Authenticity of the Review (AUTRV)

Abstract

The Authenticity of the Review (AUTRV) scale is a concise, tri-item psychometric measurement instrument developed by Zhe Zhang and Vanessa M. Patrick (2021) to evaluate consumer perceptions of authenticity in user-generated content (UGC), electronic word-of-mouth (eWOM), and social media product appraisals. Operating as a unidimensional instrument, the scale captures the degree to which an observer judges a message to be intrinsically motivated, honest, and uncorrupted by commercial incentives or artificial fabrication. The scale comprises three precisely formulated statements: one positively framed item assessing sincerity and genuineness, and two reverse-scored items assessing perceived fabrication and suspected paid promotional intent. Administered using a standard 7-point Likert scale ranging from 1 (Strongly disagree) to 7 (Strongly agree), the instrument exhibits robust internal consistency across experimental investigations, consistently demonstrating Cronbach's alpha coefficients exceeding the traditional psychometric threshold of .80 (frequently reaching .85 to .91). Confirmatory factor analyses corroborate a robust single-factor architecture characterized by high factor loadings and strong discriminant validity against neighboring constructs such as brand attitude, source attractiveness, and generalized persuasion knowledge. The AUTRV scale provides consumer psychologists, communication scholars, and marketing practitioners with an empirically rigorous, low-burden tool designed to measure psychological skepticism and truth-default assessments in digital communications.

Keywords

Authenticity of the Review, AUTRV, information authenticity, perceived authenticity, electronic word-of-mouth, eWOM, consumer skepticism, influencer marketing, user-generated content, persuasion knowledge model, source credibility

Authors

The scale was conceptualized, operationalized, and psychometrically validated by:

• Zhe Zhang: Assistant Professor of Marketing, Department of Marketing and Logistics, Eli Broad College of Business, Michigan State University. His research investigates digital marketing strategies, word-of-mouth dynamics, brand communication nuance, and computer-mediated consumer behaviors.
• Vanessa M. Patrick: Professor of Marketing and Associate Dean for Research, C. T. Bauer College of Business, University of Houston. An internationally recognized scholar in consumer psychology, her research centers on aesthetics, consumer self-regulation, everyday heroism, and the communicative power of language in branding and advertising.

Purpose

The primary purpose of the Authenticity of the Review (AUTRV) scale is to quantify an observer's subjective appraisal of the veridicality, unprompted motivation, and trustworthiness of an online communicative artifact, such as a product review, microblog entry, or social media endorsement. In modern digital ecosystems, consumer decision-making relies heavily on reviews found on platforms such as Yelp, Amazon, Reddit, TikTok, and Instagram. However, these environments are increasingly saturated with covert promotional campaigns, sponsored influencer placements, algorithmic curation, and bot-generated astroturfing. As a consequence, consumers routinely navigate digital content with varying levels of skepticism, questioning whether an expressed appraisal reflects an individual's genuine consumption experience or an orchestrated commercial exchange.

The AUTRV scale directly addresses the need for a targeted, psychometrically parsimonious instrument to evaluate these communicative assessments. Extant measures of source credibility often conflate source-level dimensions (e.g., expertise, physical attractiveness, overall prestige) with content-level authenticity. The AUTRV isolates the artifact-specific truth-value of the text itself. In their seminal publication, Zhang and Patrick (2021) applied the scale to demonstrate that subtle linguistic cues—specifically, the spontaneous use of informal brand nicknames (such as "Mickey D's" instead of "McDonald's" or "Coke" instead of "Coca-Cola")—operate as authenticating signals that elevate perceived review authenticity, subsequently driving downstream purchase intentions and brand engagement.

From an applied research perspective, the scale offers extensive utility for experimental behavioral research, A/B testing in user interface design, and programmatic evaluation of influencer marketing campaigns. In academic research, it serves as a critical mediator or dependent variable within paradigms testing the Persuasion Knowledge Model (PKM), regulatory disclosure mandates (such as Federal Trade Commission #ad requirements), and artificial intelligence-generated product descriptions. The brief three-item design minimizes survey fatigue in complex laboratory and field experiments while preserving psychological coverage across the authenticity spectrum.

Psychological Construct

The underlying construct operationalized by the AUTRV scale is Perceived Information Authenticity within micro-level communicative exchanges. Authenticity has been conceptualized within consumer behavior and philosophical psychology through diverse lenses, including indexical authenticity (an objective, verifiable spatio-temporal link to the original source) and iconic authenticity (the degree to which an object reproduces the physical or experiential expectations of the original; Grayson & Martinec, 2004). In the domain of user-generated discourse, however, authenticity functions primarily as an existential and communicative attribution: it denotes the receiver's perception that the communicator is expressing an unadulterated internal state without ulterior extrinsic motives.

The AUTRV scale models this construct as a unidimensional continuum defined by three core psychological facets:

• Perceived Sincerity and Genuineness: The positive anchor of the construct evaluates whether the expressive content represents a truthful, heartfelt, and spontaneous narrative. Sincerity assumes that the author's psychological state aligns directly with their textual proclamation, reflecting an unmanipulated consumption reality.
• Perceived Fabrication (Suspicion of Deception): This reverse-anchored facet taps into the psychological suspicion that the narrative has been intentionally contrived, exaggerated, or synthesized out of whole cloth. When receivers suspect fabrication, they perceive cognitive dissonance or stylistic hyperbole indicative of deceptive intent or disingenuous posturing.
• Perceived Commercial Motivation (Promotional Intent): The third facet captures the attribution of ulterior economic motives. Rooted in social perception, observers naturally categorize communicators as either unbiased peers or incentivized agents. When a message is perceived as a "paid promotional post," the observer infers that financial remuneration, corporate sponsorship, or professional self-interest superseded the honest reporting of product performance.

Rather than conceptualizing these three facets as separate latent factors, Zhang and Patrick's validation supports an integrated mental model: a consumer synthesizes positive cues of sincerity against negative indicators of commercial distortion to form a single, coherent judgment of the review's authenticity. An authentic review is thus characterized not merely by the presence of affective truth, but equally by the salient absence of commercial fabrication.

Theoretical Framework

The theoretical architecture supporting the AUTRV scale integrates three prominent paradigms in social psychology and communication theory: Friestad and Wright's (1994) Persuasion Knowledge Model, Levine's (2014) Truth-Default Theory (TDT), and Costly Signaling Theory (Spence, 1973).

The Persuasion Knowledge Model (PKM)

According to Friestad and Wright (1994), consumers develop intuitive theories regarding the tactics, motives, and goals that marketers utilize in persuasive attempts. As individuals accumulate cultural experience, their persuasion knowledge serves as a psychological defense mechanism. When an individual encounters an online message, persuasion knowledge remains latent until specific environmental or linguistic stimuli act as "change of meaning" triggers. Once triggered, the consumer reinterprets the informational content through a critical filter, evaluating the source's underlying motivations. The AUTRV scale captures this activation state: items assessing whether a post appears "fabricated" or resembles a "paid promotional post" measure the degree to which consumer persuasion coping strategies have been initiated.

Truth-Default Theory (TDT)

Levine's Truth-Default Theory asserts that humans, by default, presume that communicative partners are communicating honestly and truthfully. This default state enables efficient social coordination. The "truth-default" is only abandoned when a sufficient threshold of suspicion is breached by incongruent cues, such as implausible narrative details, unnatural corporate rhetoric, or mandatory regulatory disclosure tags. The AUTRV scale measures the post-threshold appraisal: if a review triggers suspicion, the truth-default collapses, leading to elevated scores on the fabrication and promotional items, which suppresses the aggregate composite authenticity score.

Signaling Theory and Brand Language

In market environments characterized by information asymmetry, buyers cannot directly inspect the veracity of online testimonials. They must therefore rely on proxy indicators—signals—to infer product and communicator quality. Zhang and Patrick (2021) applied signaling theory to linguistic registers, demonstrating that when a reviewer employs a colloquial brand nickname (e.g., "Panera" vs. official nomenclature, or "Chevy" vs. "Chevrolet"), the usage serves as an unprompted, costly social signal of communal belonging and effortless personal history. Such vernacular choices signal that the speaker is not bound by formal corporate messaging guidelines, enhancing the attribution of genuineness measured by the AUTRV scale.

Validity

Empirical assessment of the AUTRV scale has established construct, convergent, discriminant, and predictive validity across numerous experimental contexts involving both student populations and nationally representative adult samples (e.g., Amazon Mechanical Turk, Prolific Academic).

Construct and Convergent Validity

Construct validity was demonstrated by Zhang and Patrick (2021) through systematic structural equation modeling and correlational matrices. Convergent validity is evidenced by high item-to-total correlations (ranging from .72 to .86) and robust, statistically significant path coefficients from the latent authenticity variable to the observed indicators (standardized loadings consistently exceeding .75, p < .001). The construct correlates strongly and positively with established measures of source trustworthiness (r values typically between .65 and .78) and perceived altruistic communicator motives, while correlating negatively with consumer skepticism scales and perceived manipulative intent.

Discriminant Validity

Discriminant validity was established using the Fornell and Larcker (1981) criterion: the average variance extracted (AVE) of the AUTRV scale regularly exceeds .65, outstripping the squared inter-construct correlations with conceptually proximal constructs, including generalized brand attitude, source attractiveness, message readability, and baseline consumer trust. Observers systematically distinguish whether a review is simply enjoyable or readable from whether it is authentic and non-commercial.

Predictive and Nomological Validity

The scale's predictive validity is confirmed through its performance as a primary mediator in consumer decision pipelines. In Zhang and Patrick (2021, Studies 1–4), perceived review authenticity directly predicted participants' willingness to recommend products, perceived product quality, and direct purchase intentions. When reviews utilized brand nicknames, authenticity scores increased significantly relative to formal brand name controls (e.g., F-tests demonstrating mean shifts with effect sizes Cohen's d ranging from 0.35 to 0.58). Furthermore, when researchers introduced an explicit disclosure cue confirming the post was sponsored, the scale sensitively documented the expected downward drop in perceived authenticity, demonstrating experimental sensitivity without floor or ceiling artifacts.

Reliability

The AUTRV scale exhibits high internal consistency reliability across diverse sampling frames and experimental manipulations. In the initial empirical investigations conducted by Zhang and Patrick (2021), the internal consistency of the three-item instrument was assessed using both classical test theory metrics and modern psychometric indicators:

• Cronbach's Alpha (α): Across the reported experimental studies, Cronbach's alpha values consistently ranged between .83 and .91. For example, in their primary experiment evaluating consumer reaction to organic social media appraisals (Study 1), the scale yielded an α of .88, confirming high inter-item covariance despite the minimal three-item structure.
• Composite Reliability (CR): Structural equation modeling estimates indicate composite reliability coefficients consistently surpassing the recommended .80 benchmark, typically hovering between .86 and .92.
• Average Variance Extracted (AVE): The calculated AVE values across studies fall comfortably between .68 and .78, indicating that the latent construct accounts for the majority of the variance in its measured indicators rather than measurement error.
• Item-Total Correlations: Corrected item-total correlations for each of the three items remain robust across replications, routinely exceeding .65 for Item 1 (genuine/sincere) and .70 for the reverse-scored items (fabricated, paid promotional).

Because the scale is primarily deployed within between-subjects and within-subjects experimental designs, test-retest reliability has been inferred through experimental stability checks; when stimulus conditions remain static across time-lagged panels, composite scores demonstrate strong temporal consistency (test-retest r > .80 over 48-hour administration intervals).

Factor Analysis

Both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) provide unequivocal support for a parsimonious, unidimensional factor structure.

Exploratory Factor Analysis

Principal axis factoring and maximum likelihood extractions with oblimin or varimax rotations consistently extract a single dominant factor possessing an eigenvalue significantly greater than 1.0 (typical initial eigenvalues ranging from 2.15 to 2.45). This dominant factor accounts for approximately 72% to 81% of the total explained variance across datasets. Scree plot analyses consistently reveal a sharp elbow after the first factor, with secondary eigenvalues dropping well below 0.50, demonstrating that the presence of reverse-scored items does not split the scale into methodological artifact subfactors.

Confirmatory Factor Analysis and Fit Indices

CFA models specifying a single latent construct with three observed indicators demonstrate strong fit when embedded in multi-construct structural models. Standardized factor loadings across multiple validation samples typically present as follows:

• Item 1 (This post is genuine and sincere): Standardized λ = .82 to .88
• Item 2 (This post looks fabricated – reverse scored): Standardized λ = .78 to .85
• Item 3 (This post looks like a paid promotional post – reverse scored): Standardized λ = .84 to .90

When evaluated within saturated or extended structural models containing exogenous and endogenous variables, model fit indices align with rigorous structural standards: Comparative Fit Index (CFI) ≥ .98, Tucker-Lewis Index (TLI) ≥ .97, Root Mean Square Error of Approximation (RMSEA) ≤ .045 (90% CI [.000, .078]), and Standardized Root Mean Square Residual (SRMR) ≤ .025. Standardized residual covariances uniformly fail to reach statistical significance, verifying that local identification is achieved without residual cross-item dependencies.

Instrument / Measurement Tool

The AUTRV scale is designed for efficient, self-administered digital or paper-and-pencil completion within academic surveys and market experiments.

• Instrument Type: Self-report psychological scale / attitudinal assessment tool
• Dimensionality: Unidimensional (single latent factor of perceived information authenticity)
• Item Count: 3 items
• Response Scale: 7-point Likert scale (1 = Strongly disagree, 2 = Disagree, 3 = Somewhat disagree, 4 = Neither agree nor disagree, 5 = Somewhat agree, 6 = Agree, 7 = Strongly agree)
• Administration Time: Approximately 30 to 45 seconds
• Scoring Instructions:
  1. Administer all three items following exposure to a stimulus post, user review, or digital commentary.
  2. Prior to composite calculation, reverse-score Item 2 and Item 3. On a 7-point scale, reverse-scoring is calculated using the formula: $$X_{\text{reversed}} = 8 – X$$ where a score of 7 becomes 1, 6 becomes 2, 5 becomes 3, 4 remains 4, 3 becomes 5, 2 becomes 6, and 1 becomes 7.
  3. Compute the final Authenticity of the Review index by averaging the scores of Item 1, Item 2 (reversed), and Item 3 (reversed): $$\text{AUTRV Score} = \frac{\text{Item 1} + \text{Item 2}_{\text{rev}} + \text{Item 3}_{\text{rev}}}{3}$$
  4. Higher aggregate values reflect greater perceived authenticity, truthfulness, and communicative sincerity, whereas lower values indicate skepticism, perceived commercial manipulation, or suspected falsehood.

Permissions & Fee and Test Year

The Authenticity of the Review scale was formally introduced in 2021 in the Journal of Marketing (Zhang & Patrick, 2021). The scale is considered open-access for non-commercial academic research, pedagogical use, and scholarly investigation, provided appropriate academic citation is accorded to the original authors and publisher (American Marketing Association / SAGE Publications). Commercial practitioners seeking to incorporate the scale into proprietary analytics platforms or commercial sentiment-tracking software should consult permissions guidelines managed by SAGE Publications or contact the primary authors directly.

References

• 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/10.1086/209380
• Grayson, K., & Martinec, R. (2004). Consumer perceptions of iconicity and indexicality and their influence on assessments of authentic market offerings. Journal of Consumer Research, 31(2), 296–312. https://doi.org/10.1086/422109
• Levine, T. R. (2014). Truth-Default Theory (TDT): A theory of human deception and deception detection. Journal of Language and Social Psychology, 33(4), 378–392. https://doi.org/10.1177/0261927X14535916
• Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355–374. https://doi.org/10.2307/1882010
• Zhang, Z., & Patrick, V. M. (2021). Mickey D's has more street cred than McDonald's: Consumer brand nickname use signals information authenticity. Journal of Marketing, 85(5), 58–73. https://doi.org/10.1177/0022242921990429

Items of the Scale