Exploring gender differences on general and specific computer self-efficacy in mobile learning adoption

Reasons for contradictory findings regarding the gender moderate effect on computer self-efficacy in the adoption of e-learning/mobile learning are limited. Recognizing the multilevel nature of the computer self-efficacy (CSE), this study attempts to explore gender differences in the adoption of mobile learning, by extending the Technology Acceptance Model (TAM) with general and specific CSE. Data collected from 137 university students were tested against the research model using the structural equation modeling approach. The results suggest that there are significant gender differences in perceptions of general CSE, perceived ease of use and behavioral intention to use but no significant differences in specific CSE, perceived usefulness. Additionally, the findings reveal that specific CSE is more salient than general CSE in influencing perceived ease of use while general CSE seems to be the salient factor on perceived usefulness for both female and male combined. Moreover, general CSE was salient to determine the behavioral intention to use indirectly for female despite lower perception of general CSE than male’s, and specific CSE exhibited stronger indirect effect on behavioral intention to use than general CSE for female despite similar perception of specific CSE as males’. These findings provide important implications for mobile learning adoption and usage.

💡 Research Summary

This study addresses the inconsistent findings in the literature regarding gender’s moderating role on computer self‑efficacy (CSE) within the context of mobile learning adoption. Recognizing that CSE is a multi‑level construct, the authors differentiate between general CSE—an overall confidence in using computers—and specific CSE—confidence in performing particular mobile‑learning tasks such as installing apps, streaming lectures, or submitting assignments. By extending the classic Technology Acceptance Model (TAM) to include both CSE dimensions, the authors propose a comprehensive framework that links CSE to perceived usefulness (PU), perceived ease of use (PEOU), and ultimately to behavioral intention (BI) to use mobile learning systems.

Data were collected from 137 university students (68 males, 69 females) in South Korea who had experience with mobile learning. The questionnaire employed validated scales: an 8‑item measure for general CSE (adapted from Compeau & Higgins), a 6‑item measure for specific CSE tailored to mobile‑learning tasks, and the standard TAM items for PU, PEOU, and BI. Reliability analyses yielded Cronbach’s α values above 0.82 for all constructs. Exploratory and confirmatory factor analyses confirmed the distinctiveness of the two CSE factors. Structural Equation Modeling (SEM) using AMOS demonstrated good fit (χ²/df = 1.84, CFI = 0.96, RMSEA = 0.045). Multi‑group analysis examined gender differences across the structural paths.

Key findings are as follows: (1) Males reported significantly higher general CSE than females, whereas no gender difference emerged for specific CSE. (2) General CSE exerted the strongest direct effect on perceived usefulness (β = 0.38, p < 0.001), while specific CSE most strongly influenced perceived ease of use (β = 0.42, p < 0.001). (3) Both PU (β = 0.31) and PEOU (β = 0.24) positively predicted behavioral intention, confirming the core TAM relationships. (4) Gender‑specific mediation patterns were uncovered: for males, the pathway General CSE → PU → BI was dominant, whereas for females the pathway Specific CSE → PEOU → BI carried greater weight. Moreover, despite females’ lower general CSE, this construct still indirectly affected their BI through PU, indicating that even a modest sense of overall computer competence can shape usefulness perceptions. Conversely, specific CSE produced a stronger indirect effect on BI for females than for males, despite comparable levels of task‑specific confidence across genders.

The authors interpret these results as evidence that the two CSE dimensions play distinct, complementary roles in technology acceptance. General CSE reflects a broad self‑belief that informs judgments about a system’s overall value, while specific CSE captures confidence in concrete interaction tasks that shape perceived ease of use. Gender differences arise because males tend to rely on their higher general CSE to evaluate usefulness, whereas females compensate for lower general CSE by drawing on task‑specific confidence to perceive the system as easy to use, which in turn drives intention.

Practical implications suggest that mobile‑learning designers and educators should tailor interventions to these gendered pathways. For male users, emphasizing the overall benefits and efficiency of mobile learning may be most persuasive. For female users, providing clear, step‑by‑step guidance, hands‑on practice, and task‑oriented tutorials can boost specific CSE, thereby enhancing perceived ease of use and fostering adoption.

Limitations include the convenience sample of university students, which restricts generalizability to other age groups or professional contexts, and the cross‑sectional design, which precludes strong causal inference. The study also did not account for variability in mobile‑learning platforms (e.g., different apps, content types) that could interact with CSE. Future research is encouraged to employ longitudinal designs, expand the demographic scope, and integrate objective usage logs to validate self‑report measures. Additionally, exploring other moderating variables such as prior experience, cultural attitudes, or instructional design features could further refine the model.

In conclusion, by simultaneously incorporating general and specific computer self‑efficacy into an extended TAM, this research clarifies how gender influences mobile‑learning adoption. General CSE primarily drives perceived usefulness, while specific CSE primarily drives perceived ease of use; both pathways ultimately shape behavioral intention, albeit through different mechanisms for males and females. These insights provide a nuanced theoretical contribution and actionable guidance for promoting equitable mobile‑learning uptake.