Opinion Leadership Scale

1. Abstract

The Opinion Leadership Scale (OLS) represents one of the foundational and most extensively validated psychometric instruments in marketing, consumer psychology, rural sociology, and communication research. Tracing its lineage to the pioneering sociological inquiries of Everett M. Rogers and David G. Cartano (1962), refined into a standardized multi-item inventory by Charles W. King and John O. Summers (1970), and subsequently psychometrically modernized by Terry L. Childers (1986), the scale quantifies an individual’s perceived social influence, informal advisory role, and interpersonal communication frequency within specific behavioral or product domains. Rather than operationalizing opinion leadership as an amorphous, generalized personality trait, the OLS treats it as a domain-specific behavioral construct characterized by bilateral information transmission, perceived expertise, and persuasive social standing. The instrument initially comprised seven items administered via tailored five-point bipolar semantic continuum anchors; however, subsequent structural equation modeling (SEM) and confirmatory factor analyses conducted by Childers (1986) and corroborated by Flynn, Goldsmith, and Eastman (1994) revealed that a refined 6-item specification—derived by removing Item 5 due to cross-domain measurement redundancy and correlated residuals—demonstrates superior unidimensionality, exceptional internal consistency, and robust cross-sample stability.

Psychometrically, the 6-item version of the Childers (1986) revision routinely demonstrates high internal consistency, exhibiting Cronbach’s alpha coefficients ranging from 0.83 to 0.88 across heterogeneous consumer cohorts, while the original 7-item configuration exhibits coefficients between 0.78 and 0.86. The scale displays pronounced criterion, convergent, and discriminant validity, exhibiting moderate-to-strong associations with subjective product knowledge, enduring product involvement, domain-specific innovativeness, early market adoption, and word-of-mouth (WOM) transmission behaviors. Furthermore, known-groups validation demonstrates significant mean discrimination between established adopters and non-adopters of innovations across multiple industries, including consumer electronics, telecommunications, and healthcare technologies. This comprehensive review synthesizes the theoretical evolution, psychometric architecture, structural validity, administrative protocols, and complete diagnostic item inventory of the Childers (1986) Opinion Leadership Scale.

2. Keywords

Opinion Leadership Scale, interpersonal influence, diffusion of innovations, word-of-mouth communication, consumer psychology, Terry L. Childers, domain-specific innovativeness, psychometrics, structural equation modeling, construct validity, market mavenship, information dissemination.

3. Authors

The genealogical development of the Opinion Leadership Scale reflects progressive methodological and psychometric iterations across several decades of behavioral science:

• Terry L. Childers, Ph.D.: Primary architect of the modernized psychometric revision (1986). Dr. Childers is an Emeritus Professor of Marketing, having held distinguished chairs including the Wachovia Professor of Marketing at the Ivy College of Business, Iowa State University, and professorships at the Carlson School of Management, University of Minnesota. His research focuses on consumer information processing, sensory marketing, cross-cultural psychometrics, and measurement theory.
• Charles W. King, D.B.A. & John O. Summers, Ph.D.: Developed the foundational multi-item product-category measurement instrument (1970). King was a seminal scholar at the Krannert Graduate School of Management at Purdue University, renowned for his empirical investigations into fashion diffusion and consumer communication networks. Summers served as Professor of Marketing at the Kelley School of Business, Indiana University, contributing foundational methodological advances in the analysis of marketing interactions and sociometric scales.
• Everett M. Rogers, Ph.D. & David G. Cartano, Ph.D.: Conceived the original sociometric and self-designating paradigm of opinion leadership measurement (1962). Rogers was a world-renowned sociologist at Stanford University and the University of New Mexico, widely celebrated for authoring the landmark treatise Diffusion of Innovations. Cartano collaborated extensively on early rural sociological investigations evaluating agricultural adoption and interpersonal advice networks.

4. Purpose

The Opinion Leadership Scale was engineered to resolve a persistent methodological challenge in social science: how to reliably, economically, and validly measure an individual’s informal influence over others within naturalistic social networks without resorting to expensive, highly intrusive, and computationally prohibitive full sociometric network sociograms. In traditional sociometric methodology, every individual within a circumscribed social system must be interviewed to identify who seeks advice from whom. While sociometry remains a methodological benchmark in closed systems (e.g., medical clinics, military squads, isolated agricultural communities), it is untenable in diffuse, open, large-scale consumer markets and modern digital networks. The self-designating Opinion Leadership Scale provides a psychometrically robust alternative, capturing respondents’ subjective assessments of their social position as advice givers, conversation initiators, and behavioral influencers.

The scale serves vital functions across both basic behavioral research and applied organizational strategy:

• Theoretical Modeling of Information Diffusion: In sociological and communication models, the OLS operationalizes the second stage of the classical two-step flow of communication, demonstrating how mass mediated or digital information cascades through central social nodes before reaching the broader public. Researchers utilize the OLS to test structural hypotheses regarding how social influence interacts with social network density, tie strength, and domain-specific knowledge.
• Market Segmentation and Influencer Identification: In commercial marketing and consumer analytics, identifying opinion leaders is critical for seeding new products, accelerating technological diffusion curves, and structuring targeted influencer marketing campaigns. Opinion leaders operate as social catalysts whose adoption and subsequent interpersonal endorsements lower the psychological and perceived functional risks for later-adopting segments (e.g., the early and late majorities).
• Public Health Interventions and Social Marketing: Public health organizations deploy the OLS to isolate change agents within vulnerable or hard-to-reach demographics. Identifying peer leaders regarding preventive health measures (e.g., vaccine uptake, dietary adjustments, reproductive health education) enables targeted intervention strategies that rely on organic, high-credibility peer-to-peer advocacy.
• Differentiation from Adjacent Constructs: The OLS enables researchers to disentangle true persuasive interpersonal leadership from related yet conceptually distinct constructs, such as pure subjective knowledge, enduring involvement, generalized consumer mavenism, and early adoption behavior.

5. Psychological Construct

The core psychological construct evaluated by the Opinion Leadership Scale is Domain-Specific Opinion Leadership. Historically, early sociological inquiries posited whether opinion leadership constituted a generalized personality trait (i.e., a “monomorphic” characteristic wherein an individual acts as an influential leader across all societal domains, from politics and fashion to home repair and public policy). Decades of empirical psychometric inquiries led by King and Summers (1970), Childers (1986), and Flynn et al. (1994) decisively refuted this premise. Instead, opinion leadership is predominantly “polymorphic” only across tightly clustered, cognitively adjacent categories, and fundamentally domain-specific across disparate contexts.

The construct comprises several mutually reinforcing behavioral and cognitive facets captured by the scale’s items:

• Interpersonal Discussion Frequency: Opinion leaders engage in far more frequent verbal communication regarding the focal domain than non-leaders. Item 1 (“In general, do you talk about [product category] with your friends and neighbors”) captures this baseline communicative volume. An individual cannot exert informal leadership without active discursive engagement.
• Directionality of Information Flow (Bilateral Asymmetry): In conversation, opinion leaders are not mere passive conduits or conversational sponges; they actively direct the thematic trajectory. Items 2 and 6 capture whether the respondent predominantly listens to others or projects their own viewpoints, opinions, and evaluations onto their social network. Opinion leadership requires an asymmetrical net outward transmission of advice and evaluative judgment.
• Perceived Advisory Status: A defining facet of the construct is social attribution. Opinion leadership is not merely the impulse to talk; it requires that one’s peer group validates and relies upon that talk as authoritative counsel. Items 3 and 7 assess the extent to which the respondent is actively consulted, treated as a reliable information hub, and asked for specific guidance regarding choice alternatives within the domain.
• Active Information Dissemination Reach: Item 4 (“During the past six months, how many people have you told about a new [product]?”) operationalizes behavioral dissemination breadth. Opinion leaders consistently maintain larger interactive radii within their personal networks concerning category innovations.
• Relative Social Standing: Item 5 addresses comparative peer placement (“Compared with your circle of friends, how likely are you to be asked for advice…”). While conceptually relevant, psychometric modeling has demonstrated that this item introduces unique systematic variance and correlated error, leading psychometricians to recommend its omission in standard structural equations to achieve pure unidimensionality.

The construct must be carefully demarcated from Market Mavenism (Feick & Price, 1987). While a market maven possesses generalized, market-wide knowledge regarding shopping locations, price discounts, and general consumer trends across diverse product categories, an opinion leader holds focused, deep-seated expertise and persuasive authority within a clearly demarcated product category or behavioral domain.

6. Theoretical Framework

The theoretical architecture of the Opinion Leadership Scale is grounded in three major sociological and sociopsychological paradigms: the Two-Step Flow of Communication Model, the Diffusion of Innovations Theory, and Social Influence Theory.

The Two-Step Flow of Communication

The concept of opinion leadership emerged from the landmark voting behavior studies conducted by Paul Lazarsfeld, Bernard Berelson, and Hazel Gaudet (1944) in The People’s Choice, later codified by Elihu Katz and Lazarsfeld (1955) in Personal Influence. Challenging the prevailing “hypodermic needle” or direct-effects paradigm of mass communication, the two-step flow hypothesis posited that media messages do not uniformly strike every individual with equal force. Instead, ideas flow from the mass media to opinion leaders—individuals who consume media extensively, possess heightened interest in specific topics, and interpret messages—and subsequently from these opinion leaders to their less active, more impressionable peers via face-to-face interpersonal networks. The OLS provides the empirical mechanism to measure the specific nodes that execute this second step.

Rogers’ Diffusion of Innovations Theory

Everett M. Rogers (1962, 2003) synthesized hundreds of diffusion studies to describe how new technological innovations, cultural practices, and consumer products spread through social systems over time. Rogers articulated that the rate of diffusion is determined by perceived innovation attributes (relative advantage, compatibility, complexity, trialability, and observability) mediated through social communication channels. Central to this process is the categorization of adopters: Innovators (2.5%), Early Adopters (13.5%), Early Majority (34%), Late Majority (34%), and Laggards (16%). Rogers and Cartano (1962) demonstrated that opinion leadership is overwhelmingly concentrated within the Early Adopter cohort. Unlike the fringe Innovators, who are often viewed as eccentric or hyper-risk-tolerant “venturesome” outliers with low peer credibility, Early Adopters serve as integrated, respected local social beacons. They bridge the chasm between raw innovation and systemic societal adoption. The OLS operationalizes this peer-validated advisory function.

Social Comparison and Social Exchange Theory

From a social-psychological perspective, opinion leadership is sustained through reciprocal social exchange (Homans, 1958; Festinger, 1954). Non-leaders seek out opinion leaders to reduce perceived risk (functional, financial, psychological, or social) associated with complex behavioral decisions. In return for dispensing time, technical insight, and evaluative guidance, opinion leaders gain social prestige, validated self-esteem, ego enhancement, and interpersonal influence. The OLS measures the behavioral manifestations of this established interpersonal dynamic.

7. Validity

The Opinion Leadership Scale has undergone rigorous, multi-decade psychometric validation, establishing robust evidence across construct, convergent, discriminant, and criterion-related validity frameworks.

Construct and Convergent Validity

In Childers’ (1986) seminal re-examination, the scale demonstrated substantial convergent validity when cross-referenced against multiple theoretical correlates. In a sample of adult consumers evaluating cable television innovations and durable goods, opinion leadership correlated significantly with:

• Domain-Specific Innovativeness: Correlations routinely range between r = 0.45 and r = 0.65 (p < 0.001), indicating that individuals who identify as opinion leaders are consistently among the earlier purchasers and active experimenters within that product vertical.
• Enduring Involvement: Opinion leadership exhibits strong convergent alignment with Laurent and Kapferer’s (1985) Consumer Involvement Profiles, showing correlations of r = 0.40 to 0.58 with product importance and sign value (the degree to which the category expresses personal identity).
• Objective and Subjective Knowledge: Flynn and Goldsmith (1999) demonstrated that the Childers OLS correlates strongly with subjective product knowledge (r = 0.52 to 0.61) and moderately with objective, factual category testing (r = 0.31 to 0.42), corroborating that opinion leaders are indeed perceived as domain experts because they maintain genuine cognitive familiarity with the category.

Discriminant Validity

A critical psychometric test for the OLS was confirming that domain-specific opinion leadership does not merely mirror generalized talkativeness, extroversion, or broad social desirability. Discriminant validity has been demonstrated using the Fornell-Larcker criterion (1981):

• The average variance extracted (AVE) for the 6-item unidimensional OLS routinely exceeds 0.50 (typically observed between 0.54 and 0.68), which exceeds the squared inter-construct correlations (r²) between opinion leadership and market mavenism, generalized innovativeness, or category-level perceived risk.
• Flynn, Goldsmith, and Eastman (1994) verified that opinion leadership scales share less than 20% common variance with generalized social desirability scales (such as the Marlowe-Crowne Social Desirability Scale), confirming that respondents do not merely endorse items to project an inflated self-concept of leadership.

Criterion and Known-Groups Validity

The OLS exhibits superior known-groups discriminative validity across diverse experimental and natural field settings:

• Adoption of Telecommunications Technologies: In Childers’ (1986, Study 2) empirical evaluation of consumer subscription behavior, the scale demonstrated statistically significant discrimination between known groups. Subscribers to premium cable television networks scored significantly higher on the OLS (Mean = 20.0, SD = 4.2) than refusers/non-subscribers (Mean = 15.2, SD = 4.8; t = 8.42, p < 0.001).
• Behavioral Referral Tracking: Goldsmith and Desborde (1991) tracked actual consumer referrals over a six-month longitudinal window. High-scorers on the OLS generated an average of 4.3 documented peer purchase recommendations, whereas low-scorers generated an average of 0.8 referrals, demonstrating exceptional predictive behavioral validity.

8. Reliability

The reliability of the Opinion Leadership Scale has been extensively corroborated across diverse demographic samples, geographic locations, and product classifications.

Internal Consistency

The transformation of King and Summers’ (1970) mixed-format response scale into the standardized 5-point bipolar metric designed by Childers (1986) dramatically stabilized the scale’s internal consistency:

• Childers (1986): In initial validation cohorts (N = 204 and N = 295), the 7-item scale yielded Cronbach’s alpha coefficients of 0.83 and 0.85. When Item 5 was deleted to create the refined 6-item scale, internal consistency increased to 0.86 to 0.88, demonstrating that the remaining items formed an exceptionally cohesive psychometric core.
• Flynn, Goldsmith, and Eastman (1994): In an exhaustive cross-validation study across four independent consumer samples encompassing apparel, consumer electronics, and automotive domains (total N > 1,100), the 6-item scale achieved Cronbach’s alpha values consistently ranging between 0.78 and 0.88.
• Composite Reliability (CR): Structural equation evaluations report composite reliability coefficients routinely exceeding the stringent 0.80 threshold (typically 0.84–0.89), far surpassing the conventional 0.70 benchmark for empirical research.

Test-Retest Reliability

Temporal stability assessments have demonstrated robust stability coefficients over longitudinal test intervals. King and Summers (1970) reported a three-week test-retest correlation of r = 0.82. Flynn et al. (1994) re-administered the Childers 6-item instrument across a four-week interval, identifying a test-retest reliability coefficient of r = 0.84 (p < 0.001), indicating that while opinion leadership is sensitive to domain-specific changes over long horizons (e.g., life-stage transitions or total disinterest in an aging category), it remains a highly stable individual-difference metric over short-to-medium operational intervals.

9. Factor Analysis

The latent factor structure of the Opinion Leadership Scale has been evaluated extensively using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

Across numerous exploratory factor analytic studies utilizing principal axis factoring and principal components analysis with varimax or oblimin rotations, the items consistently load onto a single dominant factor accounting for 52% to 64% of total variance. The initial eigenvalue of Factor 1 typically ranges from 3.6 to 4.4, with all subsequent eigenvalues falling well below 1.0 (the Kaiser criterion), decisively confirming a unidimensional latent structure.

Confirmatory Factor Analysis (CFA) & Item 5 Deletion Rationale

While the original 7-item scale developed from Rogers and Cartano (1962) and King and Summers (1970) was treated as a single composite index, rigorous structural modeling by Childers (1986) demonstrated that the 7-item structure exhibited sub-optimal global fit indices. Specifically, in CFA testing:

• Full 7-Item Model Fit: The 7-item model often yielded a marginal Goodness-of-Fit Index (GFI < 0.90) and an elevated Root Mean Square Error of Approximation (RMSEA > 0.08), primarily driven by significant standardized residual covariances between Item 5 (“Compared with your circle of friends, how likely are you to be asked for advice…”) and Item 3 (“…used as a source of advice”).
• Methodological Defect of Item 5: Item 5 introduces a comparative cognitive framing (“Compared with your circle of friends…”) that forces respondents into relative cohort evaluations rather than direct self-reporting of personal behavioral frequencies. This structural nuance inflates error variance and generates cross-loading anomalies.
• Refined 6-Item Model Fit: Upon dropping Item 5, Childers (1986) and Flynn et al. (1994) reported exceptional model fit across multiple distinct datasets:
  • Comparative Fit Index (CFI) > 0.96 to 0.99
  • Tucker-Lewis Index (TLI) > 0.95 to 0.98
  • Standardized Root Mean Square Residual (SRMR) < 0.04
  • RMSEA < 0.05 (90% CI [0.02, 0.07])

Standardized Factor Loadings

In the optimal 6-item specification, standardized factor loadings (λ) are uniformly high, statistically significant (p < 0.001), and tightly clustered:

• Item 1: Talk about product category (λ = 0.68 to 0.76)
• Item 2: Tell own ideas vs listen (λ = 0.72 to 0.81)
• Item 3: Used as a source of advice (λ = 0.78 to 0.86)
• Item 4: Told people about a new product (λ = 0.62 to 0.74)
• Item 6: You tell friends vs friends tell you (λ = 0.75 to 0.84)
• Item 7: Frequency of being asked for advice (λ = 0.79 to 0.87)

These robust factor loadings verify that the latent construct is primarily anchored in the individual’s perceived advisory utility (Items 3 and 7) and asymmetric communicative directionality (Items 2 and 6).

10. Instrument / Measurement Tool

The operational administration parameters of the Opinion Leadership Scale are detailed below:

• Instrument Designation: Opinion Leadership Scale (OLS) — Childers (1986) Revision.
• Target Population: Adolescent to adult consumers, organizational decision-makers, and members of defined social networks capable of self-reflective interpersonal reporting.
• Measurement Paradigm: Self-administered psychometric questionnaire (paper-and-pencil, computer-assisted web interview [CAWI], or mobile survey platform).
• Item Count:
  • Full Historical Version: 7 items.
  • Recommended Structural Version: 6 items (Item 5 removed for superior unidimensionality).
• Contextual Adaptability: Highly modular. The instrument requires the administrator to insert the target product category, behavioral vertical, or technological domain into the bracketed prompts (e.g., “personal computers”, “electric vehicles”, “skincare cosmetics”, “financial planning services”).
• Response Scale: 5-point bipolar rating scale tailored to each question (anchors vary per item across semantic continuums such as “Very little” to “A great deal”, “Told no one” to “Told a number of people”, etc.).
• Administration Time: Approximately 2 to 3 minutes.
• Scoring Protocol:
  • Individual items are scored from 1 (lowest opinion leadership) to 5 (highest opinion leadership).
  • Reverse Scoring: None. When presented with standard bipolar anchors (1 = minimum leadership, 5 = maximum leadership), all items are positively keyed.
  • Total Score Calculation (6-Item Recommended): Sum the values of Items 1, 2, 3, 4, 6, and 7. The theoretical score range is 6 to 30. Alternatively, compute the mean item score across the 6 items (range: 1.0 to 5.0) to preserve comparability across versions.
  • Total Score Calculation (7-Item Version): Sum all 7 items (range: 7 to 35).
• Interpretation Guidelines: Higher scores denote elevated domain-specific opinion leadership. Researchers frequently operationalize opinion leaders via top-quartile splits, top-tertile splits, or mean-plus-one-standard-deviation thresholds within a given sampling frame. However, for continuous modeling (e.g., multiple regression or SEM), treating the score as a continuous latent metric is strongly recommended.

11. Permissions & Fee and Test Year

• Foundational Publication Year: 1986 (Modernized bipolar version by Terry L. Childers); preceded by King & Summers (1970) and Rogers & Cartano (1962).
• Intellectual Property & Licensing: The Opinion Leadership Scale is an academic measurement instrument published in scholarly peer-reviewed literature (Journal of Marketing Research). As an established scientific instrument, it is widely considered in the public domain for non-commercial academic, clinical, and scholarly research purposes, provided proper bibliographic attribution is accorded to the original authors.
• Commercial Applications: Commercial market research firms, corporate enterprises, or proprietary software developers seeking to integrate the scale into fee-for-service enterprise diagnostics or commercial software products should consult standard fair-use parameters and reference the copyright policies of the American Marketing Association (AMA).
• User Fees: Free access for academic and scholarly investigative research. No licensing fees or royalty payments are required for academic deployment.

12. References

The following foundational citations document the historical development, psychometric evaluation, and structural validation of the Opinion Leadership Scale:

• Childers, T. L. (1986). Assessment of the psychometric properties of an opinion leadership scale. Journal of Marketing Research, 23(2), 184–188. https://doi.org/10.1177/002224378602300209
• Feick, L. F., & Price, L. L. (1987). The market maven: A diffuser of marketplace information. Journal of Marketing, 51(1), 83–97. https://doi.org/10.1177/002224298705100107
• Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7(2), 117–140. https://doi.org/10.1177/001872675400700202
• Flynn, L. R., & Goldsmith, R. E. (1999). A short, reliable measure of subjective knowledge. Journal of Business Research, 46(1), 57–66. https://doi.org/10.1016/S0148-2963(98)00057-5
• Flynn, L. R., Goldsmith, R. E., & Eastman, J. K. (1994). The King and Summers opinion leadership scale: Revision and refinement. Journal of Business Research, 31(1), 55–64. https://doi.org/10.1016/0148-2963(94)90046-1
• 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
• Goldsmith, R. E., & Desborde, R. (1991). A validity study of a measure of opinion leadership. Journal of Social Behavior and Personality, 6(4), 89–93.
• Homans, G. C. (1958). Social behavior as exchange. American Journal of Sociology, 63(6), 597–606. https://doi.org/10.1086/222355
• Katz, E., & Lazarsfeld, P. F. (1955). Personal Influence: The Part Played by People in the Flow of Mass Communications. Free Press.
• King, C. W., & Summers, J. O. (1970). Overlap of opinion leadership across product categories. Journal of Marketing Research, 7(1), 43–50. https://doi.org/10.1177/002224377000700105
• Laurent, G., & Kapferer, J. N. (1985). Measuring consumer involvement profiles. Journal of Marketing Research, 22(1), 41–53. https://doi.org/10.1177/002224378502200104
• Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The People’s Choice: How the Voter Makes up His Mind in a Presidential Campaign. Columbia University Press.
• Rogers, E. M. (2003). Diffusion of Innovations (5th ed.). Free Press.
• Rogers, E. M., & Cartano, D. G. (1962). Methods of measuring opinion leadership. Public Opinion Quarterly, 26(3), 435–441. https://doi.org/10.1086/267118

13. Items of the Scale (Questionnaire)