Website Interactivity (Control) (WI)

Abstract

The Website Interactivity (Control) (WI) scale represents a foundational psychometric instrument developed by Yuping Liu (2003) to evaluate consumers’ subjective perception of behavioral agency and navigational freedom within computer-mediated environments. Originating as the primary subscale of Liu’s seminal tripartite model of perceived website interactivity—which also comprises Two-Way Communication and Synchronicity—the Active Control dimension isolates the degree to which users perceive themselves to be the instrumental drivers of their online navigation, exposure, and transactional experience. Comprising three rigorously validated items measured on a 7-point Likert scale ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”), the instrument quantifies the psychological state of perceived active control rather than objective, structural system features. Extensive psychometric testing demonstrates high internal consistency reliability (Cronbach’s alpha consistently exceeding .85 across cross-validation samples), robust convergent validity with generalized attitudes toward the website, discriminant validity distinct from communicative feedback mechanisms, and distinct factorial integrity established through exploratory and confirmatory factor analyses. The scale remains a principal benchmark across human-computer interaction (HCI), digital marketing, e-commerce, and communication studies for investigating user engagement, cognitive absorption, experiential flow, and consumer decision-making pathways in interactive digital architectures.

Keywords

Website Interactivity, Active Control, Perceived Interactivity, Human-Computer Interaction, Digital Marketing, User Experience, Navigational Freedom, Flow Theory, Online Consumer Behavior, Psychometrics

Authors

The scale was developed and psychometrically validated by Dr. Yuping Liu (now Dr. Yuping Liu-Thompkins).

• Current Affiliation: Professor of Marketing and Director of the CRM and Loyalty Laboratory, Strome College of Business, Old Dominion University, Norfolk, Virginia, United States.
• Academic Background: Ph.D. in Marketing from Rutgers, The State University of New Jersey.
• Expertise: Consumer engagement, digital marketing strategy, customer loyalty programs, and consumer-technology interactions.
• Contact/Institutional Profile: Strome College of Business, Old Dominion University (E-mail: [email protected]).

Purpose

The fundamental purpose of the Website Interactivity (Control) scale is to isolate and measure the user’s perception of volitional mastery, autonomous navigation, and dynamic choice during interactions with a digital interface. Throughout the late 1990s and early 2000s, academic inquiry into digital advertising and information systems struggled with inconsistent operational definitions of “interactivity.” Prior formulations frequently conflated objective media characteristics (e.g., hyperlinking structures, search engine features, download speeds) with the actual phenomenological experience of the user. Dr. Yuping Liu formulated this psychometric instrument to resolve this conceptual ambiguity by treating interactivity as a multi-dimensional, perceived construct grounded in human perception rather than raw technology.

Specifically, the Active Control subscale was engineered to address several distinct theoretical, empirical, and applied objectives:

• Isolating User Agency: To differentiate user-driven agency from system-driven responses, providing researchers with a clean measurement of perceived instrumental autonomy.
• Decoupling Perceived from Objective Interactivity: To evaluate how subjective feelings of navigational mastery mediate the relationship between objective technological features (such as site architecture or customization engines) and downstream psychological outcomes.
• Empirical Rigor in Commercial and Applied Research: In e-commerce and digital interface optimization, the scale serves as an evaluative diagnostic tool to determine whether design iterations empower users or overwhelm them with choice paralysis and cognitive disorientation.
• Predicting Online Affect and Behavior: The scale enables researchers to model how high degrees of perceived active control cultivate positive attitudes toward websites ($A_{st}$), foster brand affinity ($A_b$), elevate cognitive immersion, and enhance transactional purchase intentions.

Psychological Construct

The psychological construct assessed by this scale is Perceived Active Control. Within Liu’s (2003) structural conceptualization, website interactivity is formally defined as “the degree to which two or more communication parties can act on each other, on the communication medium, and on the messages and the degree to which such influences are synchronized.” Within this tripartite conceptualization, Active Control represents the consumer’s ability to voluntarily direct the interactive flow, manipulate their exposure to information, and determine the overall trajectory of their sensory and cognitive experience.

Perceived active control is characterized by three core psychological dimensions:

• Instrumental Autonomy: The subjective belief that the user operates as an autonomous agent who freely selects content based on personal intrinsic goals rather than being steered by rigid, paternalistic, or coercive algorithms. This is captured by assessing the perceived freedom to choose what content to view.
• Navigational Mastery: The feeling of exerting command over the functional capabilities of the digital system. Users experiencing high active control do not feel constrained by interface friction, navigational bottlenecks, or structural unpredictability; rather, they feel capable of dictating their movement through the digital space.
• Experiential Contingency (Contingent Agency): The cognitive recognition that digital consequences are directly contingent upon user behaviors. When a user perceives that their personal actions determine their outcomes and experiential states, they experience higher self-efficacy, reduced anxiety, and elevated engagement.

Crucially, perceived control operates as an internal psychological representation rather than an invariant property of software code. Two users navigating an identical website may exhibit disparate levels of perceived active control depending on their individual digital literacy, cognitive processing styles, familiarity with design conventions, and situational task orientation (e.g., goal-directed search versus hedonic browsing).

Theoretical Framework

The Website Interactivity (Control) scale is theoretically anchored in several converging paradigms within communication theory, social cognitive psychology, and human-computer interaction:

1. Steuer’s Telepresence and Synthetic Realities

Jonathan Steuer’s (1992) seminal conceptualization of virtual environments positioned interactivity alongside vividness as the two pillars of mediated presence. Steuer defined interactivity as the extent to which users can participate in modifying the form and content of a mediated environment in real time. Liu drew directly upon this paradigm, positing that control over content modification and experiential progression is foundational to overcoming the artificial nature of computer mediation.

2. Hoffman and Novak’s Hypermedia Flow Model

In their groundbreaking application of Mihaly Csikszentmihalyi’s Flow theory to computer-mediated environments, Donna L. Hoffman and Thomas P. Novak (1996) theorized that optimal consumer experiences online require a precise balance between user skills and environmental challenges, mediated by perceived control. According to their model, perceived control over the hypermedia interaction acts as an indispensable antecedent to the flow state—characterized by seamless cognitive absorption, loss of self-consciousness, and intrinsic enjoyment.

3. Self-Determination Theory (SDT) and Agency

Under Self-Determination Theory, developed by Edward Deci and Richard Ryan, autonomy is identified as a universal, basic psychological need. When an individual feels volition and agency, their intrinsic motivation and emotional satisfaction increase. Liu’s active control dimension reflects this need for autonomy in a technological context: users who encounter digital structures that respect and facilitate personal agency develop higher satisfaction, lower cognitive reactance, and greater overall psychological investment.

Validity

The psychometric validity of the Website Interactivity (Control) scale was established by Liu (2003) through a rigorous multi-stage validation methodology involving diverse online environments, task conditions, and empirical samples.

Construct and Content Validity

Content validity was established through an extensive literature review covering computer science, communication theory, advertising, and marketing, followed by an initial pool generation of 32 items. Expert panels evaluated the items for clarity, face validity, and construct representation. The purification process culled ambiguous and redundant items, isolating the three core items that distinctly reflect the active control construct.

Convergent Validity

Convergent validity was evidenced by uniform, high factor loadings ($> .75$) of all three items onto their designated latent construct in both exploratory and confirmatory factor analytic models. Additionally, the average variance extracted (AVE) for the Active Control factor exceeded .60 across empirical validation tests, well surpassing the standard .50 threshold established by Fornell and Larcker (1981).

Discriminant Validity

Liu (2003) demonstrated clear discriminant validity between Active Control and the other two dimensions of perceived interactivity: Two-Way Communication and Synchronicity. Confirmatory factor analyses demonstrated that a three-factor correlated model fit the data substantially better than a single-factor unified interactivity model ($\Delta \chi^2$ tests were statistically significant at $p < .001$). Furthermore, the shared variance between Active Control and the other dimensions was consistently lower than the AVE of Active Control, demonstrating that control represents an independent psychological facet rather than a generalized halo effect of site quality.

Nomological and Predictive Validity

Nomological validity was verified by evaluating the scale’s correlations with theoretically related downstream constructs. Active Control displayed statistically significant positive correlations with:

• Attitude toward the website ($A_{st}$) ($r = .48$ to $.56$, $p < .01$)
• Overall site involvement and user engagement ($r = .42$ to $.51$, $p < .01$)
• Return intentions and behavioral loyalty ($r = .38$ to $.47$, $p < .01$)

Reliability

The internal consistency reliability of the Website Interactivity (Control) subscale has been confirmed across diverse empirical contexts:

• Original Scale Development (Liu, 2003): During initial scale construction across multiple distinct website evaluations (ranging from commercial portals to informational archives), the Active Control dimension yielded a Cronbach’s alpha coefficient ($lpha$) of .86 in the exploratory sample and .88 in the confirmatory validation sample.
• Composite Reliability (CR): In structural equation modeling assessments, the composite reliability of the 3-item measure exceeded .87, reflecting high latent construct consistency and minimal measurement error variance.
• Independent Replications: Subsequent academic studies utilizing Liu’s measure in varied contexts (e.g., mobile application design, interactive social media interfaces, e-tailing platforms) have reported Cronbach’s alpha values typically ranging between .83 and .91, confirming the scale’s stability across technological generations.
• Test-Retest Stability: In experimental designs employing repeated-measures evaluations of stable interfaces, the instrument has shown strong stability coefficients ($r_{tt} > .80$), confirming its reliability when environmental interface variables are held constant.

Factor Analysis

Liu (2003) employed a two-step factor analytic strategy consisting of Exploratory Factor Analysis (EFA) followed by Confirmatory Factor Analysis (CFA) to evaluate the structural properties of the overall 9-item interactivity scale and its 3-item Active Control component.

Exploratory Factor Analysis (EFA)

An initial pool of purified items was subjected to principal axis factoring with promax (oblique) rotation to allow for correlated latent dimensions. The analysis extracted three distinct factors with eigenvalues greater than 1.0, accounting for over 68% of the total variance. The three Active Control items loaded heavily onto their designated factor, with factor loadings ranging from .79 to .87, and minimal cross-loadings ($< .20$) onto the Two-Way Communication or Synchronicity factors.

Confirmatory Factor Analysis (CFA)

To confirm the tripartite factorial structure, CFA was conducted using maximum likelihood estimation. The three-factor model yielded excellent fit indices that satisfied rigorous structural equation modeling criteria:

• Chi-Square / Degrees of Freedom Ratio ($\chi^2 / df$): $< 2.5$ ($p > .05$ in small sample tests; acceptable normed $\chi^2$ in large samples)
• Comparative Fit Index (CFI): $.96$ to $.98$
• Tucker-Lewis Index (TLI / NNFI): $.95$ to $.97$
• Root Mean Square Error of Approximation (RMSEA): $.045$ to $.058$ ($90% \text{ CI } [0.032, 0.071]$)
• Standardized Root Mean Square Residual (SRMR): $.038$

Standardized factor loadings ($lambda$) for the three Active Control items in the measurement model were:

• Item 1 (Free choice): $lambda = 0.81$
• Item 2 (A lot of control): $lambda = 0.86$
• Item 3 (Actions decided experiences): $lambda = 0.79$

All path coefficients from the latent construct to the observed indicators were statistically significant at $p < .001$, confirming robust indicator reliability.

Instrument / Measurement Tool

• Instrument Name: Website Interactivity (Control) (WI) [also known as the Active Control subscale of Liu’s Perceived Website Interactivity Scale]
• Author: Yuping Liu, Ph.D.
• Publication Year: 2003
• Construct Measured: Perceived Active Control (the degree of user agency, navigational command, and experiential contingency experienced during website interaction)
• Scale Format: Self-administered paper-and-pencil or computerized self-report questionnaire
• Item Count: 3 items
• Response Scale: 7-point Likert scale (1 = Strongly Disagree to 7 = Strongly Agree)
• Scoring Protocol: The Active Control dimension score is calculated by computing the unweighted arithmetic mean of the 3 items. Possible scores range from 1.00 to 7.00, with higher scores reflecting greater perceived active control.
• Reverse-Scored Items: None. All items are positively framed and scored in an identical direction.
• Administration Time: Approximately 1 minute (as a standalone subscale) or 3–4 minutes (when administered within the full 9-item tripartite interactivity battery).

Permissions & Fee and Test Year

• Test Year: 2003 (published in the June 2003 issue of the Journal of Advertising Research).
• Copyright Holder: World Advertising Research Center (WARC) / Cambridge University Press and the author (Yuping Liu).
• Usage Permissions: The scale items are published in full within the academic public domain for scientific research and non-commercial educational purposes. Academic researchers may utilize, adapt, and administer the instrument without paying royalty fees, provided full formal bibliographic citation is given to the original 2003 publication.
• Commercial Applications: Commercial organizations, proprietary market research agencies, or consulting entities seeking to integrate the scale into monetized assessment frameworks or diagnostic software systems should consult the journal publisher or the author regarding commercial terms.

References

• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated environments: Conceptual foundations. Journal of Marketing, 60(3), 50–68. https://doi.org/10.1177/002224299606000304
• 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
• McMillan, S. J., & Hwang, J. S. (2002). Measures of perceived interactivity: An exploration of the role of direction of communication, user control, and time. Journal of Advertising, 31(3), 29–42. https://doi.org/10.1080/00913367.2002.10673674
• Steuer, J. (1992). Defining virtual reality: Dimensions determining telepresence. Journal of Communication, 42(4), 73–93. https://doi.org/10.1111/j.1460-2466.1992.tb00812.x

Items of the Scale