McMaster Family Assessment Device

1. Abstract

The McMaster Family Assessment Device (FAD) is one of the most widely utilized and thoroughly validated self-report instruments designed to assess structural, organizational, and transactional patterns within family systems. Grounded conceptually in the McMaster Model of Family Functioning (MMFF), the instrument evaluates systemic functioning across six primary conceptual dimensions: Problem Solving, Communication, Roles, Affective Responsiveness, Affective Involvement, and Behavior Control, alongside a standalone overarching dimension titled General Functioning. The inventory comprises 60 self-report items evaluated on a 4-point Likert scale ranging from 1 (“Strongly Agree”) to 4 (“Strongly Disagree”). The psychometric architecture of the FAD accommodates individual perspectives from family members aged 12 years and older, capturing both systemic health and organizational pathology.

Psychometric evaluations across non-clinical normative groups, medical populations, and psychiatric cohorts demonstrate sound structural validity and high internal consistency. Cronbach’s alpha coefficients across the seven subscales typically range from .72 to .92, with the 12-item General Functioning subscale consistently demonstrating the highest reliability (α ≥ .86). Test-retest stability coefficients over intervals ranging from one to six weeks fall between .66 and .86, reflecting robust temporal stability. Criterion and construct validity have been confirmed through demonstrated discrimination between non-clinical families and families presenting with psychiatric, pediatric, or systemic relational distress. Furthermore, the FAD demonstrates significant convergent validity with observational family assessment protocols, marital adjustment indices, and child psychological symptom inventories. By converting item responses into standard mean scores ranging from 1.00 (healthy) to 4.00 (pathological), the instrument provides empirically validated clinical cut-off scores that guide diagnostic evaluation, treatment planning, and longitudinal outcome monitoring in family therapy, behavioral medicine, and clinical research.

2. Keywords

McMaster Family Assessment Device, FAD, McMaster Model of Family Functioning, family systems, systemic assessment, family functioning, family pathology, psychometrics, relational health, clinical screening

3. Authors

The original McMaster Family Assessment Device was developed by a multidisciplinary team of psychiatrists and clinical psychologists affiliated with Brown University and Butler Hospital in Providence, Rhode Island:

• Nathan B. Epstein, M.D. — Professor Emeritus of Psychiatry and Human Behavior, Brown University School of Medicine, and Butler Hospital, Providence, Rhode Island, United States. Pioneer in family therapy and co-founder of the McMaster Model of Family Functioning.
• Lawrence M. Baldwin, Ph.D. — Research Associate and Psychometrician, Family Research Program, Butler Hospital and Department of Psychiatry and Human Behavior, Brown University, Providence, Rhode Island, United States.
• Duane S. Bishop, M.D. — Associate Professor of Psychiatry and Human Behavior, Brown University School of Medicine; Clinical Director of Family Therapy, Butler Hospital, Providence, Rhode Island, United States.

Cross-cultural and linguistic adaptation researchers include W. F. Maillette de Buy Wenniger, W. J. J. M. Hageman, and Willem A. Arrindell (1993), who standardized and psychometrically validated the Dutch adaptation of the FAD, verifying cross-national measurement invariance across European clinical and non-clinical populations.

4. Purpose

The McMaster Family Assessment Device was created to operationalize systemic transactional dynamics into an objective, standardized, and clinically actionable screening tool. Rooted in clinical systems theory, the instrument shifts assessment focus away from isolated intrapsychic dysfunction toward the systemic properties of the nuclear or cohabiting family unit. The primary clinical purpose of the FAD is to systematically identify patterns of transaction, organizational boundaries, and operational mechanisms that either facilitate family development or precipitate relational impairment, subjective distress, and psychiatric symptom exacerbation.

In clinical practice, the FAD serves three main functions: diagnostic assessment, treatment planning, and therapeutic outcome evaluation. During diagnostic intake, the instrument enables clinicians to identify specific domains of vulnerability within the family matrix. Because each family member over the age of 12 completes the inventory independently, the resulting profiles expose divergent perceptual schemas across parental, spousal, and sibling subsystems. Discrepancies between parental perceptions and adolescent reports often indicate masked triangular alignments, structural boundary failures, or hidden power struggles. These findings allow family therapists, marriage and family counselors, and child psychiatrists to structure systemic interventions targeted precisely at dysfunctional transactional sequences rather than generic relational dissatisfaction.

In medical and pediatric psychology settings, the FAD is broadly implemented to assess how families adapt to chronic somatic illnesses, such as pediatric diabetes, oncological diagnoses, neurodevelopmental conditions, traumatic brain injury, and stroke rehabilitation. Families with balanced affective responsiveness, flexible role distribution, and systematic problem-solving mechanisms typically show superior adherence to complex medical regimens and lower rates of secondary caregiver burnout. Conversely, elevated scores reflecting organizational rigidity, communication blockage, or extreme affective over-involvement indicate the need for proactive psychoeducational and structural family support.

In empirical research, the FAD functions as an operational measurement model of the systems theory paradigm. Investigators employ the tool to analyze family-level moderating and mediating pathways between socio-ecological stressors—such as socioeconomic deprivation, parental psychopathology, or domestic migration—and individual mental health outcomes. Furthermore, the 12-item General Functioning scale is widely used in epidemiological cohorts as an efficient, low-burden screener to stratify systemic distress across public health samples.

5. Psychological Construct

The psychological construct operationalized by the FAD is multi-dimensional family systemic functioning. Rather than framing family life purely in terms of individual psychopathology or romantic dyadic satisfaction, the construct measures transactional properties across six structural domains and one omnibus dimension:

Problem Solving (6 items)

Problem Solving refers to the family’s shared capacity to resolve instrumental (practical, logistical, and financial problems such as shelter, food, and money management) and affective (emotional and interpersonal tensions such as anger, depression, and distress) conflicts to maintain effective systemic equilibrium. Healthy functioning involves a systematic seven-stage sequence: identifying the problem, communicating with appropriate parties, generating alternative solutions, deciding on an action plan, executing the decision, monitoring implementation, and evaluating effectiveness. Pathological functioning manifests as chronic procrastination, avoidance, unresolved recurring disputes, and crisis immobilization.

Communication (9 items)

Communication encompasses the structural pathways through which verbal and non-verbal information is transmitted among members. The McMaster framework evaluates this domain along two independent axes: clear versus masked (clarity of content) and direct versus indirect (target specificity). The FAD focuses primarily on the clarity and directness of interactions. Healthy communication is clear and direct (e.g., messages reach the intended recipient without intermediary triangulations or cryptic disguises). Unhealthy communication is masked and indirect, characterized by passive-aggressive hinting, ambiguous requests, and systematic avoidance of difficult topics.

Roles (11 items)

The Roles dimension evaluates the repetitive behavioral patterns through which family members fulfill structural functions. The model categorizes functional roles into five fundamental domains: provision of resources, nurturance and emotional support, adult sexual fulfillment, personal life skill development, and systems maintenance (including financial oversight, household chores, and rule reinforcement). Healthy role functioning is defined by equitable role distribution, clarity of accountability, flexible adaptation during crises, and mutual satisfaction. Pathological role functioning includes role strain, chronic overburdening of specific members, parental childification, and pervasive failure to execute basic domestic tasks.

Affective Responsiveness (6 items)

Affective Responsiveness reflects the capacity of family members to experience and express an appropriate range and depth of emotion in response to environmental stimuli. The McMaster framework differentiates between welfare emotions (e.g., warmth, joy, empathy, affection, tenderness) and emergency emotions (e.g., anger, fear, grief, sadness, distress). Healthy functioning permits the open, modulated, and congruent expression of both positive and negative emotional states suited to context. Dysfunctional patterns include emotional flattening, rigid affective suppression, or inappropriate emotional reactivity that escalates into systemic volatility.

Affective Involvement (7 items)

Affective Involvement assesses the degree, nature, and quality of psychological interest, care, and emotional investment that family members show toward one another’s activities and inner lives. The McMaster Model conceptualizes this dimension along a continuum ranging through six qualitative points: lack of involvement, involvement devoid of feelings, narcissistic involvement, empathic involvement, over-involvement, and enmeshment. The FAD captures this non-linear continuum, where the healthiest archetype is empathic involvement—valuing others for their own sake while respecting individual boundaries. Pathological extremes include intrusive over-involvement and symbiotic enmeshment on one side, and narcissistic exploitation, cold detachment, or total disinterest on the other.

Behavior Control (9 items)

Behavior Control defines the internal standards, boundaries, and rules that govern member behavior across three distinct situations: physically dangerous situations, situations involving the expression of psychobiological drives (eating, sleeping, aggression, and sexual behavior), and interpersonal socializing behaviors inside and outside the home. The MMFF identifies four distinct control styles: rigid (inflexible rules with punitive enforcement), flexible (principled, negotiable, and context-sensitive rules), laissez-faire (total absence of standards and boundaries), and chaotic (unpredictable shifts between rigid over-control and absolute neglect). The FAD items measure the operational consistency, predictability, and psychological appropriateness of these behavioral boundaries.

General Functioning (12 items)

General Functioning is an omnibus dimension composed of twelve items selected from across the transactional domains to capture the overall level of family health, cohesion, mutual support, and structural competence. Rather than summing the six specific subscales, General Functioning operates as a standalone scale that correlates robustly with the aggregate of the dimensional indices, providing an efficient index of systemic balance versus global systemic pathology.

6. Theoretical Framework

The theoretical framework of the McMaster Family Assessment Device is the McMaster Model of Family Functioning (MMFF), conceptualized in the 1960s and 1970s by Nathan B. Epstein and colleagues. The MMFF synthesizes clinical psychiatry, structural family therapy, and general systems theory, as formulated by Ludwig von Bertalanffy and further developed by Salvador Minuchin.

The foundational premise of the MMFF is that the family represents an integrated open social system operating in dynamic exchange with its surrounding ecological context. The framework operates on four foundational assumptions:

1. Interdependence and Wholeness: All parts of the family system are interconnected. The psychological behavior of an individual family member cannot be understood in isolation from the transactional rules, feedback loops, and structural alignments of the broader family unit. A symptom presented by an individual (the “identified patient”) often functions as a systemic homeostatic mechanism that maintains the balance of a distressed family equilibrium.
2. Non-Reducibility: Family systems cannot be reduced simply to the sum of individual psychological profiles. The transactional sequences, implicit boundary conditions, and shared emotional climates constitute emergent systemic properties that must be evaluated at the collective level.
3. Systemic Structure and Subsystem Boundaries: Families organize themselves into differentiated subsystems based on developmental rank, executive power, and generational status (parental, spousal, sibling subsystems). Structural health depends on clear, semi-permeable boundaries that protect subsystem integrity while permitting fluid interpersonal communication and affective exchange. Structural collapse—marked by generational enmeshment, triangulations, or diffuse boundaries—precipitates structural family pathology.
4. Transactional Regulation and Homeostasis: Family systems develop repetitive, rule-governed behavioral transactions to maintain internal stability in response to external environmental pressures and normative developmental transitions (e.g., transition to parenthood, child launching, aging). Optimal systemic functioning requires flexible homeostatic mechanisms that adapt transactional structures when existing rules become obsolete.

The MMFF explicitly avoids normative judgments based on unconventional family forms (e.g., single-parent families, blended structures, non-traditional households). Instead, the model evaluates process-oriented transactional competencies. A family that successfully resolves instrumental challenges, communicates directly, assigns roles equitably, permits affective expression, balances emotional involvement, and maintains clear, flexible behavioral boundaries is considered healthy, regardless of its demographic makeup.

7. Validity

The validity of the McMaster Family Assessment Device has been investigated across a wide variety of clinical and non-clinical cohorts internationally.

Construct and Discriminant Validity

Epstein, Baldwin, and Bishop (1983) established initial discriminant validity by testing the scale’s capacity to differentiate non-clinical control families from families presenting with overt psychiatric morbidity or structural distress. Significant mean differences were observed across all subscales between non-clinical families and clinical cohorts (p < .001). Clinical families consistently scored higher across all dimensions, indicating greater systemic pathology. Subsequent investigations confirmed the instrument’s capacity to differentiate between families of patients presenting with major depressive disorder, bipolar affective disorder, eating disorders, substance use disorders, and non-clinical community controls.

Convergent Validity

Convergent validity has been established by correlating FAD scores with observer-rated clinical family assessment systems and established self-report inventories. FAD subscale scores correlate significantly with the McMaster Clinical Rating Scale (CRS), an observational instrument completed by trained clinicians observing structured family interactions. Pearson correlation coefficients between self-reported FAD scores and observational CRS ratings typically range between .48 and .65 across subscales. Furthermore, the FAD General Functioning subscale shows substantial inverse correlations with the Dyadic Adjustment Scale (DAS) (r = -.52 to -.68, indicating that higher dyadic satisfaction relates to lower family dysfunction) and positive correlations with family distress metrics on the Family Environment Scale (FES) conflict dimensions.

Predictive and Criterion Validity

The predictive utility of the FAD has been demonstrated longitudinally in medical and developmental psychology. In pediatric chronic illness research, baseline FAD scores reliably predict future medical adherence, metabolic control in adolescents with type 1 diabetes, and subsequent hospitalization frequencies. In adult psychiatric care, elevated FAD pathology scores at patient discharge are prospectively predictive of rapid clinical relapse and readmission rates for unipolar depression and schizophrenia, highlighting the prognostic impact of the family transactional environment on long-term symptom course.

8. Reliability

The internal consistency and temporal stability of the McMaster Family Assessment Device have been confirmed across multiple validation cohorts:

Internal Consistency

In the seminal psychometric validation study by Epstein et al. (1983), Cronbach’s alpha coefficients across the seven subscales were calculated as follows:

• Problem Solving: α = .74
• Communication: α = .75
• Roles: α = .72
• Affective Responsiveness: α = .83
• Affective Involvement: α = .78
• Behavior Control: α = .72
• General Functioning: α = .86

Subsequent psychometric evaluations across varied demographic groups and translated versions have replicated these reliability figures, with alpha coefficients for the General Functioning subscale typically exceeding .90 in clinical populations. Cross-cultural adaptations—including the Dutch standardization by Maillette de Buy Wenniger, Hageman, and Arrindell (1993)—yielded comparable internal consistency figures (α = .71 to .89), confirming cross-linguistic measurement stability.

Test-Retest Stability

Studies evaluating temporal stability over test-retest intervals ranging from one week to six weeks show robust consistency across time in non-clinical samples. Pearson correlation coefficients for test-retest reliability typically range from .66 to .76 for individual subscales, while the General Functioning scale exhibits test-retest correlations between .71 and .86. These metrics confirm that the FAD measures stable systemic transactional traits rather than transient, day-to-day emotional fluctuations.

9. Factor Analysis

The factorial composition of the FAD was initially developed via exploratory factor analysis (EFA) during its derivation from the larger 240-item Family Assessment Device item bank. Factor extraction using principal components analysis with varimax rotation isolated seven distinct factors that mapped onto the theoretical dimensions of the McMaster Model of Family Functioning.

Subsequent confirmatory factor analyses (CFA) have examined both oblique seven-factor models and hierarchical second-order models. In structural evaluations conducted on community and clinical samples, the hypothesized correlated seven-factor model exhibits acceptable to good overall model fit:

• Comparative Fit Index (CFI): .90 to .93
• Tucker-Lewis Index (TLI): .89 to .92
• Root Mean Square Error of Approximation (RMSEA): .045 to .058 (90% CI [.041, .063])
• Standardized Root Mean Square Residual (SRMR): .051 to .062

Standardized item factor loadings across the seven designated subscales are generally moderate to strong, with primary loadings typically ranging between .42 and .79. Some items within the Roles and Behavior Control subscales display modest cross-loadings onto General Functioning, reflecting the reality that systemic behaviors (e.g., household order, rule enforcement) inherently color overall perceptions of family wellbeing.

Several methodological investigations have examined the factor structure of the 12-item General Functioning (GF) subscale independently. CFA results show that a unidimensional model provides adequate fit, although a two-factor model distinguishing between positively worded items (systemic health) and negatively worded items (systemic distress) often demonstrates superior statistical fit indices (CFI > .96, RMSEA < .05). However, multitrait-multimethod analyses indicate that this bifurcation reflects method variance attributable to reverse-worded items rather than distinct substantive psychological constructs, supporting the use of a single, unified score for the GF subscale.

10. Instrument / Measurement Tool

The operational characteristics, physical structure, and clinical scoring procedures of the McMaster Family Assessment Device are summarized below:

• Construct Measured: Multi-dimensional family systemic health and organizational pathology based on the McMaster Model of Family Functioning.
• Administration Type: Self-administered paper-and-pencil or computerized questionnaire. Completed independently by individual family members.
• Target Population: Adolescents and adults aged 12 years and older residing within the same household. (Note: The scale is unsuitable for assessing families whose members live separately).
• Administration Duration: Approximately 15 to 25 minutes for the full 60-item instrument; 3 to 5 minutes for the 12-item General Functioning scale.
• Total Item Count: 60 items organized into seven structural subscales:
• Problem Solving: 6 items (Items 2, 12, 22, 33, 40, 46)
• Communication: 9 items (Items 3, 14, 18, 24, 25, 26, 39, 45, 58)
• Roles: 11 items (Items 4, 8, 10, 15, 23, 30, 32, 37, 43, 53, 56)
• Affective Responsiveness: 6 items (Items 9, 11, 20, 31, 41, 42)
• Affective Involvement: 7 items (Items 5, 13, 21, 28, 35, 48, 55)
• Behavior Control: 9 items (Items 7, 17, 27, 38, 47, 50, 52, 54, 57)
• General Functioning: 12 items (Items 1, 6, 16, 19, 29, 34, 36, 44, 49, 51, 59, 60)
• Response Format: 4-point Likert scale:
  • 1 = Strongly Agree
  • 2 = Agree
  • 3 = Disagree
  • 4 = Strongly Disagree
• Scoring and Transformation Rules:
  • Items are categorized into healthy descriptions (reflecting adaptive functioning) and unhealthy descriptions (reflecting systemic pathology).
  • To adhere to standard McMaster scoring where higher scores reflect greater pathology / unhealthy functioning, responses are scored as follows:
  • Unhealthy Items: Scored directly (1 = Strongly Agree, 2 = Agree, 3 = Disagree, 4 = Strongly Disagree). (Items: 2, 4, 6, 8, 9, 11, 12, 13, 14, 17, 20, 22, 24, 26, 29, 31, 33, 35, 38, 40, 41, 43, 45, 46, 49, 51, 53, 55, 57, 59).
  • Healthy Items (Reverse-Scored): Reverse scored (1 = 4, 2 = 3, 3 = 2, 4 = 1). In standard clinical scoring, the items requiring score reversal include: 1, 3, 5, 7, 10, 15, 16, 18, 19, 21, 23, 25, 27, 28, 30, 32, 34, 36, 37, 39, 42, 44, 47, 48, 50, 52, 54, 56, 58, 60.
  • Subscale Calculation: The score for each subscale is computed by dividing the sum of the transformed item responses by the total number of items endorsed on that subscale. This yields an index ranging from 1.00 (completely healthy) to 4.00 (severely unhealthy).
• Clinical Cut-off Scores (Pathology Thresholds): Scores exceeding the following validated clinical cut-offs indicate significant systemic impairment requiring clinical evaluation or family intervention:
  • Problem Solving: > 2.20
  • Communication: > 2.20
  • Roles: > 2.30
  • Affective Responsiveness: > 2.20
  • Affective Involvement: > 2.10
  • Behavior Control: > 1.90
  • General Functioning: > 2.00

11. Permissions & Fee and Test Year

The McMaster Family Assessment Device was originally published in 1983 by Nathan B. Epstein, Lawrence M. Baldwin, and Duane S. Bishop. The primary intellectual property rights associated with the FAD instrument and the McMaster Model of Family Functioning are managed through the Family Research Program at Butler Hospital and the Department of Psychiatry and Human Behavior at the Warren Alpert Medical School of Brown University.

For independent academic research, scholarly inquiry, and non-commercial educational use, the FAD is broadly made accessible without royalty fees, provided the test items are reproduced accurately without unauthorized modifications and that academic citations are appropriately credited. Commercial diagnostic software integrations, sponsored clinical trial applications, or corporate testing initiatives typically require written permission and formal licensing agreements from the copyright holders or institutional representatives at Butler Hospital/Brown University.

12. References

Epstein, N. B., Baldwin, L. M., & Bishop, D. S. (1983). The McMaster Family Assessment Device. Journal of Marital and Family Therapy, 9(2), 171–180. https://doi.org/10.1111/j.1752-0606.1983.tb01497.x

Kabacoff, R. I., Miller, I. W., Bishop, D. S., Epstein, N. B., & Keitner, G. I. (1990). A factor analytic study of the McMaster Family Assessment Device. Journal of Marital and Family Therapy, 16(1), 85–97. https://doi.org/10.1111/j.1752-0606.1990.tb00827.x

Maillette de Buy Wenniger, W. F., Hageman, W. J. J. M., & Arrindell, W. A. (1993). De McMaster Family Assessment Device (FAD). Gedrag & Gezondheid: Tijdschrift voor Psychologie en Gezondheid, 21(2), 79–86.

Miller, I. W., Epstein, N. B., Bishop, D. S., & Keitner, G. I. (1985). The McMaster Family Assessment Device: Reliability and validity. Journal of Marital and Family Therapy, 11(4), 345–356. https://doi.org/10.1111/j.1752-0606.1985.tb00028.x

Miller, I. W., Ryan, C. E., Keitner, G. I., Bishop, D. S., & Epstein, N. B. (2000). The McMaster Approach to Families: theory, assessment, treatment and research. Journal of Family Therapy, 22(2), 168–189. https://doi.org/10.1111/1467-6427.00145

Ryan, C. E., Epstein, N. B., Keitner, G. I., Miller, I. W., & Bishop, D. S. (2005). Evaluating and treating families: The McMaster approach. Routledge. https://doi.org/10.4324/9780203956106

13. Items of the Scale