What promotes the mobile payment behavior of the elderly?

The study addresses why older adults adopt or resist mobile payment systems in the context of widespread smartphone use and rapid growth of mobile commerce, particularly in China. Despite mobile payment’s convenience and accelerated uptake during COVID-19, adoption among the elderly lags behind younger cohorts. Given China’s aging population and the elderly’s importance as consumers, understanding their intention to use mobile payments is both practically and socially important to bridge the digital divide. The paper aims to identify determinants of elderly users’ mobile payment intention by extending the Technology Acceptance Model (TAM) with additional constructs relevant to mobile financial services and by integrating elements from the Information Systems Success Model (ISSM). The research questions focus on how perceived ease of use, perceived usefulness, information/system/service quality, perceived trust, perceived risk, and social influence shape elderly users’ intention to use mobile payments.

The paper builds on TAM (Davis et al., 1989), where perceived ease of use (PEOU) influences perceived usefulness (PU) and behavioral intention, and on ISSM (DeLone & McLean, 1992; 2003), which emphasizes information quality (IQ), system quality (SYQ), and service quality (SEQ) as drivers of use and satisfaction. Recognizing TAM’s limitations (e.g., lack of negative beliefs and some social/contextual factors), the study extends TAM by incorporating perceived risk (PR), perceived trust (PT), and social influence (SI), and draws ISSM constructs to capture platform qualities salient to mobile payments. Prior research shows mixed evidence regarding SI’s role and highlights trust and risk as central in fintech adoption. The authors propose twelve hypotheses: H1 PEOU→PU (+); H2 PEOU→mobile payment intention (MP) (+); H3 PU→MP (+); H4 IQ→PT (+); H5 IQ→MP (+); H6 SYQ→PT (+); H7 SYQ→MP (+); H8 SEQ→PT (+); H9 SEQ→MP (+); H10 PT→MP (+); H11 SI→MP (+); H12 PR→MP (−). The conceptual model integrates TAM (PEOU, PU), ISSM (IQ, SYQ, SEQ), and additional variables (PT, PR, SI) to predict elderly MP.

Design: Cross-sectional survey with Partial Least Squares Structural Equation Modeling (PLS-SEM) for model estimation and hypothesis testing. Sample: 316 Chinese elderly (≥60 years) from Zhanjiang, Guangdong Province, all able to travel independently and with mobile payment experience. Gender: 52.5% male (n=166), 47.5% female (n=150). Age: 60–69 (51.9%), 70–79 (44.9%), ≥80 (3.2%). Education: elementary or below (33.2%), junior high (37.0%), high school (26.3%), college+ (3.5%). Mobile payment experience: <1 year (31.6%), 1–3 years (41.5%), 4–6 years (24.4%), >6 years (2.5%). Instruments: Nine constructs measured with items adapted from prior validated scales: PU & PEOU (Davis, 1989), IQ/SYQ/SEQ (Zhou, 2011), PT (Jain et al., 2022), SI (Venkatesh et al., 2012), PR (Habib & Hamadneh, 2021), MP (Maduku & Thusi, 2023). Demographics collected. The questionnaire wording was refined with feedback from two elderly readers. Procedure and analysis: Data analyzed using SPSS 25 and PLS-SEM (appropriate for complex models and smaller samples). Common method variance mitigated by simple language and anonymity; Harman’s single-factor test showed the first factor explained 37.50% (<50%) indicating CMV not severe. Reliability and validity were assessed via factor loadings, Cronbach’s alpha, composite reliability, AVE, Fornell–Larcker criterion, and HTMT ratios. Measurement quality: All constructs had loadings ≥0.767, CA and CR >0.7, and AVE ≥0.659, indicating good reliability and convergent validity; Fornell–Larcker and HTMT (<0.85) supported discriminant validity.

Model fit and explanatory power: SRMR=0.065 (<0.08) indicates good fit; GOF=0.554 (>0.36) indicates high fit. R²: MP=0.590, PT=0.364, PU=0.272. Significant paths (standardized coefficients, p-values): - PEOU→PU: β=0.522, p<0.001 (H1 supported). - PEOU→MP: β=0.262, p<0.001 (H2 supported). - PU→MP: β=0.158, p<0.01 (H3 supported). - IQ→PT: β=0.454, p<0.001 (H4 supported). - IQ→MP: β=0.103, p<0.05 (H5 supported). - SEQ→PT: β=0.231, p<0.001 (supports hypothesized positive effect). - SEQ→MP: β=0.156, p<0.001 (supports hypothesized positive effect). - PT→MP: β=0.262, p<0.001 (H11 supported). - PR→MP: β=−0.148, p<0.001 (H12 supported; negative effect). Non-significant paths: - SYQ→PT: β=−0.023, p=0.669 (H6 not supported). - SYQ→MP: β=−0.044, p=0.226 (H7 not supported). - SI→MP: β=0.111, p=0.129 (H12 in text refers to PR; SI hypothesis not supported). Overall, elderly mobile payment intention is driven positively by PEOU, PU, PT, IQ, and SEQ, and negatively by PR; SYQ and SI showed no direct significant effects on intention.

Findings indicate that trust is central to elderly users’ intention to use mobile payments; enhancing perceived security, reliability, and privacy increases PT and thus intention. PR significantly deters intention, consistent with elderly users’ higher sensitivity to financial and privacy risks. Contrary to some prior work, SI did not significantly affect elderly intention, possibly because payment decisions involve personal finances and sensitive data where older adults act cautiously despite social cues. From the ISSM perspective, IQ and SEQ bolster both trust and intention, highlighting the importance of accurate, timely transaction information and responsive, personalized service for older users. From TAM, PEOU directly elevates both PU and intention, and PU further contributes to intention, underscoring the need for simple, efficient payment experiences. Together, results address the research question by showing that improving ease of use, usefulness, information and service quality, and trust—while reducing risk—are key to promoting elderly mobile payment adoption.

The study integrates TAM and ISSM to construct and validate a mobile payment acceptance model for elderly users. Using PLS-SEM on 316 elderly participants, the model explains 59% of the variance in intention and exhibits good fit. Key contributions include demonstrating the pivotal roles of perceived trust, ease of use, usefulness, information quality, and service quality, as well as the deterrent effect of perceived risk; system quality and social influence showed no direct impact on intention. Practical implications: developers and designers should prioritize age-friendly, simple interfaces and flows; deliver accurate, timely transaction information; provide reliable, responsive, and personalized services; and implement robust security and privacy protections to build trust and reduce risk. Policymakers and regulators can support these aims through guidelines and consumer protections tailored to older adults. Future research should examine cross-national and cultural contexts, consider additional psychosocial and cultural variables, compare rural–urban populations, and employ longitudinal designs to capture changes over time.

The sample is region-specific (Zhanjiang, Guangdong, China) with few participants aged 80+ and few with college education or higher, limiting generalizability to other regions and educational strata, especially in developed countries. The cross-sectional design precludes assessing changes over time or with experience. Cultural and social context factors (e.g., collectivism/individualism, detailed family and intergenerational dynamics) were not deeply modeled. Future studies should broaden and diversify samples (including cross-country and rural–urban comparisons), incorporate additional cultural and social constructs, and use longitudinal approaches.