Trust as a Mediator of Perceived Ease of Use and Promotion in Shopeepay Transaction Decisions in Banyumas Regency

Aulia Fitria Salmah (Management Study Program, Faculty of Social Sciences, Harapan Bangsa University)
Faizal Rizky Yuttama (Management Study Program, Faculty of Social Sciences, Harapan Bangsa University)
Puspita Lianti Putri (Management Study Program, Faculty of Social Sciences, Harapan Bangsa University)

Indonesian Journal of Multidisciplinary Sciences
Vol. 5 No. 2 (2026): Indonesian Journal of Multidisciplinary Sciences (IJoMS)

The rapid development of digital technology has accelerated the adoption of e-wallets, particularly ShopeePay, as a primary non-cash payment method. This study analyzes the structural relationships between perceived ease of use, promotion, user trust, and transaction decisions among ShopeePay users in Banyumas Regency. Adopting a quantitative survey design, data were collected via structured self-report questionnaires from a purposive sample of 200 active users aged 18–50 years residing in the region. Data were processed using IBM SPSS Statistics 27. In the mediator model (Z =Trust), ease of use (beta = 0.197, p < 0.001) and promotion (beta = 0.563, p < 0.001) significantly predicted user trust, explaining 63.5% of its variance (R2 = 0.635). In the outcome model (Y = Transaction Decision), trust (beta = 0.500, p < 0.001), ease of use (beta = 0.134, p = 0.001), and promotion (beta= 0.286, p < 0.001) exerted significant direct effects, accounting for 63.5% of the total variance (R^2 = 0.635, Adjusted  R^2 = 0.630). Path analysis indicated indirect effects of ease of use (ab = 0.150) and promotion (ab = 0.293) on transaction decisions through trust. Although traditional Sobel tests indicated statistical significance (Z = 3.140, p = 0.002 and Z = 6.626, p < 0.001, respectively), these results assume normal sampling distributions and require cautious interpretation due to detected heteroscedasticity in the outcome model (p < 0.05). Assuming non-normal error distribution, bias-corrected bootstrap confidence intervals (95% CI) confirmed that trust serves as a significant partial mediator for both paths. These findings emphasize that user trust, driven by strategic promotions and seamless usability, acts as a primary psychological mechanism in shaping digital transaction behavior. However, the study’s generalizability is constrained by its cross-sectional design, reliance on self-reported perceptual data, non-probability purposive sampling, potential common-method bias, and the absence of actual transaction log data.