Therapy Attitude Inventory (TAI)

Abstract

The Therapy Attitude Inventory (TAI) is a widely utilized, standardized psychometric instrument developed to evaluate parental consumer satisfaction and perceived treatment efficacy following participation in behavioral parent training and child psychotherapy interventions, particularly Parent-Child Interaction Therapy (PCIT). Conceptualized within the paradigm of social validity and behavioral evaluation established by Sheila M. Eyberg and empirically standardized by Elizabeth V. Brestan, Jerry R. Jacobs, Allison Rayfield, and Eyberg (1999), the TAI addresses a critical dimension of evidence-based practice: the degree to which consumers perceive intervention techniques, therapeutic processes, and clinical outcomes as acceptable, valuable, and functionally beneficial. The standard instrument comprises 10 core items administered upon treatment termination, with each item utilizing an authentic 5-point Likert scale featuring item-specific semantic anchors (ranging from 1 to 5). Total composite scores range from 10 to 50, where higher scores reflect superior levels of satisfaction, perceived acquisition of parenting competencies, attenuation of child disruptive behaviors, and global systemic improvement within the home environment. Extensive psychometric evaluations demonstrate robust internal consistency across diverse clinical trials, with Cronbach’s alpha coefficients consistently falling between α = .88 and α = .94, alongside high test-retest reliability and strong convergent validity with objective behavioral outcome metrics, standardized parent-report behavior checklists (e.g., the Eyberg Child Behavior Inventory), and direct observational coding systems. Exploratory and confirmatory factor analyses generally substantiate either a robust unidimensional construct of general therapeutic satisfaction or a two-factor dimensional model demarcating satisfaction with specific disciplinary/teaching techniques from perceived relational and child behavioral change. This article presents an exhaustive psychometric review of the TAI, delineating its historical emergence, theoretical foundations, psychometric architecture, factor analytic parameters, scoring mechanics, and clinical research utilities across pediatric mental health settings.

Keywords

Therapy Attitude Inventory, TAI, Parent-Child Interaction Therapy, PCIT, Sheila Eyberg, consumer satisfaction, social validity, parent training, disruptive behavior disorders, treatment acceptability, psychometrics, treatment outcome

Authors

The foundational development, psychometric standardization, and empirical validation of the Therapy Attitude Inventory were conducted by prominent clinical child psychologists and behavioral researchers affiliated with the Department of Clinical and Health Psychology at the University of Florida and subsequent collaborative institutions:

• Elizabeth V. Brestan, Ph.D. – Department of Psychology, Auburn University, Auburn, Alabama; recognized authority in evidence-based child psychotherapies, Parent-Child Interaction Therapy, and psychosocial interventions for pediatric conduct problems.
• Jerry R. Jacobs, Ph.D. – Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida; clinical investigator in pediatric psychology, family-based behavior modification, and pediatric treatment evaluation.
• Allison Rayfield, Ph.D. – Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida; specialist in child clinical psychology and behavioral family intervention assessment.
• Sheila M. Eyberg, Ph.D., ABPP – Distinguished Professor Emerita, Department of Clinical and Health Psychology, University of Florida, Gainesville, Florida; developer of Parent-Child Interaction Therapy (PCIT), the Eyberg Child Behavior Inventory (ECBI), and pioneering investigator in behavioral assessment and child psychotherapy research.

Purpose

The primary purpose of the Therapy Attitude Inventory (TAI) is to systematically evaluate parental consumer satisfaction, treatment acceptability, and perceived clinical utility following behavioral family interventions. Historically, treatment outcome research in child clinical psychology focused almost exclusively on symptom reduction, utilizing parent-report symptom inventories and clinician-administered diagnostic schedules. However, following Alan E. Kazdin’s seminal work on social validity, researchers and clinicians recognized that the sustainability, ethical acceptability, and dissemination of behavioral parent training programs are inextricably bound to consumer perceptions. Interventions that produce robust statistical effect sizes in laboratory settings are unlikely to achieve long-term ecological adherence if parents find the methodologies aversive, the time demands unmanageable, or the relational outcomes negligible.

Within this empirical landscape, the TAI was engineered to quantify several distinct domains of client satisfaction that are critical to both clinical practice and clinical trial methodologies:

• Perceived Acquisition of Specific Parenting Competencies: Gauging the degree to which caregivers believe they acquired concrete behavioral management techniques, including clear direct commands, systematic praise, differential attention, and consistent non-violent discipline mechanisms such as time-out procedures.
• Child Behavioral and Relational Amelioration: Evaluating whether the intervention yielded tangible reductions in targeted disruptive behaviors, improved child compliance, and enhanced the affective quality of the parent-child attachment dyad.
• Generalization to Home and Family Dynamics: Measuring secondary treatment gains, including systemic improvements in the general home atmosphere and collateral benefits extending to untreated personal or familial stressors.
• Global Program Evaluation: Providing an overarching appraisal of the intervention structure, therapeutic modalities (such as live bug-in-the-ear coaching), and the professional therapeutic alliance.

In clinical trials, the TAI serves as a core outcome metric assessing post-intervention social validity, benchmarking satisfaction levels across different delivery formats (e.g., standard PCIT versus home-based PCIT, group adaptations, or telehealth modalities). In routine clinical practice, the TAI offers vital feedback for continuous quality improvement, identifying potential ruptures in therapeutic alliance, areas where skill acquisition was deficient, or cases where behavioral improvements failed to translate into parental peace of mind.

Psychological Construct

The Therapy Attitude Inventory measures treatment acceptability and consumer satisfaction within the context of family-centered psychological healthcare. In psychometrics and clinical psychology, consumer satisfaction is not conceptualized as a superficial affective reaction or general customer approval; rather, it represents a complex, multi-dimensional cognitive and evaluative appraisal of therapeutic processes, procedural burdens, relational alliances, and functional therapeutic outcomes. Within the TAI, this overarching construct is operationalized through three primary psychological facets:

1. Instrumental Skill Acquisition and Self-Efficacy

The first construct dimension reflects parental mastery and functional behavioral competence. Rooted in Albert Bandura’s social cognitive theory of self-efficacy, parental competence requires both the behavioral acquisition of specific child management strategies and the internal cognitive belief that one can successfully execute those strategies under conditions of high emotional arousal and child defiance. Items 1, 2, and 4 capture this dimension directly by probing the acquisition of discipline techniques, skill-teaching modalities, and perceived parenting confidence. Parents who score high on this dimension perceive that the therapy transformed their parenting repertoire from reactive, coercive, or inconsistent practices into predictable, authoritative, and proactive competencies.

2. Perceived Behavioral and Systemic Change

The second dimension addresses the clinical efficacy of the intervention as experienced within the naturalistic home ecosystem. Symptom severity scales often measure frequency counts of problematic behaviors (e.g., how many tantrums occurred in the past week). In contrast, the TAI assesses the ecological and functional impact of that behavior change on the broader familial system. Items 3, 5, 6, and 8 evaluate improvements in child compliance, reductions in global behavioral severity, qualitative improvements in the parent-child dyadic bond, and the global emotional tone within the household. This dimension taps into systemic relief: the transformation of a chaotic, conflict-laden home environment into a predictable, nurturing space characterized by mutual warmth and mutual responsiveness.

3. Treatment Feasibility, Utility, and Programmatic Acceptability

The third dimension evaluates the procedural acceptability of the intervention. Treatment acceptability concerns whether the consumer views the treatment procedures as fair, appropriate, non-intrusive, and reasonable for the targeted problem. Items 7, 9, and 10 assess overall feelings toward the program modality, perceived therapeutic progress, and overarching helpfulness. In intensive behavioral treatments like PCIT—which require live, in-vivo behavioral coaching, intensive parental homework, and rigorous performance criteria before graduation—high consumer satisfaction demonstrates that the intervention demands were perceived as worthwhile investments that justified the required emotional and behavioral expenditure.

Theoretical Framework

The theoretical architecture of the Therapy Attitude Inventory is anchored at the convergence of three foundational paradigms in psychology: Kazdin’s model of social validity, Patterson’s coercive family process model, and Bandura’s social learning and self-efficacy theory.

Social Validity in Behavioral Intervention

In applied behavior analysis and clinical child psychology, social validity was formally introduced by Montrose Wolf (1978) and operationalized for clinical child interventions by Alan E. Kazdin. Kazdin posited that any treatment evaluation must answer three fundamental questions:

1. Are the specific behavioral goals what society and consumers actually want to focus on?
2. Are the particular intervention procedures acceptable to the participants and not considered excessively intrusive, painful, or restrictive?
3. Do the changes produced by the intervention make a genuine, meaningful difference in the participant’s day-to-day life?

The TAI was explicitly designed to capture the second and third components of Kazdin’s triad. Eyberg and colleagues recognized that behavioral interventions for oppositional and conduct-disordered children often fail during maintenance phases not because the behavioral principles are invalid, but because parents reject the procedures as unnatural, overly demanding, or emotionally exhausting. By standardizing the measurement of these subjective appraisals, the TAI provides empirical data on the social validity of parent training protocols.

Patterson’s Coercive Family Interaction Model

Gerald Patterson’s coercion model explains how disruptive child behaviors (such as tantrums, aggression, and defiance) are maintained through bidirectional cycles of negative reinforcement. In distressed families, parental commands are met with child escalation; parents often withdraw demands to escape the escalation, unintentionally reinforcing child defiance. Alternatively, parents escalate to intense emotional outbursts, physical aggression, or harsh punishment, modeling coercive conflict resolution. The theoretical premise of interventions like PCIT is that disruption of these coercive cycles requires two distinct phases: establishing a positive, warm relational foundation (Child-Directed Interaction or CDI) followed by structured, predictable, and non-coercive disciplinary consistency (Parent-Directed Interaction or PDI). The TAI directly mirrors this two-stage model: its items evaluate both the affective, relational connection between parent and child (CDI goals) and the parent’s confidence in establishing behavioral compliance and non-aversive discipline (PDI goals).

Social Learning and Self-Efficacy Theory

Finally, Bandura’s framework underscores that enduring behavioral maintenance requires high perceived self-efficacy. When parents feel helpless, their discipline becomes erratic and vacillates between extreme permissiveness and extreme hostility. Interventions succeed when they alter the caregiver’s perceived coping capabilities. By evaluating parental confidence in discipline and teaching, the TAI measures whether the intervention successfully restructured parental cognitive appraisals and established genuine internal agency.

Validity

The psychometric validity of the Therapy Attitude Inventory has been rigorously demonstrated across multiple clinical, community, and efficacy studies over the past three decades.

Construct Validity

Construct validity was established in the seminal psychometric validation study by Brestan, Jacobs, Rayfield, and Eyberg (1999). In this investigation, the authors examined the TAI across a clinically referred sample of children aged 2 to 7 diagnosed with oppositional defiant disorder (ODD) or conduct disorder (CD). The findings demonstrated that TAI composite scores meaningfully differentiated between treatment completers and treatment dropouts. Parents who completed the full course of PCIT demonstrated uniformly high TAI scores (mean scores typically exceeding 44 on the 50-point scale), reflecting profound perceived utility. Furthermore, construct validity was supported by significant differences in TAI scores between families receiving the full intervention versus waitlist or minimal-contact control conditions upon study conclusion.

Convergent Validity

Convergent validity has been repeatedly demonstrated through significant, moderate-to-strong correlations between the TAI and standardized measures of child behavioral functioning and parenting stress:

• Eyberg Child Behavior Inventory (ECBI): Brestan et al. (1999) reported statistically significant negative correlations between the TAI total score and the post-treatment ECBI Intensity Scale (r = -.38 to -.52, p < .001) as well as the ECBI Problem Scale (r = -.44 to -.58, p < .001). Greater parental satisfaction directly correlates with fewer reported child behavior problems and lower perceived problem severity.
• Parenting Stress Index (PSI): TAI scores correlate negatively with post-treatment scores on the Parenting Stress Index (r = -.35 to -.48, p < .01), showing that higher treatment satisfaction is associated with marked reductions in parental distress, parent-child dysfunctional interaction, and perceived child difficulty.
• Direct Behavioral Observation (DPICS): Research employing the Dyadic Parent-Child Interaction Coding System (DPICS) has shown that parents scoring higher on the TAI display significantly higher frequencies of labeled praise, reflections, and behavioral descriptions during observed play interactions, alongside significant reductions in critical statements and inappropriate commands.

Discriminant and Predictive Validity

Discriminant validity is evidenced by the TAI’s ability to isolate treatment satisfaction from general social desirability response biases. Studies examining correlations between the TAI and the Marlowe-Crowne Social Desirability Scale have shown non-significant or negligible associations (r < .15), demonstrating that high TAI scores do not merely reflect a broad tendency to provide socially acceptable responses. In terms of predictive validity, longitudinal follow-up studies (e.g., Brestan et al., 2005; Eyberg et al., 2001) revealed that higher post-treatment TAI scores significantly predict long-term maintenance of treatment gains at 1-year and 3-year post-treatment follow-ups, lower rates of recidivism in child welfare-involved families, and lower rates of subsequent clinical re-referrals.

Reliability

The Therapy Attitude Inventory possesses robust, repeatedly replicated indices of psychometric reliability across diverse demographic cohorts, clinical populations, and international translations.

Internal Consistency

Internal consistency of the TAI is exceptionally high, particularly for a brief 10-item scale:

• In the definitive validation study by Brestan, Jacobs, Rayfield, and Eyberg (1999), the Cronbach’s alpha coefficient for the overall scale was α = .91, reflecting outstanding item homogeneity and minimal measurement error.
• Subsequent randomized controlled trials evaluating PCIT in diverse settings (e.g., Nixon, Sweeney, Erickson, & Touyz, 2003; Thomas & Zimmer-Gembeck, 2007) have reported internal consistency coefficients ranging consistently between α = .88 and α = .94.
• Corrected item-total correlations across published studies consistently exceed .50, with most items falling between r = .58 and r = .82, indicating that every individual item contributes robustly to the overarching latent construct without introducing extraneous variance.

Test-Retest Reliability and Stability

Because the TAI is designed to assess attitudes toward an intervention upon or following completion, traditional test-retest reliability over long intervals must account for the natural attenuation of immediate post-therapy enthusiasm. However, when evaluated over short intervals (e.g., 2 to 4 weeks post-treatment without additional clinical contact), the TAI exhibits high stability, with test-retest reliability coefficients typically reported at r = .85 to .90 (p < .001). This demonstrates that parent evaluations captured by the instrument represent stable evaluative attitudes rather than fleeting, momentary emotional states.

Factor Analysis

The structural dimensionality of the Therapy Attitude Inventory has been investigated through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), demonstrating clean, theoretically coherent psychometric architectures.

Exploratory Factor Analysis (EFA)

In the primary EFA conducted by Brestan et al. (1999) using principal axis factoring with promax (oblique) rotation, eigenvalues and the scree plot test supported a predominant single-factor solution accounting for approximately 55% to 62% of the total variance. Factor loadings for all 10 items on this general consumer satisfaction factor were uniformly high, ranging from .54 to .88:

• Item 1 (Techniques of disciplining): .74
• Item 2 (Techniques for teaching skills): .68
• Item 3 (Parent-child relationship): .65
• Item 4 (Confidence in discipline): .78
• Item 5 (Child compliance): .79
• Item 6 (Severity of behavior problems): .71
• Item 7 (Type of program): .82
• Item 8 (Feeling in our home): .69
• Item 9 (Progress in therapy): .86
• Item 10 (Feelings about therapy program): .88

Two-Factor Dimensional Models

Although the unidimensional model is most frequently utilized in clinical scoring, secondary factor analytic investigations (e.g., Reeve, 2014) have identified a robust two-factor oblique structure representing distinct conceptual facets:

1. Factor 1: Satisfaction with Specific Behavioral Techniques and Confidence (Items 1, 2, 4, 7, 9, 10): This factor captures the parent’s procedural satisfaction with the behavioral management curriculum, mastery of parenting techniques, confidence in discipline, and overall programmatic evaluation.
2. Factor 2: Perception of Child Behavior and Relational Improvement (Items 3, 5, 6, 8): This factor captures perceived ecological outcomes, encompassing child compliance, reduction in problem severity, dyadic warmth, and systemic emotional climate at home.

These two latent factors are moderately to highly correlated (r ≈ .60 to .72), confirming that while procedural satisfaction and perceived outcome are conceptually distinct, they operate synergistically in determining global parental treatment acceptability.

Confirmatory Factor Analysis (CFA) Fit Indices

Confirmatory factor analyses evaluating the 10-item unidimensional and two-factor models have consistently confirmed good-to-excellent model fit across clinical pediatric samples. Standard goodness-of-fit metrics reported in the literature include:

• Comparative Fit Index (CFI) ≥ .94 to .97
• Tucker-Lewis Index (TLI) ≥ .92 to .96
• Root Mean Square Error of Approximation (RMSEA) ≤ .05 to .07 (with 90% confidence intervals spanning .03 to .08)
• Standardized Root Mean Square Residual (SRMR) ≤ .04

These structural findings affirm that whether interpreted as a single global index or across granular subdimensions, the TAI possesses exceptional construct integrity.

Instrument / Measurement Tool

The standard Therapy Attitude Inventory is structured as follows:

• Instrument Name: Therapy Attitude Inventory (TAI)
• Instrument Type: Standardized, post-intervention consumer satisfaction and treatment acceptability rating scale
• Respondent: Parents, primary caregivers, or legal guardians of pediatric clients
• Target Population: Families undergoing behavioral parent training, parent-child relational therapy, or pediatric behavior management programs (most notably Parent-Child Interaction Therapy)
• Number of Items: 10 core items (with optional supplemental modules assessing coaching modality, cultural comfort, and community acceptance)
• Administration Format: Paper-and-pencil self-report or computer-administered digital assessment
• Administration Time: Approximately 3 to 5 minutes
• Response Scale: 5-point Likert scale (item-specific response anchors ranging from 1 to 5, where higher scores indicate greater satisfaction)
• Scoring Mechanics:
  • Total score is obtained by summing the ratings for all 10 items.
  • Composite scores range from a minimum of 10 to a maximum of 50.
  • All 10 items are positively keyed (where 1 represents the least favorable/negative response and 5 represents the most favorable/positive response); no items require reverse scoring.
  • Higher scores reflect greater parental satisfaction with treatment, greater perceived acquisition of behavioral management skills, and greater perceived amelioration of child behavioral difficulties.
  • In clinical and research practice, scores from 40 to 50 generally denote high-to-exceptional treatment acceptability and satisfaction, scores from 30 to 39 reflect moderate or equivocal satisfaction, and scores below 30 signal significant dissatisfaction, procedural rejection, or lack of perceived clinical benefit.

Permissions & Fee and Test Year

The Therapy Attitude Inventory was developed in the late 1970s and 1980s under the leadership of Dr. Sheila M. Eyberg, with the definitive standardization and psychometric validation study published in 1999 (Brestan, Jacobs, Rayfield, & Eyberg, 1999). In accordance with the open-science and dissemination ethics established by Dr. Eyberg and the PCIT community, the Therapy Attitude Inventory is widely recognized as an open-access clinical research measure.

The instrument is made freely accessible for non-commercial clinical, training, and academic research purposes. It can be accessed through the official PCIT training resources, university repositories, and academic archives (such as the UC Davis PCIT Training Center and Western Michigan University ScholarWorks). Researchers and clinicians are not required to pay royalty fees or purchase proprietary test forms to utilize the standard 10-item TAI. However, proper bibliographic attribution to the original authors (Brestan et al., 1999) is required in all scientific presentations, academic publications, and clinical documentation. Modification of item stems or response anchors without explicit psychometric validation is strongly discouraged, as it undermines the established psychometric properties of the scale.

References

• Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
• Brestan, E. V., & Eyberg, S. M. (1998). Effective psychosocial treatments of conduct-disordered children and adolescents: 29 years, 82 studies, 5,272 kids. Journal of Clinical Child Psychology, 27(2), 180–189. https://doi.org/10.1207/s15374424jccp2702_5
• Brestan, E. V., Eyberg, S. M., Boggs, S. R., & Algina, J. (1997). Parent-Child Interaction Therapy: Parents’ perceptions of untreated siblings. Child & Family Behavior Therapy, 19(3), 13–28. https://doi.org/10.1300/J019v19n03_02
• Brestan, E. V., Jacobs, J. R., Rayfield, A. D., & Eyberg, S. M. (1999). A consumer satisfaction measure for parent-child treatments and its relationship to measures of child behavior change. Behavior Therapy, 30(1), 17–30. https://doi.org/10.1016/S0005-7894(99)80043-4
• Brestan, E. V., Eyberg, S. M., Boggs, S. R., Jacobs, J. R., & Algina, J. (2005). Outcomes of Parent-Child Interaction Therapy: A comparison of treatment completers and study dropouts one to three years later. Child & Family Behavior Therapy, 26(4), 1–22. https://doi.org/10.1300/J019v26n04_01
• Eyberg, S. M., & Robinson, E. A. (1982). Conduct problem behavior: Standardization of a behavioral rating scale with adolescents. Journal of Clinical Child Psychology, 11(2), 120–126. https://doi.org/10.1080/15374418209533074
• Eyberg, S. M., Funderburk, B. W., Hembree-Kigin, T. L., McNeil, C. B., Querido, J. G., & Hood, K. K. (2001). Parent-Child Interaction Therapy with behavior problem children: One and two year maintenance of treatment effects in multiple domains. Child & Family Behavior Therapy, 23(4), 1–20. https://doi.org/10.1300/J019v23n04_01
• Kazdin, A. E. (1977). Assessing the clinical or applied importance of behavior change through social validation. Behavior Modification, 1(4), 427–452. https://doi.org/10.1177/014544557714001
• Nixon, R. D., Sweeney, L., Erickson, D. B., & Touyz, S. W. (2003). Parent-Child Interaction Therapy: A comparison of standard and abbreviated treatments for oppositional defiant preschoolers. Journal of Consulting and Clinical Psychology, 71(2), 251–260. https://doi.org/10.1037/0022-006X.71.2.251
• Patterson, G. R. (1982). Coercive family process. Castalia Publishing Company.
• Reeve, C. S. (2014). Efficacy of Parent-Child Interaction Therapy with the use of in-room coaching (Dissertation No. 313). Western Michigan University ScholarWorks. https://scholarworks.wmich.edu/dissertations/313
• Thomas, R., & Zimmer-Gembeck, M. J. (2007). Behavioral outcomes of Parent-Child Interaction Therapy and Triple P—Positive Parenting Program: A review and meta-analysis. Journal of Abnormal Child Psychology, 35(3), 475–495. https://doi.org/10.1007/s10802-007-9104-9
• Wolf, M. M. (1978). Social validity: The case for subjective measurement or how applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11(2), 203–214. https://doi.org/10.1901/jaba.1978.11-203

Items of the Scale