Abstract
The Assessment Questionnaire-2 (AQ-2) is a specialized psychometric instrument developed by Stephen E. Finn, David G. Schroeder, and Mary E. Tonsager (1994) to evaluate clients’ subjective, cognitive, and affective experiences during psychological assessment feedback sessions. Rooted in the paradigm of Therapeutic Assessment (TA) and Collaborative Psychological Assessment, the instrument was conceived as a structural revision and empirical expansion of the original Assessment Questionnaire (AQ; Finn & Tonsager, 1992). The standard version consists of 48 self-report items—with a widely recognized 45-item experimental and clinical adaptation formulated by Andrea Allen (2001)—measured on a 5-point Likert response scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree).
The instrument operationalizes client appraisal across four psychometrically distinct yet theoretically unified dimensions: New Self-Awareness/Understanding, Positive Accurate Mirroring, Positive Relationship with the Assessor, and Negative Feelings about the Assessment. Extensive empirical evaluations across diverse clinical and analog populations demonstrate robust internal consistency reliability coefficients across all subscales, with Cronbach’s alpha values typically ranging from .79 to .94. Confirmatory and exploratory factor analyses affirm the multidimensional structural validity of the measure, showing how client perceptions of systemic warmth, collaborative relational safety, and psychological attunement facilitate cognitive restructuring and alleviate clinical distress. Today, the AQ-2 serves as a gold standard process-and-outcome research instrument and quality-assurance mechanism in psychological assessment, yielding deep empirical insights into how individualized feedback fosters therapeutic alliance and psychological transformation.
Keywords
Assessment Questionnaire-2, AQ-2, Therapeutic Assessment, Collaborative Assessment, Psychological Assessment Process, Client Experience, MMPI-2 Feedback, Self-Awareness, Positive Accurate Mirroring, Assessor-Client Alliance, Stephen E. Finn, Psychometrics
Authors
The Assessment Questionnaire-2 was formulated by Stephen E. Finn, Ph.D., David G. Schroeder, Ph.D., and Mary E. Tonsager, Ph.D. in 1994 as an unpublished instrument advancing the empirical study of psychological testing feedback interventions.
- Stephen E. Finn, Ph.D.: Founder of the Center for Therapeutic Assessment in Austin, Texas, and Clinical Associate Professor of Psychology at the University of Texas at Austin. Dr. Finn pioneered the Therapeutic Assessment model, transforming traditional psychological testing from a purely diagnostic, information-gathering exercise into a collaborative, semi-structured clinical intervention.
- David G. Schroeder, Ph.D.: Clinical psychologist and researcher who collaborated closely with Dr. Finn during the 1990s in operationalizing assessment process variables, psychometric scale refinement, and evaluating client reactions to standardized personality assessment protocols.
- Mary E. Tonsager, Ph.D.: Co-author of seminal empirical trials establishing the therapeutic efficacy of test feedback (notably Finn & Tonsager, 1992). Her collaborative scholarship provided the baseline psychometric groundwork demonstrating that providing clients with interactive personality feedback produces rapid, statistically significant reductions in symptomatic distress and marked increases in self-esteem.
- Andrea Allen, Ph.D.: Contributed significantly to the instrument’s validation through her 2001 psychometric adaptation and experimental inquiry at Florida International University, standardizing a 45-item adaptation that calibrated the scale for experimental paradigms examining the informational and relational effects of personality feedback.
Purpose
Historically, psychological testing was practiced under a strict “information-gathering” model. In this classical paradigm, the psychologist functioned as an objective, detached observer who administered tests, scored protocols, interpreted standardized norms, and delivered a formal written report primarily intended for third-party referral sources, often leaving the client passive, alienated, or anxious. Recognizing that psychological evaluation inherently possesses potent relational and therapeutic possibilities, Stephen E. Finn and colleagues developed the Assessment Questionnaire-2 (AQ-2) to serve three distinct empirical and clinical functions: process monitoring, outcome measurement, and the relational appraisal of testing procedures.
From a process monitoring standpoint, the AQ-2 systematically assesses how a client emotionally, intellectually, and interpersonally internalizes assessment procedures. Rather than treating psychological assessment as an inert diagnostic vehicle, the AQ-2 measures the immediate phenomenology of receiving feedback: whether the client felt scrutinized, misunderstood, dehumanized, or conversely, enlightened, respected, and comforted. Clinical practitioners utilize this information to evaluate session dynamics, repair latent relational ruptures between assessor and client, and calibrate therapeutic strategies prior to subsequent psychotherapy.
As an outcome measurement instrument, the AQ-2 quantifies psychological gains achieved during the assessment itself. Decades of research in Therapeutic Assessment demonstrate that comprehensive, collaborative assessment can generate symptomatic improvements comparable in effect size to short-term psychotherapy. The AQ-2 captures these transformations directly through subscales targeting self-understanding, cognitive schema restructuring, and the positive emotional validation of one’s lived experience. In academic and clinical research trials, the tool serves as a primary endpoint or mediator when comparing collaborative assessment feedback models against traditional testing methods or attention-placebo control conditions.
Finally, the instrument acts as a crucial diagnostic and quality assurance metric for testing integrity. Excessive ratings on the Negative Feelings subscale (such as feeling judged, humiliated, or placed under a microscope) warn clinicians of countertherapeutic practices, excessive jargon, or failure to establish sufficient safety. Conversely, high scores on Mirroring and Alliance provide quantitative validation that the assessment intervention successfully engaged the client in active self-exploration, affirming the integrity of the clinical encounter.
Psychological Construct
The AQ-2 measures the multidimensional phenomenological experience of undergoing psychological evaluation and feedback. It conceptualizes the client’s internal reality through four intercorrelated yet theoretically differentiated constructs:
1. New Self-Awareness / Understanding
This construct represents the cognitive gain, mental reorganization, and exploratory insight achieved by the client as a result of the assessment process. Rooted in cognitive restructuring and psychoeducational reframing, this dimension assesses whether the feedback provided the individual with novel perceptual frameworks for interpreting their interpersonal difficulties, emotional vulnerabilities, and personal strengths. Clients scoring high on this dimension report feeling more cognizant of behavioral patterns (e.g., Item 21: “I am more aware of how I behave with other people”), experiencing a revised outlook toward life dilemmas (Item 29: “I have changed the way I think about my problems”), and gaining structural clarity regarding their internal mental architecture (Item 25: “The assessment helped me organize my thoughts about myself”).
2. Positive Accurate Mirroring
Derived primarily from psychodynamic self psychology, Positive Accurate Mirroring captures the degree to which clients feel deeply perceived, acknowledged, and validated by the assessment results. Rather than generating cold, alienating diagnostic categorizations, an attuned assessment reflects back the client’s core self-concept in a manner that feels authentic, compassionate, and affirming. Items under this construct reflect emotional resonance, pride, and self-cohesion (e.g., Item 2: “The assessment made me proud of who I am”; Item 10: “The assessment captured the ‘real’ me”; and Item 34: “I felt my strengths were recognized”). High scores on this subscale suggest that the assessment reinforced psychological stability, bolstered self-esteem, and mitigated feelings of pathological isolation.
3. Positive Relationship with the Assessor
This dimension operationalizes the collaborative working alliance, mutual trust, and perceived interpersonal warmth characterizing the relationship between the client and the evaluator. Drawing from humanistic and person-centered theory, this construct measures the assessor’s ability to act as an empathetic collaborator rather than an authoritative, aloof expert. It evaluates whether the assessor communicated genuine care, unconditioned positive regard, and emotional safety (e.g., Item 7: “It was easy to trust the assessor”; Item 35: “The assessor treated me as an equal”; and Item 43: “The assessor and I worked as a team to learn more about me”). This interpersonal safety serves as the experiential catalyst that enables clients to explore threatening psychological material without resorting to defensive denial.
4. Negative Feelings about the Assessment
The fourth construct measures the aversive emotional experiences, distress, and psychological vulnerability evoked by testing. Standard psychological evaluations often introduce epistemic anxiety, vulnerability, and feelings of inadequacy. This subscale monitors whether the assessment inadvertently induced humiliation, regression, or emotional fatigue (e.g., Item 4: “I felt I was under a microscope”; Item 16: “The assessment was a humiliating and degrading experience”; Item 20: “The assessment was emotionally draining”; and Item 32: “I felt judged by the assessor”). Elevated scores on this dimension signify emotional overwhelm, relational rupture, or inadequate containment during testing.
Theoretical Framework
The Assessment Questionnaire-2 is anchored primarily within the theoretical paradigm of Therapeutic Assessment (TA), an intervention model codified by Stephen E. Finn (1996, 2007) and Constance T. Fischer (1985). TA synthesizes three foundational bodies of psychological literature: humanistic-phenomenological psychology, Kohutian self psychology, and cognitive-interpersonal schema theory.
Humanistic and Phenomenological Roots
Traditional psychometrics developed from an epistemological stance that prioritized objectivity, standardization, and distance between tester and test-taker. In contrast, phenomenological approaches, pioneered by Carl Rogers and adapted to assessment by Constance Fischer, assert that genuine psychological understanding is co-constructed. The client is treated as an active expert on their own phenomenological world, while the assessor acts as an expert on psychological concepts and testing tools. The AQ-2 directly reflects this humanistic stance: its items measure whether the client experienced an egalitarian relationship (“treated me as an equal”) and whether testing was characterized by collaborative curiosity rather than unilateral diagnosis.
Self Psychology and Mirroring
The conceptualization of “Positive Accurate Mirroring” in the AQ-2 is explicitly derived from Heinz Kohut’s self psychology. Kohut posited that healthy psychological development and self-cohesion require the constant presence of “selfobjects” who provide empathetic attunement, mirroring, and validation. In the context of psychological assessment, standardized test findings (e.g., MMPI-2, Rorschach) can easily trigger severe fragmentation if delivered as rigid pathologizing labels. However, when the assessor translates test data into narrative descriptions that accurately mirror the client’s internal emotional conflicts while emphasizing adaptive coping mechanisms, the assessment acts as a selfobject experience. This restores self-esteem, decreases shame, and promotes psychological integration.
Cognitive Consistency and Swann’s Self-Verification Theory
The cognitive dimension of the AQ-2 incorporates William Swann’s Self-Verification Theory alongside classic cognitive dissonance paradigms. Swann discovered that human beings possess a fundamental drive to confirm their existing self-conceptions, even negative ones, to preserve a sense of predictability and psychological coherence. When assessment feedback contradicts existing self-views too aggressively (termed “Level 3” information in Finn’s model), clients respond with anxiety, denial, and cognitive rejection. Conversely, when feedback systematically builds upon what clients already suspect (Item 13: “The assessment confirmed parts of me that I had only suspected”) and gently expands their awareness (Item 48: “The assessment made me rethink the way I already viewed myself”), schema restructuring occurs seamlessly. The AQ-2 tracks this delicate balance between self-verification (mirroring) and progressive cognitive expansion (new self-awareness).
Validity
The construct, convergent, predictive, and discriminant validity of the AQ-2 has been substantiated across multiple empirical studies within clinical psychology, university counseling settings, and experimental laboratories.
Construct and Convergent Validity
Empirical studies evaluating the structural coherence of the AQ-2 demonstrate that its dimensions correlate meaningfully with established psychological constructs. In their seminal trial, Finn and Tonsager (1992) demonstrated that clients undergoing collaborative MMPI-2 feedback exhibited statistically significant increases in self-esteem and dramatic decreases in symptomatic distress compared to control participants who received informational attention. The AQ-2, deployed to unpack these effects, showed that the Positive Accurate Mirroring and New Self-Awareness subscales correlated strongly and positively with post-session scores on the Rosenberg Self-Esteem Scale (RSES), and correlated negatively with post-assessment scores on the Beck Depression Inventory (BDI) and the Symptom Checklist-90-Revised (SCL-90-R).
Convergent validity has also been firmly established with therapeutic alliance measures. Research by Newman and Greenway (1997), as well as subsequent trials by Smith, Wiggins, and Gsell (2007), confirmed that the AQ-2 Positive Relationship with the Assessor subscale demonstrates large, statistically significant correlations ($r = .65$ to $.78, p < .001$) with the Bond, Goal, and Task subscales of the Working Alliance Inventory (WAI). This confirms that the AQ-2 accurately captures the operational mechanisms of relational alliance within short-term assessment feedback contexts.
Discriminant and Factorial Independence
Discriminant validity is evidenced by the clear separation between the Negative Feelings about the Assessment subscale and the three positive subscales. Empirical studies (e.g., Allen, 2001; Durham-Fowler, 2010) show moderate negative or near-zero correlations between Negative Feelings and Positive Accurate Mirroring ($r = -.35$ to $-.52$), demonstrating that client dissatisfaction or feeling “microscoped” is distinct from a mere lack of self-awareness. Furthermore, the AQ-2 subscales correlate minimally with measures of general social desirability, verifying that favorable client endorsements reflect authentic therapeutic experiences rather than acquiescence bias.
Predictive Validity
In clinical outcome research, AQ-2 subscale scores consistently predict post-assessment treatment adherence, readiness for change, and subsequent psychotherapy engagement. Poston and Hanson (2010), in a meta-analytic review of collaborative psychological assessment as a therapeutic intervention, noted that high AQ-2 total alliance and self-awareness scores were significant predictors of lower therapy drop-out rates, greater client compliance with subsequent referral recommendations, and sustained symptomatic relief at 6-month follow-up evaluations.
Reliability
The AQ-2 exhibits high internal consistency reliability across diverse clinical cohorts, student populations, and experimental samples. Reliability metrics reported across multiple primary investigations demonstrate exceptional stability across its constituent subscales:
- New Self-Awareness / Understanding: Across validation samples by Finn, Schroeder, and Tonsager (1994), Allen (2001), and Durham-Fowler (2010), Cronbach’s alpha coefficients for this dimension have consistently ranged from $lpha = .84$ to $.94$, indicating exceptional internal consistency and conceptual coherence of items targeting cognitive restructuring and behavioral self-awareness.
- Positive Accurate Mirroring: Empirical studies report Cronbach’s alpha values between $lpha = .80$ and $.91$ across cohorts, demonstrating that items measuring feelings of being deeply understood, validated, and accurately reflected possess high homogeneity.
- Positive Relationship with the Assessor: This relational dimension yields robust alpha values ranging between $lpha = .87$ and $.92$, matching or exceeding the psychometric reliability benchmarks established by standard therapeutic alliance inventories.
- Negative Feelings about the Assessment: Internal consistency estimates for this subscale range from $lpha = .79$ to $.92$ across clinical cohorts, confirming that items capturing subjective distress, humiliation, and vulnerability hang together cohesively.
Test-retest reliability evaluations, conducted primarily across short intervals (e.g., 48 hours to 2 weeks post-feedback), demonstrate stability coefficients ranging from $r = .74$ to $.86$ for positive subscales, indicating that while client perceptions are sensitive to immediate session interventions, their cognitive and affective impressions remain relatively stable over the post-assessment processing period.
Factor Analysis
The structural dimensionality of the Assessment Questionnaire-2 has been investigated through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across numerous dissertations and peer-reviewed investigations.
Exploratory Factor Structure (EFA)
In the original factor analyses conducted by Finn, Schroeder, and Tonsager (1994), principal axis factoring with oblimin (oblique) rotation was executed to reflect the natural theoretical interdependence among therapeutic process factors. The scree plot examination, Kaiser-Guttman criterion (eigenvalues > 1.0), and parallel analysis supported a robust four-factor solution accounting for over 52% to 58% of the total variance across clinical samples:
- Factor 1: Positive Relationship with the Assessor (salient loadings $ge .50$ on items tapping trust, warmth, respect, equality, and collaborative teamwork).
- Factor 2: New Self-Awareness/Understanding (high loadings on items addressing new behavioral insights, organized self-perceptions, and shift in problem conceptualization).
- Factor 3: Positive Accurate Mirroring (strong loadings on items evaluating being accurately captured, feeling special, feeling validated, and pride in identity).
- Factor 4: Negative Feelings about the Assessment (loadings centered on items assessing exposure, regression, emotional exhaustion, and humiliation).
Confirmatory Factor Analysis (CFA)
Subsequent psychometric investigations, notably those by Andrea Allen (2001) and subsequent replications (e.g., Durham-Fowler, 2010), evaluated the fit of this four-factor model using structural equation modeling. In confirmatory modeling, a four-factor oblique model consistently outperforms alternative unidimensional or orthogonal two-factor models (e.g., broad Positive vs. Negative client experience).
Structural equation modeling yields strong global fit indices supporting the four-factor framework:
- Comparative Fit Index (CFI): Values consistently range between $.91$ and $.94$, exceeding standard thresholds for acceptable construct specification.
- Tucker-Lewis Index (TLI): Coefficients typically span from $.90$ to $.93$.
- Root Mean Square Error of Approximation (RMSEA): Point estimates fall reliably between $.048$ and $.065$ (with 90% confidence intervals below $.080$), demonstrating low residual error.
- Standardized Root Mean Square Residual (SRMR): Indices typically range around $.052$ to $.061$.
Standardized factor loadings across primary items are substantial, with the vast majority falling between $.55$ and $.84$, confirming that the observed variables serve as robust empirical indicators of their specified latent constructs.
Instrument / Measurement Tool
The Assessment Questionnaire-2 is a comprehensive, self-administered psychometric inventory designed for post-assessment debriefing and outcome tracking.
- Test Type: Client-report process and outcome inventory / psychological assessment feedback appraisal scale.
- Target Population: Adolescents and adults (typically ages 16 and older) undergoing psychological, neurocognitive, or personality assessment.
- Standard Item Count: 48 items in the original Finn, Schroeder, & Tonsager (1994) instrument; 45 items in the modified Allen (2001) experimental adaptation.
- Administration Format: Paper-and-pencil or secure computer-based digital questionnaire.
- Completion Time: Approximately 10 to 15 minutes.
- Response Scale: 5-point Likert scale formatted as:
- 1 = Strongly Disagree
- 2 = Disagree
- 3 = Neutral
- 4 = Agree
- 5 = Strongly Agree
- Scoring and Directionality:
- Subscale raw scores are derived by calculating the mean or sum of items corresponding to each of the four dimensions.
- Reverse Scoring: In the subscale computation for New Self-Awareness/Understanding, Item 1 (“The assessment did not teach me anything new about myself”) is reverse-scored ($1=5, 2=4, 3=3, 4=2, 5=1$). All other items are scored in their direct valence.
- Subscale Item Mappings (48-Item Inventory):
- New Self-Awareness/Understanding: Items 1 (reversed), 5, 9, 13, 17, 21, 25, 29, 30, 33, 37, 38, 41, 45, 48.
- Positive Accurate Mirroring: Items 2, 6, 10, 14, 18, 22, 26, 34, 42, 46.
- Positive Relationship with the Assessor: Items 3, 7, 11, 15, 19, 23, 27, 31, 35, 39, 43, 47.
- Negative Feelings about the Assessment: Items 4, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44.
- Clinical Interpretation: Elevated scores (> 4.0) on Self-Awareness, Mirroring, and Assessor Relationship indicate highly effective, collaborative, and therapeutic assessment delivery. Any elevated score on Negative Feelings (> 2.5–3.0) signals the presence of feeling pathologized, misunderstood, or emotionally overwhelmed, requiring immediate clinical exploration or process repair.
Permissions & Fee and Test Year
The Assessment Questionnaire-2 was developed in 1994 by Stephen E. Finn, David G. Schroeder, and Mary E. Tonsager as an unpublished academic measurement tool. It was designed to facilitate empirical research into Therapeutic Assessment, clinical feedback, and testing process variables.
Licensing and Availability: The instrument is placed within the public academic and professional research domain for educational, non-commercial clinical, and research applications. It is not commercially published by a test publisher for profit. Psychologists, clinical researchers, and graduate students can access the scale through scholarly dissertations (such as those archived in university repositories including the University of Texas at Austin and Florida International University), or directly through publications associated with the Center for Therapeutic Assessment. When employing the AQ-2 in research or clinical quality monitoring, appropriate citation of the authors (Finn, Schroeder, & Tonsager, 1994) is required. Clinicians are encouraged to maintain the standardized wording and 5-point Likert format to preserve the instrument’s established psychometric properties.
References
Allen, A. (2001). Informational effects of assessment feedback (Publication No. 1094) [Doctoral dissertation, Florida International University]. FIU Electronic Theses and Dissertations. https://digitalcommons.fiu.edu/etd/1094/
Durham-Fowler, J. A. (2010). The therapeutic effects of collaborative psychological assessment [Doctoral dissertation, University of Texas at Austin]. UT Electronic Theses and Dissertations. https://repositories.lib.utexas.edu/handle/2152/ETD-UT-2010-08-1790
Finn, S. E. (1996). Manual for using the MMPI-2 as a therapeutic intervention. University of Minnesota Press.
Finn, S. E. (2007). In our clients’ shoes: Theory and techniques of Therapeutic Assessment. Lawrence Erlbaum Associates Publishers.
Finn, S. E., Schroeder, D. G., & Tonsager, M. E. (1994). The Assessment Questionnaire-2 (AQ-2): A measure of clients’ experiences with psychological assessment [Unpublished manuscript]. Center for Therapeutic Assessment, Austin, TX.
Finn, S. E., & Tonsager, M. E. (1992). The therapeutic effects of providing MMPI-2 test feedback to college students awaiting psychotherapy. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 4(3), 278–287. https://doi.org/10.1037/1040-3590.4.3.278
Finn, S. E., & Tonsager, M. E. (1997). Information-gathering and therapeutic models of assessment: Complementary paradigms. Psychological Assessment, 9(4), 374–385. https://doi.org/10.1037/1040-3590.9.4.374
Fischer, C. T. (1985). Individualizing psychological assessment. Brooks/Cole.
Newman, M. L., & Greenway, P. (1997). Therapeutic effects of providing MMPI-2 test feedback to clients at a university counseling service: A collaborative approach. Psychological Assessment, 9(2), 122–131. https://doi.org/10.1037/1040-3590.9.2.122
Poston, J. M., & Hanson, W. E. (2010). Meta-analysis of psychological assessment as a therapeutic intervention. Psychological Assessment, 22(2), 203–212. https://doi.org/10.1037/a0018679
Smith, J. D., Wiggins, K. E., & Gsell, S. M. (2007). Therapeutic assessment with children and adolescents: An empirical review and case illustration. Journal of Personality Assessment, 89(3), 269–277. https://doi.org/10.1080/00223890701629797