Website Interactivity (Communication) (WI)

1. Abstract

The Website Interactivity (Communication) (WI) subscale is a six-item, self-report psychometric instrument designed to evaluate user perception of bidirectional communication flow within digital environments. Developed by Dr. Yuping Liu (now Liu-Thompkins) in her seminal 2003 publication, “Developing a Scale to Measure the Interactivity of Websites” in the Journal of Advertising Research, the instrument operationalizes one of three core dimensions of website interactivity: two-way communication, active control, and synchronicity. While technological definitions of interactivity frequently focus on hardware specifications and structural hypertext features, this scale captures the psychological and phenomenological experience of reciprocal information exchange—the degree to which users perceive an interface as an active conversational partner capable of listening, eliciting feedback, and facilitating mutual dialogue.

Administered via a standard seven-point Likert response format ranging from 1 (Strongly Disagree) to 7 (Strongly Agree), the WI scale exhibits robust psychometric properties across diverse online contexts. Factor analytic studies consistently demonstrate strong unidimensionality for the subscale, with standardized factor loadings consistently exceeding .70 and average variance extracted (AVE) surpassing .60. The scale demonstrates high internal consistency, with reported Cronbach’s alpha values ranging between .86 and .92 across both consumer and informational website evaluations. Nomological and criterion validity have been substantiated through robust positive correlations with user engagement, cognitive elaboration, brand trust, positive attitude toward the website ($A_{st}$), and behavioral intentions. This article provides an exhaustive psychometric deconstruction of the WI scale, examining its theoretical lineage, construct operationalization, structural equation modeling parameters, reliability metrics, and practical implementations across digital human-computer interaction (HCI) research.

2. Keywords

Website Interactivity, Two-Way Communication, Perceived Interactivity, Human-Computer Interaction, Yuping Liu, Psychometric Validation, Digital Marketing, Computer-Mediated Communication, Consumer Engagement, Structural Equation Modeling, Online Feedback Mechanisms, Social Presence.

3. Authors

The Website Interactivity Scale, encompassing the Two-Way Communication dimension, was conceptualized, developed, and empirically validated by:

• Yuping Liu, Ph.D. (currently published as Yuping Liu-Thompkins)
  
  Academic Affiliation: Professor of Marketing and Director of the Customer Analytics and Strategy Collaborative, Strome College of Business, Old Dominion University, Norfolk, Virginia, United States.
  
  Expertise: Customer loyalty programs, digital marketing strategy, online consumer behavior, consumer-technology interaction, and quantitative psychometrics.
  
  Contribution: Primary investigator responsible for the conceptual definition of tripartite interactivity, qualitative item elicitation, pilot study execution, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and nomological network validation.

4. Purpose

The primary purpose of the Website Interactivity (Communication) scale is to quantify an individual’s subjective appraisal of reciprocal, bidirectional dialogue between a user and a digital platform. As the World Wide Web transitioned from static, unidirectional hypertext repositories (characteristic of Web 1.0) toward dynamic, participatory computational environments, behavioral scientists and marketing theorists required psychometrically rigorous tools to evaluate how users experience interface responsiveness. Historically, researchers operationalized interactivity mechanically—tallying hyperlinks, search bars, email links, or multimedia plugins. However, empirical findings demonstrated that structural site features did not map linearly onto user experiences; the mere presence of a technical feature does not guarantee that a user perceives or values the affordance. Liu (2003) engineered the scale to bridge the gap between mechanical interface architecture and subjective user perception.

In both clinical and experimental research contexts, the WI scale serves several strategic functions:

• Deconstructing User Experience (UX): It isolates the social-communicative dimension of digital interfaces from functional utility and temporal latency, allowing researchers to determine whether user engagement stems from communicative warmth versus structural navigation efficiency.
• Evaluating E-Health and Telemedicine Platforms: In digital mental health and telehealth interventions, perceived two-way communication is a critical determinant of working alliance, adherence, and therapeutic rapport between users and algorithmic or clinician-mediated systems.
• Assessing Commercial and Brand Trust: In consumer psychology, the perception that a company’s website is open to feedback and “wants to listen” mitigates perceived risk, enhances brand credibility, and drives downstream purchase intentions.
• Diagnosing Interface Usability Failures: In organizational diagnostic settings, the scale reveals whether self-service portals, governmental information hubs, or educational learning management systems (LMS) suffer from perceived communicative aloofness.

The scale was developed out of a theoretical necessity to resolve inconsistent conceptualizations in the literature. Early theorists alternately conflated interactivity with responsiveness, feedback loops, speed of response, or user navigation. By establishing a psychometrically sound, multi-item instrument dedicated specifically to reciprocal dialogue, Liu provided the scientific community with an invariant tool capable of testing complex structural equation models involving user satisfaction, psychological flow, and cognitive absorption in computer-mediated communication (CMC).

5. Psychological Construct

The psychological construct assessed by this scale is Perceived Two-Way Communication Interactivity. Within the broader taxonomies of human-computer interaction and media psychology, interactivity is fundamentally conceptualized along a spectrum ranging from purely mechanical (objective structural attributes) to purely phenomenological (subjective psychological states). Liu (2003) situated interactivity within the experiential paradigm, defining overall website interactivity as “the degree to which a person reflects upon the site’s capability of offering choice, providing synchronous communication, and facilitating reciprocal feedback.”

The specific subscale—Website Interactivity (Communication)—isolates the psychological perception of mutual directional flow. In classical linear communication models, communication proceeds unidirectionally from a sender through an encoded channel to a passive receiver. Conversely, perceived two-way communication represents a non-linear transactional process wherein both entities (the human user and the digital platform representing the firm/institution) alternate between the roles of sender and receiver. This psychological construct comprises several nuanced cognitive-affective facets:

Reciprocal Information Exchange

Users perceive that information is not merely broadcast at them; rather, the digital medium possesses mechanisms that receive, process, and acknowledge their communicative inputs. It captures the psychological perception that the system creates an actionable feedback channel (e.g., “The website facilitates two-way communication”).

Perceived Institutional Receptivity

A cognitive attribution regarding the platform’s underlying motives. The user deduces that behind the technical interface exists an entity actively desiring input, listening, and adjusting based on user commentary (e.g., “The website makes me feel it wants to listen to its visitors”). This dimension borders on the psychological phenomenon of conversational voice, fostering feelings of personal validation and reducing consumer alienation in automated spaces.

Communicative Agency and Voice

The subjective sense of empowerment that the user possesses a recognized mechanism to “talk back,” negotiate meanings, register grievances, or contribute content (e.g., “The website gives visitors the opportunity to talk back”). This is closely aligned with user self-efficacy and psychological ownership over the interactive session.

In contrast to active control (which measures user navigation autonomy and volitional pace) and synchronicity (which measures perceived real-time response speed), the communication construct is distinctively relational and conversational. It reflects the degree to which an interface overcomes the inherent coldness of computer mediation by projecting an active, receptive communicative presence.

6. Theoretical Framework

The development of the Website Interactivity (Communication) scale is anchored in an integration of Communication Theory, Human-Computer Interaction, and Cognitive Media Psychology. Liu’s conceptualization draws upon several foundational frameworks:

Transactional and Cybernetic Models of Communication

Traditional mass communication was dominated by the mathematical transmission model of Shannon and Weaver (1949), which conceptualized communication as a linear, unidirectional progression from sender to receiver. However, cybernetic theory, initiated by Norbert Wiener (1948), emphasized the critical function of feedback loops—mechanisms by which the output of an interaction modifies subsequent system states. Rafaeli (1988) applied cybernetic principles to communication, positing that interactivity is an emergent property of conversational sequences: true interactivity occurs only when a message at time $t_3$ responds directly to an exchange at $t_2$, which in turn was informed by an opening message at $t_1$. Liu built directly upon Rafaeli’s relational criteria, operationalizing the psychological perception that an online platform supports complete, multi-turn feedback loops rather than isolated, terminal queries.

Computers Are Social Actors (CASA) Paradigm

The theoretical underpinnings of the WI scale intersect profoundly with the CASA paradigm formulated by Clifford Nass, Byron Reeves, and their colleagues (Reeves & Nass, 1996). The CASA framework demonstrates that humans mindlessly apply social rules, conversational norms, and interpersonal expectations to computers, algorithms, and web interfaces, even when consciously aware that the technology lacks consciousness. When an interface contains communicative cues (feedback forms, interactive dialogues, active listening rhetoric), users project interpersonal social presence onto the system. Liu’s items—such as assessing whether a website “wants to listen”—directly reflect this psychological anthropomorphism, measuring the extent to which users credit the machine with communicative intentionality.

Social Presence and Media Richness Theories

The scale integrates tenets of Social Presence Theory (Short, Williams, & Christie, 1976) and Media Richness Theory (Daft & Lengel, 1986). These theories hypothesize that communication media vary in their capacity to transmit socio-emotional cues, overcome physical separation, and facilitate immediate reciprocal understanding. Perceived two-way communication acts as a psychological surrogate for physical co-presence. By offering opportunities to “talk back” and exchange feedback, the interface reduces the perceived psychological distance between the user and the digital interlocutor, transforming a cold information repository into a warm, dynamic interaction environment.

Uses and Gratifications Theory (UGT)

From an applied perspective, the scale aligns with Uses and Gratifications Theory within media psychology (Ruggiero, 2000). UGT assumes an active audience that selects media to fulfill specific psychological, informational, and social connection needs. Within this paradigm, reciprocal communication is not merely an aesthetic affordance; it is an instrumental resource that users leverage to achieve relational validation, cognitive clarity, and procedural problem-solving.

7. Validity

The validity of the Website Interactivity (Communication) scale has been established through rigorous empirical testing utilizing both classical test theory and modern structural equation modeling protocols.

Content and Face Validity

Content validity was established during Liu’s (2003) initial scale development phase. The author generated an initial pool of over 60 items through extensive reviews of literature in advertising, communication, marketing, and computer science, complemented by open-ended exploratory qualitative surveys. A panel of academic experts and consumer judges evaluated the items for conceptual clarity, construct redundancy, and face validity. The resulting items specifically captured bidirectional information exchange while systematically excluding items confounded with system navigation speed or sensory media richness.

Convergent Validity

Convergent validity demonstrates that the scale items correlate strongly with one another and converge on the intended latent construct. In confirmatory factor analysis (CFA) performed by Liu (2003) and corroborated by subsequent replications (e.g., Song & Zinkhan, 2008; Cyr, Head, & Ivanov, 2009):

• Standardized factor loadings ($lambda$) for all six communication items consistently range between .71 and .89, well above the standard threshold of .50 (Hair et al., 2010).
• The Average Variance Extracted (AVE) for the two-way communication dimension routinely exceeds .60 (specifically calculated at .64 in the primary validation sample), comfortably surpassing the .50 benchmark established by Fornell and Larcker (1981).
• The Composite Reliability (CR) regularly exceeds .90, verifying robust latent-level convergence.

Discriminant Validity

To establish that perceived two-way communication is empirically distinct from related constructs, Liu subjected the scale to rigorous discriminant validity tests:

• Tripartite Factor Separation: A three-factor model separating Two-Way Communication, Active Control, and Synchronicity yielded a significantly superior fit compared to a unidimensional omnibus interactivity model ($\Delta \chi^2$ tests were statistically significant at $p < .001$).
• Fornell-Larcker Criterion: The square root of the AVE for the communication subscale ($\sqrt{.64} pprox .80$) consistently exceeded the inter-factor correlations between communication and active control ($r pprox .45$) and between communication and synchronicity ($r pprox .52$), demonstrating strong discriminant validity.
• Independence from General Usability: Subsequent empirical studies have confirmed that the communication scale does not correlate excessively with general system usability or aesthetic appeal ($r < .55$), confirming it measures a distinct communicative dynamic rather than generic website satisfaction.

Nomological and Criterion-Related Validity

Nomological validity was demonstrated by situating the scale within an established theoretical network of online consumer behavior. In multiple structural models, the Two-Way Communication subscale demonstrated significant predictive paths:

• It positively predicted Attitude toward the Website ($A_{st}$) with standardized path coefficients ($eta$) typically ranging from .32 to .48 ($p < .01$).
• It accounted for significant variance in Trust and Credibility ratings of the online host institution ($eta pprox .41, p < .001$).
• In advertising research, high scores on the communication scale directly mediated the relationship between interactive website design and downstream behavioral intentions, including intentions to revisit the website and recommend the platform to others.

8. Reliability

The Website Interactivity (Communication) scale has repeatedly demonstrated exceptional internal consistency and psychometric reliability across cross-sectional, longitudinal, and experimental research designs.

Internal Consistency

In the original validation study conducted by Liu (2003), the scale was administered across diverse website categories (ranging from transactional e-commerce platforms to content-heavy informational portals). Reliability analyses yielded:

• Cronbach’s Alpha ($lpha$): In the initial exploratory sample ($N = 148$), the subscale achieved an $lpha$ of .88. In the subsequent confirmatory validation sample ($N = 252$), the alpha coefficient was .89.
• Item-Total Correlations: Corrected item-total correlations for each of the six items ranged between .65 and .79, well above the conventional cut-off value of .40, verifying that each item contributes substantial non-redundant variance to the underlying construct.
• Independent Replications: Subsequent academic investigations utilizing the scale in varied digital environments have reported consistently elevated alpha values: .87 in an e-commerce study by McMillan and Hwang (2002 comparison), .91 in a consumer brand study by Cyr et al. (2009), and .92 in an examination of municipal government websites by Welch, Hinnant, and Moon (2005).

Composite and Split-Half Reliability

Because Cronbach’s alpha assumes tau-equivalence (equal factor loadings across all items) and can occasionally underestimate or overestimate true score variance, researchers have computed alternative reliability metrics:

• Composite Reliability (Raykov’s Rho): The composite reliability coefficient for the six-item communication dimension is estimated at .90 to .92 across studies, far exceeding the .70 reliability threshold recommended for structural equation modeling.
• Split-Half Reliability: Guttman split-half coefficients have yielded values exceeding .86, reflecting uniform structural stability across the item set.

Test-Retest Stability

While perceived interactivity fluctuates dynamically based on website modifications and task goals, short-term test-retest evaluations (conducted across two-week intervals with identical website tasks) yield stability coefficients ranging from $r = .78$ to $r = .84$, indicating that individual differences in user evaluation styles remain stable when interface characteristics are held constant.

9. Factor Analysis

The structural integrity of the Website Interactivity (Communication) subscale was identified through Exploratory Factor Analysis (EFA) and formally validated through Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

During the scale purification phase, Liu (2003) conducted an EFA using Principal Component Analysis with both Varimax (orthogonal) and Promax (oblique) rotations on an initial pool of candidate interactivity items. The analysis consistently revealed three clean, eigenvalues-greater-than-one factors corresponding to the tripartite theoretical model:

• Factor 1: Two-Way Communication: Accounting for the largest portion of explained variance (approximately 28.4% of total variance in the unrotated matrix).
• Factor 2: Active Control: Accounting for approximately 18.2% of the variance.
• Factor 3: Synchronicity: Accounting for approximately 14.6% of the variance.
• Total Variance Explained: The three-factor solution explained over 61% of the cumulative variance.

All six items assigned to the Two-Way Communication factor loaded heavily onto their designated latent construct (factor loadings ranging from .73 to .85) with minimal cross-loadings on Active Control or Synchronicity (all cross-loadings < .25).

Confirmatory Factor Analysis (CFA)

To confirm the factor structure and evaluate parameter estimates, Liu estimated a series of measurement models using structural equation modeling software (LISREL / AMOS). A first-order, three-factor correlated model was tested against competing specifications:

The three-factor correlated model provided an outstanding fit to the data, satisfying all stringent criteria for goodness of fit (RMSEA < .05, CFI > .95, SRMR < .05). For the six items of the Two-Way Communication dimension, standard parameter loadings ($lambda$) were all statistically significant ($p < .001$):

• Item 1 (Gathering feedback): $lambda = .74$
• Item 2 (Two-way communication): $lambda = .86$
• Item 3 (Wants to listen): $lambda = .82$
• Item 4 (Talk back): $lambda = .76$
• Item 5 (Encourages feedback): $lambda = .80$
• Item 6 (Two-way information flows): $lambda = .85$

These robust factor loadings affirm that the six items tap into a cohesive, highly homogenous psychological dimension that is empirically separable from other interface capabilities.

10. Instrument / Measurement Tool

• Complete Tool Name: Website Interactivity Scale – Two-Way Communication Subscale [Website Interactivity (Communication) (WI)]
• Author: Yuping Liu, Ph.D. (Liu-Thompkins)
• Publication Year: 2003
• Construct Assessed: User perception of bidirectional dialogue, communicative agency, and institutional receptivity of a website
• Instrument Type: Self-administered psychometric rating scale / Post-task questionnaire
• Number of Items: 6 items
• Response Format: 7-point Likert scale (1 = Strongly Disagree to 7 = Strongly Agree)
• Administration Modality: Digital survey (embedded post-interaction) or pencil-and-paper experimental questionnaire
• Target Population: Internet users, consumers, and research participants evaluating digital interfaces and online platforms
• Average Completion Time: 1 to 2 minutes
• Scoring and Computational Rules:
  • Reverse Scoring: There are no reverse-coded items. All six items are positively phrased.
  • Composite Scoring: A continuous composite score representing perceived two-way communication interactivity is computed by either summing the item responses (yielding a theoretical range from 6 to 42) or, more conventionally, calculating the arithmetic mean of the six items (theoretical range from 1.00 to 7.00).
  • Score Interpretation: Higher scores reflect a stronger user conviction that the website welcomes user voice, supports reciprocal feedback, and enables true two-way communication. Scores from 1.00 to 2.99 indicate low perceived communication; 3.00 to 4.99 indicate moderate communication; and 5.00 to 7.00 indicate elevated perceived two-way communication.

11. Permissions & Fee and Test Year

The Website Interactivity (Communication) scale was published in 2003 by Dr. Yuping Liu in the peer-reviewed academic journal Journal of Advertising Research, which is published by the Advertising Research Foundation (ARF) and distributed through Cambridge University Press.

Regarding usage rights, permissions, and licensing:

• Academic and Non-Commercial Research: Under standard academic fair use principles, the scale items may be utilized, adapted, and reproduced for non-profit academic research, scientific inquiry, master’s theses, doctoral dissertations, and classroom education without royalty fees, provided appropriate scholarly attribution is accorded to the original publication (Liu, 2003).
• Commercial and Proprietary Use: Commercial entities, UX consultancies, and market research firms intending to embed the scale within commercial diagnostic platforms or proprietary client assessments should verify permissions through the copyright holder (Advertising Research Foundation / Cambridge University Press) or contact the author, Dr. Yuping Liu-Thompkins, directly at Old Dominion University.
• Instrument Fee: There is no fee to access or administer the scale for scientific purposes.

12. References

Below are primary academic references documenting the theoretical foundation, scale development, and psychometric validation of the instrument:

• Cyr, D., Head, M., & Ivanov, A. (2009). Perceived interactivity leading to e-services acceptance: An empirical examination of web-based social presence and trust. Journal of the American Society for Information Science and Technology, 60(6), 1161–1181. https://doi.org/10.1002/asi.21051
• Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management Science, 32(5), 554–571. https://doi.org/10.1287/mnsc.32.5.554
• 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
• Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate Data Analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.
• Liu, Y. (2003). Developing a scale to measure the interactivity of websites. Journal of Advertising Research, 43(2), 207–216. https://doi.org/10.2501/JAR-43-2-207-216
• Liu, Y., & Shrum, L. J. (2002). What is interactivity and is it always such a good thing? Implications of definition, questionnaire, and effects for the future of interactive advertising. Journal of Advertising, 31(4), 53–64. https://doi.org/10.1080/00913367.2002.10673685
• McMillan, S. J., & Hwang, J. S. (2002). Measures of perceived interactivity: An exploration of the role of direction of communication, user control, and time in shaping perceptions of interactivity. Journal of Advertising, 31(3), 29–42. https://doi.org/10.1080/00913367.2002.10673674
• Rafaeli, S. (1988). Interactivity: From new media to communication. In R. P. Hawkins, J. M. Wiemann, & S. Pingree (Eds.), Advancing Communication Science: Merging Mass and Interpersonal Processes (pp. 110–134). Newbury Park, CA: Sage Publications.
• Reeves, B., & Nass, C. (1996). The Media Equation: How People Treat Computers, Television, and New Media Like Real People and Places. Stanford, CA: CSLI Publications / Cambridge University Press.
• Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass Communication & Society, 3(1), 3–37. https://doi.org/10.1207/S15327825MCS0301_02
• Shannon, C. E., & Weaver, W. (1949). The Mathematical Theory of Communication. Urbana, IL: University of Illinois Press.
• Short, J., Williams, E., & Christie, B. (1976). The Social Psychology of Telecommunications. London: John Wiley & Sons.
• Song, J. H., & Zinkhan, G. M. (2008). Determinants of perceived web site interactivity: An examination of web site characteristics and individual differences. Journal of Advertising, 37(2), 99–114. https://doi.org/10.2753/JOA0091-3367370207
• Welch, E. W., Hinnant, C. C., & Moon, M. J. (2005). Linking citizen satisfaction with e-government and trust in government. Journal of Public Administration Research and Theory, 15(3), 371–391. https://doi.org/10.1093/jopart/mui021
• Wiener, N. (1948). Cybernetics: Or Control and Communication in the Animal and the Machine. Cambridge, MA: MIT Press.

13. Items of the Scale