Endorser Similarity (ES)

1. Abstract

The Endorser Similarity (ES) scale is a psychometric instrument developed by Lawrence Feick and Robin A. Higie (1992) to evaluate the perceived similarity between a consumer and an advertising endorser or source. Formulated within the broader context of source effects, social influence, and consumer persuasion research, the scale quantifies the degree to which an individual perceives an endorser as sharing salient personal, demographic, experiential, or psychographic characteristics with oneself. Perceived source similarity represents a foundational dimension of source credibility and attractiveness models, operating as a pivotal cognitive heuristic and affective cue in advertising evaluation, attitude formation, and behavioral intentions.

The ES scale consists of a parsimonious three-item unidimensional measure evaluated along a 7-point Likert response format ranging from 1 (Strongly Disagree) to 7 (Strongly Agree). It incorporates both positively phrased items and a reverse-scored item to attenuate acquiescence response bias. Psychometric evaluations across multiple experimental and field validation studies demonstrate robust internal consistency reliability, with Cronbach’s alpha coefficients consistently exceeding standard thresholds (α ≥ .85 to .92). Exploratory and confirmatory factor analyses confirm a strictly unidimensional latent factor structure accounting for high percentages of variance, with strong convergent, discriminant, and predictive validity in relation to constructs such as source expertise, source trustworthiness, brand attitude, purchase intention, and identification.

By capturing consumer evaluations of endorser likeness, the scale serves as an indispensable tool across advertising psychology, marketing communications, influencer marketing, health communication campaigns, and political persuasion research. This article provides a comprehensive academic review of the Endorser Similarity scale, detailing its theoretical foundations in social comparison theory, source credibility models, and reference group influence, alongside exhaustive analyses of its psychometric properties, factor structure, administration parameters, and experimental applications.

2. Keywords

Endorser similarity, source characteristics, source credibility, advertising effectiveness, social comparison theory, consumer persuasion, psychometric validation, perceived similarity, reference group theory, attitude change, influencer marketing, source attractiveness

3. Authors

The Endorser Similarity (ES) scale was formulated and validated by Lawrence F. Feick and Robin A. Higie (subsequently Robin Higie Coulter) in their landmark investigation titled “The Effects of Preference Heterogeneity and Source Characteristics on Ad Processing and Judgments About Endorsers,” published in the Journal of Advertising (1992).

• Lawrence F. Feick, Ph.D.
  Affiliation: Professor of Business Administration and Marketing, Joseph M. Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, Pennsylvania, United States.
  Research Focus: Consumer information search, word-of-mouth communications, market mavens, source credibility, pricing perception, and international marketing strategy.
• Robin A. Higie (Coulter), Ph.D.
  Affiliation: Professor of Marketing, School of Business, University of Connecticut, Storrs, Connecticut, United States.
  Research Focus: Advertising processing, consumer psychology, brand relationships, cross-cultural consumer behavior, and visual communication.

4. Purpose

The primary purpose of the Endorser Similarity (ES) scale is to capture and quantify the psychological construct of perceived similarity that a target viewer or consumer attributes to an endorser featured within promotional messages, public service announcements, or experimental communication stimuli. Understanding how consumers process advertising claims requires examining not merely the semantic content of the message or the logical strength of arguments, but also the structural and relational attributes of the message source. While traditional advertising models historically emphasized source expertise and physical attractiveness, Feick and Higie (1992) identified a critical gap: consumers evaluate sources along dimensions of personal identification and shared perspective, particularly when evaluating subjective, experiential, or preference-heterogeneous product and service categories.

In consumer research and experimental psychology, the ES scale functions as both a key independent manipulation check and a focal mediator. In laboratory and online experiments, researchers systematically manipulate endorser portrayals—contrasting ordinary consumers, professional experts, peer referents, and celebrity figures. The ES scale provides an empirically rigorous measurement tool to confirm that experimental manipulations of endorser type successfully induce distinct levels of perceived likeness without confounding other source dimensions such as prestige or perceived competence.

Beyond manipulation checks, the scale serves vital diagnostic and predictive functions in applied communication research:

• Predicting Attitude Change and Message Acceptance: The instrument assesses how perceived likeness facilitates internalization and identification processes, directly predicting attitudes toward the advertisement (A_ad), attitudes toward the brand (A_b), and behavioral intentions.
• Evaluating Peer and Micro-Influencer Campaigns: In modern social media and digital marketing environments, the rise of creator-led branding and peer endorsements has intensified the need for validated metrics that gauge relatable peer perception versus distant aspirational figures.
• Health and Risk Communication: Public health researchers utilize the scale to evaluate whether target populations perceive testimonial speakers in vaccination, anti-smoking, mental health, or addiction campaigns as genuine peers, directly moderating threat appraisals and self-efficacy beliefs.
• Testing Boundary Conditions of Preference Heterogeneity: As conceptualized by Feick and Higie (1992), when products involve high preference heterogeneity (where personal tastes, lifestyles, and aesthetic values govern choice rather than objective technical standards), similarity to the endorser becomes significantly more persuasive than technical expertise.

5. Psychological Construct

The psychological construct captured by the Endorser Similarity scale is perceived source similarity—the subjective judgment made by an individual receiver that an information source shares fundamental attributes, life circumstances, preferences, values, or experiential backgrounds with the self. Rather than measuring objective, demographic demographic concordance (such as verified age, gender, or income parity), the construct captures the phenomenological reality of the receiver: the perceived psychological distance between the self-concept and the cognitive representation of the endorser.

Dimensions of Similarity in Persuasion Psychology

Within psychometric literature and communication theory, source similarity manifests across several interconnected dimensions:

• Demographic Similarity: The perception that the source shares visible or categorical identity markers, including age, gender, ethnicity, socioeconomic status, and geographic origin.
• Attitudinal and Value Similarity: The perception that the endorser holds beliefs, ethical outlooks, lifestyle choices, and taste structures congruent with the receiver’s personal identity.
• Experiential or Situational Similarity: The perception that the endorser faces identical problems, consumption dilemmas, or life challenges (e.g., managing a hectic family schedule, suffering from a chronic health condition, or navigating academic stress).

The ES scale deliberately operationalizes endorser similarity at a global, holistic level. By utilizing items such as “The person in the ad is a lot like me” and “The person in the ad seems to be similar to me,” the instrument allows respondents to integrate their own idiosyncratic weighting of demographic, attitudinal, and situational cues into a unified assessment of self-referential alignment. This global formulation prevents the measurement artifacts that arise when researchers presuppose which specific attributes (e.g., age versus lifestyle) drive perceived likeness in a given context.

Psychological Mechanisms of Perceived Similarity

When an individual perceives high endorser similarity, several psychological mechanisms are activated:

1. Enhanced Self-Referencing: Similarity cues stimulate cognitive elaboration wherein message arguments and product usage imagery are processed in relation to one’s own self-concept and autobiographical memory (Self-Reference Effect).
2. Attribution of Shared Tastes and Needs: Receivers infer that because the endorser shares fundamental traits or lifestyles, the endorser’s satisfaction with a brand or solution serves as a diagnostic proxy for their own prospective satisfaction.
3. Reduced Counterarguing and Psychological Reactance: Dissimilar or overtly commercially driven communicators frequently trigger skepticism and defensive resistance. In contrast, similar communicators are categorized as in-group members or trusted peers, leading to lower counterarguing and heightened affective warmth.

6. Theoretical Framework

The Endorser Similarity scale is grounded in three major psychological paradigms: Social Comparison Theory, Kelman’s Processes of Opinion Change, and Source Credibility/Attractiveness Models.

Festinger’s Social Comparison Theory

Leon Festinger’s (1954) Social Comparison Theory posits that human beings possess an innate drive to evaluate their opinions, beliefs, and abilities. When objective, physical standards of evaluation are unavailable, individuals evaluate themselves by comparison with others. Crucially, Festinger postulated that the drive to compare oneself with another specific person decreases as the perceived difference between one’s own opinion or ability and that of the other person increases. Individuals preferentially seek comparisons with similar others because similar others provide diagnostic, relevant, and valid reference benchmarks.

In consumer contexts, objective performance criteria rarely dictate whether a movie is enjoyable, whether a vacation destination is satisfying, or whether a clothing style is appealing. Under conditions of high preference heterogeneity, consumers rely on lateral social comparisons with similar peers rather than downward or upward comparisons with technical experts or unattainable celebrities. Feick and Higie (1992) embedded this logic directly into their conceptual model, demonstrating that perceived similarity is the primary driver of normative and informational social influence when consumer preferences are subjective.

Kelman’s Processes of Opinion Change

Herbert Kelman’s (1958, 1961) classic tripartite taxonomy of social influence delineates three distinct mechanisms through which an individual accepts influence from a communicator:

• Compliance: The receiver adopts the induced behavior not because of private belief, but to achieve a favorable reaction, reward, or avoid punishment from an agent possessing social or structural power.
• Internalization: The receiver accepts influence because the content of the communication is intrinsically congruent with their value system and objective reality, typically elicited by source credibility and expertise.
• Identification: The receiver adopts attitudes or behaviors because they are associated with a satisfying self-defining relationship to another person or group. Identification is elicited by source attractiveness and perceived similarity.

The Endorser Similarity scale operationalizes the cognitive appraisal that underpins Kelman’s identification pathway. When an endorser is perceived as “a lot like me,” persuasion occurs through an affiliative process: adopting the endorser’s brand preference allows the consumer to maintain, express, or reinforce their own actual or aspirational self-identity.

Source Attractiveness and Source Credibility Frameworks

Classic source models (e.g., Hovland & Weiss, 1951; McGuire, 1985; Ohanian, 1990) categorized source characteristics into two broad domains: credibility (comprising expertise and trustworthiness) and attractiveness (comprising physical beauty, familiarity, and similarity). While early advertising research prioritized physical attractiveness and technical expertise, Feick and Higie established that similarity represents an orthogonal construct. A peer endorser may lack formal institutional credentials (low technical expertise) and professional modeling aesthetics (moderate physical attractiveness), yet achieve superior persuasive impact due to high perceived similarity.

7. Validity

The psychometric validity of the Endorser Similarity scale has been established across multiple experimental investigations, cross-sectional surveys, and meta-analytic reviews in marketing, consumer behavior, and communication studies.

Construct and Content Validity

Content validity was established through systematic conceptualization grounded in social psychological literature on source characteristics. Feick and Higie (1992) conducted preliminary pretests to evaluate how consumers naturally describe and differentiate endorsers in print advertisements. The items were crafted to represent the core latent construct of self-endorser congruence directly, avoiding context-specific qualifiers (such as income or occupation) that could attenuate construct coverage across diverse demographic cohorts.

Convergent Validity

Convergent validity is evidenced by high item-to-total correlations and substantial factor loadings across validation samples. In Feick and Higie’s (1992) primary study involving 288 participants evaluating print advertisements across distinct service categories (hair salons and automotive repair), standardized factor loadings for all three items exceeded .80, indicating that the items share substantial variance with the underlying latent construct. Subsequent replications (e.g., Coulter et al., 2002; Munnukka et al., 2016) demonstrated that the average variance extracted (AVE) for the scale consistently exceeds .70, comfortably surpassing the standard .50 threshold established by Fornell and Larcker (1981).

Discriminant Validity

Discriminant validity requires proof that the scale measures a construct distinct from related source dimensions, such as source expertise, source physical attractiveness, and general brand familiarity. In their factor analyses, Feick and Higie (1992) demonstrated that the ES scale items loaded exclusively on an independent endorser similarity factor, exhibiting negligible cross-loadings (< .20) on factors capturing source expertise (measured via competence, training, and knowledge items) or source trustworthiness. Furthermore, inter-construct correlations between similarity and expertise rarely exceed .35, confirming that consumers maintain clear cognitive demarcations between someone who is competent and someone who is like themselves.

Predictive and Nomological Validity

Nomological validity is demonstrated through predictable relationships with theoretical antecedents and outcomes:

• Preference Heterogeneity Moderation: In support of Feick and Higie’s core hypotheses, endorser similarity significantly predicted cognitive ad processing, positive cognitive responses, and attitudes toward the endorser when product preference heterogeneity was high, whereas source expertise dominated when preference heterogeneity was low.
• Attitude toward the Brand (A_b) and Purchase Intentions: Structural equation models across empirical marketing literature regularly confirm direct positive paths from perceived similarity to brand attitudes (β values typically between .25 and .45) and indirect paths to purchase intent mediated through brand trust and ad credibility.
• Peer Endorser Manipulations: The scale consistently produces statistically significant differences across experimental treatment conditions (e.g., target consumer peer vs. celebrity endorser vs. technical specialist), confirming its diagnostic efficacy as a manipulation check instrument (F-values routinely demonstrating p < .001).

8. Reliability

The Endorser Similarity scale exhibits high levels of internal consistency reliability across varied participant demographics, experimental manipulations, and media modalities.

Internal Consistency Metrics

In the foundational investigation by Feick and Higie (1992), the three-item instrument yielded robust internal consistency statistics across multiple experimental conditions:

• Original Scale Validation: Cronbach’s alpha coefficients ranged between α = .88 and α = .91 across independent service categories and endorser experimental conditions.
• Subsequent Replications: In follow-up investigations utilizing identical or slightly adapted versions in consumer and digital advertising contexts (e.g., Coulter & Puntoni, 2003; Munnukka, Uusitalo, & Toivonen, 2016), Cronbach’s alpha coefficients have ranged between .85 and .93.
• Composite Reliability (CR): Structural equation modeling analyses evaluating composite reliability consistently report values exceeding .88, confirming that the scale is not compromised by item-specific variance or random measurement error.

Item-Total Correlations and Attenuation Correction

Corrected item-to-total correlations for each of the three items systematically exceed .70. Item 1 (“The person in the ad is a lot like me”) and Item 2 (“The person in the ad seems to be similar to me”) demonstrate item-total correlations typically between .75 and .85. Item 3, which is reverse-scored (“The person in the ad is not like me at all”), demonstrates item-total correlations between .68 and .78. The slightly lower correlation for Item 3 reflects typical psychometric patterns associated with reverse-worded items, yet its inclusion provides essential protection against response set artifacts and acquiescence bias.

Test-Retest Stability

While endorser similarity is fundamentally a state-like, situational evaluation tied to a specific advertisement stimulus rather than a stable personality trait, test-retest reliability has been evaluated across short-interval laboratory sessions (e.g., 48-hour to 1-week intervals with identical stimulus materials). Re-test correlation coefficients consistently hover between r = .78 and r = .84, confirming measurement stability in the absence of confounding experimental interventions.

9. Factor Analysis

Extensive factor-analytic evaluations confirm the unidimensional construct structure of the Endorser Similarity scale.

Exploratory Factor Analysis (EFA)

During scale formulation, Feick and Higie (1992) conducted exploratory factor analyses using principal component analysis (PCA) with both orthogonal (Varimax) and oblique (Promax) rotations alongside items assessing source expertise, source trustworthiness, and physical attractiveness. The factor structure converged cleanly:

• Eigenvalues and Explained Variance: The three similarity items loaded unambiguously onto a single distinct factor with eigenvalues substantially exceeding 1.0 (typically ranging from 2.15 to 2.45), accounting for 72% to 82% of the total item variance.
• Factor Loadings: Standardized factor loadings across samples consistently demonstrate high primary loadings: Item 1 loadings range from .84 to .91; Item 2 loadings range from .86 to .93; Item 3 loadings range from -.78 to -.88 prior to reverse-scoring, and .78 to .88 when reverse-coded.
• Cross-Loadings: Cross-loadings onto adjacent source characteristic dimensions (expertise, attractiveness) remained negligible (consistently < .22), verifying clean structural independence.

Confirmatory Factor Analysis (CFA)

Contemporary psychometric evaluations employing Confirmatory Factor Analysis (CFA) within structural equation modeling (SEM) software (e.g., AMOS, Mplus, lavaan) confirm superior fit indices for a one-factor congeneric measurement model. Typical fit parameters reported in the empirical literature include:

• Chi-Square / Degrees of Freedom: χ²/df ≤ 2.10 (frequently non-significant, p > .05, due to minimal degrees of freedom).
• Comparative Fit Index (CFI): Values regularly exceed .98 to 1.00.
• Tucker-Lewis Index (TLI): Values consistently exceed .97.
• Root Mean Square Error of Approximation (RMSEA): Values typically range from .000 to .048, with 90% confidence intervals encompassing zero.
• Standardized Root Mean Square Residual (SRMR): Values systematically below .025.

Attempts to fit multi-factor structures or split positively and negatively worded items into distinct latent sub-constructs yield inferior or degenerate solutions, demonstrating that the reverse-scored item functions effectively as a structural component of the single latent similarity dimension rather than reflecting a substantive second factor.

10. Instrument / Measurement Tool

The Endorser Similarity (ES) instrument is structured as an efficient, reliable, self-report psychometric test. The key operational attributes of the instrument are detailed below:

• Test Type: Self-administered psychometric rating scale; perceptual evaluation instrument.
• Target Respondent: Consumers, study participants, viewers, or media recipients exposed to an advertisement, public service announcement, commercial video, or influencer post.
• Format: Paper-and-pencil or computerized self-report questionnaire.
• Item Count: Exactly 3 items.
• Response Scale: 7-point Likert scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Somewhat Disagree
  • 4 = Neither Agree nor Disagree (Neutral)
  • 5 = Somewhat Agree
  • 6 = Agree
  • 7 = Strongly Agree
• Scoring Rules:
  • Item 1: Direct scoring (1 = 1, 2 = 2, 3 = 3, 4 = 4, 5 = 5, 6 = 6, 7 = 7).
  • Item 2: Direct scoring (1 = 1, 2 = 2, 3 = 3, 4 = 4, 5 = 5, 6 = 6, 7 = 7).
  • Item 3: Reverse-scored (1 = 7, 2 = 6, 3 = 5, 4 = 4, 5 = 3, 6 = 2, 7 = 1).
• Index Calculation: Items are averaged or summed to form an overall index of perceived endorser similarity. When averaged, the composite score ranges from 1.0 to 7.0. Higher scores indicate greater perceived similarity to the endorser.
• Administration Time: Less than 1 minute (typically 30 to 45 seconds).

11. Permissions & Fee and Test Year

The Endorser Similarity scale was originally published in 1992 in the peer-reviewed academic article:

Feick, Lawrence, and Robin A. Higie. (1992). “The Effects of Preference Heterogeneity and Source Characteristics on Ad Processing and Judgments About Endorsers.” Journal of Advertising, 21(2), 9–24.

Licensing and Academic Access:

• Fee: Free for scholarly, non-commercial research, academic theses, and university-sponsored scientific investigations. No licensing fees or royalty payments are charged for academic usage.
• Permissions: Standard fair use principles under academic copyright law permit researchers to cite and administer the scale items provided appropriate bibliographic attribution is given to the original authors (Feick & Higie, 1992) and the Journal of Advertising.
• Commercial Applications: Commercial organizations, proprietary market research agencies, or corporate entities seeking to embed the scale within commercial diagnostic platforms should consult the copyright policies of Taylor & Francis (the current publisher of the Journal of Advertising on behalf of the American Academy of Advertising).

12. References

Below is a comprehensive list of scholarly references cited in this article, formatted in accordance with the American Psychological Association (APA 7th edition) standard:

• Coulter, R. A., & Puntoni, S. (2003). The effects of endorser ethnicity and product category on consumer judgments. Journal of Advertising Research, 43(4), 385–398. https://doi.org/10.2501/JAR-43-4-385-398
• Feick, L., & Higie, R. A. (1992). The effects of preference heterogeneity and source characteristics on ad processing and judgments about endorsers. Journal of Advertising, 21(2), 9–24. https://doi.org/10.1080/00913367.1992.10673364
• Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117–140. https://doi.org/10.1177/001872675400700202
• 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
• 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
• Kelman, H. C. (1958). Compliance, identification, and internalization: Three processes of attitude change. Journal of Conflict Resolution, 2(1), 51–60. https://doi.org/10.1177/002200275800200106
• Kelman, H. C. (1961). Processes of opinion change. Public Opinion Quarterly, 25(1), 57–78. https://doi.org/10.1086/266996
• McGuire, W. J. (1985). Attitudes and attitude change. In G. Lindzey & E. Aronson (Eds.), Handbook of Social Psychology (3rd ed., Vol. 2, pp. 233–346). Random House.
• Munnukka, J., Uusitalo, O., & Toivonen, H. (2016). Credibility of a peer endorser and advertising effectiveness. Journal of Consumer Marketing, 33(3), 182–192. https://doi.org/10.1108/JCM-11-2014-1221
• 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

13. Items of the Scale