Source Credibility (Trustworthiness) — Compilation (SCCRED)

1. Abstract

The Source Credibility (Trustworthiness) — Compilation (SCCRED) represents an operationalized psychometric reservoir of bipolar semantic differential adjective pairs designed to quantify perceived communicator integrity, believability, and moral uprightness across social psychological, organizational, mass communication, and marketing contexts. Rooted in foundational research from the Yale Communication and Attitude Change Program led by Carl Hovland and systematically formalized in consumer behavior paradigms by MacKenzie and Lutz (1989) and Ohanian (1990), the compilation captures the fundamental evaluative axis along which human recipients process message transmission: the degree of confidence in an information source’s communicative intent. While source credibility is recognized as a multidimensional meta-construct typically spanning competence (expertise), attractiveness, and trustworthiness, the SCCRED pool focuses primarily on the critical ethical and veridical dimensions of Trustworthiness / Integrity alongside its operational intersections with Competence / Credibility.

The full item compilation comprises a curated pool of 13 bipolar semantic differential adjective pairs evaluated along standardized 7-point rating continua (e.g., untrustworthy/trustworthy, dishonest/honest, insincere/sincere). Psychometric evaluations across diverse empirical investigations demonstrate exceptional internal consistency reliability, with Cronbach’s alpha (α) coefficients routinely exceeding .90, composite reliability (CR) values surpassing .88, and test-retest stabilities over 2- to 4-week intervals displaying intraclass correlation coefficients (ICC) ranging between .78 and .86. Exploratory and confirmatory factor analyses consistently substantiate either a clean unidimensional configuration when isolated solely for epistemic honesty, or a distinct two-factor orthogonal or oblique structural model when combined with perceptual competence items. Factor loadings for primary trustworthiness markers typically span from .75 to .93, accompanied by average variance extracted (AVE) estimates routinely eclipsing .65, firmly establishing both convergent and discriminant validity against related constructs such as source attractiveness, brand equity, and argument quality.

2. Keywords

Source credibility, trustworthiness, integrity, semantic differential, advertising pretesting, persuasion, Hovland Yale model, psychometrics, attitude toward the ad, cognitive response theory

3. Authors

The Source Credibility (Trustworthiness) Compilation does not derive from a single, isolated psychological test developer, but reflects an empirical synthesis and standardization of psychometric scales developed across prominent persuasion and consumer psychology milestones. Key foundational contributors include:

• Scott B. MacKenzie, Ph.D. — Professor Emeritus of Marketing, Kelley School of Business, Indiana University Bloomington. Renowned for structural equation modeling, advertising effectiveness, and measurement artifact detection in consumer research.
• Richard J. Lutz, Ph.D. — J.C. Penney Professor of Marketing, Warrington College of Business, University of Florida. Past president of the Association for Consumer Research (ACR) and former editor of the Journal of Consumer Research.
• Roobina Ohanian, D.B.A. — Pioneer in the tri-component operationalization of celebrity endorser credibility (Trustworthiness, Expertise, and Attractiveness), formerly affiliated with the Department of Marketing, J. Mack Robinson College of Business, Georgia State University.
• Carl I. Hovland, Ph.D. (Historical Antecedent) — Former Chair of the Department of Psychology at Yale University, whose experimental investigations with Walter Weiss and Irving Janis established the empirical separation of source expertise and trustworthiness.

Contemporary psychometric inquiries regarding the adaptation and application of these compilations can be directed to academic literature repositories and peer-reviewed works curated in the Journal of Marketing, Journal of Marketing Research, and Journal of Consumer Research.

4. Purpose

The primary purpose of the Source Credibility (Trustworthiness) Compilation (SCCRED) is to provide empirical researchers and clinical or applied practitioners with an ecologically valid, psychometrically sound, and flexible tool to measure recipient perceptions of communicator authenticity, ethical motivation, and objective truthfulness. Across human communication systems, message reception does not occur in a cognitive vacuum; receivers continuously scrutinize the transmitter of information to assess whether the communicated proposition mirrors objective reality or reflects ulterior, manipulative motives.

Theoretical and Conceptual Rationale

In theoretical frameworks such as the Elaboration Likelihood Model (ELM) and the Heuristic-Systematic Model (HSM), source credibility acts alternately as a peripheral cue, a cognitive heuristic, an argument in itself, or a catalyst that biases systematic cognitive elaboration. When cognitive capacity or motivation is constrained, individuals rely heavily on heuristic assumptions, such as “statements made by trustworthy sources can be trusted.” Conversely, under conditions of high elaboration, the perceived trustworthiness of an advocate determines whether recipient counterarguing is actively suppressed or stimulated. By systematically assessing the perceived ethical orientation of the source, the SCCRED pool isolates whether persuasion variance is attributable to intrinsic message persuasiveness or to the perceived integrity of the transmitting entity.

Research Applications

In experimental and field communication research, the compilation functions in several distinct capacities:

• Advertising Pretesting and Persuasion Research: In structural frameworks such as MacKenzie and Lutz’s (1989) antecedent model of attitude toward the ad ($A_{ad}$), advertiser credibility directly dictates cognitive responses, ad trust, and downstream brand attitudes ($A_b$). SCCRED enables researchers to quantify source-level variance distinct from message execution factors.
• Corporate Communications and Public Relations: Evaluating the degree of cynicism or public trust elicited by corporate spokespersons, crisis management communications, and public health officials during epidemiological interventions.
• Digital and Social Media Influencer Studies: Disentangling the commercial monetization motives of digital content creators from their perceived authenticity, objective recommendations, and intrinsic credibility.
• Political Communication and Misinformation Research: Quantifying partisan biases in source evaluation and measuring the psychological impact of “fake news” flags, fact-checking interventions, and algorithmic algorithmic transparency cues on communicator believability.

Clinical and Applied Setting Utility

Beyond commercial and mass-media persuasion, the construct of communicator trustworthiness is pivotal in therapeutic, medical, and counseling psychology. The therapeutic alliance relies fundamentally on the client’s perception of the therapist as an empathic, authentic, and thoroughly trustworthy practitioner. In medical decision-making paradigms, patient compliance with complex pharmacological regimens directly correlates with perceived physician credibility. The SCCRED pool, adapted for relational dyads, facilitates granular clinical research assessing how clinician communication styles, institutional affiliations, and disclosure of conflicts of interest affect patient trust and treatment adherence.

5. Psychological Construct

Source credibility is widely conceptualized in social psychology as an attributional construct formulated in the mind of the receiver rather than an absolute, static property residing solely within the communicator. Psychometric modeling has long recognized that source credibility consists of distinct, intercorrelated factors, among which Trustworthiness occupies the preeminent affective and ethical locus.

The Core Dimension: Trustworthiness and Integrity

The trustworthiness dimension captures the recipient’s degree of psychological confidence in, and acceptance of, the communicator and their message. Crucially, trustworthiness refers to the perceived willingness of the source to tell the truth, untainted by hidden agendas, opportunistic deception, or commercial bias. A communicator may possess world-class technical competence, yet if the recipient suspects opportunistic intentions, the overall credibility of the message collapses. Trustworthiness encapsulates several critical cognitive appraisals:

• Perceived Honesty and Truthfulness: The baseline appraisal that the assertions made by the source reflect factual reality rather than calculated fabrications.
• Sincerity and Benevolence: The perception that the source genuinely believes the advocacy and is motivated by a prosocial desire to inform or protect the recipient, rather than purely exploitative self-interest.
• Objectivity and Lack of Bias: The degree to which the communicator is perceived as intellectually independent, fair-minded, and free of undue external influence or unstated sponsorships.
• Dependability and Moral Integrity: The characterological assessment that the communicator operates within a coherent, pro-normative ethical framework, rendering their future statements reliable and consistent over time.

The Secondary Dimension: Competence and Credibility

While the SCCRED compilation isolates trustworthiness as its focal operational target, several items in the pool bridge the evaluative gap between ethical character and cognitive competence. The competence (or expertise) dimension represents the perceived ability of the source to make valid assertions. It encompasses:

• Authoritativeness and Knowledgeability: The degree to which the communicator possesses formal education, procedural training, professional mastery, or deep experiential familiarity with the topic under consideration.
• Technical Capability and Sound Judgment: The receiver’s subjective confidence that the source has executed rigorous due diligence prior to formulating their advocacy, minimizing the probability of cognitive error.

In structural path models, trustworthiness and competence frequently exhibit substantial positive covariance ($r \approx .50\text{ to }.70$), yet they represent empirically and functionally distinct vectors. For instance, an expert witness may be perceived as extremely competent (possessing specialized forensic expertise) but deeply untrustworthy (suspected of providing bought-and-paid-for biased testimony). Conversely, an ordinary peer or consumer reviewer may be viewed as remarkably trustworthy (completely sincere, altruistic, and well-intentioned) yet relatively low in technical competence. The SCCRED compilation enables investigators to pinpoint whether communicative resistance stems from doubts regarding the source’s moral character versus doubts regarding their technical qualifications.

6. Theoretical Framework

The conceptual architecture of the Source Credibility (Trustworthiness) Compilation is grounded in several interrelated paradigms of cognitive psychology, attribution theory, and persuasion science.

The Yale Communication and Attitude Change Approach

The foundational paradigm dates to the post-World War II experiments conducted by Carl Hovland and colleagues at Yale University. Hovland, Janis, and Kelley (1953) formalized the classic matrix question: “Who says what to whom through what channel with what effect?” Hovland and Weiss (1951) systematically manipulated the identity of message sources while holding the literal textual arguments invariant. They observed that identical arguments generated dramatically greater immediate attitude change when attributed to sources high in credibility (e.g., Robert J. Oppenheimer discussing atomic submarines) versus sources low in credibility (e.g., the Soviet newspaper Pravda). Hovland and colleagues explicitly bisected source credibility into two operational components: (1) expertness, defined as the extent to which a communicator is perceived to be capable of making correct assertions, and (2) trustworthiness, defined as the degree of confidence the listener has in the communicator’s intent to communicate the assertions he considers most valid.

Attribution Theory and Knowledge/Reporting Biases

The psychological mechanism through which trustworthiness functions was substantially refined through Attribution Theory, specifically through the work of Eagly, Chaiken, and Wood (1978). When exposed to a persuasive message, recipients generate causal attributions regarding why the communicator is taking a particular stance. Eagly et al. delineated two primary cognitive biases that recipients routinely attribute to communicators:

• Knowledge Bias: The recipient suspects that the source’s background, training, or selective exposure has resulted in a distorted view of objective reality. This directly undermines the competence dimension.
• Reporting Bias: The recipient suspects that the source knows the truth, but is unwilling to state it honestly due to external constraints, financial incentives, social pressures, or strategic deception. This directly undermines the trustworthiness dimension.

When recipients perceive a strong reporting bias, the perceived trustworthiness score on the SCCRED plummets, triggering intense cognitive resistance, cynical counterarguing, or total discounting of the presented evidence.

The Persuasion Knowledge Model (PKM)

In modern advertising and marketing scholarship, the Persuasion Knowledge Model developed by Friestad and Wright (1994) provides a complementary cognitive framework. Consumers gradually amass intuitive beliefs and psychological heuristics regarding how agents construct persuasion attempts. When consumers recognize persuasion tactics, their persuasion knowledge is activated. If the advertiser or source is perceived as utilizing manipulative, deceptive, or coercive maneuvers, the recipient engages in “change of meaning” coping mechanisms, downgrading the perceived honesty, sincerity, and ethical standing of the sponsor. The SCCRED pool provides the direct psychometric gauge for evaluating the outcome of these coping episodes.

Dual-Process Cognitive Architectures: ELM and HSM

In Petty and Cacioppo’s Elaboration Likelihood Model, the role of perceived trustworthiness varies predictably along the elaboration continuum. Under low cognitive involvement or high distraction, source trustworthiness operates as a non-thoughtful peripheral cue, triggering immediate heuristic acceptance or rejection via basic affective heuristics. Under moderate elaboration, source trustworthiness serves as an informational catalyst, deciding whether the individual expends effort analyzing the message. Under high elaboration, trustworthiness serves as an interpretive frame that biases the valence of elaboration, determining whether ambiguous evidence is processed with charitable openness or aggressive skepticism.

7. Validity

The Source Credibility (Trustworthiness) Compilation has been subjected to extensive psychometric scrutiny across more than four decades of empirical persuasion literature, demonstrating exemplary construct, convergent, discriminant, and predictive validity.

Construct and Factorial Validity

Construct validity has been affirmed through repeated exploratory and confirmatory factor analyses. Ohanian (1990) subjected an extensive candidate pool of source characteristics to rigorous scale purification across multiple independent consumer samples evaluating diverse celebrity spokespersons. Confirmatory factor analysis (CFA) revealed that trustworthiness items loaded cleanly onto an independent latent factor separate from physical attractiveness and professional expertise, with standardized factor loadings consistently exceeding .80 ($p < .001$) and explaining substantial portions of common variance.

Convergent Validity

Convergent validity is established when items postulated to reflect a single theoretical construct demonstrate high statistical communality and share substantial variance. In structural equation modeling studies utilizing the SCCRED pool, Average Variance Extracted (AVE) values for the trustworthiness dimension routinely exceed the accepted .50 threshold established by Fornell and Larcker (1981), frequently reaching values between .68 and .82. Furthermore, composite reliability coefficients regularly exceed .90, demonstrating that the disparate bipolar adjectives (e.g., honest, sincere, dependable, trustworthy) converge cohesively on the underlying latent construct of communicator integrity.

Discriminant Validity

A rigorous test of any credibility metric is its ability to differentiate moral/ethical evaluations from adjacent social-evaluative dimensions, such as aesthetic appeal, likeability, and cognitive capability. Across empirical studies:

• Distinction from Source Attractiveness: Inter-factor correlations between trustworthiness and source attractiveness typically range between $.20$ and $.45$, well below the threshold that would suggest collinearity. Fornell-Larcker tests consistently confirm that the square root of the AVE for trustworthiness exceeds its inter-factor correlations with attractiveness.
• Distinction from Source Expertise: Although trustworthiness and competence share moderate covariance ($r = .50\text{ to }.68$), nested model comparisons (constraining the factor correlation to 1.0 versus freely estimating it) yield statistically significant chi-square differences ($\Delta\chi^2(1) > 100.0, p < .0001$), firmly confirming that receivers maintain distinct mental representations for communicator capability versus communicator honesty.
• Heterotrait-Monotrait Ratio of Correlations (HTMT): Contemporary reassessments using the HTMT criterion routinely yield values below .85, confirming strict discriminant validity against related marketing constructs including brand familiarity, cognitive ad elaboration, and general positive affect.

Predictive and Nomological Validity

Nomological validity is demonstrated through the scale’s predictable integration into established theoretical networks. In MacKenzie and Lutz’s (1989) benchmark structural equation model, advertiser credibility (measured using trustworthiness and expertise markers) demonstrated a powerful, statistically significant direct positive path to attitude toward the ad ($A_{ad}$, $\beta = .41, p < .01$). Downstream, $A_{ad}$ accounted for substantial variance in attitude toward the brand ($A_b$) and subsequent purchase intentions ($PI$). Similarly, in public health communication trials, higher baseline scores on the SCCRED trustworthiness scale significantly predicted vaccine acceptance, adherence to medical quarantine advisories, and decreased susceptibility to conspiratorial counter-attitudinal communications.

8. Reliability

The psychometric reliability of the Source Credibility (Trustworthiness) Compilation has been documented extensively across experimental social psychology, communication science, and marketing research.

Internal Consistency Reliability

Internal consistency metrics for the SCCRED trustworthiness scales have consistently surpassed standard psychometric benchmarks across diverse demographic groups, cultural contexts, and experimental designs:

• Cronbach’s Alpha ($lpha$): In original scale purification work by Ohanian (1990), the internal consistency for the 5-item core trustworthiness subscale yielded a Cronbach’s alpha of $.92$ in exploratory samples and $.91$ in cross-validation samples. MacKenzie and Lutz (1989) observed an alpha of $.88$ for their concise operationalization within structural equation models. Subsequent research pooling subsets from the 13-item reservoir routinely report alpha coefficients spanning from $.89$ to $.96$.
• Composite Reliability ($CR$ / McDonald’s $\omega$): In structural equation modeling frameworks that correct for unequal factor loadings (tau-equivalence violations), McDonald’s omega hierarchical ($\omega_h$) and composite reliability coefficients routinely range between $.91$ and $.95$, proving that the item pool provides exceptionally precise measurement across the latent continuum.
• Item-Total Correlations: Corrected item-total correlations for individual bipolar pairs (such as untrustworthy / trustworthy, dishonest / honest, and insincere / sincere) consistently fall within the $.72$ to $.88$ range, indicating that no single redundant or divergent item degrades the scale’s internal cohesion.

Test-Retest Stability

Evaluating the temporal stability of source credibility metrics necessitates distinguishing between transient situational source evaluations and enduring communicator reputation. In longitudinal paradigms evaluating established, enduring communicators (e.g., reputable news anchors, familiar institutional spokespersons) over 2- to 4-week test-retest intervals without intervening persuasive crises, the SCCRED demonstrates high temporal stability, with intraclass correlation coefficients (ICC) ranging between $.78$ and $.86$. However, when experimental interventions (such as news exposures revealing corporate corruption or conflict-of-interest revelations) are introduced, the scale exhibits sensitive, immediate, and statistically significant responsiveness, confirming that while the tool measures stable evaluations under steady-state conditions, it remains acutely sensitive to new evidentiary inputs.

9. Factor Analysis

The latent structure of the Source Credibility Compilation has been repeatedly scrutinized using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

Early psychometric purification studies utilizing principal axis factoring (PAF) or maximum likelihood extraction with oblique (e.g., Promax or Oblimin) rotation on the broader pool of source credibility markers consistently reveal eigenvalues greater than 1.0 corresponding to distinct conceptual dimensions. When the 13 items comprising the trustworthiness and integrity compilation are analyzed alongside general source competence items, a clear two-factor structure emerges:

• Factor 1: Trustworthiness / Moral Integrity: Accounts for the largest share of explained variance (typically between $45%$ and $58%$). Items such as untrustworthy/trustworthy, dishonest/honest, insincere/sincere, and undependable/dependable load cleanly on this factor, with primary loadings spanning from $.76$ to $.92$, and cross-loadings onto the competence factor typically remaining below $.25$.
• Factor 2: Competence / Expert Credibility: Accounts for an additional $12%$ to $18%$ of explained variance. Items evaluating knowledge, skill, and authoritative credibility cluster tightly here, with primary loadings between $.70$ and $.88$.

Confirmatory Factor Analysis (CFA) and Model Fit Indices

To confirm structural integrity, researchers routinely test competing factor models using maximum likelihood structural estimation. The standard measurement model specifies a designated number of continuous latent variables with correlated error terms strictly fixed to zero.

As illustrated in the comparative table above, the Two-Factor Oblique Model demonstrates superior goodness-of-fit indices across all standard criteria: the ratio of chi-square to degrees of freedom ($\chi^2/df = 1.86$) falls comfortably below the conservative benchmark of 2.0; the Comparative Fit Index (CFI = .981) and Tucker-Lewis Index (TLI = .976) both far exceed the .95 threshold; the Root Mean Square Error of Approximation (RMSEA = .042) meets the rigorous standard for close model fit; and the Standardized Root Mean Square Residual (SRMR = .029) is substantially below .05. These results conclusively demonstrate that while communicator competence and moral trustworthiness are correlated dimensions of overall source credibility, forcing them into a single undifferentiated global factor results in significant model mis-specification.

10. Instrument / Measurement Tool

The Source Credibility (Trustworthiness) Compilation (SCCRED) operates as an adaptable psychometric inventory. In standard operational deployments, researchers select either the complete 13-item reservoir or validated 3- to 5-item sub-permutations depending on survey administration length constraints and analytical requirements.

• Test Type: Bipolar Semantic Differential Questionnaire / Rating Scale.
• Target Population: Adolescents through adults (reading grade level approximately 6th grade and above). Validated for diverse general public, organizational, and consumer populations.
• Item Format: Paired opposing adjectives positioned at the opposing polar anchors of a continuous multi-point horizontal continuum.
• Total Item Pool: 13 bipolar adjective pairs specifically cataloged across the trustworthiness, honesty, integrity, and veridical credibility spectrum.
• Response Scale: Typically administered as a 7-point bipolar rating continuum (ranging from 1 to 7, or scored from -3 to +3). Midpoints (e.g., 4) denote neutral, ambivalent, or unformed evaluations.
• Administration Time: Approximately 1 to 3 minutes for selected subsets (3 to 5 items); 3 to 5 minutes for the full 13-item compilation.
• Scoring Procedure:
  • Polarity Coding: Anchor orientations must be standardized prior to scoring. Positively valenced adjectives (e.g., Honest, Trustworthy, Sincere) are assigned a score of 7, whereas negatively valenced adjectives (e.g., Dishonest, Untrustworthy, Insincere) are assigned a score of 1. If physical presentation alternates direction to mitigate acquiescence and straight-lining response sets, items must be reverse-coded accordingly.
  • Composite Score Computation: An overall Trustworthiness index is computed either by calculating the arithmetic mean (sum of scores divided by the number of completed items) or by summing raw item scores. Composite means range from 1.0 (complete perceived untrustworthiness / severe perceived deceit) to 7.0 (absolute perceived trustworthiness / uncompromised integrity).
  • Latent Variable Modeling: In SEM investigations, raw indicators are directly used to load onto the latent factor with one indicator fixed to 1.0 for scale identification or set freely via standardized latent variance constraints.

11. Permissions & Fee and Test Year

• Test Year: 1989 (formalized within structural advertising frameworks by MacKenzie & Lutz) and 1990 (purified and validated within endorser credibility scales by Ohanian), with theoretical origins tracing back to Hovland & Weiss (1951).
• Accessibility & Academic Status: Publicly accessible for academic, scientific, and instructional research. The compilation consists of standard semantic differential adjective descriptors established across foundational social psychological and marketing literature.
• Licensing & Fees: No licensing fees, royalty payments, or formal organizational permissions are mandated for non-commercial academic investigations. Researchers may freely adapt the bipolar pairs for empirical studies provided proper academic attribution is maintained.
• Commercial Applications: Commercial market research firms, corporate pretesting platforms, and proprietary testing agencies utilizing specific commercial implementations or proprietary diagnostic software based on these dimensions may be subject to proprietary copyright or trademarks established by specific market research entities. Academic citations must reference the primary publications by MacKenzie and Lutz (1989) and Ohanian (1990).

12. References

Below are primary academic works and empirical validations foundational to the operationalization and psychometric modeling of the Source Credibility (Trustworthiness) Compilation:

• Eagly, A. H., Chaiken, S., & Wood, W. (1978). An attribution analysis of the effect of communicator bias on persuasion: An overview and test of a new theoretical approach. Journal of Personality and Social Psychology, 36(4), 424–440. https://doi.org/10.1037/0022-3514.36.4.424
• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• 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/10.1086/209378
• Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion: Psychological studies of opinion change. Yale University Press.
• Hovland, C. I., & Weiss, W. (1951). The influence of source credibility on communication effectiveness. Public Opinion Quarterly, 15(4), 635–650. https://doi.org/10.1086/266350
• MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing, 53(2), 48–65. https://doi.org/10.1177/002224378902600105
• 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
• Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of five decades’ evidence. Journal of Applied Social Psychology, 34(2), 243–281. https://doi.org/10.1111/j.1559-1816.2004.tb02547.x

13. Items of the Scale