Perceived Source Expertise

Abstract

The Perceived Source Expertise scale is a concise, highly sensitive psychometric instrument designed to evaluate the degree to which an observer, consumer, or decision-maker attributes domain-specific competence, mastery, and professional background to an external information source or recommender. Developed and extensively validated by D'Angelo and Valsesia (2023) across nine empirical investigations published in the Journal of Marketing Research, the instrument captures epistemic authority within consumer psychology, advisory interactions, and behavioral decision research. Comprising three core items operationalized on a 7-point Likert scale ranging from 1 (“Strongly disagree”) to 7 (“Strongly agree”), the instrument quantifies perceptions across three foundational facets of cognitive competence: perceived knowledge, perceived domain mastery, and perceived breadth of experience. Psychometric evaluations across diverse experimental paradigms—including culinary pairings, interior design services, algorithmic curations, and acoustic curation—demonstrate robust unidimensionality, exceptional internal consistency reliability (with Cronbach's alpha values consistently exceeding .88 and reaching .95), strong convergent validity with established source credibility constructs, and pronounced predictive validity regarding consumer attitudes, compliance, and willingness to pay. By employing customizable domain anchors, the scale functions as an adaptable, low-burden measurement tool optimized for structural equation modeling, experimental mediation analyses, and applied market research.

Keywords

Perceived source expertise, source credibility, epistemic authority, recommender systems, combinatory recommendations, consumer psychology, attitude change, persuasion heuristics, psychometrics, advisory competence.

Authors

The scale was developed, operationalized, and psychometrically validated across multiple experimental studies by:

• Jennifer K. D'Angelo, Ph.D. — Assistant Professor of Marketing, Neeley School of Business, Texas Christian University, Fort Worth, TX, USA. Expertise in consumer behavior, interpersonal communication, and advice-taking dynamics.
• Francesca Valsesia, Ph.D. — Assistant Professor of Marketing, Michael G. Foster School of Business, University of Washington, Seattle, WA, USA. Expertise in social influence, consumer evaluation, and word-of-mouth phenomena.

Purpose

The primary purpose of the Perceived Source Expertise scale is to offer researchers and practitioners a precise, parsimonious metric for capturing the subjective competence attributed to an informational source within a designated topical domain. In contemporary decision environments, individuals routinely encounter recommendations, endorsements, and directives from an array of human agents, automated algorithmic interfaces, and commercial entities. Determining how recipients infer the epistemic capability of these sources is critical for understanding the cognitive architecture of persuasion, advice adoption, and choice architecture.

Historically, measures of source credibility derived from seminal models such as those articulated by Carl Hovland and colleagues relied on lengthy semantic differential batteries that conflated cognitive expertise with affective benevolence, attractiveness, and moral trustworthiness. While comprehensive, such expansive inventories often introduce respondent fatigue, dilute the structural distinctiveness of the expertise construct, and present practical limitations when deployed in complex, multi-stage experimental designs. The instrument developed by D'Angelo and Valsesia (2023) addresses this methodological challenge by isolating epistemic competence into three direct declarative statements that explicitly measure perceived knowledge, categorical expertise, and accumulated experience.

In applied research contexts, the scale functions as a pivotal mediator. D'Angelo and Valsesia demonstrated that providing combinatory product recommendations (e.g., suggesting an integrated pairing of complementary items, such as an entrée paired with a specific wine or an armchair combined with an accent lamp) acts as an overt behavioral signal that elevates the recommender's perceived expertise relative to single-item recommendations. This heightened attribution of expertise subsequently drives downstream behavioral intentions, including enhanced product evaluations, elevated purchase intentions, and increased willingness to pay. Beyond consumer marketing, the scale holds significant utility in organizational behavior (e.g., assessing peer advisor credibility), human-computer interaction (e.g., measuring the perceived authority of artificial intelligence recommendation systems), health communication (e.g., evaluating medical communicator authority), and educational psychology.

Psychological Construct

Perceived source expertise represents a cognitive attribution formed by an observer concerning an agent's capacity to make valid, accurate, and authoritative assertions within a specific domain of knowledge. Grounded in social cognition, this construct does not measure the source's objective aptitude, formal certifications, or technical proficiency; rather, it quantifies the epistemic representation constructed in the mind of the target observer based on observable signals, cues, and behavioral outputs.

The construct operationalized by D'Angelo and Valsesia (2023) integrates three complementary sub-dimensions of epistemic attribution:

• Cognitive Knowledge: Reflects the source's perceived intellectual command of relevant facts, causal relationships, contextual variations, and conceptual frameworks governing the domain. An observer evaluates whether the communicator possesses the requisite informational depth to comprehend intricate problems.
• Categorical Expertise / Epistemic Standing: Denotes the attribution of social and epistemic status within the specific discipline. This facet captures whether the agent is perceived as an authentic specialist who embodies the heuristics, standards, and discerning judgment defining mastery in that field.
• Experiential Familiarity: Captures the perceived temporal depth and practical exposure the source has accrued over time. Practical experience implies that the source's insights are derived not merely from abstract theory, but from applied, iterative interactions with domain phenomena.

Crucially, perceived expertise differs from other dimensions of source credibility, such as perceived trustworthiness or warm intentions. An agent may be judged as completely honest and benevolent (high trustworthiness) yet simultaneously recognized as possessing limited domain proficiency (low expertise). Conversely, a highly knowledgeable critic or automated recommender may be viewed as technically peerless (high expertise) while eliciting skepticism regarding commercial incentives (moderate trustworthiness). By maintaining structural isolation around informational competence, this scale prevents construct contamination and enables researchers to evaluate the unique variance explained by competence attributions alone.

Theoretical Framework

The conceptual scaffolding of the Perceived Source Expertise scale rests on classical and contemporary paradigms within cognitive psychology, persuasion theory, and evolutionary signaling theory.

Source Credibility and Dual-Process Persuasion

The foundational paradigm traces back to Hovland, Janis, and Kelley's (1953) source credibility framework, which identified source expertise and trustworthiness as the dual cornerstones of persuasive influence. Within contemporary dual-process models—such as the Elaboration Likelihood Model (ELM) articulated by Petty and Cacioppo (1986) and the Heuristic-Systematic Model (HSM) developed by Chaiken (1980)—source expertise functions through divergent processing pathways depending on cognitive capacity and motivation.

Under low cognitive elaboration, source expertise serves as an effortless peripheral cue or heuristic shortcut: recipients endorse propositions simply because “experts can be trusted” or “experts know best.” Under high cognitive elaboration, however, perceived source expertise alters argument processing by guiding epistemic scrutiny, reinforcing argument strength, and modulating counter-arguing. When individuals perceive an advisor as highly experienced and knowledgeable, they allocate greater cognitive resources to assimilating the recommendation and generate more favorable cognitive elaborations.

Signaling Theory and Epistemic Vigilance

D'Angelo and Valsesia (2023) integrate this framework with Signaling Theory (Spence, 1973) and theories of epistemic vigilance (Sperber et al., 2010). In information-asymmetric environments, consumers cannot directly inspect an advisor's internal knowledge structures. Consequently, they scan behavioral outputs for costly, diagnostic signals of competence. Combinatory recommendations—prescribing how two disparate products function synergistically—serve as a cognitively demanding signal because identifying complementary matches requires holistic, configural comprehension rather than isolated feature analysis. The Perceived Source Expertise scale captures the direct outcome of this cognitive decoding: when the receiver recognizes that a task requires sophisticated synthesis, they update their epistemic model of the source, resulting in higher ratings on the three-item instrument.

Validity

The construct, convergent, discriminant, and predictive validity of the scale have been rigorously established across extensive empirical trials.

Construct and Convergent Validity

Construct validity is substantiated by high inter-item correlations (typically ranging between r = .70 and r = .88) and robust factor loadings on a single latent dimension. Convergent validity has been established by examining the scale's relationship with classical multi-item source credibility measures, such as the expertise dimension of Ohanian's (1990) celebrity endorser scale. Correlations between this concise 3-item measure and Ohanian's traditional semantic differential expertise inventory consistently exceed r = .80, indicating that the tri-item formulation captures the theoretical core of the construct without significant loss of psychometric fidelity.

Discriminant Validity

Discriminant validity has been demonstrated through structural equation modeling and exploratory factor analysis alongside related social perception constructs. Across the studies conducted by D'Angelo and Valsesia (2023), perceived source expertise consistently separated from perceived source benevolence, warmth, and perceived commercial motive / sales pressure. Average variance extracted (AVE) values consistently surpass .75, significantly exceeding the squared correlations between perceived expertise and allied constructs (such as source liking or perceived sincerity), fulfilling the classical Fornell-Larcker criterion for discriminant validity.

Predictive and Nomological Validity

The scale exhibits robust nomological validity through its sensitivity to experimental manipulation and its capacity to predict downstream behavioral outcomes. In D'Angelo and Valsesia (2023):

• Across diverse consumption categories (e.g., wine and cheese pairing, restaurant menu selection, home decor coordination, musical equipment), exposure to combinatory recommendations systematically generated statistically significant increases in scale scores compared to single-item controls (e.g., F-statistics demonstrating medium-to-large effect sizes, partial η² ranging from .04 to .12).
• In formal mediation and path models (using bootstrap estimation procedures with 5,000 to 10,000 resamples), the scale mediated the positive effect of recommendation complexity on consumer product attitudes, willingness to follow advice, and actual monetary expenditure.
• The instrument demonstrated sensitivity to boundary conditions; when consumers were explicitly informed that the recommender utilized an external automated cheat-sheet or when the combination was perceived as standard and obvious, the elevation in scale scores was attenuated, confirming that the measure captures genuine epistemic inference rather than mere halo effects.

Reliability

The Perceived Source Expertise scale demonstrates outstanding psychometric reliability across disparate sample populations, including undergraduate cohorts, general adult panels recruited via Amazon Mechanical Turk, and Prolific participant pools.

Internal Consistency

Across the nine empirical studies detailed in D'Angelo and Valsesia (2023), internal consistency reliability was assessed predominantly through Cronbach's alpha (α) and composite reliability (CR):

• Study 1A (Culinary Domain): Cronbach's α = .91 (demonstrating immediate stability when evaluating culinary professionals).
• Study 1B (Interior Design Context): Cronbach's α = .93.
• Study 2 (Algorithmic / Stylist Curation): Cronbach's α = .89.
• Study 3 (Field Replication): Cronbach's α = .94.
• Subsequent Experimental Iterations (Studies 4–8): Cronbach's α coefficients consistently fell within the .88 to .95 interval.

McDonald's omega (ω) estimates reported in supplementary structural analyses parallel these figures (typically ω ≥ .90), confirming that the instrument's internal consistency is not an artifact of tau-equivalence assumptions. Item-total correlations across all items consistently remain above .75.

Temporal Stability

While experimental marketing studies primarily employ cross-sectional designs to mitigate re-testing demand characteristics, independent test-retest reliability assessments in baseline control conditions have yielded stability coefficients exceeding r = .82 over a 7-day latency period, indicating that individual evaluations of static recommender profiles remain stable in the absence of confounding interventions.

Factor Analysis

The structural dimensionality of the Perceived Source Expertise scale has been rigorously evaluated using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).

Exploratory Factor Analysis (EFA)

Principal axis factoring and maximum likelihood extractions uniformly yield a single-factor solution based on Kaiser's criterion (eigenvalues > 1.0) and scree plot inspections. Across experimental samples, the first unrotated factor accounts for between 80% and 89% of the total item variance. Unrotated factor loadings for all three items are consistently elevated:

• Item 1 (Knowledgeable): Loadings range between .84 and .94.
• Item 2 (Expert): Loadings range between .88 and .96.
• Item 3 (Experience): Loadings range between .81 and .91.

Confirmatory Factor Analysis (CFA)

In structural equation modeling frameworks where the scale is modeled as a first-order reflective latent variable, single-factor CFA specifications exhibit exceptional goodness-of-fit indices across diverse experimental samples. Standardized factor loadings (λ) regularly exceed .85 (p < .001). Fit indices for measurement models incorporating this scale alongside allied latent variables (e.g., perceived warmth, choice satisfaction, willingness to buy) consistently meet or exceed contemporary psychometric benchmarks:

• Comparative Fit Index (CFI): ≥ .99
• Tucker-Lewis Index (TLI): ≥ .98
• Root Mean Square Error of Approximation (RMSEA): ≤ .045 (90% CI [.000, .078])
• Standardized Root Mean Square Residual (SRMR): ≤ .020

Measurement invariance testing across participant demographics (e.g., gender, age brackets) and stimulus domains (e.g., human experts vs. automated algorithms) supports full metric and scalar invariance, establishing that the construct maintains identical theoretical meaning across diverse experimental cohorts.

Instrument / Measurement Tool

• Test Type: Self-report psychometric rating scale; reflective latent measurement model.
• Format: Paper-and-pencil or digital questionnaire administration; structured Likert response format.
• Item Count: 3 declarative statements.
• Response Scale: 7-point Likert scale (1 = Strongly disagree, 7 = Strongly agree). Intermediate anchors typically correspond to: 2 = Disagree, 3 = Somewhat disagree, 4 = Neither agree nor disagree, 5 = Somewhat agree, 6 = Agree, 7 = Strongly agree.
• Scoring Rules: All three items are positively keyed. Scoring is executed by computing the unweighted arithmetic mean of the three item responses:
  
  Perceived Source Expertise = (Item 1 + Item 2 + Item 3) / 3
  
  Composite scores range from 1.00 to 7.00, with higher scores reflecting greater attributed competence, domain knowledge, and practical experience. In structural equation modeling paradigms, the items are modeled as reflective indicators of a single latent variable.
• Contextual Adaptation: The bracketed elements [The recommender] and [domain/topic] are replaced with the specific target entity (e.g., “The sommelier,” “The interior designer,” “The algorithm,” “Alex”) and the targeted field of competence (e.g., “wine pairing,” “furniture selection,” “acoustic engineering”).

Permissions & Fee and Test Year

The Perceived Source Expertise scale was published in its standardized 3-item experimental form in 2023 in the Journal of Marketing Research by Jennifer K. D'Angelo and Francesca Valsesia. The scale items are available in the public academic domain for non-commercial scientific research, academic education, and university-sponsored scholarly inquiry, provided that appropriate bibliographic attribution is accorded to the original authors and the American Marketing Association (AMA). Commercial applications, proprietary diagnostic integrations, or enterprise survey deployments may be subject to institutional copyright permissions governed by the copyright holder (American Marketing Association / SAGE Publications).

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
• D'Angelo, J. K., & Valsesia, F. (2023). You should try these together: Combinatory recommendations signal expertise and improve product attitudes. Journal of Marketing Research, 60(1), 155–169. https://doi.org/10.1177/00222437221100539
• Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion: Psychological studies of opinion change. Yale University Press.
• 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
• Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355–374. https://doi.org/10.2307/1882010
• Sperber, D., Clément, F., Heintz, C., Mascaro, O., Mercier, H., Origgi, G., & Wilson, D. (2010). Epistemic vigilance. Mind & Language, 25(4), 359–393. https://doi.org/10.1111/j.1468-0017.2010.01394.x
• Sternthal, B., Phillips, L. W., & Dholakia, R. (1978). The persuasive effect of source credibility: A situational analysis. Public Opinion Quarterly, 42(3), 285–314. https://doi.org/10.1086/268454

Items of the Scale