The research addresses a distinctive commercial and psychological puzzle. Online paid courses, or OPCs, require consumers to evaluate two things simultaneously: the educational value of the content and the financial commitment needed to access it. Unlike general e-commerce, where the product is tangible and often returnable, a course is an intangible service whose quality can only be judged after purchase. Unlike free online learning, it carries real monetary risk. The authors argue that this dual burden of value assessment and financial exposure makes the purchase decision for paid digital education a unique behavioral case that existing single-theory models struggle to explain fully on their own.

To capture that complexity, the researchers built their framework from three established theoretical traditions. The technology acceptance model, originally formulated by Fred Davis in the 1980s, holds that adoption of a technology depends chiefly on perceived usefulness, the belief that the system will improve performance, and perceived ease of use, the belief that using it will be free of excessive effort. The theory of planned behavior, developed by Icek Ajzen, adds that behavior is predicted by intention, which is shaped by attitude toward the behavior, subjective norms, the perceived social pressure from important others, and perceived behavioral control. Trust was incorporated because digital transactions inherently involve uncertainty, and prior work in e-commerce has shown that trust shapes how consumers evaluate and adopt online services.

Methodologically, the study followed a quantitative design. Data were collected from 200 consumers using a structured questionnaire, and the proposed model was tested with structural equation modeling, a statistical technique that allows researchers to estimate networks of relationships among latent constructs that cannot be observed directly, such as usefulness or trust, while accounting for measurement error. Structural equation modeling is particularly well suited to this kind of theory integration because it evaluates whether the overall pattern of covariances in the data is consistent with the hypothesized causal structure, rather than testing each relationship in isolation.

The results are striking for what they confirm and, perhaps even more so, for what they contradict. Perceived usefulness, perceived ease of use and subjective norm all exerted positive and significant effects on both consumers’ attitudes toward online paid courses and their purchase intentions. In other words, when Bangladeshi consumers believe a course will genuinely improve their skills, when the platform is easy to navigate, and when family, friends or professional peers endorse the purchase, they both feel more positively about the courses and are more likely to buy them. This alignment between technology-related beliefs, social influence and intention suggests that the core engines of adoption in this market are practical and social rather than abstract.

The surprises emerge in the trust and attitude pathways. Trust significantly improved consumers’ attitudes toward online paid courses, confirming its theorized role as a belief that softens skepticism about online transactions. However, neither trust nor attitude translated directly into purchase intention. This finding runs against the intuitive assumption, common in e-commerce research, that favorable attitudes and trust reliably convert into buying behavior. In the Bangladeshi paid-course context, it appears that a consumer can trust a platform and hold a positive view of online education yet still hesitate to pay, with the decision hinging more directly on perceived usefulness, ease of use and social endorsement than on emotional or relational factors.

One plausible interpretation, consistent with the authors’ framing, is that in a developing economy where discretionary spending on education competes with many household priorities, the purchase decision for digital learning is treated as a calculated investment rather than an expression of sentiment. Consumers may weigh the expected performance gains from a course, the friction involved in accessing it, and the opinions of their social circle far more heavily than generalized feelings of warmth or confidence toward the seller. Trust, in this account, does its work earlier in the psychological chain, shaping whether a consumer views the category favorably at all, while the final intention to pay is governed by instrumental and normative judgments.

The findings carry substantial practical implications for e-learning providers operating in Bangladesh and in comparable emerging markets. First, improving platform usability is not cosmetic; perceived ease of use had a significant effect on both attitude and intention, meaning that intuitive interfaces, smooth enrollment processes, reliable video delivery and responsive support can directly move the bottom line. Second, providers should work to enhance the perceived value of their courses, since perceived usefulness was among the strongest levers in the model. Concrete demonstrations of skill outcomes, career relevance, certificates with labor-market recognition and transparent curricula can strengthen the belief that a course will genuinely improve performance. Third, although trust did not directly drive intention, it still shaped attitudes, and positive attitudes remain part of the adoption pathway, so visible credentials, secure payment systems, clear refund policies and credible instructor profiles retain strategic value.

The significance of subjective norm deserves particular attention from marketers in this market. Social influence, meaning the encouragement of friends, family, colleagues and peers, significantly affected both attitude and purchase intention. In a collectivist cultural context such as Bangladesh, where purchasing decisions are often discussed within families and social networks, word-of-mouth recommendations, testimonials from relatable peers, referral incentives and community-building around courses may be among the most cost-effective growth strategies available to course providers. The study suggests that a satisfied learner is not merely a repeat customer but a node in a social network that transmits adoption pressure to others.

Beyond its commercial relevance, the study extends academic theory by demonstrating that the integrated TAM-TPB-trust framework applies to the digital education context, a domain that sits awkwardly between pure technology adoption and consumer e-commerce. The finding that technology-related beliefs and attitudinal factors dominate while trust operates indirectly refines the relative weighting of constructs in this framework for paid-learning settings. As Bangladesh’s online education sector matures, the authors argue that improving usability, enhancing perceived value and establishing confidence will together promote course adoption and support the sustainable development of a market that has become a critical channel for skills formation in a rapidly digitizing economy. The research was self-funded by the authors and received ethics approval from the Research Management Committee of the Department of Marketing at Patuakhali Science and Technology University, with informed consent obtained from all survey participants.

The study’s sampling approach deserves brief explanation. Snowball sampling, a technique formalized in the social sciences in the early 1960s, relies on initial participants to recruit others from their own networks, making it a pragmatic choice when no comprehensive sampling frame of online course consumers exists. While this limits the statistical generalizability of the estimates to the broader Bangladeshi population, it is a common and accepted strategy for exploratory studies of emerging consumer segments, and the authors note that the underlying data will be made available on reasonable request to support scrutiny and replication.

The choice of structural equation modeling also connects the study to a substantial methodological literature. Researchers working with latent variable models typically distinguish between covariance-based and partial least squares approaches, and contemporary guidance emphasizes that the selection between them should follow considerations of model complexity, sample size and the explanatory versus predictive goals of the analysis. By situating their work within these established conventions, the authors align the study with a body of information systems research stretching back to foundational work on trust and technology acceptance in online shopping, which first demonstrated that trust and the classic acceptance beliefs operate as related but distinct drivers of consumer behavior.

The findings also resonate with recent research on the economics of paid online learning. Prior work on massive open online courses has shown that payment itself changes user engagement, with paying learners behaving differently from free participants, which underscores why purchase intention in this domain is a meaningful outcome in its own right rather than a proxy for adoption. Other scholarship has documented how pandemic-era policy responses accelerated the democratization of online learning, expanding access across socioeconomic groups and creating the demand conditions that markets like Bangladesh are now experiencing.

For theory, the indirect role of trust observed here echoes earlier integrated models in e-commerce, where trust frequently shapes attitudes and reduces perceived risk without always exerting a direct effect on intention. The Bangladeshi results suggest that in low-trust or high-uncertainty payment environments, trust may function primarily as a gatekeeping belief that determines whether consumers engage with the category at all. Future research could test whether these pathways hold across other paid digital education markets in South Asia, and whether constructs such as perceived behavioral control or price fairness, which the present framework did not center, add explanatory power as the market matures.

Subject of Research: Consumer purchase intention toward online paid courses in Bangladesh using an integrated technology acceptance, planned behavior and trust framework

Article Title: Predicting purchase intention towards online paid courses in Bangladesh: an integrated TAM-TPB-trust framework

Article References: Dipta, A. D., & Hossain, A. (2026). Predicting purchase intention towards online paid courses in Bangladesh: an integrated TAM-TPB-trust framework. SN Social Sciences, 6(9), Article 422. https://doi.org/10.1007/s43545-026-01715-y

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01715-y

Keywords: online paid courses, purchase intention, Bangladesh, technology acceptance model, theory of planned behavior, trust, e-learning, structural equation modeling, consumer behavior, digital education, perceived usefulness, subjective norm