Source Credibility Measures

1. Abstract

The Source Credibility Measures, developed by James C. McCroskey and Jason J. Teven in 1999, represent an authoritative, psychometrically validated operationalization of ethos and source evaluation within human communication research. Rooted in classical Aristotelian rhetorical theory, the instrument captures a receiver’s perceptual assessment of a communicator across three distinct, correlated dimensions: Competence, Trustworthiness (character), and Goodwill (perceived caring). The instrument consists of 18 semantic differential items presented along a 7-point bipolar continuum, with each dimension indexed by six item pairs. Psychometric evaluations across diverse interpersonal, instructional, and organizational settings demonstrate high internal consistency, with Cronbach’s alpha coefficients typically ranging from .80 to .94 across the three subscales. Confirmatory and exploratory factor analyses utilizing oblique rotation verify a correlated three-factor structural model, overcoming the mathematical and theoretical limitations of earlier orthogonal models from the 1960s and 1970s. This article presents an exhaustive review of the instrument’s historical development, theoretical lineage, psychometric indices, scoring architecture, and practical administration across instructional, organizational, and mediated communication paradigms.

2. Keywords

Source Credibility Measures, Ethos, Source Credibility, Perceived Caring, Goodwill, Competence, Trustworthiness, Semantic Differential, James C. McCroskey, Communication Theory, Interpersonal Perception, Organizational Communication

3. Authors

The contemporary 18-item Source Credibility Measure was formulated and validated by:

• James C. McCroskey, Ed.D. (1938–2012): Professor Emeritus of Communication Studies at West Virginia University. Dr. McCroskey was one of the most prolific scholars in the communication discipline, pioneering empirical measurement in communication apprehension, nonverbal immediacy, instructional communication, and ethos.
• Jason J. Teven, Ed.D.: Professor of Human Communication Studies at California State University, Fullerton. Dr. Teven’s scholarship focuses extensively on instructional communication, supervisor-subordinate relationships, interpersonal influence, and source evaluation metrics.

4. Purpose

The fundamental purpose of the Source Credibility Measures is to provide an empirically sound, theoretically authentic index of an individual’s perceived credibility as judged by an audience or interlocutor. In applied psychological, instructional, and organizational settings, human behavior and message receptivity are rarely governed solely by the objective logical merits of a message. Rather, cognitive processing, attitude change, and behavioral compliance are profoundly mediated by the perceived characteristics of the message source. McCroskey and Teven formulated this instrument to address theoretical omissions and measurement artifacts present in earlier source evaluation instruments, specifically re-establishing the critical construct of goodwill (perceived caring) alongside competence and character.

In instructional environments, the scale is routinely deployed to assess student evaluations of teaching, examining how teacher credibility correlates with student affective learning, cognitive retention, and student motivation. In corporate and organizational research, the tool serves to quantify how managerial competence, moral integrity, and perceived empathy drive employee job satisfaction, organizational citizenship behaviors, employee burnout, and turnover intentions. In public relations, political communication, and health communication, researchers implement the measure to evaluate spokesperson viability, public trust in institutional representatives, and patient adherence to medical practitioners’ recommendations. By decomposing credibility into three distinct sub-constructs, the scale offers targeted diagnostic insights, identifying precisely whether a source’s communicative vulnerability stems from deficits in technical knowledge, perceived honesty, or perceived lack of empathy.

5. Psychological Construct

Source credibility represents a receiver-based perception rather than an objective trait residing inherently within a speaker. The Source Credibility Measures assess three interconnected psychometric facets:

Competence

Competence reflects the degree to which a communicator is perceived by the receiver to be knowledgeable, expert, intelligent, trained, and capable within a given domain. Rather than measuring actual cognitive capability or verified mastery, this dimension captures the receiver’s subjective attribution of domain-specific qualification and general intelligence. A speaker may possess high formal qualifications, but if their communicative presentation is disorganized or hesitant, their perceived competence will diminish. Conversely, polished rhetorical delivery can elevate perceived competence even in the absence of verified subject-matter expertise. Exemplary item polarities include Intelligent / Unintelligent, Trained / Untrained, and Expert / Inexpert.

Trustworthiness

Trustworthiness evaluates the perceived character, honesty, integrity, and ethical baseline of the communicator. Receivers continuously scan verbal and nonverbal cues to determine whether a communicator possesses honorable intentions and whether their assertions represent authentic truth. When trustworthiness is low, receivers enact defensive cognitive filtering, scrutinizing messages for manipulative subtexts or dismissing informational validity altogether. Trustworthiness is operationalized through bipolar adjective pairs such as Honest / Dishonest, Trustworthy / Untrustworthy, and Moral / Immoral.

Goodwill / Perceived Caring

Goodwill, historically identified as eunoia, represents the receiver’s perception that the source has their best interests at heart, genuinely cares about their well-being, and is not purely self-centered or exploitative. McCroskey and Teven demonstrated that receivers separate technical competence and moral honesty from interpersonal concern. A source can be acknowledged as brilliant (competent) and principled (trustworthy), yet be perceived as indifferent, callous, or self-absorbed. Goodwill serves as an interpersonal bridge; when audiences perceive that a communicator genuinely values them, psychological resistance decreases markedly. Adjective anchors include Cares about me / Doesn’t care about me, Has my interests at heart / Doesn’t have my interests at heart, and Not self-centered / Self-centered.

6. Theoretical Framework

The theoretical architecture of the Source Credibility Measures traces its lineage directly to classical rhetoric, specifically Aristotle’s Rhetoric (composed in the 4th century BCE). Aristotle posited that persuasion generated through speech relies on three artistic proofs: logos (rational argumentation), pathos (emotional appeal), and ethos (the personal character and perceived qualities of the orator). Crucially, Aristotle asserted that ethos is comprised of three essential elements: phronesis (practical wisdom or competence), arete (moral excellence or virtue), and eunoia (goodwill toward the audience).

During the mid-twentieth century, early empirical communication researchers and social psychologists—most notably Carl Hovland and colleagues in the Yale Communication and Attitude Change Program—sought to quantify ethos. However, initial factor-analytic investigations by McCroskey (1966) and contemporaries repeatedly collapsed source credibility into two primary dimensions: Authoritativeness (Competence) and Character (Trustworthiness). Goodwill frequently failed to emerge as an independent factor in these early studies. McCroskey and Teven (1999) hypothesized that this was not due to the absence of goodwill in receiver cognition, but rather was an artifact of item selection and methodological constraints, specifically the reliance on orthogonal (varimax) factor rotation which artificially constrained factors to be uncorrelated. By reintroducing explicit perceived-caring items and employing oblique rotation, McCroskey and Teven empirically validated Aristotle’s tripartite model of ethos, demonstrating that receiver attribution processes simultaneously evaluate expertise, integrity, and relational benevolence.

7. Validity

Extensive empirical investigation supports the construct, convergent, discriminant, and criterion-related validity of the Source Credibility Measures:

• Construct Validity: Confirmatory factor analyses across diverse demographic and relational samples support a three-factor correlated model. Goodness-of-fit indices consistently meet conventional thresholds (CFI > .95, RMSEA < .06), demonstrating that the three dimensions measure distinct yet interrelated facets of ethos rather than an undifferentiated global impression.
• Convergent Validity: Subscale scores correlate strongly with conceptually related relational constructs. For instance, the Goodwill/Caring subscale correlates significantly with Virginia Richmond’s Nonverbal Immediacy Scale ($r = .50$ to $.70$), affirming that communicators who exhibit warm, approachable nonverbal behaviors are perceived as possessing higher goodwill. Similarly, the Competence subscale correlates robustly with measures of cognitive learning and communicator clarity.
• Discriminant Validity: Although the three dimensions are intercorrelated (factor correlations typically ranging between $.40$ and $.70$), multi-trait multi-method matrices and differential predictive patterns verify that they are not redundant. For example, in instructional contexts, Goodwill often predicts student motivation and affect toward the instructor more strongly than Competence, whereas Competence more heavily predicts perceived course rigor and formal learning outcomes.
• Predictive / Criterion Validity: The scale reliably predicts real-world organizational and relational outcomes. In workplace settings, supervisor credibility scores account for substantial variance in subordinate job satisfaction, affective organizational commitment, and willingness to voice concerns. In health communication, physician credibility predicts patient compliance with therapeutic regimens.

8. Reliability

The internal consistency of the Source Credibility Measures has been demonstrated across hundreds of independent empirical studies spanning over two decades:

• Internal Consistency (Cronbach’s Alpha): Across empirical literature, alpha estimates consistently fall between $.80$ and $.94$ for all three subscales:
  • Competence: $\alpha = .85$ to $.93$
  • Trustworthiness: $\alpha = .86$ to $.92$
  • Goodwill / Caring: $\alpha = .88$ to $.94$
• Composite Reliability: Structural equation modeling studies routinely report composite reliability (CR) values exceeding $.88$ for each latent dimension, indicating high shared variance among observed indicators.
• Test-Retest Stability: While source credibility is inherently state-sensitive and subject to shifting communicative inputs, stable relational contexts (e.g., long-standing employee-supervisor dyads) yield two- to four-week test-retest reliability coefficients ranging from $.78$ to $.86$, reflecting adequate temporal stability when contextual factors remain constant.

9. Factor Analysis

The structural integrity of the 18-item measure has been substantiated through exploratory (EFA) and confirmatory factor analyses (CFA):

Exploratory Factor Analysis

When subjected to principal axis factoring or maximum likelihood extraction with oblique (promax or direct oblimin) rotation, the 18 items cleanly resolve into three primary factors accounting for over $60%$ to $70%$ of the total variance. Unlike early orthogonal models that forced factors into artificial 90-degree independence, oblique rotation permits inter-factor correlations (typically $r = .45$ to $.68$), directly reflecting real-world cognitive assessments where people who are perceived as highly moral are also frequently seen as caring and capable. Factor loadings for primary items routinely exceed $.65$, with minimal cross-loadings rarely exceeding $.25$.

Confirmatory Factor Analysis

CFA studies comparing unidimensional, two-factor (Competence and Character), and three-factor oblique models demonstrate unambiguous superiority for the three-factor model. Typical fit indices include:

• $\chi^2 / \text{df} < 2.5$
• Comparative Fit Index (CFI) $ge .96$
• Tucker-Lewis Index (TLI) $ge .95$
• Root Mean Square Error of Approximation (RMSEA) $le .055$ (90% CI [.042, .068])
• Standardized Root Mean Square Residual (SRMR) $le .045$

10. Instrument / Measurement Tool

• Construct Measured: Source Credibility (perceived ethos in interpersonal and public communication contexts).
• Dimensions / Subscales: Three correlated subscales of 6 items each:
  • Competence: Items 1, 2, 7, 11, 13, 16
  • Goodwill / Perceived Caring: Items 3, 5, 8, 9, 15, 18
  • Trustworthiness: Items 4, 6, 10, 12, 14, 17
• Item Count: 18 bipolar semantic differential item pairs.
• Response Format: 7-point semantic differential scale anchored by bipolar adjectives (1 and 7 represent very strong feelings, 2 and 6 represent strong feelings, 3 and 5 represent fairly weak feelings, and 4 represents undecided).
• Administration Time: Approximately 3 to 5 minutes.
• Target Population: Adolescents and adults evaluating a specified communicator (e.g., manager, instructor, public speaker, peer, physician).
• Scoring Architecture:
  • Items are balanced to prevent response set bias. Negatively keyed items where the positive anchor appears on the right must be reverse-coded (recode $1=7, 2=6, 3=5, 4=4, 5=3, 6=2, 7=1$).
  • Specifically, items 2, 6, 7, 8, 13, 14, 15, 17, and 18 require reverse recoding when calculating dimension scores so that higher scores consistently reflect higher perceived credibility.
  • Subscale scores are obtained by summing the six corresponding recoded items for each dimension, producing a score range of 6 to 42 for Competence, 6 to 42 for Goodwill, and 6 to 42 for Trustworthiness.
  • A total Source Credibility composite score may be derived by summing all three dimensions (range: 18 to 126), though reporting distinct dimension scores is strongly recommended for diagnostic fidelity.

11. Permissions & Fee and Test Year

The modern 18-item Source Credibility Measure was published in 1999 by James C. McCroskey and Jason J. Teven, expanding upon McCroskey’s foundational 1966 scales. In accordance with the longstanding academic policy established by Dr. McCroskey and the Department of Communication Studies at West Virginia University, this instrument is placed in the public domain for educational, instructional, and non-commercial scholarly research purposes. No licensing fees or formal permissions are required for non-commercial academic use, provided that appropriate scholarly attribution is cited. Organizations utilizing the instrument within proprietary commercial assessments or commercial software platforms should contact the rights holders or relevant institutional copyright representatives.

12. References

• Aristotle. (1991). On rhetoric: A theory of civic discourse (G. A. Kennedy, Trans.). Oxford University Press.
• McCroskey, J. C. (1966). Scales for the measurement of ethos. Speech Monographs, 33(1), 65–72. https://doi.org/10.1080/03637756609375482
• 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
• McCroskey, J. C., & Young, T. J. (1981). Ethos and credibility: The construct and its measurement after three decades. Central States Speech Journal, 32(1), 24–34. https://doi.org/10.1080/10510978109368075
• Teven, J. J., & McCroskey, J. C. (1997). The relationship of perceived teacher caring to student learning and teacher evaluation. Communication Education, 46(1), 1–9. https://doi.org/10.1080/03634529709379069

13. Items of the Scale