The Credibility (Expertise) (CREDEXP)<\/strong> scale is a specialized, four-item psychometric subscale developed by Stephen J. Newell and Ronald E. Goldsmith (2001) as an integral dimension of their foundational Corporate Credibility Scale. Designed to operationalize the degree to which an organization or corporate entity is perceived by stakeholders, consumers, and the public as experienced, technically competent, proficient, and authoritative within its operational domain, the CREDEXP subscale addresses a critical gap in organizational measurement. Rooted in classical source credibility theory<\/a> and consumer psychology, the instrument isolates technical competence from affective or moral judgments (such as corporate trustworthiness). Administered via a standard seven-point Likert-type response format ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”), the four items assess institutional track record, operational capability, subject-matter expertise, and experiential deficits via one reverse-coded item.<\/p>\n Extensive psychometric evaluations conducted across diverse consumer samples indicate that the CREDEXP scale exhibits exceptional psychometric stability. Confirmatory factor analyses (CFA) consistently demonstrate robust single-factor dimensionality for the subscale and clear discriminant divergence from the companion Trustworthiness (CREDTRU) dimension, with structural equation modeling fit indices frequently satisfying stringent thresholds (CFI<\/a> > .95, RMSEA<\/a> < .06). Reliability metrics, including Cronbach’s alpha coefficients typically exceeding α = .88 and composite reliability values well above the .70 criterion, affirm strong internal consistency. Convergent validity is evidenced through substantial factor loadings (> .75) and average variance extracted (AVE<\/a>) surpassing .60. Predictive validity analyses show that perceived corporate expertise significantly influences consumer attitude toward the advertisement<\/span> ($A_{ad}$<\/span>), attitude toward the brand ($A_b$<\/span>), corporate reputation, and behavioral purchase intentions ($PI$<\/span>). Consequently, CREDEXP serves as an authoritative, parsimonious instrument in marketing science, strategic communication, organizational behavior, and public relations research.<\/p>\n Corporate Credibility, Perceived Expertise, Source Credibility Theory, Psychometrics, Newell and Goldsmith, Scale Validation, Brand Equity, Organizational Competence, Structural Equation Modeling, Persuasion Knowledge<\/p>\n The CREDEXP scale was conceptualized, operationalized, and psychometrically validated by:<\/p>\n Scholarly correspondence regarding the original scale development can be directed to the Department of Marketing, Western Michigan University, or via archival academic records indexed in the Journal of Business Research<\/a>.<\/p>\n In modern market environments characterized by information asymmetry, product proliferation, and hyper-competitive organizational landscapes, stakeholders must routinely evaluate corporate messages without having complete technical knowledge of the firm’s manufacturing processes, operational algorithms, or supply chain integrity. Consequently, stakeholders rely on cognitive heuristics and reputational signals to evaluate corporate claims. The primary purpose of the Credibility (Expertise) (CREDEXP) scale is to furnish researchers, corporate strategists, and industrial psychologists with a reliable, empirically validated, and highly parsimonious measurement tool that captures perceived corporate expertise independently from confounding social, emotional, or moral evaluations.<\/p>\n Historically, research examining source credibility originated within interpersonal communication frameworks (e.g., public speaking, political rhetoric, and individual spokesperson endorsement). However, transferring these interpersonal dimensions directly to complex corporate entities created considerable theoretical and methodological ambiguity. Corporate bodies possess collective institutional identities that operate differently from human spokespersons. Newell and Goldsmith recognized that while a firm might be viewed as benevolent or ethical (trustworthy), it may simultaneously be perceived as technologically outmoded, operationally clumsy, or amateurish (lacking expertise). Conversely, a multinational monopoly might be recognized as exceptionally capable, technologically sophisticated, and industrially expert, despite severe public skepticism regarding its motives or honesty. The CREDEXP scale was specifically engineered to measure this cognitive, ability-based dimension of corporate credibility.<\/p>\n In applied and empirical contexts, the CREDEXP instrument serves multiple critical functions:<\/p>\n The construct measured by the CREDEXP scale is Perceived Corporate Expertise<\/em>. Psychometrically defined, perceived corporate expertise represents the extent to which a focal organization is judged by an observer as possessing the specialized knowledge, technical competencies, accumulated experience, operational skills, and intellectual capabilities necessary to consistently fulfill its stated claims, deliver high-quality outputs, and solve complex domain-specific problems. Unlike objective corporate competence—which might be quantified using patent counts, balance sheet liquidity, ISO certifications, or manufacturing defect rates—perceived corporate expertise is an evaluative, perceptual construct situated within the cognitive psychology of the perceiver.<\/p>\n In the tripartite conceptualization of attitudes (cognitive, affective, and conative), perceived expertise is fundamentally cognitive<\/strong>. It reflects propositional beliefs held by an individual concerning a firm’s ability-related attributes. These beliefs crystallize through direct personal interactions (e.g., past product usage), indirect social communications (e.g., word-of-mouth, journalism, independent reviews), and firm-controlled signaling (e.g., corporate advertising, technical whitepapers, and public leadership disclosures). When individuals process messages from a corporate source, the cognitive dimension of expertise serves as an anchor determining the source’s informational authority.<\/p>\n Within Newell and Goldsmith’s (2001) conceptual model, corporate credibility is bifurcated into two distinct yet correlated latent constructs:<\/p>\n To fully appreciate the conceptual depth of CREDEXP, it must be decomposed into its three constitutive indicators:<\/p>\n Crucially, CREDEXP treats corporate expertise as a unidimensional sub-construct within the broader multi-factor framework of corporate reputation. While an entity may demonstrate varying degrees of expertise across disparate operational divisions (e.g., an electronics conglomerate possessing high expertise in displays but low expertise in audio engineering), the CREDEXP scale measures the generalized, top-of-mind gestalt perception of competence that consumers assign to the overarching corporate brand.<\/p>\n The conceptual foundation of the CREDEXP scale rests at the intersection of classical communication theory, cognitive information processing, and economic signaling paradigms. The development of the scale synthesized several foundational theoretical models:<\/p>\n The earliest formal psychometric examinations of credibility emerged from the Yale Communication and Attitude Change Program directed by Carl Hovland and colleagues (Hovland, Janis, & Kelley, 1953<\/a>; Hovland & Weiss, 1951). Investigating social persuasion, Hovland identified two fundamental pillars that determine whether an audience accepts a communicator’s message: expertness<\/em> (the communicator’s perceived qualification to make valid assertions) and trustworthiness<\/em> (the audience’s confidence that the communicator intends to communicate valid assertions without bias or deception). Subsequent interpersonal research by McCroskey and Jenson (1975) expanded these dimensions into multidimensional arrays. Newell and Goldsmith (2001) systematically established that despite the proliferation of complex multidimensional models in speech communication, corporate communication robustly condenses back into Hovland’s parsimonious two-factor paradigm: expertise and trustworthiness.<\/p>\n Under Signaling Theory<\/a> (Spence, 1973<\/a>; Kirmani & Rao, 2000), markets are characterized by asymmetric information: sellers possess comprehensive knowledge regarding the true quality of their offerings, whereas buyers possess incomplete, imperfect information. To alleviate market failure caused by adverse selection (Akerlof, 1970), high-quality organizations deploy credible, cost-intensive signals to demonstrate their true underlying capabilities. A firm’s accumulated expertise represents an unobservable, highly valuable asset. By cultivating perceived corporate expertise, an enterprise projects an unobservable quality signal that reduces perceived purchase risk for consumers. The CREDEXP scale captures the empirical efficacy of these corporate signals in elevating consumer subjective confidence.<\/p>\n According to the Elaboration Likelihood Model (Petty & Cacioppo, 1986<\/a>), persuasion occurs along two distinct processing routes: the central route and the peripheral route. The functional role of source expertise dynamically fluctuates depending on an individual’s cognitive motivation and ability to scrutinize message arguments:<\/p>\n In modern strategic marketing, CREDEXP interfaces with corporate brand equity models (Keller, 1993; Aaker, 1996<\/a>) and the corporate reputation framework (Fombrun, 1996). Corporate expertise forms the cognitive foundation of organizational legitimacy and brand stature. Without demonstrable expertise, brand equity remains vulnerable to competitive obsolescence, whereas high perceived expertise erects sustainable psychological barriers to entry.<\/p>\n The psychometric validation of the CREDEXP scale by Newell and Goldsmith (2001) followed rigorous, multi-stage construct validation protocols aligned with psychometric standards recommended by Churchill (1979) and Gerbing and Anderson (1988).<\/p>\n Content validity was established during initial item generation through an exhaustive review of the extant literature on communicator credibility, advertiser credibility, and organizational reputation. An initial pool of potential items was evaluated by expert panels comprising marketing professors and psychometricians. Judges assessed each item for clarity, domain representativeness, semantic redundancy, and construct purity. Items that conflated corporate expertise with general admiration, financial size, or moral integrity were eliminated, resulting in a refined set of candidate items explicitly mapping onto skill, experience, and domain knowledge.<\/p>\n Factorial validity was established across independent data collections utilizing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). In an initial development sample ($N = 197$<\/span>), principal components and principal axis factoring with oblique rotation yielded a clean two-factor solution corresponding directly to Corporate Expertise and Corporate Trustworthiness. The four CREDEXP items loaded substantially on their hypothesized latent construct (loadings ranging from .76 to .91) with negligible cross-loadings on the Trustworthiness factor (< .20). Subsequent CFA on an independent validation sample<\/span> ($N = 372$<\/span>) confirmed that the two-factor model provided a superior fit over a single-factor “generalized credibility” model, establishing structural construct validity.<\/p>\n Convergent validity evaluates the extent to which the indicators of a latent construct share a high proportion of common variance. In Newell and Goldsmith’s (2001) structural models, all standardized factor loadings for the four CREDEXP items were statistically significant ($p < .001$<\/span>) and exceeded the recognized threshold of .70. The Average Variance Extracted (AVE) for the CREDEXP subscale consistently surpassed the .50 benchmark (typically exceeding .65), confirming that the latent construct explains far more variance in its observed indicators than measurement error does.<\/p>\n A central psychometric objective of Newell and Goldsmith was ensuring that Perceived Expertise was empirically distinct from Perceived Trustworthiness. Discriminant validity was empirically substantiated through several diagnostic tests:<\/p>\n The utility of any psychometric tool hinges on its ability to forecast theoretically linked outcomes. Newell and Goldsmith (2001) tested the predictive efficacy of CREDEXP within a structural equation model evaluating consumer responses to advertisements. CREDEXP accounted for significant, direct proportions of variance in:<\/p>\n The internal consistency and measurement stability of the CREDEXP scale have been documented across numerous studies. Psychometric theory dictates that reliable instruments minimize random measurement error, thereby yielding consistent scores across repeated administrations and within-scale items.<\/p>\n In the foundational scale development studies conducted by Newell and Goldsmith (2001), the CREDEXP subscale demonstrated internal consistency metrics that comfortably exceed established psychometric benchmarks:<\/p>\n Subsequent independent studies in consumer behavior, service marketing, and corporate communications have repeatedly confirmed the scale’s high internal consistency. Replications across diverse industries—ranging from airline carriers and telecommunications providers to financial institutions and pharmaceutical manufacturers—consistently report Cronbach’s alphas between .85 and .93. Cross-cultural adaptations of the scale (translated into Spanish, German, Chinese, and Korean) have maintained internal consistency reliability estimates well above the standard .70 or .80 academic research thresholds.<\/p>\n The four-item structure achieves an optimal balance between reliability and brevity. With an Average Variance Extracted (AVE) routinely above .65, over 65% of the variance captured by the measure reflects true score construct variance rather than idiosyncratic item error. The inclusion of one negatively worded (reverse-scored) item (“[Company] does not have much experience”) prevents systematic acquiescence response bias without introducing substantial method-factor artifact, provided participants are attentive.<\/p>\n The structural composition of the CREDEXP scale was isolated and confirmed using exploratory and confirmatory factor analytic procedures.<\/p>\n During the initial derivation phase, candidate items reflecting corporate credibility were subjected to common factor analysis (principal axis factoring) with both orthogonal (Varimax) and oblique (Promax) rotations to allow for expected inter-factor correlations between expertise and trustworthiness. Kaiser-Meyer-Olkin (KMO) measures of sampling adequacy were consistently observed above .85, and Bartlett’s Test of Sphericity yielded significant results ($\\chi^2, p < .001$<\/span>), confirming data factorability.<\/p>\n The eigenvalues-greater-than-one (Kaiser) criterion, paired with Cattell’s scree plot analysis, distinctly identified a two-factor latent structure. The first factor accounted for the majority of the common variance, representing Corporate Trustworthiness, while the second factor represented Corporate Expertise (or vice versa, depending on the sample stimulus). The four items designated for the CREDEXP subscale loaded heavily on their intended latent dimension:<\/p>\n2. Keywords<\/h2>\n
3. Authors<\/h2>\n
\n
4. Purpose<\/h2>\n
\n
5. Psychological Construct<\/h2>\n
\n
\n
6. Theoretical Framework<\/h2>\n
The Yale Communication Program and Source Credibility Theory<\/h3>\n
Signaling Theory under Information Asymmetry<\/h3>\n
Dual-Process Models of Persuasion: The Elaboration Likelihood Model (ELM)<\/h3>\n
\n
Corporate Reputation and Brand Equity Frameworks<\/h3>\n
7. Validity<\/h2>\n
Content and Face Validity<\/h3>\n
Construct and Factorial Validity<\/h3>\n
Convergent Validity<\/h3>\n
Discriminant Validity<\/h3>\n
\n
Criterion-Related and Predictive Validity<\/h3>\n
\n
8. Reliability<\/h2>\n
Internal Consistency Metrics<\/h3>\n
\n
Replication and Cross-Cultural Reliability<\/h3>\n
Scale Parsimony and Error Variance<\/h3>\n
9. Factor Analysis<\/h2>\n
Exploratory Factor Analysis (EFA)<\/h3>\n