Anxiety (Technological) (ANXT1)

1. Abstract

The Technological Anxiety Scale (ANXT1), developed by Joel E. Collier and Daniel L. Sherrell in their seminal 2010 investigation into customer adoption of self-service technologies (SSTs), represents a psychometrically validated, parsimonious, context-specific self-report instrument designed to quantify situational apprehension, trepidation, and psychological resistance provoked by interaction with discrete technological systems. Comprising three unidimensional items, the scale explicitly addresses consumer hesitation rooted in the catastrophic fear of committing irreversible execution errors, generalized cognitive intimidation induced by machine interfaces, and self-doubt concerning task-specific operational self-efficacy. Unlike legacy psychometric batteries that evaluate broad, dispositional technophobia or generalized computer anxiety as stable personality traits, the ANXT1 features a modular referent structure—denoted by the placeholder bracket [technology]—enabling researchers and practitioners to isolate user responses to target hardware, software, automated retail kiosks, mobile digital platforms, and artificial intelligence interfaces.

The ANXT1 employs a standard 7-point Likert response continuum ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”), with higher aggregate scores indicating elevated technological anxiety. Psychometric evaluations across multiple empirical field studies demonstrate exceptional internal consistency reliability, yielding Cronbach’s alpha coefficients consistently exceeding .85 (specifically .88 in the primary validation sample) and composite reliability indices surpassing .87. Confirmatory factor analysis confirms a robust single-factor architecture characterized by high standardized factor loadings (> .80), strong average variance extracted (AVE > .70), and distinct discriminant validity against related consumer constructs such as perceived control, perceived convenience, performance risk, and service technology self-efficacy. The scale provides marketing researchers, human-computer interaction (HCI) engineers, and organizational psychologists with an efficient, minimally burdensome, and methodologically sound metric for modeling the psychological barriers impeding digital adoption.

2. Keywords

technology anxiety, computer anxiety, self-service technology, human-computer interaction, perceived control, digital divide, technophobia, consumer behavior, psychometrics, scale validation, interface intimidation, technology adoption

3. Authors

The Technological Anxiety Scale (ANXT1) was conceived, operationalized, and psychometrically validated by researchers specializing in marketing, retail technology, and consumer decision-making:

• Joel E. Collier, Ph.D.: Tommy and Victoria Baker Endowed Chair in Business, Department of Marketing, Quantitative Analysis, and Business Law, College of Business, Mississippi State University, Mississippi State, MS, USA. Dr. Collier’s research centers on service marketing, customer satisfaction, self-service technology interfaces, structural equation modeling methodology, and consumer-technology interactions.
• Daniel L. Sherrell, Ph.D.: Professor Emeritus of Marketing, Department of Marketing and Supply Chain Management, Fogelman College of Business and Economics, University of Memphis, Memphis, TN, USA. Dr. Sherrell’s scholarship spans retail strategy, services marketing, consumer processing of electronic service environments, and structural measurement of behavioral intention.

The primary validation of the ANXT1 scale was published in the Journal of the Academy of Marketing Science in 2010. Institutional inquiries regarding the conceptualization of self-service control models and applied survey deployment may be directed to the corresponding author, Dr. Joel E. Collier, through the academic portal of Mississippi State University.

4. Purpose

The primary objective underlying the development of the ANXT1 was to resolve a persistent methodological and conceptual mismatch in human-computer interaction and consumer research: the reliance on broad, generalized technophobia inventories to predict interactions with discrete, localized technology interfaces. Historically, instruments designed to gauge technological discomfort—such as the Computer Anxiety Rating Scale (CARS) or the Computer Anxiety Index (CAIN)—measured an enduring, trait-like psychological aversion toward computing in the abstract. However, as digital self-service technologies (such as automated airline ticketing kiosks, self-checkout terminals, automated teller machines, digital health kiosks, and web-based transaction platforms) proliferated throughout commercial environments, researchers observed that consumer reluctance could not be adequately explained by broad computer phobia alone. Consumers who demonstrated high comfort with smartphones or personal computers frequently exhibited acute avoidance, performance stress, and cognitive freeze when confronted with unfamiliar, public self-service kiosks.

To overcome this theoretical and empirical limitation, Collier and Sherrell (2010) formulated the ANXT1 as an intentionally tailored, context-specific measure. By isolating the cognitive appraisals of intimidation, error vulnerability, and operational incompetence triggered by a specific designated technology, the instrument enables investigators to assess situational anxiety with pinpoint accuracy. The scale explicitly investigates why a user hesitates to engage with a novel interface, focusing on three core behavioral inhibitors:

• Anticipatory Execution Fear: The prospective psychological terror that one will commit an operational error that cannot be undone, rectified, or canceled, thereby trapping the user in a high-stakes failure state or creating financial, logistical, or social embarrassment.
• Interface Intimidation: The subjective assessment that the complexity, form factor, or input requirements of the machine interface are psychologically overpowering, foreign, or unapproachable.
• Task-Specific Self-Doubt: A localized deficit in perceived operational competence, wherein the user doubts their immediate capacity to execute the sequence of motor and cognitive steps required to complete a digital transaction.

In applied organizational and clinical settings, the ANXT1 serves as a diagnostic audit tool. Organizations investing capital into digital transformation and automated customer contact points can deploy the ANXT1 during pilot usability testing to identify interfaces that generate prohibitive cognitive friction. In ergonomic engineering, human-factor specialists utilize the scale to evaluate user responses across redesign iterations. Furthermore, in clinical and gerontological contexts, the instrument aids behavioral scientists in quantifying situational anxiety among aging cohorts or marginalized populations who experience disproportionate cognitive strain during societal transitions toward digital-only public services.

5. Psychological Construct

Technological anxiety, as operationalized within the ANXT1 instrument, represents a state-level, affective-cognitive reaction characterized by tension, apprehension, and intrusive worries concerning one’s current or prospective interaction with a specific technological device. Rather than functioning as an immutable personality dimension, this construct operates as a dynamic, situationally contingent psychological state governed by the perceived balance between user competence and situational demands.

Within contemporary psychometric theory, anxiety encompasses both physiological arousal and cognitive appraisal. In the context of technological encounters, the cognitive appraisal component dominates. Collier and Sherrell (2010) structured the ANXT1 to reflect the multidimensional cognitive facets of situational technological distress within a lean, unified measurement model. The three primary facets captured by the scale items include:

Fear of Irreversible Errors (Anticipatory Catastrophizing)

The first dimension, encapsulated by Item 1 (“I hesitate to use [technology] for fear of making mistakes I cannot correct”), addresses cognitive catastrophizing regarding operational failure. In computerized environments, non-expert users frequently hold an inaccurate mental model of software and system architecture, assuming that missteps will trigger irreversible penalties—such as financial loss, account lockouts, corrupted data, or public service failure. This fear prompts behavioral hesitation, latency, or complete task abandonment. In self-service environments, this psychological barrier is amplified by social presence: committing an uncorrectable mistake in front of waiting bystanders generates intense public self-consciousness and social anxiety.

Interface Intimidation (Affective Apprehension)

The second dimension, tapped by Item 2 (“[Technology] is somewhat intimidating to me”), isolates the primary affective response evoked by the external characteristics of the system. Intimidation occurs when an interface presents perceptual cues that signal excessive complexity, unfriendliness, or technological elitism. Factors such as screen clutter, obscure menu taxonomies, aggressive visual feedback, or physical ergonomics can induce a sense of alienation. Intimidation functions as an affective gatekeeper; when a technological interface appears intimidating, the consumer instinctively engages an avoidance defense mechanism rather than an exploratory behavioral approach.

Operational Insecurity and Self-Efficacy Deficit

The third dimension, captured by Item 3 (“I feel insecure about my ability to use [technology]”), maps directly onto the perceived competence and self-efficacy appraisals of the user. Rooted in social cognitive theory, self-efficacy reflects an individual’s belief in their capability to execute courses of action required to manage prospective situations. Insecurity represents the inverse affective manifestation of low self-efficacy. When an individual lacks confidence in their digital procedural fluency, they anticipate personal incompetence, triggering acute performance anxiety. This insecurity diminishes the user’s perceived behavioral control, lowering motivation and reinforcing avoidant coping strategies.

6. Theoretical Framework

The conceptual foundation of the ANXT1 scale is anchored within several prominent psychological and behavioral paradigms, most notably Lazarus and Folkman’s (1984) Transactional Model of Stress and Coping, Bandura’s (1977, 1986) Social Cognitive Theory, and the Technology Acceptance Model (TAM) developed by Davis (1989), alongside its extended variants incorporating perceived control.

The Transactional Model of Stress and Coping

According to Lazarus and Folkman (1984), psychological stress emerges from a dual-stage cognitive appraisal process: primary appraisal, during which an individual determines whether an environmental encounter is benign, positive, or threatening; and secondary appraisal, during which the individual evaluates their available coping resources to mitigate that threat. The ANXT1 directly operationalizes this appraisal mechanism within a human-computer interaction framework. Confronted with a novel technological artifact (such as an automated checkout counter), the consumer evaluates the system as potentially threatening (primary appraisal of intimidation and error severity). Concurrently, the consumer assesses their procedural skills as inadequate (secondary appraisal of personal insecurity). The cognitive outcome of this transactional imbalance is technology anxiety, which triggers behavioral avoidance as an emotion-focused coping strategy.

Self-Efficacy and Perceived Behavioral Control

Bandura’s (1977) social cognitive construct of self-efficacy posits that individuals actively avoid tasks and environments that exceed their perceived coping capabilities. In technological domains, computer self-efficacy (Compeau & Higgins, 1995) dictates whether a user approaches or withdraws from a system. When situational anxiety is high, it operates as a physiological and affective signal that degrades efficacy beliefs. Collier and Sherrell (2010) integrated this dynamic into a broader structural model of perceived control, demonstrating that technological anxiety directly undermines a consumer’s sense of procedural and outcome control over a service encounter. When users perceive that an error cannot be reversed (low outcome control) and that their ability is compromised (low behavioral control), anxiety peaks, suppressing transaction convenience and perceived service quality.

Integration with Technology Acceptance Frameworks

Within classic technology adoption literature, technological anxiety is frequently conceptualized as an external antecedent that exerts an indirect negative effect on system usage intentions via perceived ease of use and perceived usefulness (Venkatesh, 2000). Collier and Sherrell advanced this paradigm by demonstrating that in self-service retail settings, situational technology anxiety interacts directly with process convenience and speed. Users experiencing elevated anxiety evaluate the time spent at a terminal not as convenient or empowering, but as mentally taxing and risky. Thus, the ANXT1 serves as a critical moderating and mediating variable in structural models explaining why technically functional systems experience user rejection.

7. Validity

The psychometric validity of the ANXT1 scale has been empirically verified through rigorous methodological procedures, including construct validity, convergent validity, discriminant validity, and predictive/nomological validity across diverse empirical samples.

Construct and Convergent Validity

Construct validity was established by Collier and Sherrell (2010) utilizing a large-scale consumer sample assessing self-service technology usage across retail and ticketing environments. Convergent validity—the extent to which the three indicators share a high proportion of common variance—was evaluated via structural equation modeling and confirmatory factor analysis (CFA). All three standardized factor loadings on the technological anxiety latent construct were exceptionally high, exceeding .80 (ranging from .82 to .89, p < .001). Furthermore, the Average Variance Extracted (AVE) for the ANXT1 construct exceeded .70, comfortably surpassing the standard .50 threshold established by Fornell and Larcker (1981). This confirms that the variance explained by the underlying construct is substantially greater than the variance attributable to measurement error.

Discriminant Validity

To establish that the ANXT1 captures a unique psychological phenomenon distinct from related constructs, Collier and Sherrell (2010) subjected the scale to rigorous discriminant validity tests against theoretical neighbors, including:

• Perceived Control: The user’s perceived mastery and autonomy over the transaction environment.
• Perceived Convenience: Evaluations of time savings and physical ease associated with self-service delivery.
• Technology Readiness / Self-Efficacy: General optimism and perceived functional capability.
• Performance Risk: The perceived probability that the service will fail to deliver the expected objective outcome.

Applying the Fornell-Larcker criterion, the square root of the AVE for the ANXT1 construct was found to be markedly higher than its highest inter-construct correlation with any other variable in the structural model. Furthermore, nested chi-square difference tests comparing an unconstrained model to a model constraining the correlation between anxiety and related constructs to unity demonstrated statistically significant increases in chi-square (Δχ², p < .001), confirming distinct construct boundaries.

Predictive and Nomological Validity

The nomological validity of the scale was corroborated by testing its hypothesized paths within structural equation models. In Collier and Sherrell (2010), technological anxiety exhibited a statistically significant negative relationship with perceived control (β = -.34, p < .001) and exerted a significant negative indirect effect on user satisfaction and repeat usage intention. Subsequent empirical investigations applying the ANXT1 across various customer-facing technologies—such as mobile payment systems, artificial intelligence kiosks, and telemedicine portals—have replicated these findings, verifying that high ANXT1 scores systematically predict longer transaction times, increased operational errors, higher perceived cognitive workload, and a marked preference for human service alternatives.

8. Reliability

The reliability of the ANXT1 scale has been repeatedly validated, establishing high internal consistency, stability, and minimal standard error of measurement across varied experimental and field settings.

Internal Consistency Reliability

In the original validation study conducted by Collier and Sherrell (2010), the technological anxiety scale achieved an internal consistency reliability estimate of:

• Cronbach’s Alpha (α): .88
• Composite Reliability (CR): .89

These values comfortably exceed the conventional benchmark of .70 recommended by Nunnally and Bernstein (1994) for psychometric research and pass the stringent .80 standard suggested for basic research instruments. Subsequent independent studies employing the ANXT1 have documented consistently robust internal consistency scores, with Cronbach’s alpha values typically ranging between .85 and .92 across diverse digital domains (e.g., self-checkout counters, automated hotel check-in systems, and mobile banking applications).

Item-Total Correlations and Split-Half Reliability

Corrected item-total correlations across validation samples exceed .70 for all three items, indicating that each statement contributes substantially to the measurement of the underlying latent construct without redundancy. Inter-item correlations consistently fall within the ideal range of .65 to .78, demonstrating high coherence without indicating hyper-collinearity or tautological wording. Split-half reliability coefficients, where calculated, exceed .85, confirming internal stability.

Test-Retest Stability

Although the ANXT1 is engineered to measure situational anxiety toward a specific target technology, short-interval test-retest assessments (e.g., two-week intervals without intervention or technological exposure) yield stability coefficients ranging from r = .78 to .84. This demonstrates that while the measure responds dynamically to structural redesigns or training interventions, baseline evaluations of unfamiliar systems remain stable across short temporal windows.

9. Factor Analysis

The structural dimensionality of the ANXT1 was comprehensively explored and confirmed through exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) during its initial scale construction and validation phases.

Exploratory Factor Analysis (EFA)

During preliminary scale development, items were subjected to exploratory factor extraction using maximum likelihood estimation with oblimin and varimax rotations alongside concurrent items measuring control, convenience, and transaction risk. Across extraction protocols, the three ANXT1 items loaded cleanly onto a single, dominant factor exhibiting an eigenvalue substantially greater than 1.0 (typical initial eigenvalues > 2.2). The single factor accounted for over 75% of the total variance among the items. All factor loadings were uniformly high (above .80), and no cross-loadings onto secondary constructs exceeded the conventional threshold of .25.

Confirmatory Factor Analysis (CFA)

Collier and Sherrell (2010) validated the unidimensional measurement model using CFA within a covariance-based structural equation modeling framework (AMOS/LISREL). The measurement model fit was evaluated using multiple fit criteria:

• Comparative Fit Index (CFI): .98 – .99
• Tucker-Lewis Index (TLI): .97 – .98
• Standardized Root Mean Square Residual (SRMR): < .03
• Root Mean Square Error of Approximation (RMSEA): < .05 (with narrow 90% confidence intervals)

Standardized factor loadings (λ) for the three individual indicators were observed as follows:

• Item 1 (Hesitation / Fear of mistakes): λ ≈ .82 – .85
• Item 2 (Intimidating): λ ≈ .86 – .89
• Item 3 (Insecurity regarding ability): λ ≈ .84 – .88

All standardized loadings were statistically significant at p < .001, confirming that each item reflects the latent technological anxiety construct with minimal measurement residual. The unconstrained single-factor specification demonstrated superior parsimony and goodness-of-fit over competing multidimensional structures.

10. Instrument / Measurement Tool

The ANXT1 is structured as a brief, modular, self-administered survey tool. Its formal administrative characteristics are outlined below:

• Instrument Name: Technological Anxiety Scale (ANXT1)
• Authors: Joel E. Collier and Daniel L. Sherrell (2010)
• Construct Measured: Situational anxiety, intimidation, error catastrophizing, and operational insecurity associated with a designated technological system
• Instrument Type: Self-report psychological scale / psychometric questionnaire
• Administration Format: Paper-and-pencil, computer-assisted personal interviewing (CAPI), or web-based online survey
• Item Count: 3 items
• Referent Adaptation: The term [technology] is replaced by the specific target artifact under evaluation (e.g., “self-checkout kiosks,” “mobile banking app,” “AI diagnostic portal,” “automated passport gates”)
• Authentic Response Scale: 7-point Likert scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Somewhat Disagree
  • 4 = Neither Agree nor Disagree
  • 5 = Somewhat Agree
  • 6 = Agree
  • 7 = Strongly Agree
• Scoring and Indexing:
  • There are no reverse-scored items. All three items are positively worded in the direction of the anxiety construct.
  • Scores can be aggregated by calculating the mean of the three items (ranging from 1.00 to 7.00) or by summing raw responses (ranging from 3 to 21).
  • Higher numerical values reflect greater technological anxiety, elevated subjective intimidation, and stronger avoidance tendencies.
• Completion Time: Under 1 minute (approximately 30 to 45 seconds)

11. Permissions & Fee and Test Year

The ANXT1 was published in 2010 in the peer-reviewed academic article:

Collier, J. E., & Sherrell, D. L. (2010). Examining the influence of control and convenience in a self-service setting. Journal of the Academy of Marketing Science, 38(4), 490–509.

In accordance with standard academic conventions, the instrument items are available for scholarly research and non-commercial educational purposes without licensing fees, provided that appropriate formal academic citation is rendered to the original authors and the Journal of the Academy of Marketing Science. Commercial deployment, proprietary inclusion within commercial customer experience (CX) auditing software, or broad distribution within fee-charging consulting batteries may require copyright permission from the publisher (Springer Science+Business Media) or direct authorization from the scale authors.

12. References

• Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
• Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.
• Collier, J. E., & Sherrell, D. L. (2010). Examining the influence of control and convenience in a self-service setting. Journal of the Academy of Marketing Science, 38(4), 490–509. https://doi.org/10.1007/s11747-009-0179-4
• Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19(2), 189–211. https://doi.org/10.2307/249688
• Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008
• 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
• Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.
• Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.
• Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into the technology acceptance model. Information Systems Research, 11(4), 342–365. https://doi.org/10.1287/isre.11.4.342.11872

13. Items of the Scale