Disposition to Trust Scale (DTT)

1. Abstract

The Disposition to Trust Scale (DTT) is an established psychometric instrument designed to evaluate an individual’s generalized propensity to trust others across varied social, organizational, and technological domains. Conceptualized within the foundational trust framework formulated by D. Harrison McKnight, Larry L. Cummings, and Norman L. Chervany (1998) and empirically operationalized in subsequent structural validation studies (McKnight, Choudhury, & Kacmar, 2002), the instrument assesses trust as a cross-situational, trait-like individual difference. The standard validated operationalization features an 8-item self-report questionnaire bifurcated into two distinct yet complementary subconstructs: Faith in Humanity (4 items), which captures the baseline cognitive conviction that people are generally well-meaning, benevolent, and dependable; and Trusting Stance (4 items), which reflects a strategic, pragmatic behavioral orientation wherein an actor initiates interactions assuming others are trustworthy until empirical evidence indicates otherwise.

Responses are recorded on an authentic 7-point Likert scale ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”), incorporating balanced reverse-coded items to control for acquiescence response bias. Extensive psychometric evaluations across information systems, organizational psychology, human-computer interaction (HCI), and consumer behavior demonstrate robust psychometric performance. Internal consistency reliability estimates consistently range from acceptable to excellent, with Cronbach’s alpha coefficients routinely documented between α = .78 and .91 for Faith in Humanity, and between α = .75 and .89 for Trusting Stance. Confirmatory factor analyses (CFA) repeatedly validate the two-dimensional oblique structural model over unidimensional alternatives, demonstrating optimal goodness-of-fit indices (CFI > .95, TLI > .94, RMSEA < .06, SRMR < .05). The instrument exhibits sound convergent, discriminant, and predictive validity, operating as the premier predictor of initial trust formation in novel, ambiguous environments characterized by asymmetric information, such as e-commerce transactions, digital privacy negotiation, autonomous system interactions, and human-artificial intelligence teaming.

2. Keywords

disposition to trust, faith in humanity, trusting stance, initial trust formation, McKnight, generalized trust, psychometrics, human-computer interaction, technology adoption, organizational trust, social cognition, trait trust

3. Authors

The conceptual taxonomy and operational measurement parameters of the Disposition to Trust construct were developed by a prominent group of organizational behavior and management information systems scholars:

• D. Harrison McKnight, Ph.D. — Professor Emeritus of Management Information Systems at the College of Business, Florida State University, Tallahassee, Florida, USA. Dr. McKnight is an internationally recognized expert in organizational trust, e-commerce trust models, and human-technology trust interfaces.
• Larry L. Cummings, Ph.D. (1937–1997) — Formerly the Carlson Professor of Management and Organization at the Carlson School of Management, University of Minnesota, Minneapolis, Minnesota, USA. Dr. Cummings was a past President of the Academy of Management and a seminal contributor to organizational behavior, trust theory, and individual performance modeling.
• Norman L. Chervany, Ph.D. — Professor Emeritus of Information and Decision Sciences at the Carlson School of Management, University of Minnesota, Minneapolis, Minnesota, USA. Dr. Chervany’s research specialized in decision support systems, technology management, and empirical modeling of organizational relationships.

4. Purpose

The primary purpose of the Disposition to Trust Scale is to quantify an individual’s generalized, cross-situational tendency to depend on, believe in, and display confidence in other entities, independent of specific contextual cues or historical relationship records. In interpersonal and organizational psychology, trust has historically been conceptualized either as a dynamic, relational state developed through cumulative reciprocal interactions or as a broad personality disposition. The DTT was formulated to resolve the theoretical paradox known as the “initial trust paradox”: why individuals consistently demonstrate moderate to high levels of trust toward completely unfamiliar actors, institutions, automated technologies, or websites during initial encounters, even in the absolute absence of reputational data, direct personal experience, or institutional guarantees (McKnight et al., 1998).

From an applied research perspective, the DTT fulfills several critical functions. In digital commerce and technological adoption studies, dispositional trust functions as the fundamental baseline variable determining early-stage user engagement. When consumers encounter novel e-commerce platforms, decentralized finance applications, or artificial intelligence algorithms, contextual cues regarding provider integrity and technical competence are frequently ambiguous or completely hidden. In such high-entropy scenarios, situational cues exert weak cognitive control, allowing the respondent’s underlying dispositional trust to dominate behavioral intentions. Individuals with elevated dispositional trust exhibit significantly reduced perceived risk, elevated psychological safety, increased propensity to disclose personal and biometric information, and accelerated transaction execution.

In organizational psychology and leadership research, the DTT serves as a foundational diagnostic baseline for measuring new employee onboarding, cross-functional team assembly, virtual collaborative work, and post-merger integration. It allows researchers to control for pre-existing individual differences when examining how specific managerial behaviors, psychological contract fulfillment, or organizational justice dimensions actively construct or destroy situated interpersonal trust. In clinical, counseling, and developmental contexts, the scale provides insight into underlying interpersonal schemas, attachment security, and general social cynicism, offering empirical granularity regarding whether social reticence originates from a fundamental disbelief in human goodwill (impaired Faith in Humanity) or a defensive behavioral policy against betrayal (impaired Trusting Stance).

5. Psychological Construct

The psychological construct of disposition to trust represents a pervasive, trait-like orientation that shapes an individual’s cognitive interpretations and behavioral intentions regarding the reliability and intentions of humanity at large. Rather than treating dispositional trust as a monolithic, one-dimensional trait, McKnight and colleagues conceptualized it as a multidimensional construct comprising two interrelated yet theoretically distinct cognitive-behavioral facets:

Faith in Humanity

Faith in Humanity is a belief-based, cognitive construct reflecting an individual’s assumption that people are generally well-intentioned, benevolent, caring, competent, and honest. Grounded in cognitive schema theory, this facet operates as an internal representational model of generic human nature. A person high in Faith in Humanity subscribes to the fundamental attribution that other social actors are naturally cooperative and altruistic rather than fundamentally predatory, self-serving, or deceitful. When confronted with ambiguous social behavior, individuals with elevated Faith in Humanity systematically engage in benign interpretations, attributing lapses in communication or execution to situational constraints rather than malicious intent. Conversely, low scores indicate generalized social cynicism, misanthropy, and an expectation of opportunism, consistent with Machiavellian or defensive worldview constructs.

Trusting Stance

Trusting Stance is a strategy-based, pragmatic, and behavioral orientation that reflects an individual’s conscious conviction that treating others as if they are trustworthy yields superior personal and social outcomes, regardless of what one actually believes about human nature. It represents an operational personal policy: “I will act toward others with trust until they explicitly demonstrate that they are unworthy of that trust.” Trusting Stance is fundamentally pragmatic, rooted in rule-utilitarian social philosophy. An individual may intellectually acknowledge that some proportion of the population is dishonest or self-interested, yet deliberately adopt a trusting posture because the transaction costs of hyper-vigilance, defensive avoidance, and perpetual suspicion are viewed as self-defeating and psychologically exhausting. Trusting Stance provides the behavioral impetus to initiate cooperation, extend early credit, and grant social actors the benefit of the doubt.

The distinction between these two subscales is empirically and theoretically profound. An individual may possess low Faith in Humanity (believing that human beings are fundamentally flawed and driven by narrow self-interest) while simultaneously maintaining a high Trusting Stance (deliberately extending trust to build reciprocal alliances, lower coordination frictions, and uncover productive social opportunities). By decomposing dispositional trust into cognitive beliefs and strategic postures, the DTT offers superior descriptive granularity and predictive accuracy compared to unidimensional social trust indicators.

6. Theoretical Framework

The theoretical framework undergirding the Disposition to Trust Scale integrates classical personality psychology, social cognitive theory, and structural models of organizational trust formation. The primary conceptual foundation is derived from Julian Rotter’s (1967, 1971) pioneering work on generalized interpersonal trust. Rotter defined interpersonal trust as an expectancy held by an individual or a group that the word, promise, verbal, or written statement of another individual or group can be relied upon. Rotter postulated that when individuals enter novel, unstructured, or ambiguous situations where situational cues are weak or unfamiliar, behavioral choices are governed by generalized expectancies abstracted from historical developmental experiences.

McKnight, Cummings, and Chervany (1998) expanded upon Rotter’s generalized expectancy theory by integrating Walter Mischel’s (1977) seminal paradigm of “strong versus weak situations.” In a “strong situation” (e.g., highly regulated institutional settings with clear legal penalties, structured workflows, and extensive reputational histories), explicit rules and environmental constraints exert powerful control over human behavior, substantially dampening the influence of individual personality differences. Conversely, in a “weak situation” (e.g., encountering a newly launched digital interface, engaging with a novel AI system, or interacting with an unfamiliar organizational representative), situational guidelines are ambiguous, behavioral scripts are undefined, and formal monitoring mechanisms are absent or unverified. In these weak situations, disposition to trust acts as the primary cognitive filter through which an individual generates initial trust attitudes and behavioral intentions.

Furthermore, the DTT draws directly on George Kelly’s (1955) Personal Construct Theory, which posits that humans navigate reality by testing mental hypotheses about social actors. Faith in Humanity reflects the structural content of these core constructs, while Trusting Stance represents the procedural heuristic chosen to interact with the external world. McKnight et al. positioned Disposition to Trust alongside Institution-Based Trust (structural assurances and situational normality) and Cognitive Trust Processes (categorization, stereotyping, and unit grouping) within their comprehensive Model of Initial Trust Formation. Within this framework, dispositional trust directly fuels “Trusting Beliefs” (perceptions of specific partner competence, benevolence, and integrity) and directly stimulates “Trusting Intentions” (the operational willingness to depend upon the counterpart), establishing the causal architecture that bridges basic individual differences with definitive social and economic action.

7. Validity

The validity of the Disposition to Trust Scale has been substantiated through rigorous psychometric testing across diverse organizational, cross-cultural, and human-technology contexts.

Construct and Structural Validity

Construct validity is evidenced by the consistent replicate stability of the two-factor oblique configuration. In the comprehensive empirical validation study conducted by McKnight, Choudhury, and Kacmar (2002) involving web-based consumer cohorts (N = 1,403), confirmatory factor analysis verified that the hypothesized two-dimensional model exhibited superior fit relative to rival unidimensional models. Both Faith in Humanity and Trusting Stance demonstrated high average variance extracted (AVE > .50), establishing robust convergent validity at the latent construct level. Invariance testing has further confirmed metric and scalar equivalence across demographic groups and varied digital interaction modalities.

Convergent Validity

The DTT exhibits statistically significant convergent validity correlations with theoretically aligned psychological constructs. Faith in Humanity correlates positively and moderately with the Agreeableness dimension of the Big Five personality model (r = .38 to .52, p < .001) and generalized positive affectivity. It demonstrates expected strong negative associations with generalized social cynicism, misanthropy (r = -.54 to -.68, p < .001), and the Dark Triad traits, specifically Machiavellianism (r = -.45 to -.61, p < .001). Trusting Stance correlates moderately with generalized self-efficacy (r = .28 to .39, p < .01) and risk tolerance, confirming its identity as an active, agentic strategy rather than mere passive gullibility.

Discriminant Validity

Discriminant validity between Faith in Humanity and Trusting Stance is confirmed by inter-construct correlations that consistently fall within the moderate range (r = .42 to .64), well below the threshold that would suggest collinearity or conceptual redundancy. The square root of the AVE for each construct routinely exceeds the inter-construct correlations, meeting the rigorous Fornell-Larcker criterion. Additionally, both dispositional subscales discriminate clearly from situational trust dimensions (e.g., institution-based structural assurances and specific trustee benevolence), showing distinct factor loadings and cross-loading suppression below .20 in multi-trait multi-method structural equation models.

Predictive and Nomological Validity

Nomological validity is demonstrated across dozens of empirical studies in technology adoption, consumer behavior, and human-agent interaction. In their 2002 study, McKnight et al. demonstrated that Disposition to Trust directly and positively predicted Institution-Based Trust (β = .34, p < .001), Trusting Beliefs regarding an unfamiliar vendor (β = .22, p < .001), and direct Willingness to Depend / Trusting Intentions (β = .18, p < .01). In autonomous vehicle and artificial intelligence research, baseline DTT scores significantly predict initial reliance, decreased monitoring behaviors, and accelerated system adoption, validating the scale’s sensitivity to modern socio-technical interactions.

8. Reliability

The psychometric reliability of the Disposition to Trust Scale has been thoroughly evaluated across multiple independent empirical investigations, showing strong internal consistency and temporal stability:

• Internal Consistency Reliability: In the baseline studies by McKnight et al. (2002), the internal consistency for Faith in Humanity yielded a Cronbach’s alpha of α = .86, with composite reliability (CR) recorded at .87. The Trusting Stance subscale demonstrated a Cronbach’s alpha of α = .81, with a composite reliability of .82. Subsequent organizational and e-service studies have consistently replicated these values, documenting alpha ranges between α = .78 and .91 for Faith in Humanity, and α = .75 and .89 for Trusting Stance. Composite reliabilities consistently exceed the conventional .70 threshold, and McDonald’s omega (ω) coefficients typically range from .80 to .92, confirming strong structural reliability across diverse respondent samples.
• Item-Total Correlations: Corrected item-total correlations across all 8 items routinely exceed the standard .40 psychometric cutoff, typically clustering between r = .52 and r = .76. The reverse-coded items (Items 3 and 4) maintain robust corrected item-total correlations (ranging from .48 to .69) following appropriate mathematical reversal, confirming that cognitive attention checks and acquiescence controls do not compromise construct coherence.
• Test-Retest Reliability: Consistent with its conceptualization as a stable, personality-like disposition, longitudinal and multi-wave panel studies demonstrate high temporal stability over time. In test-retest assessments across intervals ranging from 4 to 8 weeks, intraclass correlation coefficients (ICC) and Pearson product-moment correlations have remained consistently stable (r = .74 to .83, p < .001), demonstrating that baseline dispositional trust resists short-term situational fluctuations while remaining receptive to major cumulative life changes.

9. Factor Analysis

The internal latent architecture of the Disposition to Trust Scale has been extensively evaluated using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) within structural equation modeling frameworks.

Exploratory Factor Analysis Findings

Initial exploratory factor extractions (using Principal Axis Factoring and Maximum Likelihood with Promax oblique rotation) consistently demonstrate a clean, two-factor solution based on Kaiser’s eigenvalue criterion (λ > 1.0) and scree plot inspections. Factor 1 (Faith in Humanity) typically accounts for approximately 38% to 44% of the total variance, while Factor 2 (Trusting Stance) accounts for an additional 14% to 18% of the explained variance, culminating in an aggregate explained variance routinely exceeding 55% to 62%. Factor loadings for individual items onto their primary designated latent factors are robust, consistently ranging from .62 to .88, with cross-loadings onto non-target factors uniformly remaining well below the .25 threshold.

Confirmatory Factor Analysis and Model Fit Indices

In confirmatory testing using maximum likelihood estimation with robust standard errors (MLR), researchers have consistently established the superiority of the two-factor oblique model over alternative structural configurations. Table 1 summarizes representative model fit indices comparing the hypothesized two-factor model against competing structural specifications based on empirical literature (e.g., McKnight et al., 2002; Gefen, Karahanna, & Straub, 2003):

Standardized factor loadings in the optimal two-factor model typically emerge as follows: Item 1 (λ = .78), Item 2 (λ = .84), Item 3 (λ = .72), and Item 4 (λ = .69) loading onto Faith in Humanity; Item 5 (λ = .76), Item 6 (λ = .82), Item 7 (λ = .85), and Item 8 (λ = .79) loading onto Trusting Stance. Standardized error variances are uniformly within acceptable ranges without Heywood cases or improper solutions, confirming that both subconstructs maintain distinct conceptual boundaries while sharing a meaningful covariance path (typically φ ≈ .50 to .58).

10. Instrument / Measurement Tool

• Instrument Name: Disposition to Trust Scale (DTT)
• Primary Conceptual Origin: McKnight, Cummings, & Chervany (1998)
• Empirical Validation Source: McKnight, Choudhury, & Kacmar (2002)
• Instrument Type: Psychometric self-report survey scale
• Target Construct: Generalized disposition to trust others across interpersonal, organizational, and technological interfaces
• Subscale Architecture: 2 core subscales
  • Faith in Humanity: Items 1, 2, 3, and 4
  • Trusting Stance: Items 5, 6, 7, and 8
• Total Item Count: 8 authentic items
• Authentic Response Scale: 7-point Likert scale (1 = Strongly Disagree, 7 = Strongly Agree)
• Scoring Rules & Protocols:
  • Items 1, 2, 5, 6, 7, and 8 are positively scored (1 = 1 to 7 = 7).
  • Items 3 and 4 are reverse-scored according to the standard 7-point conversion formula: Reversed Score = 8 − Raw Score (i.e., 1 → 7, 2 → 6, 3 → 5, 4 → 4, 5 → 3, 6 → 2, 7 → 1).
  • The Faith in Humanity subscale score is computed as the arithmetic mean of Items 1, 2, 3 (reversed), and 4 (reversed).
  • The Trusting Stance subscale score is computed as the arithmetic mean of Items 5, 6, 7, and 8.
  • The total composite Disposition to Trust score is computed as the overall arithmetic mean across all 8 items (with Items 3 and 4 reversed). Higher mean scores reflect a stronger dispositional propensity to trust.
• Completion Time: Approximately 2 to 4 minutes

11. Permissions & Fee and Test Year

The theoretical framework for the Disposition to Trust Scale was formally introduced in 1998 in the Academy of Management Review, with definitive operational validation published in 2002 in Information Systems Research. As an academic psychometric tool originating within peer-reviewed management and information systems literature, the scale is generally accessible for non-commercial academic research, pedagogical purposes, and university-based scholarly investigations without royalty fees, provided full scholarly citation and attribution are given to the original authors (McKnight, Cummings, & Chervany, 1998; McKnight, Choudhury, & Kacmar, 2002).

Commercial applications, proprietary corporate diagnostic deployments, organizational talent assessments, or integration into commercial digital platforms may be subject to institutional copyright governance held by the publisher (INFORMS or Academy of Management) or the authors’ estates. Researchers seeking to implement modifications, translation into other languages, or commercial enterprise applications should ensure appropriate institutional compliance and contact the primary authors or respective academic journals for formal licensing and clearance.

12. References

• Gefen, D., Karahanna, E., & Straub, D. W. (2003). Trust and TAM in online shopping: An integrated model. MIS Quarterly, 27(1), 51–90. https://doi.org/10.2307/30036519
• Kelly, G. A. (1955). The psychology of personal constructs (Vols. 1 & 2). W. W. Norton & Company.
• Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734. https://doi.org/10.5465/amr.1995.9508080332
• McKnight, D. H., Choudhury, V., & Kacmar, C. (2002). Developing and validating trust measures for e-commerce: An integrative typology. Information Systems Research, 13(3), 334–359. https://doi.org/10.1287/isre.13.3.334.81
• McKnight, D. H., Cummings, L. L., & Chervany, N. L. (1998). Initial trust formation in new organizational relationships. Academy of Management Review, 23(3), 473–490. https://doi.org/10.5465/amr.1998.926622
• Mischel, W. (1977). The interaction of person and situation. In D. Magnusson & N. S. Endler (Eds.), Personality at the crossroads: Current issues in interactional psychology (pp. 333–352). Lawrence Erlbaum Associates.
• Rotter, J. B. (1967). A new scale for the measurement of interpersonal trust. Journal of Personality, 35(4), 651–665. https://doi.org/10.1111/j.1467-6494.1967.tb01454.x
• Rotter, J. B. (1971). Generalized expectancies for interpersonal trust. American Psychologist, 26(5), 443–452. https://doi.org/10.1037/h0031464

13. Items of the Scale