Source Trustworthiness Scale (STWS)

1. Abstract

The Source Trustworthiness Scale (STWS) is a psychometric instrument designed to assess perceived communicator integrity, moral character, and benevolence within interpersonal, organizational, and mediated communication contexts. Developed and operationalized in experimental consumer psychology by Marilynn Touré-Tillery and Ann L. McGill (2015), the instrument isolates perceived trustworthiness from the broader multidimensional construct of source credibility, separating ethical character and genuine intent from epistemic competence (expertise) and social attractiveness. The scale comprises four semantic differential adjective pairs evaluated across a 7-point continuum: Dishonest – Honest, Unethical – Ethical, Fake – Genuine, and Untrustworthy – Trustworthy. Psychometric investigations reveal robust unifactorial structural validity, high internal consistency across diverse experimental conditions (Cronbach’s α typically ranging between .88 and .95), and pronounced construct, convergent, and predictive validity. Empirical evidence demonstrates that the scale sensitively captures variations in perceived persuasion intent, anthropomorphic attributions, and source authenticity. While originally validated to compare the differential persuasiveness of human versus anthropomorphized messengers, the STWS functions as an agile, psychometrically rigorous instrument widely adopted across social psychology, marketing communication, public health messaging, human-computer interaction (HCI), and political communication. Its parsimonious four-item architecture minimizes participant fatigue while providing superior diagnostic sensitivity to moral integrity and perceived communicative benevolence.

2. Keywords

Source Trustworthiness Scale, source credibility, trustworthiness, persuasion, anthropomorphism, moral integrity, benevolence, semantic differential, consumer psychology, communicator credibility, interpersonal trust, epistemic vigilance, marketing communications

3. Authors

The Source Trustworthiness Scale was formalized and validated by Marilynn Touré-Tillery and Ann L. McGill in their landmark 2015 empirical paper published in the Journal of Marketing.

• Marilynn Touré-Tillery, Ph.D. — Associate Professor of Marketing at the Kellogg School of Management, Northwestern University, Evanston, Illinois, USA. Her research focuses on consumer motivation, self-perception, moral behavior, and the psychological mechanisms underlying judgment and decision-making.
• Ann L. McGill, Ph.D. — Sears Roebuck Professor of General Management, Emerita, at the Booth School of Business, University of Chicago, Chicago, Illinois, USA. Her scholarly inquiry centers on causal reasoning, comparative thinking, and the psychological antecedents and consequences of anthropomorphism.

4. Purpose

The primary purpose of the Source Trustworthiness Scale (STWS) is to provide a focused, psychometrically rigorous, and parsimonious assessment of the perceived moral uprightness, honesty, and sincerity of a communication source. In classical psychometric operationalizations of source credibility, measurement instruments frequently connote a composite of epistemic competence (expertise), social attractiveness, status, and integrity. While comprehensive credibility inventories (e.g., Ohanian, 1990) yield multidimensional profiles, they frequently confound cognitive appraisals of whether a communicator possesses accurate knowledge (expertise) with affective and moral judgments of whether the communicator intends to tell the truth (trustworthiness).

Touré-Tillery and McGill (2015) identified a vital theoretical need to isolate trustworthiness as an independent psychological construct. When audiences process persuasive messaging, the communicator’s perceived intent to manipulate or deceive triggers defensive psychological processing, including psychological reactance and counter-arguing. The STWS was specifically engineered to detect fine-grained variance in perceived communicator motives, moral character, and communicative authenticity without the confounding variance of perceived intelligence, technical knowledge, or physical attractiveness.

In applied and experimental settings, the STWS serves several essential functions:

• Persuasion and Advertising Research: It evaluates consumer appraisals of corporate spokespeople, brand ambassadors, social media influencers, and automated agents. It captures whether audiences perceive marketing communications as deceptive or ethically grounded.
• Anthropomorphism and Artificial Intelligence: A central methodological contribution of the scale is measuring trust attributions across non-human entities. By testing whether consumers impute ulterior motives and intentionality to anthropomorphized products, conversational agents, or algorithmic advisory tools, the STWS assesses the boundaries of social cognition.
• Public Health and Risk Communication: During health crises, policy implementation, or medical controversies, adherence to behavioral recommendations is heavily dictated by perceived communicator trustworthiness. The scale provides public health researchers with a rapid diagnostic tool to evaluate whether community liaisons or institutional figures are perceived as honest and genuine.
• Political Psychology and Crisis Management: The STWS captures public reactions to organizational crises, political debates, and public apologies, isolating changes in perceived honesty and ethical conduct following transgressions or scandals.

5. Psychological Construct

The psychological construct assessed by the STWS is perceived source trustworthiness, defined as the receiver’s subjective belief that an information source possesses moral integrity, exhibits benevolent intentions, and communicates authentic, unbiased assertions without deliberate intent to deceive. Within social cognitive architecture, trustworthiness represents the moral-evaluative pillar of social perception, conceptually aligned with the “warmth” and “morality” dimensions of the Stereotype Content Model (Fiske, Cuddy, & Glick, 2007).

Core Facets of the Construct

Perceived trustworthiness comprises distinct psychological dimensions operationalized across the four semantic pairs:

• Moral Integrity and Ethical Grounding (Unethical – Ethical): Refers to the perceived adherence of the source to normative ethical standards, moral obligations, and fundamental principles of fairness. A source perceived as ethical is judged to operate under intrinsic moral constraints that discourage opportunistic deception, manipulation, or exploitation of information asymmetries.
• Veracity and Honesty (Dishonest – Honest): Represents the receiver’s expectation of factual accuracy and truthful communication. Honesty reflects the absence of calculated misrepresentation, omission of material facts, or deliberate fabrication. Unlike expertise (which denotes whether the source knows the truth), honesty denotes the source’s willingness to articulate reality as they understand it.
• Authenticity and Sincerity (Fake – Genuine): Involves appraisals of communicative motives and emotional authenticity. In modern communication environments, communicators may appear technically truthful while acting inauthentically or artificially. The “genuine” dimension measures whether the communicator’s outward presentation reflects underlying core beliefs, values, and unfeigned goodwill, as opposed to scripted, performative, or mercenary motives.
• Global Trustworthiness and Reliability (Untrustworthy – Trustworthy): Serves as an integrative, overarching appraisal of psychological safety and dependability. It captures the receiver’s readiness to be vulnerable to the communicator’s claims, reflecting confidence that relying on the source’s recommendations will not lead to betrayal, harm, or cognitive exploitation.

Divergence from Related Constructs

It is vital to distinguish perceived trustworthiness from related constructs:

• Trustworthiness vs. Expertise: Expertise captures cognitive competence, formal training, intelligence, and epistemic authority. A communicator can be exceptionally competent yet profoundly untrustworthy (e.g., a predatory financial advisor). Conversely, a communicator can be deeply honest yet inexpert (e.g., an authentic peer). The STWS isolates moral intent from technical ability.
• Trustworthiness vs. General Likability: A communicator may be charming, humorous, and socially attractive while being perceived as deceitful. Trustworthiness captures an evaluative moral judgment rather than an affective reaction of social pleasantness.

6. Theoretical Framework

The STWS is embedded in classical and contemporary models of social cognition, persuasion processing, and attribution theory.

The Yale Communication Paradigm and Source Credibility

The theoretical foundations of source credibility trace back to Carl Hovland and colleagues at Yale University (Hovland, Janis, & Kelley, 1953). In their early experimental investigations of message persuasion, Hovland et al. postulated that source credibility consists primarily of two orthogonal dimensions: expertness (the extent to which a communicator is perceived to be capable of making correct assertions) and trustworthiness (the degree of confidence in the communicator’s intent to communicate the assertions he or she considers most valid). Hovland emphasized that an audience’s perception of a communicator’s motives directly dictates message acceptance; when a communicator is suspected of having personal motives to persuade, perceived trustworthiness diminishes, activating epistemic defenses.

Dual-Process Models of Persuasion

Within dual-process frameworks, specifically the Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1986) and the Heuristic-Systematic Model (HSM; Chaiken, 1980), perceived source trustworthiness operates across multiple processing pathways depending on cognitive capacity and elaboration likelihood:

• Under low elaboration, perceived trustworthiness serves as a simple peripheral or heuristic cue (e.g., “statements by honest people can be accepted without scrutiny”), generating immediate message assent.
• Under high elaboration, perceived trustworthiness influences systematic processing by biasing the generation of cognitive responses. High perceived trustworthiness suppresses counter-arguing and cynical scrutiny, whereas perceived dishonesty activates intense counter-argumentation and search for ulterior motives.

The Persuasion Knowledge Model (PKM)

Friestad and Wright’s (1994) Persuasion Knowledge Model provides the direct cognitive rationale for the STWS. The PKM posits that over time, targets develop intuitive knowledge about the goals, tactics, and strategies used by persuaders. When consumers identify a persuasion attempt (“change of meaning” heuristic), their persuasion knowledge is activated. They question the communicator’s motives, wondering whether claims reflect authentic product value or self-serving interests. The STWS measures the psychological outcome of this appraisal: when persuasion knowledge reveals aggressive, disguised, or mercenary tactics, scores on the STWS decline dramatically.

Anthropomorphism and Social Attribution Theory

Touré-Tillery and McGill (2015) framed the STWS around anthropomorphism theory (Epley, Waytz, & Cacioppo, 2007). Human communication is characterized by the presumption of intentional agency; human speakers possess mental states, personal motives, and moral agency. Consequently, human messengers are inherently scrutinized for manipulative intent. Conversely, non-human or anthropomorphized agents (e.g., an animated device or stylized brand character) are frequently perceived as lacking conscious self-serving deception, or alternatively, lack the moral capacity to be genuine. The STWS provides the empirical baseline to capture how mind perception and intentionality attributions govern evaluations of honesty and authenticity across human and non-human sources.

7. Validity

The psychometric validity of the Source Trustworthiness Scale has been demonstrated across multiple empirical studies, laboratory experiments, and cross-sectional investigations.

Construct and Structural Validity

Construct validity has been established through confirmatory factor analytic procedures across varied experimental stimuli. The four items load uniformly onto a single dominant latent factor representing perceived moral and communicative integrity. Standardized factor loadings consistently exceed .80 across diverse sample populations, demonstrating that each adjective pair taps into the overarching construct without significant residual multidimensionality.

Convergent Validity

Convergent validity is evidenced by high correlations between the STWS and established subscales of legacy source credibility batteries. Specifically, the STWS correlates robustly with the trustworthiness dimensions of Ohanian’s (1990) Tripartite Source Credibility Scale (r > .75, p < .001) and McCroskey and Teven’s (1999) Character/Goodwill dimensions (r > .70, p < .001). Furthermore, the scale exhibits positive, statistically significant associations with measures of consumer brand trust (Chaudhuri & Holbrook, 2001) and interpersonal benevolence (Mayer, Davis, & Schoorman, 1995).

Discriminant Validity

Crucially, the STWS demonstrates clean discriminant validity against adjacent communicator attributes:

• Versus Expertise: In structural equation models, the correlation between the STWS and source expertise scales (e.g., expert, experienced, knowledgeable, qualified) remains moderate (typically r = .30 to .45), demonstrating that variance in perceived moral integrity is statistically distinct from variance in technical competence. Average Variance Extracted (AVE) for the STWS exceeds the square of its correlation with expertise dimensions, satisfying the Fornell-Larcker criterion.
• Versus Physical and Social Attractiveness: The scale shares low variance with physical attractiveness metrics (r < .25), verifying that respondents differentiate aesthetic appreciation from moral trust.
• Versus Message Quality: When exposed to logically strong arguments delivered by an insincere communicator, STWS scores accurately register negative appraisals of the source despite high ratings of argument cogency.

Predictive and Criterion Validity

In Touré-Tillery and McGill’s (2015) experimental series, the STWS demonstrated exceptional predictive utility:

• Mediation of Persuasive Impact: In Study 1 and Study 2, perceived source trustworthiness fully mediated the interaction effect between source type (human vs. anthropomorphized entity) and persuasion outcome. For human communicators promoting self-serving products, activated persuasion knowledge suppressed STWS scores, which in turn predicted decreased purchase intent and adherence.
• Moderated Behavioral Compliance: In field and laboratory contexts, high STWS scores significantly predict behavioral compliance, including willingness to donate, click-through rates on digital campaigns, and adherence to preventive health behaviors.

8. Reliability

The STWS exhibits high internal consistency and measurement precision across independent investigations and experimental manipulations.

Internal Consistency

In the foundational investigations by Touré-Tillery and McGill (2015), the four-item scale demonstrated high internal consistency across multiple studies and distinct experimental contexts:

• Across diverse experimental conditions evaluating human spokespersons, the scale demonstrated a Cronbach’s alpha ranging from α = .88 to α = .94.
• In comparative evaluations evaluating anthropomorphized messengers and automated communicative agents, internal consistency remained equally robust, yielding Cronbach’s alpha values between α = .89 and α = .95.
• Subsequent replication studies in digital marketing, influencer credibility, and corporate social responsibility (CSR) messaging consistently report Cronbach’s alpha coefficients exceeding .90, and McDonald’s omega (ω) values exceeding .91.

Item-Total Statistics and Inter-Item Correlations

Standardized item-total correlations across the four semantic pairs consistently exceed .72, with no individual item deletion resulting in an increase in the composite Cronbach’s alpha. Average inter-item correlations typically fall within the optimal range of .65 to .80, indicating high construct coherence without excessive item redundancy.

Test-Retest Stability

While the STWS is frequently administered as a state-level measure sensitive to experimental manipulation and immediate message exposure, test-retest evaluations over brief intervals (e.g., 48 to 72 hours) in the absence of conflicting informational inputs show substantial temporal stability (intra-class correlation coefficient, ICC > .82), verifying that baseline assessments of known communicators remain stable over time.

9. Factor Analysis

Empirical analyses of the STWS confirm a strictly unifactorial (one-dimensional) latent structure.

Exploratory Factor Analysis (EFA)

Principal Axis Factoring (PAF) and Principal Component Analysis (PCA) conducted on the four items consistently reveal a single dominant eigenvalue well above the Kaiser criterion cutoff (λ > 1.0), typically accounting for 74% to 83% of the total variance. The second eigenvalue consistently falls below 0.40, and scree plot inspections confirm an unequivocal one-factor solution.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic investigations evaluating the single-factor model demonstrate fit indices meeting or exceeding standard psychometric criteria:

• Comparative Fit Index (CFI): ≥ .985 (indicating strong comparative fit against the baseline independence model).
• Tucker-Lewis Index (TLI): ≥ .975.
• Root Mean Square Error of Approximation (RMSEA): ≤ .045 (90% CI [.000, .078]), indicating close approximation to the population covariance structure.
• Standardized Root Mean Square Residual (SRMR): ≤ .020.

Multi-group invariance testing confirms metric and scalar invariance across experimental conditions (e.g., across human communicator conditions versus artificial/anthropomorphic conditions), confirming that the construct meaning remains uniform regardless of communicator morphology.

10. Instrument / Measurement Tool

The operational administration guidelines and psychometric structural parameters of the Source Trustworthiness Scale are detailed below:

• Test Type: Self-report psychometric rating scale; semantic differential technique.
• Construct Measured: Perceived communicator trustworthiness, moral integrity, honesty, and authenticity.
• Number of Items: 4 bipolar adjective pairs.
• Response Format: 7-point semantic differential scale (ranging from 1 = extreme negative anchor to 7 = extreme positive anchor, with 4 representing the neutral midpoint).
• Target Population: General adult populations, consumer samples, organizational stakeholders, and experimental participants in laboratory or online research environments.
• Administration Time: Extremely brief, typically requiring less than 60 seconds to complete.
• Scoring and Index Calculation: Responses to the four items are scored from 1 (negative pole: Dishonest, Unethical, Fake, Untrustworthy) to 7 (positive pole: Honest, Ethical, Genuine, Trustworthy). The composite index is calculated by computing the arithmetic mean across all four items:

Source Trustworthiness Score = (Item 1 + Item 2 + Item 3 + Item 4) / 4

• Interpretation of Scores: Composite scores range continuously from 1.00 to 7.00. Higher numerical values reflect greater perceived trustworthiness, moral integrity, and communicator authenticity. A score of 4.00 represents a neutral appraisal, scores below 4.00 signify perceived deceit, manipulative intent, or inauthenticity, and scores above 4.00 indicate perceived benevolence, ethical credibility, and reliability.

11. Permissions & Fee and Test Year

• Year of Formal Publication: 2015.
• Primary Citation: Touré-Tillery, M., & McGill, A. L. (2015). Who or what to believe: Trust and the differential persuasiveness of human and anthropomorphized messengers. Journal of Marketing, 79(4), 94–110.
• Copyright Holder: American Marketing Association (AMA) / SAGE Publications (for the original published article).
• Permissions and Academic Licensing: The Source Trustworthiness Scale is broadly accessible for non-commercial academic research, pedagogical purposes, and non-profit scientific investigation without requiring formal permission fees, provided the original work is properly attributed in all scholarly outputs. Commercial deployment within proprietary market research, commercial diagnostic software, or consulting audits may require formal copyright clearance from the American Marketing Association or SAGE Publications.

12. References

• Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39(5), 752–766. https://doi.org/10.1037/0022-3514.39.5.752
• Chaudhuri, A., & Holbrook, M. B. (2001). The chain of effects from brand trust and brand affect to brand performance: The role of brand loyalty. Journal of Marketing, 65(2), 81–93. https://doi.org/10.1509/jmkg.65.2.81.18255
• Epley, N., Waytz, A., & Cacioppo, J. T. (2007). On seeing human: A three-factor theory of anthropomorphism. Psychological Review, 114(4), 864–886. https://doi.org/10.1037/0033-295X.114.4.864
• Fiske, S. T., Cuddy, A. J., & Glick, P. (2007). Universal dimensions of social cognition: Faces and traits, faces and races. Trends in Cognitive Sciences, 11(2), 77–83. https://doi.org/10.1016/j.tics.2006.11.005
• Friestad, M., & Wright, P. (1994). The Persuasion Knowledge Model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1–31. https://doi.org/10.1086/209380
• Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion: Psychological studies of opinion change. Yale University Press.
• 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
• McCroskey, J. C., & Teven, J. J. (1999). Goodwill: A reexamination of the construct and its measurement. Communication Monographs, 66(1), 90–103. https://doi.org/10.1080/03637759909376464
• Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness, and attractiveness. Journal of Advertising, 19(3), 39–52. https://doi.org/10.1080/00913367.1990.10673191
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. In Communication and Persuasion (pp. 1–24). Springer, New York, NY. https://doi.org/10.1007/978-1-4612-4964-1_1
• Touré-Tillery, M., & McGill, A. L. (2015). Who or what to believe: Trust and the differential persuasiveness of human and anthropomorphized messengers. Journal of Marketing, 79(4), 94–110. https://doi.org/10.1509/jm.13.0521

13. Items of the Scale (Questionnaire)