Perceived Ease of Use in Technology Acceptance

Updated 22 February 2026

PEOU is defined as a user's belief that using a technology requires minimal mental and physical effort, serving as a key determinant in adoption models.
Empirical studies employing Likert-scale instruments reveal that high PEOU significantly boosts perceived usefulness and intention to use, with varying effect sizes across contexts.
Practitioners use PEOU insights to streamline system interfaces and align task demands, thereby enhancing user satisfaction and overall technology acceptance.

Perceived Ease of Use (PEOU) refers to a core construct within the Technology Acceptance Model (TAM), originally formulated by Davis (1989), denoting “the degree to which a person believes that using a particular system would be free of effort.” PEOU has retained this foundational definition as it has been operationalized and empirically examined across domains such as software engineering, education, e-commerce, autonomous vehicles, and AI-driven systems. PEOU is considered a key predictor (often, but not always, a primary antecedent) of Perceived Usefulness (PU) and behavioral outcomes such as Intention to Use (ITU).

1. Theoretical Definition and Role in Acceptance Models

PEOU is universally formalized as a user's subjective belief about the ease with which a technology can be learned and operated, minimizing both mental and physical effort (Jalali, 2021, Nguyen et al., 2023, Zakharov et al., 29 Apr 2025, Setälä et al., 2 Jan 2025, Garcia et al., 2021, Revythi et al., 2017, Nabot et al., 2021, Shen et al., 2024, Misirlis et al., 2023). In TAM and its successors (e.g., TAM2, Unified Theory of Acceptance and Use of Technology), PEOU is typically an exogenous construct that:

Directly predicts Intention to Use (BI or ITU)
Indirectly predicts Intention by enhancing Perceived Usefulness (PU)
May itself be shaped by upstream factors (e.g., Task–Technology Fit, Self-Efficacy, Social Norms)
In some models, exhibits moderating or mediated effects on outcomes such as Trust, Attitude, or enjoyment-driven variables

Key relationships generally take the form: $PU = \beta_1 \cdot PEOU + \epsilon_1$

$ITU = \gamma_1 \cdot PU + \gamma_2 \cdot PEOU + \epsilon_2$

where path coefficients and significance may vary by context (Harryanto et al., 2019, Harizi et al., 2022, AlSoufi et al., 2014, Choudhury et al., 18 Feb 2025).

2. Operationalization, Instruments, and Psychometric Properties

Measurement of PEOU draws upon standardized subscales from the TAM literature, often comprising 3–11 items per construct. Commonly, items are rated on Likert-type scales—5, 7, or occasionally 4 points—addressing dimensions such as ease of learning, clarity of interaction, required mental effort, and skill acquisition (Jalali, 2021, Nguyen et al., 2023, Zakharov et al., 29 Apr 2025, Garcia et al., 2021, Revythi et al., 2017, Setälä et al., 2 Jan 2025, Misirlis et al., 2023, Shen et al., 2024, Choudhury et al., 18 Feb 2025).

A representative item set includes:

“It is easy for me to become skillful at using [the system].”
“I find [the system] easy to use.”
“Learning to operate [the system] is easy for me.”
“My interaction with [the system] is clear and understandable.”
“Using [the system] does not require a lot of mental effort.”

Instrument reliability is consistently evaluated by Cronbach’s α, Composite Reliability (CR), and AVE, with values typically: $\alpha_{\mathrm{PEOU}} \geq 0.75; \quad CR_{\mathrm{PEOU}} \geq 0.80; \quad AVE_{\mathrm{PEOU}} \geq 0.50$ Factor loadings usually exceed 0.70; discriminant and convergent validity are generally confirmed via Fornell–Larcker and cross-loading analyses (Jalali, 2021, Shen et al., 2024, Zakharov et al., 29 Apr 2025, Setälä et al., 2 Jan 2025).

3. Structural Models, Empirical Findings, and Effect Sizes

Empirical studies deploying (P)LS-SEM, GSCA, or CFA underline several general patterns:

PEOU exerts a positive and (usually) significant direct effect on PU and, often, ITU (Nguyen et al., 2023, Nabot et al., 2021, Garcia et al., 2021, AlSoufi et al., 2014, Harizi et al., 2022, Shen et al., 2024, Choudhury et al., 18 Feb 2025).
Contextual moderators (e.g., Experience, Task–Technology Fit) and exogenous antecedents (Self-Efficacy, Compatibility, Social Norm) can either intensify or attenuate PEOU’s direct effects (Harryanto et al., 2019, Revythi et al., 2017, Nguyen et al., 2023, AlSoufi et al., 2014, Misirlis et al., 2023, Sui et al., 2023).
In some domains or contexts, PEOU → PU or PEOU → ITU fails to reach significance, with mediation/indirect or cultural factors dominating (e.g., MetaEducation, blockchain prototypes, Finnish secondary mathematics, cross-country comparisons) (Setälä et al., 2 Jan 2025, Shrestha et al., 2019, Misirlis et al., 2023, Tarhini et al., 2015).

Observed standardized path coefficients for PEOU to core outcomes:

Outcome	Coefficient Range	Reference
PU	0.08–0.92	(Garcia et al., 2021, AlSoufi et al., 2014, Shen et al., 2024)
ITU/BI	0.07–0.61	(Nabot et al., 2021, Harizi et al., 2022, AlSoufi et al., 2014, Choudhury et al., 18 Feb 2025)
Trust	~0.36	(Choudhury et al., 18 Feb 2025)
Attitude	~0.10–0.20	(Garcia et al., 2021, Tarhini et al., 2015)
Indirect via mediators	Mixed	(Yao et al., 23 Mar 2025, Choudhury et al., 18 Feb 2025)

In “AI in Software Engineering: Perceived Roles and Their Impact on Adoption,” a richer mental model of AI (i.e., as both tool and expert) substantially boosts PEOU (r = 0.56, p < 0.001), with a mean PEOU of ≈4.0 (SD = 0.7) among developers (Zakharov et al., 29 Apr 2025). In the autonomous vehicle context, PEOU exhibits both a strong direct effect on ITU (β = 0.396, p = 0.003) and an extremely high effect on PU (β = 0.920, p < 0.001), although ease of use can increase perceived risk if not paired with control features (Shen et al., 2024).

4. Antecedents and Determinants of PEOU

Several independent variables systematically predict PEOU:

Self-Efficacy: Higher technological or task self-efficacy predicts greater PEOU, both directly and via mediated paths (Revythi et al., 2017, AlSoufi et al., 2014, Sui et al., 2023, Misirlis et al., 2023).
Task–Technology Fit: Alignment between user tasks and system capabilities increases PEOU (β = 0.636, p < .05 for Dialogflow) (Nguyen et al., 2023).
Social Norms: Peer, instructor, and institutional influences especially affect PEOU in nascent technology settings and educational use of the metaverse (β = 0.91–1.00) (Misirlis et al., 2023, Revythi et al., 2017).
Compatibility: Systems matching existing workflows or user routines have higher PEOU (β = 0.460 in mobile banking; small gains in upper-secondary GenAI) (AlSoufi et al., 2014, Setälä et al., 2 Jan 2025).
Technology-Readiness Motivators: Optimism and Innovativeness in students/users increase PEOU for educational chatbots (Hasan et al., 2023).
User Experience: For high-experience users, PEOU exerts far stronger effects on intention than for novices, implying interaction/moderation (Harryanto et al., 2019).

5. Contextual Factors and Domain-Specific Observations

PEOU’s structural function and practical importance exhibit context dependence:

Educational Platforms and AI: In LMS and AI tool adoption, PEOU’s impact on PU is often moderate (β ≈ 0.17–0.42), sometimes non-significant; its direct effect on Attitude or ITU is often smaller, except in specific contexts with high interface complexity or low baseline familiarity (Garcia et al., 2021, Yao et al., 23 Mar 2025, Setälä et al., 2 Jan 2025, Revythi et al., 2017).
E-commerce, Mobile Services: PEOU tends to be the strongest or a co-equal driver of adoption intention, frequently exceeding PU in effect size (β = 0.442 vs. 0.287 for mobile banking in Bahrain; β = 0.35 vs. 0.20 for mobile commerce in Jordan) (AlSoufi et al., 2014, Nabot et al., 2021, Harizi et al., 2022).
AI/Chatbots and Risk Domains: In sensitive sectors (healthcare: LLMs), the relationship between PEOU and behavioral intention is fully mediated by Trust. Non-linear “tipping point” effects, where PEOU must surpass a threshold before sharply increasing intention, have been empirically confirmed (Choudhury et al., 18 Feb 2025).
Cross-Cultural Differences: The strength of PEOU’s effect on PU, attitude, and intention is sometimes diminished or even nullified in certain cultural contexts, necessitating tailored interventions (e.g., stronger demonstration and training in Lebanon vs. the UK for RSS adoption) (Tarhini et al., 2015).

6. Measurement Rigor, Invariance, and Recommendations

Extensive psychometric validation accompanies PEOU instrument deployment, including:

Exploratory and Confirmatory Factor Analysis
Composite Reliability, AVE, item-total correlations
Measurement invariance across groups or cultures (configural, metric, scalar, and factorial testing) (Tarhini et al., 2015)
Non-parametric and distribution-free testing when normality is violated (Jalali, 2021)

Robustness of PEOU as a latent construct supports its continued use across modeling approaches (CB-SEM, PLS-SEM, GSCA, Bayesian SEM), with structural R² values for PEOU and downstream constructs typically in the 0.2–0.8 range depending on predictors and model complexity.

7. Practical Implications and Design Guidance

Research consistently underscores several actionable recommendations:

Interface and Onboarding: Streamline workflows, provide clear navigation, minimize cognitive load, and implement just-in-time tutorials to immediately reduce perceived effort (Nguyen et al., 2023, Setälä et al., 2 Jan 2025, Misirlis et al., 2023).
Task Alignment: Ensure system feature sets closely match user activities, raising both Task–Technology Fit and PEOU (Nguyen et al., 2023).
Role Framing and Personalization: Let users select AI tool “roles” or interface modes to align with personal mental models, increasing comfort and learnability (Zakharov et al., 29 Apr 2025).
Social Cues and Training: Leverage positive social norms via peer endorsements, targeted demonstrations, and skill-building sessions, especially during technology transitions in education (Revythi et al., 2017, Misirlis et al., 2023).
Risk and Control Affordances: In high-stakes domains, pair ease-of-use with visible control options and trust-building mechanisms to avoid heightened risk perceptions (Shen et al., 2024, Choudhury et al., 18 Feb 2025).
Iterative Usability Assessment: Employ usability scales (e.g., System Usability Scale), factor analysis, and ongoing user feedback after interface changes to sustain high PEOU (Revythi et al., 2017, Garcia et al., 2021).

Overall, enhancing PEOU effectively raises likelihood of adoption and continued use across almost all studied contexts, provided the system design is attuned to users’ capabilities, workflows, and domain-specific concerns. As technology applications diversify, the role of PEOU both as a predictor and as a target for user-centered system refinement remains central to technology acceptance research and practice.

References: