Information Source Importance (Internet Sources) (ISI)

Abstract

The Information Source Importance (Internet Sources) (ISI) scale is a psychometric instrument designed to evaluate the subjective cognitive value, diagnostic utility, and behavioral reliance that consumers assign to online, web-based channels when seeking information regarding health conditions, pharmacological therapies, and medical decision-making. Developed within the empirical context of direct-to-consumer pharmaceutical advertising by Wonsuk Jerry Kim and Karen Whitehill King (2009), the instrument operationalizes the epistemic weight individuals allocate to digital external search environments relative to other informational modalities. The instrument measures consumer perceptions across diverse online channels, distinguishing between commercial brand-sponsored portals, independent scientific health repositories, and peer-to-peer interactive health forums. Typically implemented as a multi-item, Likert-type response battery, the scale evaluates items along dimensions of perceived credibility, accessibility, decision support, and informational completeness. Psychometric evaluations demonstrate strong internal consistency reliability (Cronbach’s alpha coefficients consistently exceeding α = .80 across independent consumer samples) and robust construct validity substantiated through exploratory and confirmatory factor analyses. The scale reveals robust convergent associations with objective search duration, health involvement, and epistemic curiosity, while establishing discriminant divergence from traditional interpersonal medical advice channels (such as consultations with physicians and pharmacists) and traditional mass media advertising. This article provides a comprehensive psychometric profile of the ISI, reviewing its theoretical underpinnings in the economics of information, dual-process cognitive frameworks, structural dimensionality, validity profiles, and research applications in health communication, psychopharmacology, and consumer behavior.

Keywords

Information Source Importance, Internet Sources, External Information Search, Health Information Seeking, Direct-to-Consumer Advertising, Consumer Health Informatics, Psychometrics, Scale Validation, Health Literacy, Decision-Making

Authors

The Information Source Importance (Internet Sources) scale was formulated and validated by scholars specializing in advertising theory, health communication, and consumer psychology:

• Wonsuk Jerry Kim, Ph.D. — Former doctoral researcher at the Grady College of Journalism and Mass Communication, University of Georgia, Athens, GA, USA; subsequently academic researcher and faculty contributor specializing in consumer information search strategies, pharmaceutical marketing, and interactive digital advertising environments.
• Karen Whitehill King, Ph.D. — Professor Emerita of Advertising, Department of Advertising and Public Relations, Grady College of Journalism and Mass Communication, University of Georgia, Athens, GA, USA. Dr. King is an internationally recognized scholar in direct-to-consumer (DTC) prescription drug advertising, health communication campaigns, advertising regulation, and media selection behavior.

Purpose

The primary purpose of the Information Source Importance (Internet Sources) scale is to quantify the cognitive and evaluative priority that an individual assigns to digital information channels during external information search episodes. In the modern healthcare and consumer landscape, individuals are confronted with an expansive array of competing data sources when managing symptomatic episodes, evaluating pharmaceutical alternatives, or diagnosing health states. Prior to the widespread adoption of the internet, consumers relied heavily on asymmetric informational interactions characterized by traditional offline gatekeepers: physicians, pharmacists, and print or television broadcast advertisements. The rapid growth of the digital landscape democratized health intelligence, shifting consumer behavior toward active, autonomous external search.

From an applied perspective, the scale serves critical functions across academic research, health policy, and behavioral healthcare:

• Consumer Decision-Making Analysis: It enables behavioral scientists to quantify how perceived channel importance alters the consumer decision journey, specifically predicting whether an individual will engage in collaborative decision-making, challenge clinical recommendations, or exhibit treatment compliance.
• Direct-to-Consumer Advertising (DTCA) Evaluation: The scale was engineered specifically to investigate differential search dynamics between high-involvement, high-risk categories (prescription medications, Rx) and lower-risk categories (over-the-counter medications, OTC). It illuminates how external broadcast stimuli prompt consumers to navigate to digital channels for validation and risk assessment.
• Health Informatics and Interface Design: Public health organizations, hospital systems, and regulatory agencies (e.g., the Food and Drug Administration) use this measurement framework to discern which digital source characteristics foster consumer trust, thereby optimizing the delivery of public health advisories and combatting medical misinformation online.

Psychological Construct

The Information Source Importance (Internet Sources) scale measures a multidimensional cognitive-evaluative construct rooted in subjective utility theory and external information acquisition behavior. Construct definitions within psychometrics conceptualize “source importance” not merely as the frequency of usage, but as the perceived informational diagnostic value and decision-making weight allocated to an external channel.

When an individual encounters an internal epistemic need (e.g., an unfamiliar medical symptom, a newly prescribed medication regimen, or conflicting therapeutic claims), cognitive dissonance and perceived risk stimulate external search behavior. Within the internet modality, this construct spans multiple distinct sub-dimensions:

• Perceived Diagnostic Credibility: The degree to which web-based repositories are judged to provide accurate, scientifically valid, and authoritative medical knowledge. This reflects cognitive trust in digital health portals (e.g., Mayo Clinic, WebMD, academic medical centers) over commercial narratives.
• Transactional Decision Utility: The extent to which internet sources provide actionable, pragmatic guidance that directly shapes behavioral outcomes, such as prompting an appointment with a specialist, requesting a specific branded pharmaceutical product, or adjusting dosage regimens.
• Commercial vs. Non-Commercial Source Discrimination: The psychological capacity and tendency of the consumer to calibrate the importance of manufacturer-sponsored websites (e.g., branded drug landing pages designed for promotional conversion) relative to independent, non-sponsored consumer advocacy portals or governmental repositories (e.g., the National Institutes of Health, CDC).
• Peer-Led Participatory Value: The perceived value of experiential knowledge gathered from digital peer networks, including online patient forums, social media support groups, and patient narrative repositories, where affective support and anecdotal efficacy data are prioritized.

The ISI captures the consumer’s internal weighting matrix. A consumer may exhibit extensive search volume across websites while simultaneously assigning low importance to the obtained data if they judge the medium to be saturated with bias or low in diagnostic veracity. Conversely, an individual assigning high importance to internet sources integrates digital findings directly into their self-concept, symptom evaluation schema, and physician consultation agenda.

Theoretical Framework

The ISI scale is theoretically grounded in the convergence of neoclassical economic models of search behavior, consumer psychology, and cognitive processing paradigms:

1. Economics of Information and Search Theory

Originating from the seminal work of George Stigler (1961) and later expanded by behavioral economists (e.g., Schmidt & Spreng, 1996), external information search is modeled as a rational cost-benefit calculus. An individual engages in information seeking as long as the marginal cognitive and practical benefits of obtaining additional data exceed the marginal search costs (temporal expenditure, cognitive strain, financial cost). The internet drastically compresses search costs, facilitating near-zero-cost retrieval of vast informational repositories. The ISI measures the subjective benefit component of this equation: how valuable and indispensable the consumer perceives digital data to be in resolving uncertainty.

2. The Elaboration Likelihood Model (ELM) and Dual-Process Theory

The Elaboration Likelihood Model (Petty & Cacioppo, 1986) posits two routes to persuasion: the central route (requiring systematic, high-elaboration cognitive processing of message arguments) and the peripheral route (relying on heuristic cues, source credibility, and affective framing). In the context of pharmaceutical external search, prescription medications represent high-involvement products carrying substantial physical, financial, and psychological risk. Consequently, consumers are motivated to process information via the central route. Web-based channels provide hyperlinked, detailed text, clinical trial disclosures, and contraindication data that satisfy the central route processing needs of highly involved health consumers.

3. The Health Belief Model (HBM) and Risk Perception

According to the Health Belief Model (Rosenstock, 1974), health-related actions are governed by perceived susceptibility, perceived severity, perceived benefits of preventive or therapeutic behaviors, and perceived barriers. Kim and King (2009) anchored their inquiry in the premise that category-level risk (Rx versus OTC) fundamentally shifts the perceived importance of information channels. When perceived severity and susceptibility are high (typical of prescription-level chronic or acute illnesses), consumers evaluate external information sources with heightened rigor, transforming internet searches into primary heuristic tools for corroborating or challenging health interventions.

Validity

The validity of the Information Source Importance (Internet Sources) scale has been established through empirical research in advertising, consumer psychology, and health communication:

Construct and Factorial Validity

Construct validity was established by Kim and King (2009) via exploratory and confirmatory factor analyses, demonstrating that the items measuring internet source importance load onto a discrete, structurally distinct factor separate from offline informational sources (e.g., health professionals, mass media advertising, interpersonal friend/family networks). The standardized factor loadings for internet items typically range from .71 to .88, demonstrating substantial convergent sharing of variance among the measured indicators.

Convergent Validity

Convergent validity is evidenced by significant, theoretically aligned correlations with related behavioral and cognitive constructs. Empirical investigations demonstrate that higher ISI scores correlate positively with:

• Total External Search Effort: As measured by time spent searching, number of web pages viewed, and diverse query formulation (r values typically ranging between .35 and .52, p < .001).
• Health Involvement and Epistemic Need: Significant positive correlations with the Zaichkowsky Personal Involvement Inventory (PII) adapted for health contexts.
• Direct-to-Consumer Advertising Responsiveness: Consumers placing high importance on web sources show a significantly higher likelihood of clicking online display advertisements or retrieving targeted URLs shown during televised DTC pharmaceutical commercials.

Discriminant Validity

Discriminant validity has been rigorously demonstrated via average variance extracted (AVE) evaluations surpassing shared inter-construct correlations (Fornell-Larcker criterion). The ISI successfully discriminates between:

• Professional Interpersonal Sources: Factor correlations between internet source importance and physician/pharmacist source importance remain moderate to low (r = .18 to .29), establishing that digital search importance is psychometrically distinct from dependence on medical professionals.
• Passive Mass Media Sources: Divergence from television, print magazine, and radio advertising importance scales, showing that active pull-medium evaluation (the internet) measures a distinct cognitive construct compared to push-medium exposure.

Predictive and Criterion Validity

Predictive validity is substantiated by the scale’s capacity to forecast downstream consumer health behaviors. Longitudinal and cross-sectional studies indicate that high scores on the ISI reliably predict:

1. The formulation of specific, technical questions brought to subsequent physician consultations.
2. Active requests for specific branded medications or generic therapeutic substitutions.
3. Higher rates of therapeutic non-compliance when web sources identify disputed adverse side effects.

Reliability

The reliability of the Information Source Importance (Internet Sources) scale has been demonstrated across multiple experimental and survey-based implementations:

Internal Consistency Reliability

In the foundational validation study by Kim and King (2009), the internal consistency reliability of the multi-item internet source importance scale yielded a Cronbach’s alpha coefficient exceeding the standard psychometric threshold of .70:

• Prescription Drug (Rx) Sample: Cronbach’s α = .84, demonstrating high internal homogeneity among items measuring the importance of various internet information sources for high-involvement health conditions.
• Nonprescription Drug (OTC) Sample: Cronbach’s α = .82, confirming measurement invariance across diverse risk profiles and product categories.

Subsequent studies replicating the scale across broader consumer samples (e.g., general medical consumers, elderly populations navigating Medicare Part D, chronic disease cohorts) have reported alpha coefficients ranging reliably between α = .81 and α = .89, alongside composite reliability (CR) values exceeding .85.

Test-Retest Stability

Although the ISI measures an evaluative state that can fluctuate based on emerging acute health events, short-term test-retest reliability assessments (over 2- to 4-week intervals in stable, non-acute populations) demonstrate intraclass correlation coefficients (ICC) ranging between .74 and .81, indicating satisfactory temporal stability of general source-importance attitudes.

Factor Analysis

The latent structural integrity of the Information Source Importance (Internet Sources) scale has been investigated utilizing both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA):

Exploratory Factor Analysis (EFA)

During initial scale development, Kim and King (2009) subjected a broad battery of information source items (encompassing broadcast television, print magazines, point-of-purchase displays, physicians, pharmacists, friends/family, and multiple internet modalities) to principal components analysis (PCA) and principal axis factoring with varimax and promax rotations. The analyses consistently isolated a distinct Digital/Internet Information Sources factor with an eigenvalue exceeding 1.0 (Kaiser criterion), accounting for a substantial percentage of the total variance explained (typically > 18% of unique variance within the broader external search battery). Factor loadings for the digital indicators met the rigorous cutoff criterion (> .60), demonstrating negligible cross-loadings onto interpersonal or offline commercial media factors (< .25).

Confirmatory Factor Analysis (CFA)

In structural equation modeling (SEM) frameworks validating the measurement model, CFA specifications confirm that the internet source items load robustly onto a single higher-order latent construct or a two-tiered model (distinguishing commercial pharmaceutical websites from independent health portals). Fit indices reported across empirical implementations consistently indicate good model-to-data fit:

• Chi-Square to Degrees of Freedom Ratio (χ²/df): Ranging between 1.45 and 2.30, well within the recommended threshold of < 3.0.
• Comparative Fit Index (CFI): Values routinely exceed .95 (ranging from .96 to .98), denoting excellent fit relative to a baseline null model.
• Tucker-Lewis Index (TLI): Coefficients consistently exceeding .94.
• Root Mean Square Error of Approximation (RMSEA): Estimates ranging from .038 to .055, with 90% confidence intervals remaining below the stringent .08 cutoff point.
• Standardized Root Mean Square Residual (SRMR): Values remaining below .045.

Instrument / Measurement Tool

The Information Source Importance (Internet Sources) instrument is structured as an efficient, self-administered survey scale. Below is the technical specification of the instrument architecture:

• Instrument Designation: Information Source Importance (Internet Sources) (ISI)
• Target Population: Adult consumers, patients, and health information seekers (adaptable to general consumer product domains).
• Administration Modality: Self-administered paper-and-pencil, computer-assisted web interviewing (CAWI), or mobile digital surveys.
• Administration Duration: Approximately 2 to 4 minutes.
• Response Architecture: 5-point or 7-point Likert-type response formats (commonly anchored from 1 = “Not at all important” to 5 or 7 = “Extremely important”). Alternative anchoring formats use semantic differential pairs (e.g., “Unimportant / Important”, “Worthless / Valuable”, “Irrelevant / Relevant”).
• Structural Composition: Typically operationalized as a 3- to 6-item battery evaluating distinct digital touchpoints (e.g., search engines, pharmaceutical company websites, independent health information sites, interactive patient forums).
• Scoring Protocol:
  • Composite Score Calculation: Items are summed or averaged to yield an overall Internet Source Importance index, with higher aggregate values reflecting greater reliance on web-based channels.
  • Dimensional Profiling: Alternatively, items can be separated into sub-indices: Commercial Web Importance (brand/manufacturer sites) versus Independent/Peer Web Importance (unaffiliated educational portals, social health communities).

Permissions & Fee and Test Year

The Information Source Importance (Internet Sources) scale was published in its empirical form in 2009 within the Journal of Advertising (Volume 38, Issue 1, pp. 5–19):

• Publication Year: 2009
• Copyright Holder: The American Academy of Advertising; articles published in the Journal of Advertising are distributed under the administrative auspices of Taylor & Francis Group, LLC.
• Academic and Non-Commercial Research Licensing: Under standard fair-use academic conventions, researchers may utilize the measurement scale, item phrasing, and conceptual dimensions for non-commercial scholarly research, doctoral dissertations, and institutional investigations, provided complete bibliographic attribution is accorded to Kim and King (2009).
• Commercial and Proprietary Licensing: Commercial entities, market research firms, and pharmaceutical corporations conducting proprietary consumer research must secure appropriate permissions or licensing clearances through the Copyright Clearance Center (CCC) or Taylor & Francis permissions portal.

References

• Beatty, S. E., & Smith, S. M. (1987). External search effort: An investigation across several product categories. Journal of Consumer Research, 14(1), 83–95. https://doi.org/10.1086/209095
• 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
• Kim, W. J., & King, K. W. (2009). Product category effects on external search for prescription and nonprescription drugs. Journal of Advertising, 38(1), 5–19. https://doi.org/10.2753/JOA0091-3367380101
• King, K. W., Reid, L. N., & Morrison, M. A. (2004). The direct-to-consumer prescription drug advertising research landscape: An agenda for the future. Health Communication, 16(1), 105–128. https://doi.org/10.1207/S15327027HC1601_7
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2
• Rosenstock, I. M. (1974). Historical origins of the Health Belief Model. Health Education Monographs, 2(4), 328–335. https://doi.org/10.1177/109019817400200403
• Schmidt, J. B., & Spreng, R. A. (1996). A proposed model of external consumer information search. Journal of the Academy of Marketing Science, 24(3), 246–256. https://doi.org/10.1177/0092070396243005
• Stigler, G. J. (1961). The economics of information. Journal of Political Economy, 69(3), 213–225. https://doi.org/10.1086/258464

Items of the Scale