo-
phrenic, 56%; completed therapy, 38%; no therapy, 7%), the
percentage of balanced families increased (neurotic, 8%; schizo-
phrenic, 12%; completed therapy, 38%; no therapy, 48%) as
hypothesized.

157
Figure 4 Problem families and Circumplex Model

Page 157
Circumplex Model
14/4/0 11:00 am
- Main Text
JOFT 22.2
Figure 5 Sex offenders and Circumplex Model

David H. Olson
Page 158
14/4/0 11:00 am
- Main Text
158
JOFT 22.2
A study by Carnes (1989) that used FACES II investigated the
family systems in sex offenders and found high levels of unbalanced
family types in both their family-of-origin and their current families
(see Figure 5). While 49% had unbalanced family types in their
family-of-origin and 66% of their current families were unbalanced
types, only 19% of the non-offender families were unbalanced.
Conversely, while only 11% of their family-of-origin and 19% of
their current families were balanced types, 47% of the non-offender
families were balanced.
These studies of clinical samples clearly demonstrate the discrim-
inate power of the Circumplex Model in distinguishing between
problem families and non-symptomatic families. There is strong
empirical support for the hypothesis that balanced types of families
are more functional than unbalanced family types. There is,
however, a lack of evidence that any of these symptoms are specifi-
cally linked with a specific type of family system; for example, chaot-
ically enmeshed. This was a misplaced hope of early family research
linking family symptoms (for example, schizophrenic offspring)
and family systems (Walsh and Olson, 1989).
Balanced families and communication

Another hypothesis is that balanced families will have more positive
communication skills than unbalanced families. Using data from a
national survey of 1,000 families, Barnes and Olson (1986) investi-
gated parent–adolescent communication and family functioning.
Using ‘non-problem’ families, the hypothesis that balanced families
would have better communication skills was supported when relying
on data from the parents’ reports. Rodick et al. (1986) also found
strong support for the hypothesis that balanced families have more
positive communication skills. Using observational measures of
mother–adolescent interaction, they found that mothers in the
balanced group had significantly higher rates of supportive commu-
nication, explicit information and positive affect than did mothers
in the unbalanced group.
Assessment with the circumplex instruments

In order to adequately assess the three major dimensions of the
Circumplex Model and other related concepts, Olson et al. (1986)
developed a variety of self-report instruments. The self-report

160 David H. Olson
instrument, the Circumplex Assessment Package (CAP) provides
the insider’s perspective, whereas the Clinical Rating Scale (CRS)
provides the outsider’s perspective. Both perspectives are useful,
but they often yield apparently conflicting data. Used together,
however, they help capture the complexity of marital and family
systems.
Assessment: multi-method, multi-person, multi-trait and multi-system levels

Multi-method assessment utilizes self-report scales, which provide an
‘insider’s perspective’ on their own relationship, and the therapists’
or observers’ ratings, which provide an ‘outsider’s perspective’ on
the same system. Because these two approaches give different
perspectives, this provides an important rationale why both
approaches should be used in work with families (Olson, 1986).
Multi-person assessment is also important because family members
often do not agree with each other in describing their family system
(Olson et al., 1989). Assessment using multiple family members
therefore provides a more complete picture of how each family
member views the system and the level of agreement or disagree-
ment between them.
Multi-trait assessment is based on the three central dimensions of
the Circumplex Model: cohesion, flexibility and communication.
Although other traits can be incorporated into couple and family
assessment, these three dimensions provide the foundation and
central core of these relationship systems.
Multi-system assessment ideally focuses on the individual, the mari-
tal system, parent–child system and total family-including extended
family relationships. One important question to ask family members
is who they each consider to be members of their family. It is surpris-
ing to us how often family members disagree regarding who is
currently in their family system. This raises important questions
about boundary issues and who is psychologically and/or physically
present in a given family system (Boss and Greenberg, 1984) and
especially important given the increasing diversity of family forms,
particularly the changes accompanying divorce and remarriage.
Circumplex Assessment Package (CAP) and Clinical Rating Scale (CRS)

The Circumplex Assessment Package (CAP) is the latest in a series
of self-report assessments based on the Circumplex Model. This

procedure is multi-dimensional in that it assesses the three circum-
plex dimensions of cohesion, adaptability and communication. It
also includes the satisfaction dimension, which focuses on each
person’s satisfaction with various aspects of the family system.
It is also multi-systemic in that procedures focus on both the
marital and family systems. More specifically, each of the four
dimensions are assessed at the couple and family levels. Two-parent
families (nuclear or blended) would complete both the marital and
family scales. Single-parent families would complete the family
scales and the marital scales if the single parent has a significant
other. Couples (married or cohabitating) would complete the
couple scales.
FACES II, acronym for Family Adaptability and Cohesion
Evaluation Scales – Second Edition, is designed for assessing fam-
ilies. It may be used for both clinical work and research. MACES III,
acronym for Marital Adaptability and Cohesion Evaluation Scales –
Third Edition, is designed for assessing couples. The communica-
tion dimension is assessed at the couple level using a subscale from
the ENRICH instrument (Olson et al., 1986a) and family communi-
cation is based on the Parent–Adolescent Communication Scale
(Barnes and Olson, 1986). The satisfaction dimension is assessed at
the couple level using a subscale from the ENRICH instrument, and
the Family Satisfaction Scale (Olson and Wilson, 1986) is based on
the circumplex dimensions.
For clinical work with premarital and married couples, there are
two comprehensive inventories that can be used called
PREPARE/ENRICH Inventories (Olson, 1997; Olsen et al., 1986a).
Each of these focus on the couple system (cohesion and flexibility)
and family-of-origin (cohesion and flexibility) so it is possible to
compare the couple and family system. PREPARE, for premarital
couples, has been found to predict with 80–85% accuracy which
couples will divorce. ENRICH is designed for married couples and
is able to discriminate happy, non-clinical pairs from clinical
couples with 90% accuracy. Both PREPARE and ENRICH
Inventories contain twenty content categories, are computer-scored
and have norms based on over 500,000 couples.
The reliability of these scales has been evaluated in a variety of
studies. The most comprehensive summary is provided in Family
Inventories (Olson et al., 1986b). Both the internal consistency and
test–retest reliability of these scales is consistently high (r = .80).
Therapists and researchers have evaluated the items in terms of face

162 David H. Olson
validity and find them to meet acceptable criteria. The scales also
have discriminative validity in that they distinguish between clinical
and non-clinical families (Olson, 1986).
The Clinical Rating Scale (CRS) was developed by Olson (1990)
to assess cohesion, flexibility and communication from the perspec-
tive of clinicians or researchers observing family interaction. It has
been validated in an extensive study by Thomas and Olson (1993).
The CRS describes specific indicators for each level of the three
dimensions. This scale is a useful training device for helping indi-
viduals learn more about the Circumplex Model, and its value for
family assessment and planning treatment intervention.
Clinical implications
Goals of family therapy using the Circumplex Model

While family therapists have as a central goal reduction of the
presenting problems and symptoms of family members, this is
achieved by intervention focused on changing a dysfunctional type
of couple and/or family system. The basic assumption is that the
current family system dynamics are helping to maintain sympto-
matic behaviours. Thus the current pattern of interaction in the
family needs to be changed before the symptoms or presenting
problems will be alleviated.
Table 2 summarizes the specific goals of family therapy based on
the Circumplex Model. The first goal is ultimately to reduce any
problems and symptoms. Since most dysfunctional families coming
for therapy represent mid-range or unbalanced types, change often
involves trying to shift the system one level on cohesion and one
level on adaptability towards the balanced levels. It is therefore
typically assumed that the family will function more adequately if
the marital and/or family system is moved towards the balanced
types.
Because the Model is dynamic, intervention on cohesion or
adaptability often has a ripple effect in that it influences the system
on the other dimension. In terms of cohesion, problems in families
often occur because of their inability to balance separateness
(autonomy) and togetherness (intimacy). In couples coming for
therapy, there is often a difference in the amount of separateness
and togetherness the two partners experience or desire. For exam-
ple, in disengaged couples, one or both individuals have empha-

TABLE 2 Goals of marital and family therapy based on Circumplex
Model
Goal regarding symptoms

• Reduce presenting problems and symptoms
Goals regarding system (marital and/or family)
• Change system one level on cohesion and adaptability towards balanced
types
• On cohesion, balancing separateness and togetherness
• On flexibility, balancing stability and change
• Improve communication skills
Meta-goal regarding system (preventive)
• Increase ability to negotiate system change over time
sized looking out for themselves, and thus they have not maintained
their emotional bond of intimacy.
In families, the dynamics on cohesion are often more compli-
cated. One type of problem family might have an enmeshed
mother–adolescent coalition with a disengaged father. In this case,
the marital dyad would not be emotionally close. Increasing their
marital/parental collaboration is an effective strategy for breaking
up the strong parent–child coalition.
In terms of flexibility, couples and families with problems often
have difficulty balancing stability and change. These relationships
are either too rigid or too chaotic. With rigid systems, their beha-
vioural repertoire is often very narrow. When they are confronted
with increasing stress, they tend to become more rigid and inflexi-
ble. These families can often benefit from learning and using more
democratic decision-making and better problem-solving skills. On
the other hand, chaotic relationships often need increasing struc-
ture and they can also benefit from improved problem-solving skills.
Increasing the positive communication skills of couples and fam-
ilies can also facilitate system change. Individuals in problem fam-
ilies often need to learn how to be more assertive in expressing
their wants and desires. They can usually gain from learning how to
express their feelings in a constructive manner and how to listen
and give empathic feedback to each other.
However, improving communication skills in a family is a neces-
sary but not sufficient condition for change on the dimensions of

164 David H. Olson
cohesion and flexibility. Communication skills can help increase
awareness of current needs and preferences. System change on
cohesion and adaptability is more difficult and complex. Having
good communication skills enables families to express more clearly
the type of relationship they would like to have on cohesion and
flexibility.
One desirable goal of couple and family therapy is ultimately to
teach couples not only to deal with their current issues, but also to
provide them with the necessary skills to negotiate system change
over time. It is an assumption of the Circumplex Model that couples
and families need to alter their system as their individual needs and
preferences change. Being able to articulate and negotiate these
changes on cohesion and adaptability will also enable the couple or
family to more adequately cope with stress and the other problem-
atic issues they encounter over time. This is an important preventive
goal that moves beyond dealing with the current and presenting
symptoms. Unfortunately, this meta-goal is rarely achieved in ther-
apy because most families, and even some therapists, are too
focused on only reducing the current presenting problems.
Treatment planning using the Circumplex Model

The Circumplex Model is a valuable resource in assessment-based
treatment planning with severely dysfunctional families. A major
task for research outcome is to determine which elements of inter-
vention are most appropriate and effective with which presenting
problems and elements of family functioning. FACES and the CRS
are tools that can be used for treatment planning and outcome evalu-
ation (Walsh and Olson, 1989).
The model provides a conceptual framework for assessing family
system functioning on two fundamental dimensions of family
organization: cohesion and flexibility. This descriptive typology of
transactional patterns can be used to determine a family’s current
level and style of functioning on each dimension, and to guide
treatment planning to strengthen particular components of func-
tioning towards clearly specified and realistic objectives. Thus
family therapy is not limited to reduction or interruption of
extreme dysfunctional patterns, but is directed systematically
towards promotion of more functional patterns.
For families assessed at either extreme on the dimensions, inter-
vention strategies can be targeted to fit their particular pattern of

organization and to guide change, in a stepwise progression,
towards a more balanced system. In most cases of severe and
chronic dysfunction, a reachable therapeutic goal would be the
achievement of higher functioning at the next, adjacent pattern,
such as a shift from disengaged to separated or from enmeshed to
connected. It would be unrealistic to attempt to change family
patterns to a quite different type of organization, such as pushing a
disengaged family to be strongly connected, or an enmeshed family
to become separated.
Severely dysfunctional families often assume such extreme all-or-
none positions regarding change. They are likely to alternate
between feelings of hopelessness that any change can occur and
unrealistic expectations of goals that are unlikely to be met. They
commonly fluctuate between extremes of enmeshed/disengaged
and extremes of rigidity/chaos. An enmeshed family may resist a
clinician’s efforts to promote physical separation, such as leaving
home at launching, when they hold catastrophic expectations that
any separation will result in a total cut-off.
Opposite extremes may also be found in different family sub-
systems. In many enmeshed families, some siblings may disengage
completely from the family in order to avoid fusion, assuming posi-
tions of pseudo-autonomy that dissolve in contact with the family.
Clinicians must be cautious not to collude with presuppositions of
either all-or-none position. Fears of runaway change or loss of
patterns considered to be essential to individual or family survival
are common sources of ‘resistance’ to change and therapy drop-
outs. Clinicians need to be alert to prevent extreme family oscilla-
tion that can occur much like a ‘short-circuiting’ process.
A therapist must be active in structuring and monitoring family
interaction to block or interrupt the all-or-none tendency in these
families to flip to the other extreme from their current organizational
pattern. In work with families with extreme patterns, it is essential
to set modest, concrete objectives to be reached through small
increments of change in order to reduce anxiety to a manageable
level, to prevent extreme fluctuations, and to help the family to
modulate and moderate changes than can be maintained over time.
In summary, the Circumplex Model is designed for clinical assess-
ment and treatment planning with couples and families. The assess-
ment package includes both the self-report scales of FACES and the
observer (therapist) rating called the Clinical Rating Scale. The
assessment can be used for treatment planning with the goal of

166 David H. Olson
making the system more functional and being able to more effec-
tively deal with current and future relationship problems. The ulti-
mate goal of the Circumplex Model is to bridge research, theory
and clinical practice.
References
Barnes, H. and Olson, D.H. (1986) Parent–adolescent communication scale. In
D.H. Olson, H.I. McCubbin, H. Barnes, A. Larsen, M. Muxen and M. Wilson
(eds) Family Inventories. St Paul, MN: Family Social Science, University of
Minnesota.
Beavers, W.B. and Hampson, R.B. (1990) Successful Families: Assessment and
Intervention, New York: W.W. Norton.
Benjamin, L.S. (1977) Structural analysis of a family in therapy. Journal of
Counseling Clinical Psychology, 45: 391–406.
Boss, P.G. and Greenberg, J. (1984) Family boundary ambiguity: a new variable in
family stress theory. Family Process, 23: 535–546.
Carnes, P. (1989) Contrary to Love: Helping the Sexual Addict. Minneapolis, MN:
CompCare Publications.
Carter, B. and McGoldrick, M. (eds) (1988) The Changing Family Life Cycle. New
York: Gardner Press.
Clarke, J. (1984) The family types of neurotics, schizophrenics and normals.
Unpublished doctoral dissertation. St Paul, MN: Family Social Science,
University of Minnesota.
Constantine, L. (1986) Family Paradigms. New York: Guilford Press.
Epstein, N., Baldwin, L. and Bishop, D. (1983) The McMaster Family Assessment
Device. Journal of Marital and Family Therapy, 9: 171–180.
Epstein, N., Bishop, D., Ryan, C., Miller, I. and Keitner, G. (1993) The McMaster
Model: view of healthy family functioning. In F. Walsh (ed.) Normal Family
Processes. New York: Guilford Press.
French, A.P. and Guidera, B.J. (1974) The family as a system in four dimensions: a
theoretical model. Paper presented at the American Academy of Child
Psychology, San Francisco, CA.
Gottman, J.M. (1994) Why Marriages Succeed or Fail. New York: Simon & Schuster.
Hill, R. (1970) Family Development in Three Generations. Cambridge, MA: Schenkman.
Kantor, D. and Lehr, W. (1975) Inside the Family. San Francisco, CA: Jossey Bass.
Leary, T. (1975) Interpersonal Diagnosis of Personality. New York: Ronald Press.
Lee, C. (1988) Theories of family adaptability: toward a synthesis of Olson’s
Circumplex Model and Beavers’ Systems Model. Family Process, 27: 73–85.
Leff, J. and Vaughn, C. (1985) Expressed Emotion in Families. New York: Guilford
Press.
Olson, D.H. (1986) Circumplex Model VII: validation studies and FACES III.
Family Process, 25: 337–351.
Olson, D.H. (1990) Clinical Rating Scale for Circumplex Model. St Paul, MN: Family
Social Science, University of Minnesota.
Olson, D.H. (1991) Three dimensional (3-D) Circumplex Model and revised scor-
ing of FACES. Family Process, 30: 74–79.

Olson, D.H. (1993) Circumplex Model of Marital and Family Systems. In F. Walsh
(ed.) Normal Family Processes (2nd edn). New York: Guilford Press.
Olson, D.H. (1994) Curvilinearity survives: the world is not flat. Family Process, 33:
471–178.
Olson, D.H. (1996) Clinical assessment and treatment interventions using the
Circumplex Model. In F.W. Kaslow (ed.) Handbook of Relational Diagnosis and
Dysfunctional Family Patterns. New York: John Wiley and Sons.
Olson, D.H. (1997) PREPARE/ENRICH Program: Version 2000. Minneapolis, MN:
Life Innovations.
Olson, D.H. and Wilson, M. (1986) Family satisfaction. In D.H. Olson et al. (eds)
Family Inventories. St Paul, MN: Family Social Science, University of Minnesota.
Olson, D.H., Fournier, D.G. and Druckman, J.M. (1986a) PREPARE, PREPARE-MC
and ENRICH inventories (2nd edn). Minneapolis, MN: PREPARE/ENRICH,
Incorporated.
Olson, D.H., McCubbin, H.I., Barnes, H., Larsen, A., Muxen, M. and Wilson, M.
(1986b) Family Inventories. St Paul, MN: Family Social Science, University of
Minnesota.
Olson, D.H., Russell, C.S. and Sprenkle, D.H. (1989a) Circumplex Model: Systemic
Assessment and Treatment of Families. New York: Haworth Press.
Olson, D.H., McCubbin, H.I., Barnes, H., Larsen, A., Muxen, M. and Wilson, M.
(1989b) (2nd edn) Families: What Makes them Work. Los Angeles, CA: Sage.
Parsons, T. and Bales, R.F. (1955) Family Socialization and Interaction Process.
Glencoe, IL: Free Press.
Reiss, D. (1981) The Family’s Construction of Reality. Cambridge, MA: Harvard
University Press.
Rodick, J.D., Henggler, S.W. and Hanson, C.L. (1986) An evaluation of family
adaptability, cohesion evaluation scales (FACES) and the Circumplex Model.
Journal of Abnormal Child Psychology, 14: 77–87.
Skinner, H.A., Santa-Barbara, J. and Steinhauer, P.D. (1983) The family assessment
measure. Canadian Journal of Community Mental Health, 2: 91–105.
Thomas, V. and Olson, D.H. (1993) Problem families and the Circumplex Model:
observational assessment using the Clinical Rating Scale. Journal of Marital and
Family Therapy, 19: 159–175.
Walsh, F. and Olson, D.H. (1989) Utility of the Circumplex Model with severely
dysfunctional family systems. In D.H. Olson, C.S. Russell and D.H. Sprenkle
(eds) Circumplex Model: Systemic Assessment and Treatment of Families (2nd edn).
New York: Haworth Press.

Journal of Family Therapy

Uploaded by

Copyright:

Available Formats

Journal of Family Therapy

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Family Therapy

Uploaded by

Copyright:

Available Formats

JOFT 22.

2 - Main Text 14/4/0 10:59 am Page 144

Circumplex Model of Marital and Family Systems

The Circumplex Model focuses on the three central dimensions of marital

a Family Social Science, University of Minnesota, 290 McNeal Hall, St Paul,

 2000 The Association for Family Therapy and Systemic Practice

Marital and family cohesion (togetherness)

 2000 The Association for Family Therapy and Systemic Practice

Beavers and Hampton (1990) Stylistic dimension Adaptability Affect

Marital and family flexibility

 2000 The Association for Family Therapy and Systemic Practice

Figure 1 Circumplex Model: Couple and family map

 2000 The Association for Family Therapy and Systemic Practice

Marital and family communication

 2000 The Association for Family Therapy and Systemic Practice

Three-dimensional (3-D) Circumplex Model and linear scoring

 2000 The Association for Family Therapy and Systemic Practice

Figure 2 Three-dimensional family Circumplex Model

the Circumplex Model (Olson, 1991), an idea suggested by Lee

 2000 The Association for Family Therapy and Systemic Practice

Hypotheses derived from the Circumplex Model

 2000 The Association for Family Therapy and Systemic Practice

Changes in couple and family systems over time in response to stress

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

Figure 3 Family change before and after husband’s heart attack

rigidly enmeshed), before ending up once again as a balanced

 2000 The Association for Family Therapy and Systemic Practice

Research validating the Circumplex Model

Balance versus unbalanced families

 2000 The Association for Family Therapy and Systemic Practice

Figure 4 Problem families and Circumplex Model

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

Balanced families and communication

Assessment with the circumplex instruments

 2000 The Association for Family Therapy and Systemic Practice

Assessment: multi-method, multi-person, multi-trait and multi-system levels

Circumplex Assessment Package (CAP) and Clinical Rating Scale (CRS)

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

Goals of family therapy using the Circumplex Model

 2000 The Association for Family Therapy and Systemic Practice

Goal regarding symptoms

 2000 The Association for Family Therapy and Systemic Practice

Treatment planning using the Circumplex Model

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

 2000 The Association for Family Therapy and Systemic Practice

You might also like