Client Satisfaction Inventory (CSI)

Abstract

The Client Satisfaction Inventory (CSI), developed by Steven L. McMurtry and Walter W. Hudson (1994, 2000), is a psychometrically validated, standardized instrument designed to evaluate consumer satisfaction within clinical social work, psychological counseling, behavioral healthcare, and social service environments. Constructed to overcome the severe ceiling effects, psychometric instability, and lack of standardized norming characteristic of early, idiosyncratic satisfaction measures, the CSI provides an empirically grounded index of service recipients’ subjective appraisal of care. The complete instrument comprises 25 items evaluated along a 7-point Likert-type frequency response continuum ranging from 1 (“none of the time”) to 7 (“all of the time”). Five items are negatively valenced and reverse-scored to mitigate acquiescence response bias. In addition to the full 25-item version, a psychometrically validated 9-item short form (CSI-SF) is available for high-throughput clinical evaluation and rapid administrative monitoring.

Extensive psychometric investigations indicate that the CSI exhibits exceptional internal consistency, with Cronbach’s alpha coefficients routinely exceeding .94 for the full 25-item scale and .90 for the 9-item short form. Exploratory and confirmatory factor analyses demonstrate that while the scale captures nuanced service facets—such as provider competence, perceived outcome, interpersonal warmth, empowerment, and organizational respect—it functions robustly as an overarching unidimensional construct of generalized consumer satisfaction. The instrument utilizes a standard linear transformation formula producing a normalized composite score ranging from 0 to 100, where higher scores reflect superior levels of satisfaction. The CSI possesses strong convergent validity with established measures such as the Client Satisfaction Questionnaire (CSQ-8) and working alliance inventories, and displays robust predictive validity regarding treatment compliance, premature termination, and global clinical outcomes.

Keywords

Client Satisfaction Inventory, CSI, consumer satisfaction, psychometrics, social work evaluation, human services, clinical outcomes, patient-reported outcome measures, Walter W. Hudson, Steven L. McMurtry, service quality, therapeutic alliance

Authors

The Client Satisfaction Inventory was authored by Steven L. McMurtry, Ph.D., and the late Walter W. Hudson, Ph.D.

• Steven L. McMurtry, Ph.D.: Professor Emeritus at the Helen Bader School of Social Welfare, University of Wisconsin–Milwaukee. Dr. McMurtry’s scholarly trajectory has centered on child welfare, quantitative research methodology, computerized clinical assessment, and the psychometric evaluation of social work interventions. (Correspondence contact recorded during primary dissemination: [email protected]).
• Walter W. Hudson, Ph.D. (1934–1999): Formerly the Joyce M. F. Baker and Robert D. Baker Distinguished Professor of Social Work at the School of Social Work, Florida State University, and founder of the Walmyr Publishing Company. Dr. Hudson was a foundational figure in clinical social work assessment, internationally recognized for developing the WALMYR Assessment Scales (such as the Index of Self-Esteem, Index of Marital Satisfaction, and Clinical Measurement Package), pioneering single-case evaluation metrics and computerized clinical measurement.

Purpose

The primary purpose of the Client Satisfaction Inventory (CSI) is to provide human service professionals, mental health clinicians, program evaluators, and healthcare administrators with an objective, psychometrically rigorous, and standardized methodology for measuring service recipients’ evaluations of care. Historically, human services suffered from poorly conceived satisfaction surveys characterized by skewed response distributions, ill-defined measurement properties, and non-generalizable idiosyncratic items created for specific single-agency reviews. The CSI was created to rectify these operational deficiencies by delivering a standardized assessment instrument capable of tracking satisfaction across diverse provider contexts, organizational settings, and intervention modalities.

In clinical practice, the CSI functions as a vital component of continuous quality improvement (CQI) and outcome evaluation. Because human services operate within complex interpersonal paradigms, understanding the client’s perception of the working relationship, agency climate, and service efficacy is essential. Administering the CSI allows practitioners to identify subtle therapeutic ruptures, evaluate client engagement, and pinpoint organizational bottlenecks that compromise service delivery. The tool facilitates single-system design monitoring, enabling clinicians to assess how modifications in clinical technique or agency procedures influence the consumer’s subjective evaluation across time.

Within broader health service research and administrative oversight, the CSI serves critical compliance, accreditation, and program evaluation functions. Healthcare accreditation bodies—such as The Joint Commission and the Commission on Accreditation of Rehabilitation Facilities (CARF)—mandate the systematic collection of consumer-reported outcome measures (PROMs) and service evaluations. The CSI supplies an empirically validated metric capable of cross-agency benchmarking. Because its scoring is standardized onto a 0 to 100 metric, administrative bodies and researchers can compare satisfaction levels across distinct agency units, clinical programs, demographic strata, and socioeconomic cohorts, thereby establishing systemic performance targets and evaluating the comparative effectiveness of alternative programmatic models.

Psychological Construct

The psychological construct measured by the Client Satisfaction Inventory is client satisfaction with human and behavioral health services. Within psychometric literature, client satisfaction is conceptualized as a multi-dimensional cognitive and affective appraisal made by the recipient regarding the quality, processes, interpersonal dynamics, and tangible outcomes of clinical intervention. Rather than reflecting an objective audit of clinical technique, satisfaction captures the subjective congruence between a client’s pre-treatment expectations, values, and experienced reality.

Although the CSI is structured to provide an overall global index of satisfaction, it operationalizes several distinct, interconnected domains essential to therapeutic quality:

• Interpersonal Empathy and Working Alliance: Items such as “People here really seem to care about me” (Item 2), “People here accept me for who I am” (Item 11), and “People here seem to understand how I feel” (Item 21) evaluate the core interpersonal climate. This domain taps into the practitioner’s warmth, non-judgmental acceptance, and capacity to cultivate a safe, empathic bond.
• Perceived Efficacy and Clinical Outcome: Items like “The services I get here are a big help to me” (Item 1), “I feel much better now than when I first came here” (Item 12), and “People who know me say this place has made a positive change in me” (Item 19) assess symptom amelioration, behavioral improvement, and the pragmatic utility of the therapeutic intervention.
• Empowerment and Skill Acquisition: Modern therapeutic paradigms emphasize client autonomy. The CSI directly captures this via “I have learned a lot here about how to deal with my problems” (Item 6) and “The biggest help I get here is learning how to help myself” (Item 17), distinguishing between passive client dependency and active psychological empowerment.
• Provider Competence and Professionalism: Tapped by items like “People here really know what they are doing” (Item 9), measuring the consumer’s confidence in the provider’s technical skills and clinical acumen.
• Negative Agency Experiences and Autonomy Frustration: Reverse-scored items assess paternalism and marginalization, such as “People here want to do things their way, instead of helping me find my way” (Item 7), “People here put me down when I disagree with them” (Item 16), and “People here are only concerned about getting paid” (Item 22). These items detect coercive, bureaucratic, or mercenary dynamics that undermine consumer trust.
• Global Endorsement and Loyalty: Represented by behavioral intention markers, including “I would come back here if I need help again” (Item 3) and “I would recommend this place to people I care about” (Item 8).

Theoretical Framework

The theoretical architecture of the Client Satisfaction Inventory is rooted in several converging paradigms within social work, social exchange theory, healthcare quality modeling, and person-centered clinical theory.

Foremost among these is Donabedian’s Healthcare Quality Framework (Donabedian, 1988), which conceptualizes service quality along three interlocking dimensions: Structure (the physical, financial, and organizational context), Process (the interpersonal and technical manner in which care is delivered), and Outcome (the tangible change in client health status or well-being). The CSI deliberately emphasizes the Process and Outcome components. By focusing on how providers interact with clients (process) and whether the client experiences functional relief (outcome), the CSI operationalizes Donabedian’s assertion that consumer appraisal of care represents an indispensable element of clinical validation.

A second conceptual foundation is drawn from Expectancy-Disconfirmation Theory (Oliver, 1980). This model posits that consumer satisfaction is an affective reaction generated through a cognitive comparison process: clients arrive with baseline expectations regarding therapeutic efficacy and professional conduct. When perceived performance surpasses these benchmarks, positive disconfirmation occurs, eliciting elevated satisfaction (captured by Item 24: “The help I get here is better than I expected”). Conversely, negative disconfirmation yields dissatisfaction and clinical disengagement.

Finally, the scale reflects Carl Rogers’ Person-Centered Humanistic Theory (Rogers, 1957). Central to Rogers’ framework are the core therapeutic conditions: unconditional positive regard, congruence, and empathic understanding. The CSI operationalizes these conditions directly by assessing whether clients feel treated as unique individuals rather than depersonalized statistics (Item 5), whether their perspectives are validated or disparaged (Item 16), and whether their self-determination is honored rather than overridden by provider authority (Item 7).

Validity

The psychometric validation of the Client Satisfaction Inventory (McMurtry & Hudson, 2000) established robust empirical support across multiple validity dimensions using diverse, multi-site human service samples encompassing outpatient mental health, child and family welfare, substance abuse recovery, and medical social work settings.

Construct and Structural Validity

Construct validity was demonstrated by examining the degree to which items conformed to theoretical expectations regarding client-provider relationships and perceived service efficacy. High inter-item correlations (median $r > .50$) and uniformly strong item-total correlations confirm that the items coherently measure a unified psychological construct. Reverse-worded items correlated negatively with positively phrased statements prior to recoding and mapped cleanly onto the unified construct post-transformation, confirming that the scale captures genuine variance rather than method artifact.

Convergent Validity

The CSI displays robust convergent validity with alternative measures of consumer satisfaction and clinical alliance. In concurrent validity evaluations, the CSI correlated strongly with the 8-item Client Satisfaction Questionnaire (CSQ-8; Larsen et al., 1979), with coefficients typically exceeding $r = .80$. Furthermore, the CSI demonstrates significant positive correlations with therapeutic alliance inventories—such as the Working Alliance Inventory (WAI)—particularly along the Goal and Bond subscales ($r = .65$ to $.75$). This confirms that higher scores on the CSI meaningfully reflect collaborative, supportive clinical partnerships.

Discriminant and Criterion-Related Validity

Discriminant validity has been corroborated by demonstrating that the CSI does not merely reflect generalized positive affectivity or transient mood states. Moderate, non-redundant correlations are observed between CSI scores and standardized measures of baseline psychopathology or symptom severity (e.g., the Beck Depression Inventory or Generalized Anxiety Scale), indicating that clients can experience severe symptomatology while simultaneously evaluating their services and providers as exceptionally satisfactory.

In terms of criterion-related and predictive validity, CSI scores reliably differentiate between clients who complete planned therapeutic courses and those who drop out prematurely. Longitudinal outcome tracking indicates that elevated CSI scores administered mid-treatment predict greater appointment adherence, higher rates of homework completion in cognitive-behavioral paradigms, and superior post-discharge functional status.

Reliability

The Client Satisfaction Inventory demonstrates exceptionally high reliability across independent clinical samples and diverse service populations.

Internal Consistency

In the foundational validation study by McMurtry and Hudson (2000), the 25-item full version achieved a Cronbach’s alpha ($lpha$) coefficient between .94 and .97 across distinct client cohorts. Split-half reliability coefficients, adjusted via the Spearman-Brown prophecy formula, similarly exceeded .95. Item-total correlations across the 25 items range from .48 to .83, with the vast majority falling above .60, indicating that each item contributes substantial common variance to the underlying construct.

The 9-item short form (CSI-SF), selected on the basis of optimal psychometric performance and representative item-total loadings, consistently demonstrates Cronbach’s alpha values ranging from .90 to .93. This confirms that the abbreviated version achieves rapid administration without sacrificing measurement precision.

Standard Error of Measurement and Stability

Owing to its elevated internal consistency, the CSI yields an exceptionally small Standard Error of Measurement (SEM). On the normalized 0 to 100 scale, the SEM is typically estimated at under 4.0 points, providing clinicians with narrow, reliable confidence intervals when evaluating individual client scores. Test-retest evaluations over brief stability intervals (e.g., 7 to 14 days) in stable service environments demonstrate reliability coefficients between $r_{tt} = .82$ and $.89$, establishing that the instrument captures enduring evaluative judgments rather than transient day-to-day fluctuations.

Factor Analysis

The latent dimensionality of the Client Satisfaction Inventory has been scrutinized via both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across diverse validation datasets.

Exploratory Factor Analysis (EFA)

Initial principal components and maximum likelihood exploratory factor analyses conducted by McMurtry and Hudson revealed an overwhelmingly dominant first unrotated factor. In unrotated factor solutions, the primary eigenvalue typically accounts for over 50% to 60% of the total variance, whereas subsequent factors exhibit eigenvalues dropping sharply below 1.5. In oblique (Promax) and orthogonal (Varimax) rotations, secondary factors sometimes emerge reflecting distinguishable sub-domains: Interpersonal Alliance/Respect, Problem-Solving/Outcome, and Agency Policies/Negative Experiences. However, these factors correlate intensely with one another ($r > .60$), indicating that multi-factor solutions represent sub-facets of a single overarching general factor.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations evaluating a unidimensional model have demonstrated acceptable to good model fit across clinical cohorts, supporting the aggregation of items into a single composite score. When accounting for residual covariance among identically worded negative items, structural equation modeling yields excellent goodness-of-fit indices:

• Comparative Fit Index (CFI) $ge .94$
• Tucker-Lewis Index (TLI) $ge .93$
• Root Mean Square Error of Approximation (RMSEA) $le .06$
• Standardized Root Mean Square Residual (SRMR) $le .05$

Standardized factor loadings for the 25 items on the primary satisfaction dimension are predominantly high, ranging from $lambda = .55$ to $lambda = .88$. The 9 items retained for the short form (CSI-SF) uniformly exhibit factor loadings exceeding $.70$, confirming their status as core indicators of the underlying client satisfaction construct.

Instrument / Measurement Tool

The operational characteristics and administrative guidelines for the Client Satisfaction Inventory are summarized below:

• Test Type: Standardized, self-administered Patient-Reported Outcome Measure (PROM) / Consumer Evaluative Rating Scale.
• Target Population: Adolescents and adults receiving mental health, counseling, clinical social work, medical case management, or social welfare services.
• Administration Modality: Paper-and-pencil questionnaire or computerized/electronic client survey portal. Typically completed independently by the client in privacy to minimize social desirability bias.
• Administration Time: Approximately 3 to 5 minutes for the 25-item full inventory; under 2 minutes for the 9-item short form (CSI-SF).
• Item Count:
  • Full Version (CSI): 25 items.
  • Short Form (CSI-SF): 9 items (Items 2, 3, 8, 9, 10, 11, 21, 23, and 24).
• Response Continuum: 7-point Likert-type frequency scale:
  • 1 = None of the time
  • 2 = Very rarely
  • 3 = A little of the time
  • 4 = Some of the time
  • 5 = A good part of the time
  • 6 = Most of the time
  • 7 = All of the time
• Reverse Scoring Rules: Five items on the full 25-item version reflect negative experiences and must be reverse-coded prior to composite score calculation:
  • Items to Reverse: Items 4, 7, 16, 18, and 22.
  • Transformation Algorithm: $\text{Recoded Value} = 8 – \text{Raw Response}$. (e.g., an original score of 1 becomes 7; an original score of 7 becomes 1).
• Standardized Scoring Formula: The instrument uses the standard Hudson scale linear transformation to convert raw sums into a 0 to 100 metric:

  $$S = \frac{(\sum Y – N) \times 100}{N \times 6}$$

  Where:

  • $S$ = Final standardized score (0 to 100).
  • $Y$ = Item response value after reverse-scoring recodes.
  • $\sum Y$ = Sum of all valid item responses.
  • $N$ = Number of items completed by the client (allows adjustment for missing items, provided at least 80% of items are completed).
  • $6$ = Number of scale intervals ($k – 1 = 7 – 1 = 6$).
• Clinical Interpretation: Scores range from 0 to 100. Higher scores reflect greater satisfaction. Scores above 75 generally denote solid client satisfaction, whereas scores below 50 reflect marked dissatisfaction, warranting clinical review or administrative inquiry.

Permissions & Fee and Test Year

The Client Satisfaction Inventory was authored in 1994 by Steven L. McMurtry and Walter W. Hudson and was formally detailed in Research on Social Work Practice in 2000. The scale was originally published and copyrighted by the Walmyr Publishing Company (Tallahassee, FL) as part of the WALMYR Assessment Scales portfolio developed by Walter W. Hudson.

For research, educational, and non-commercial clinical outcome monitoring, the scale has been widely made accessible via academic repositories, clinical training programs, and university consortia. Investigators wishing to use the scale in proprietary commercial applications, electronic health record (EHR) integrations, or enterprise healthcare systems should review copyright requirements and seek authorization through the Helen Bader School of Social Welfare at the University of Wisconsin–Milwaukee or through the representative estate and publishers of the Walmyr Assessment Scales.

References

• Attkisson, C. C., & Zwick, R. (1982). The Client Satisfaction Questionnaire: Psychometric properties and correlations with service utilization and psychotherapy outcome. Evaluation and Program Planning, 5(3), 233–237. https://doi.org/10.1016/0149-7189(82)90074-X
• Donabedian, A. (1988). The quality of care: How can it be assessed? JAMA, 260(12), 1743–1748. https://doi.org/10.1001/jama.1988.03410120089033
• Larsen, D. L., Attkisson, C. C., Hargreaves, W. A., & Nguyen, T. D. (1979). Assessment of client/patient satisfaction in human service programs. Evaluation and Program Planning, 2(3), 197–207. https://doi.org/10.1016/0149-7189(79)90094-4
• McMurtry, S. L. (1994). Client Satisfaction Inventory (CSI). Tallahassee, FL: Walmyr Publishing Company.
• McMurtry, S. L., & Hudson, W. W. (2000). The Client Satisfaction Inventory: Results of an initial validation study. Research on Social Work Practice, 10(5), 644–663. https://doi.org/10.1177/104973150001000506
• Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research, 17(4), 460–469. https://doi.org/10.1177/002224378001700405
• Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21(2), 95–103. https://doi.org/10.1037/h0045357

13. Items of the Scale (Questionnaire)