The sustainable livelihoods framework provides a comprehensive lens for analyzing how rural households utilize various capital assets to achieve livelihood outcomes under different vulnerability contexts and institutional structures (Rahman and Hickey, 2020). In our study context, we identify four critical capital assets that enable Danzhou farmers to participate in the digital economy.

Human capital encompasses the digital skills, entrepreneurial capabilities, and cultural knowledge cultivated through vocational innovation-entrepreneurship education. Students trained in blockchain technology applications, e-commerce operations, cultural creative design, and digital marketing constitute the human capital foundation for rural digital transformation. This extends beyond traditional agricultural skills to include competencies essential for navigating digital economy opportunities.

Cultural capital, drawing from Bourdieu’s conceptualization, represents the Danzhou Tiaosheng intangible heritage resources, including folk songs, performance traditions, musical patterns, and associated cultural narratives. The Tiaosheng tradition, with its unique antiphonal singing structure and historical significance dating back over 1,000 years, provides rich symbolic resources for product differentiation and brand storytelling. Unlike physical or financial capital, cultural capital’s value realization requires creative transformation processes that translate heritage elements into marketable products while maintaining authenticity.

Technical capital comprises the blockchain traceability infrastructure, including distributed ledger systems, IoT sensors for real-time data collection, QR code interfaces for consumer engagement, and smart contract protocols for automated verification. The AntChain BaaS platform utilized in our project reduces deployment costs and technical barriers, enabling smallholder farmers to participate in advanced digital systems.

Social capital encompasses the networks, partnerships, and trust relationships established through industry-education integration mechanisms, school-local collaborations, farmer cooperatives, e-commerce platform connections, and consumer communities. These relationships facilitate knowledge transfer, resource mobilization, market access, and collective action essential for sustainable development outcomes.

Drawing on Digital Empowerment Theory, we conceptualize digital empowerment as a multi-dimensional process whereby marginalized communities gain agency through access to digital technologies, skills, and networks (Kosasih and Sulaiman, 2024). This framework complements SLF by explaining the transformative mechanisms through which technological adoption converts latent capital assets into realized livelihood outcomes.

Bourdieu’s theory of cultural capital provides theoretical grounding for understanding how intangible cultural heritage can be strategically leveraged for economic value creation. Cultural capital exists in embodied (skills, knowledge), objectified (cultural products), and institutionalized (certifications, recognition) forms. Our intervention facilitates the conversion from embodied cultural knowledge to objectified cultural products through creative design processes, amplified by blockchain-enabled authenticity signals.

Trust Theory, particularly as applied to food systems (Palma et al., 2015), explains how information asymmetries create market failures that blockchain transparency can address. By providing verifiable provenance information, blockchain reduces consumer risk perception and enables trust-based premium pricing for credence goods.

Based on the sustainable livelihood framework and drawing on concepts of trust theory, entrepreneurship theory, and digital transformation studies, we develop three key hypotheses as shown in Figure 1.

There are three hypotheses examined based on theoretical frameworks that have been previously established. In hypothesis one, it is assumed that there is a positive relationship between blockchain traceability transparency and purchase intention of consumers of cultural creative agriculture products. According to trust theory, transparency of information enhances consumer trust and loss of risk associated with products, as blockchain technology helps address issues of asymmetry of information inherent in agriculture value chains, especially concerning non-physical or credence aspects of agriculture products that a customer cannot easily verify on their own. In hypothesis two, it is assumed that innovation-entrepreneurship vocational training intensity has a positive effect on success rates of students engaged in digital rural economy entrepreneurship ventures. According to human capital theory, formal innovation-entrepreneurship vocational training enhances human capital and, as such, enhances success rates of entrepreneurs, especially through innovation and technical skills and expertise that can be applied within digital rural economy settings and involve expertise and skills in heritage creative design, blockchain technology, and e-commerce platform operations. Hypothesis three proposes that digital empowerment composite index has a positive effect on farmers’ income growth rate. Hypothesis 3 (H3): The digital empowerment composite index positively influences farmers’ income growth rate. According to digital transformation theory (Kosasih and Sulaiman, 2024), successful digital technology implementation requires complementary alignment of technological infrastructure, human capital, and market linkages. Farmers who adopt blockchain traceability systems, partner with vocational students, and utilize e-commerce platforms can capture added value through three mechanisms: (1) premium pricing enabled by verified authenticity signals; (2) expanded market access through digital channels that bypass traditional intermediaries; and (3) improved bargaining power in agricultural value chains through direct consumer connections. The digital empowerment composite index, constructed from four dimensions (blockchain adoption, student partnerships, e-commerce capability, and marketing intensity), represents the integrated capacity of farmers to leverage these digital opportunities for income growth.

Our research used a mixed methods design that combined quantitative survey research, qualitative techniques, case studies, and quasi-experiments that included control groups. Our research took place over a period of 18 months, from September of 2023 until December of 2024, and was administered at Danzhou City, located in the province of Hainan, China. Danzhou is a city of about 950,000 people, with large rural areas devoted to traditional agriculture and culture preservation (Danzhou Municipal Statistics Bureau, www.danzhou.gov.cn).

Tiaosheng folk song is listed as one of China’s intangible cultural heritages on its National Intangible Cultural Heritage List since 2006, and it symbolizes a precious and distinctive culture that has a history of more than 1,000 years. Our project combined elements of Tiaosheng culture and three characteristic agricultural products, namely zongzi (rice dumplings made of glutinous rice, eaten on Dragon Boat Festival), dried red fish, and Danzhou coffee, which is a new high-quality crop.

Our study incorporated multiple data sources to ensure triangulation and robustness. As shown in Table 1, we collected data from diverse stakeholder groups including consumers, students, farmers, and digital platforms.

To validate the quasi-experimental design, we conducted baseline balance tests comparing treatment and control groups on key pre-intervention characteristics. As shown in Table 2, farmer participants in the blockchain traceability project (n = 28) and control group farmers (n = 14) showed no significant differences across demographic and economic variables measured in 2023. Household income, the primary outcome variable, was nearly equivalent between groups (treatment: 32,450 ± 8,230 RMB vs. control: 31,880 ± 7,540 RMB, t = 0.23, p = 0.82). Similarly, household size, farmland area, distance to town center, and age of household head showed no statistically significant differences (all p > 0.10), suggesting that observable characteristics were well-balanced prior to the intervention. Table 3 presents baseline balance tests for student participants. The experimental group (n = 60) receiving the enhanced innovation- entrepreneurship curriculum and the control group (n = 60) following the standard curriculum demonstrated equivalence on academic performance (GPA: 3.21 ± 0.54 vs. 3.18 ± 0.52, t = 0.32, p = 0.75), urban/rural background composition (45% vs. 43% urban, χ² = 0.05, p = 0.82), prior entrepreneurship exposure (18% vs. 20%, χ² = 0.07, p = 0.79), and baseline digital skills (2.3 ± 0.8 vs. 2.2 ± 0.9, t = 0.64, p = 0.52). These baseline balance tests confirm that treatment and control groups were comparable on key covariates prior to intervention, supporting the internal validity of the quasi-experimental design and strengthening causal inferences from subsequent difference-in-differences analyses.

To develop a consortium blockchain system, we used Ant Group’s AntChain Business Accelerator (BaaS) platform, which enables enterprise-grade blockchain services attainable at low costs, making it suitable for small farmer engagement (Chakravorty and Mukherjee, 2023). Our system design involves five critical nodes of the supply chain, namely (1) Raw material purchasing and production, through which farmers register crop planting or harvest information, such as geographical co-ordinates, dates, and primary quality parameters; (2) Processing and manufacturing, through which farmer co-ops register ingredient details, processing techniques, and employee certifications; (3) Quality inspection and packaging, through which third-party testing bodies register final food safety certificates, nutritional tests, and environment compliance papers; (4) Logistic and distribution, through which on-the-go temperature tracking, GPS, and time of transit can be recorded; (5) Consumer verification, through which scanning enables viewing of comprehensive production details, cultural lineage (including Tiaosheng audio recordings and culture or heritage section details).

Information on all these is cryptographically hashed and stored on the blockchain, thereby providing immutability and non-repudiation of records (Banerjee and Saha, 2025). Smart contracts are used, and these automatically validate whether all required information is available and alert on any irregularities. IoT sensor incorporation enables automatic collection of information (temperature when stored and when transported), and this system is compatible with available quality testing LIMS systems that enable easy upload of certificates into the system. Setup costs on average per farmer household participating in the system ranged from 2,800 RMB, while continued operation costs a nominal sum of 50 RMB on a monthly basis.

Based on this, we focused on Hainan Vocational and Technical College, a provincial-level college of higher vocational education, which currently has about 15,000 enrolled students.¹ They acted as our pilot college for implementing changes in their curriculum. In designing our innovation and entrepreneurship program at E-Commerce School, we included three innovation and entrepreneurship courses that are new and more advanced, such as “Intangible Heritage Creative Design” (48 class hours), “Blockchain Applications of Agricultural Supply Chains” (32 class hours), and “Rural E-Commerce Business and Digital Marketing” (40 class hours). These, of course, are for the experimental group consisting of 60 students, while the other group of 60 students followed our regular innovation and entrepreneurship course emphasizing basic e-commerce skills only, without any focus on heritage and blockchain aspects.

Student participants of the experimental group consisted of 12 teams of entrepreneurs (4–6 members each), and they made plans for business development of cultural creative agro-products. Teams got startup capital (5,000 RMB), office space provided by Innovation and Entrepreneurship Incubation Centre of the college, and one-week guidance from faculty members and entrepreneurs on all projects undertaken. They got connections with local farmers, inheritor entrepreneurs of Tiaosheng, and e-platform representatives, providing opportunities for interaction and implementation on a larger scale. Practice sites have been set up at Danzhou townships of Nada, Yangpu, and Lanyang, through which students went on-site and made studies on conventional products, approached Tiaosheng inheritor entrepreneurs, and collaborated on agro-produces development.

As shown in Table 4, we operationalized key constructs using validated scales from existing literature, adapted to our specific context.

To calculate the digital empowerment composite index (H3), we have built a multidimensional index that combines: (1) Level of blockchain technology adoption (scaled from 0, meaning no adoption, to 2, meaning full adoption, based on whether more than or equal to 90% of products have been traced); (2) Student partnerships for vocational talent cooperation (number of partnerships); (3) E-commerce capability (scaled from 0, meaning no online presence, to 2, meaning active presence on multiple platforms); (4) Intensity of digital marketing (the number of content creation and interaction activities). All four dimensions are standardized and then averaged as equal-weighted indicators of a composite index. We acknowledge that equal weighting represents a simplification. Sensitivity analyses using Principal Component Analysis-derived weights yielded highly similar results (r = 0.94 with equal-weighted index), suggesting robustness to weighting specifications.

Consumer surveys were administered via Wenjuanxing, aiming at people who had previously made online purchases of agricultural products within 6 months, and screening questions helped identify people aware of traceability systems (Corallo et al., 2020). Experiments on consumer surveys used descriptive text of products, sometimes containing and sometimes lacking blockchain technology, permitting intra-group comparisons (Çallı and Ediz, 2023). Incomplete and discordant responses, ascertained through attention checks, were eliminated, leaving us with 568 valid responses from a total of 642 completed ones. Student evaluations used a pre-test and post-test quasi-experiment on a control group, and pre-test interviews took place in September 2023, testing for self-efficacy of entrepreneurs, digital competencies, and culture, followed by post-test interviews held in June 2024. There was a structured interview format, consisting of pre-test interviews held in September 2023 and follow-up interviews held in December 2024, and income verification was done through validation of household, sales in farmer co-ops, and e-commerce payment receipts. Blockchain system sales information was collected directly from the AntChain platform, providing objective sales records, and e-commerce sales information was collected via platform analytics reporting sales, customer, and demographics of purchasing customers.

To address these questions, we used a range of analytic techniques. Descriptive statistics helped describe distributions of and information about the variables of interest (Marshall and Jonker, 2010). Correlation analyses with Pearson’s r calculated bivariate associations of key variables (Weisburd et al., 2021). Structural equation modeling via AMOS Software Version 24.0 examined pathways of interrelationships postulated by and explained in Figure 1 above (Debnath and Khatri, 2024). To assess model fitness, we followed typical guidelines that included having χ²/DF < 3, Comparative Fit Index >0.90, Tucker–Lewis Index >0.90, and Root Mean Square Error of Approximation <0.08 (Sathyanarayana and Mohanasundaram, 2024).

For H2 and H3, we used Difference-in-Differences (DID) analysis as a strategy to evaluate causal effects of interventions, conditioning on time-invariant controls and shared time trends (Ryan et al., 2019).

By definition, the DID estimator evaluates changes over time between treatment and control groups as follows (Equation 1):

To address potential lack of exogeneity, we clustered standard errors at the individual/household level, allowing for correlation within observational units (Doss and Quisumbing, 2020). Additional methods aimed at assessing model robustness consisted of propensity score matching and placebo tests on pre-period observations, testing assumptions of parallel trends (Gavilanes, 2023).

Additional methods aimed at assessing model robustness consisted of propensity score matching and placebo tests on pre-period observations, testing assumptions of parallel trends (Gavilanes, 2023). To validate the DID approach, we assessed the parallel trends assumption through two methods. First, we examined pre- intervention growth rates (2022–2023) for available historical data and found no significant differences between groups (treatment: 6.8% vs. control: 7.2%, t = 0.18, p = 0.86). Second, we conducted placebo tests using pre-treatment periods, which yielded null effects, supporting the parallel trends assumption.

Verbatim transcripts of qualitative interviews conducted on farmers and students underwent thematic analysis through the utilization of NVivo 12 software (Susanti and Dani, 2025). Coding was conducted through an iterative approach that considered both deductive codes based on theoretical concepts and thematic induction based on emerging tacit patterns within the qualitative transcripts (Fereday and Muir-Cochrane, 2006), having achieved good inter-rater reliability (κ = 0.83).

Launched in September of 2023, workshops of participatory design assembled Tiaosheng culture heritage inheritor communities, farmers, vocational college design students, and food processing coop members. Teams developed three flagship product series based on co-creative workshops, integrating elements of Tiaosheng culture. Tiaosheng Zongzi consists of glutinous rice dumplings packaged with musical score designs and QR code links for performances and zongzi-making legends of local communities through audio format interactions. Cultural Red Fish Gift Sets feature dried red fish, a Danzhou local seafood, packaged with artistic illustrations of performances and narrations of Traditional fishing communities’ ties through Tiaosheng culture, told through songs throughout fishing excursions through history. Singing Coffee products label Danzhou coffee, grown since the early 20th century, and feature Tiaosheng culture through cultured design and artistic processing and links through QR codes, providing song and coffee pairing suggestions. Heritage inheritor communities are assured of accuracy, and vocational college design students went through ethnographic studies, observed performances, and met up with senior singers for interviews, influencing their designing, connecting traditionally and modernly acceptable design choices.

System implementation took place from January 2024 after infrastructure establishment and farmer training. The AntChain BaaS platform was developed and applied for agri-traceability, establishing digital IDs for participating farmer families numbering 28, spanning three townships. Every farmer received elementary hardware consisting of mobile phones, QR code printers, and educational literature, which took six-hour cohorts through blockchain technology, entry of production details, quality documentation, and marketing outreach coaching. Entering production details began in March 2024, as farmers keyed geolocation point entry, dates of planting, application of inputs, and quantities of produce harvested. In zongzi processing, ingredient origin details included information on rice varieties, meat that was certified, and processing companies that got registered, and red fish catching details collected fishing boat details, GPS, and methods of preservation. Third-party testing was incorporated through partnerships involving China’s Hainan Province Agricultural Product Quality Safety Center and actual laboratories, and test reports were submitted online and stored as immutable records on the blockchain system itself. Between March and December of 2024, a total of 856 batches of products received blockchain-verified tests, producing a pass rate of 98.7%. Logistical connections relied on partnerships with Cainiao Network and SF Express, and IoT sensor technology provided instant temperature and humidity readings automatically transmitted and pushed into and stored as blockchain records through API connections and applications. Consumer-interfacing QR coded links provided visualization of producer-origin journey stages through images and verification marks, and contained added-value content offering Tiaosheng voice files and videos of culture holders and carriers. Performance analysis indicated that fully 63.7% of consumers accessed and scanned QR coded links, averaging 2 min and 17 s per scanning instance.

Implementation of the curriculum at Hainan Vocational and Technical College combined theoretical education and practical implementation. Topics of Intangible Heritage Creative Design included basic courses on cultural anthropology, intellectual property, and design thinking, complemented by lectures by Tiaosheng masters offering tangible culture transmission and contact through performances and ethnographic research. Assignments involved designing original concepts for cultural creative products and marketing them. Blockchain Applications in Agricultural Supply Chains taught basic concepts about blockchain technology and programming of smart contracts, complemented by practical experiences implementing blockchain networks based on Hyperledger Fabric technology and case studies of the AntChain platform used within the Danzhou project implementation. Topics of Rural E-commerce Operations and Digital Marketing included ecosystem approaches, designing marketing content for short online videos, and customer relationship management, complemented by practical experiences involving managing actual e-commerce businesses and implementing marketing campaigns that could be measured weekly based on actual outcomes. Entrepreneurial groups of students combined voluntarily based on common interests, and 12 groups came together, eight of whom successfully opened businesses that reached survival and revenue goals by December of 2024, as indicated within Table 5, adopting varied business models and market positioning techniques.

The four teams that could not attain success milestones experienced issues such as conflicts, supply chain, and market demand, offering rich learning experiences that helped modify the curriculum design for future batches. Successful teams showed remarkable improvement on a range of dimensions, and themes that emerged from qualitative interviews included empowering confidence, applications of business acumen, and overall pride of culture. Students appreciated that their familiarity and interaction experiences with Tiaosheng heritage helped them view rural development differently and offered them competitive advantages due to blockchain awareness.

Digital marketing campaigns utilized multi-channel approaches, targeting varied consumer segments. Short videos on Douyin showed great success, capitalizing on its more than 700 million daily active user base. Student groups created high-quality videos featuring Danzhou landscapes, Tiaosheng performances, production process shots, and farmer interviews, using popular music and challenges on the platform. These viral video series, ‘24 Hours of a Tiaosheng Song Performer and Farmer,’ followed a 67-year-old heritage inheritor through daily activities of coffee harvest, zongzi-making, and nighttime singing practice, racking up a total of 3.27 million views over six episodes and generating heavy e-commerce traffic and emotional responses through comments on these videos available on the platform. E-commerce live streaming enabled immediate interactions on products, occurring on a weekly basis, including introductions, cooking shows, and performances of local culture and heritage. Peak concurrent viewers of up to 5,000–15,000 viewers on successful e-commerce live streaming sessions generated sales of up to 200,000–50,000 RMB during festive seasons, though other marketing methods, namely search engine optimization and corporate gifting, explored other market segments and opportunities on these platforms.

Results from consumer survey research (N = 568) showed pronounced positive effects of blockchain traceability on trust and purchasing intention. As indicated in Table 6, products that possess traceability elements received significantly higher ratings on all tested parameters than similar products lacking traceability information.

Analysis of QR code scan behavior based on blockchain backend information indicated that 63.7% of customers scanned product codes, and of these, 34.2% did this multiple times and on different purchasing events, indicating that scan frequency was positively associated with customer lifetime value (r = 0.47, p < 0.001). Content interaction behavior indicated that interaction was focused on culture-related narratives (mean viewing time of 68 s), followed by producer profiles (51 s), as opposed to more technical details about logistics (23 s) and quality certificates (31 s), suggesting that more emotional and storied content was considered more compelling than technical content on this platform and that technical details of traced products need not be interacted with as they will be provided through other means. Analysis of QR code scan behavior from blockchain backend data indicated that 63.7% of customers scanned product codes, with 34.2% scanning multiple times across different purchases. Scan frequency was positively associated with customer lifetime value (r = 0.47, p < 0.001). Content interaction analysis revealed that consumers engaged most with cultural narratives (mean viewing time: 68 s), followed by producer profiles (51 s), compared to technical details such as logistics tracking (23 s) and quality certificates (31 s). This suggests that emotional storytelling and cultural connections resonate more strongly with consumers than technical transparency features, highlighting the complementary role of cultural capital and blockchain technology in creating consumer value.

Results of entrepreneurship education had dramatically positive effects on experimental participants, as compared to control settings. As seen in Figure 2, success rate of entrepreneurship (defined as survival of venture for more than 6 months and average monthly revenue of more than 3,000 RMB), was significantly increased (53.3% or 32 of 60 participants, as compared to control of 16.7% or 10 of 60 participants). There was a highly significant difference of 3.2-fold, as calculated by Cramér’s V statistics (χ² = 17.24, p < 0.001, Cramér’s V = 0.38).

Competency assessments using validated entrepreneurship skills scales revealed significant improvements across multiple dimensions. Table 7 presents pre-test and post-test comparisons for the experimental group, demonstrating substantial learning gains.

Financial projection of project viability for successful ventures of students showed economic viability and growth potential. As indicated by Figure 3, cumulative revenues of eight successful teams totaled 455,600 RMB for the 12-month period of operation (January–December, 2024). The average revenue for each of these teams was 56,950 RMB. There was a remarkable growth in monthly revenues, averaging 9,475 RMB per team by December, up from averages of 2,150 RMB per team for each of the first months, indicating a growth of 4.4 times. Profit margins ranged from 23 to 31% after deducting costs of products, logistics, marketing, and commission fees.

Interviews showed that these students experienced transformative learning that went beyond technical mastery and covered personal development, career goals, and societal engagement. As one of them explained, “This program has completely altered the direction of my career. Originally, I was going to work for a large corporation and be based in a large city, but now I would like to begin a social enterprise that assists rural communities. Entrepreneurship can have a beneficial social impact that is financially sustainable as well.” Another explained, “I have gained respect for traditional knowledge through interacting with the Tiaosheng inheritors, and there is a responsibility for our generation to connect tradition and modernity.”

Employment outcomes for experimental group students had better quality and relevance for regional development goals. Six months post-graduation (June 2024 cohorts evaluated as of December 2024), employment or establishment of businesses pertinent to rural e-commerce, culture and creative industries, or agriculture chain management was achieved by 78% of experimental group students, as opposed to only 34% of control group students. Starting salaries of experimental group members averaged 6,800 RMB/month, or about 35% more than that of control group members (5,050 RMB/month).

There was a significant increment in the income of participating farmers due to digital empowerment interventions. As indicated in Table 8, an average of 50.2% growth in yearly income was observed for the 28 participating families in the blockchain traceability project and cooperation with students, as against a 7% growth for the control families of 14 members, resulting in a difference of 43.2 percentage points (p < 0.01).

The DID analysis specification (Equation 2):

where Treat indicates project participation, Post indicates time period (2024 vs. 2023), X represents control variables, and β 1 captures the causal effect estimate (+13,870 RMB, p < 0.01 , robust standard errors clustered at household level).

Results from propensity score matching (PSM) on observed pretreatment variables, used as a robustness check, showed consistent estimates. When matched on household size, farmland, initial income, education, and distance to town center, the average treatment effect on the treated (ATT) was estimated at +14,230 RMB (p < 0.01), fully consistent with estimates from DID.

Income source diversification was an essential mechanism that helped drive improvement. Farmers, who earlier relied heavily on sales of their commodity products (which constituted 87% of their source of income before intervention), diversified their income sources and integrated other components, such as premiums from sales of cultural creative products, e-commerce sales that reduce middleman markup, and other sales generated through cooperation and fees for operating services through blockchain technology.

Income distribution analysis, as illustrated by Figure 4, shows that this intervention positively affected farmers across all income groups, though this was more pronounced for middle-class farmers. Less wealthy farmers had some capacity barriers that would have impeded them from fully participating (adopting technology, for instance, and producing enough products), and ceiling effects existed for top income farmers, as these farmers’ operations would already be optimal. There is a need for supplementary measures that would enhance farmers’ capacity, especially from more marginalized groups.

Qualitative interviews conducted with farmers helped reveal methods and experiences. Speaking about this issue, a 54-year-old farmer growing and processing coffee explained, “I used to sell beans for 30 RMB per kilogram of beans through middlemen, but now, thanks to online purchases that demonstrate producer story and quality, I can sell them for 68 RMB per kilogram and help consumers trust me.” Another zongzi producer explained, “Some innovations came through cooperation with college students, including teaching me about photography for advertising on social media platforms, designing attractive packaging, and giving me sales opportunities through city connections. Since then, my revenue has actually doubled, and my daughter has considered becoming a farmer.”

Job creation went beyond generating revenue for farmers. It is estimated that the project created about 68 jobs, including packaging personnel, logistics coordinators, quality controllers, content developers, and customer care personnel. These opportunities went, in many cases, to youth and women from rural areas, thereby countering urban migration and keeping families united.

Spillover effects occurred in other neighboring communities. Non-participating farmers reported increased awareness of e-commerce opportunities, and seven families began their own e-commerce ventures based on the success of project participants. Neighboring co-ops showed interest in implementing blockchain technology, and township governments asked for project expansion into other villages. These effects demonstrate the potential for economic transformation on a regional scale.

Comprehensive sustainability assessment across economic, social, and environmental dimensions revealed positive outcomes. Table 9 presents key indicators with baseline (2023) and current (2024) values.

Progress made on sustainability issues concerning the environment is of great importance, especially due to worldwide worries about the effects of agriculture on issues of climate change, loss of biodiversity, and resource depletion. Nowadays, people are increasingly focused on healthy and sustainable eating habits, and there has been progress made through this project on local ingredient sourcing, shorter transportation distances, and sustainable alternative materials used for packaging, as these have made significant changes to decrease carbon footprint. Life Cycle Assessments (LCAs) made on products following ISO 14040/14044 guidelines have helped identify and calculate present GHG emissions along the actual lifecycle of products from agriculture through consumption delivery of products. Use of local ingredients has helped decrease ‘food miles’ by as much as 69%, thereby eliminating GHG emissions due to transporting products over long distances through trucks and planes. Biodegradable packaging materials made of bamboo or cornstarch plastics have helped.

Thus, organic farming practices gained more popularity as economic benefits from premium market accessibility through blockchain-traceability encouraged sustainable practices. Also, organic practices had simplified documentation through blockchain technology, thereby removing organizational hurdles. Additionally, farmer testimonials showed that consumers’ preparedness to pay more for organic, sustainably grown products made sustainability financially attractive, as before, when margins were low, sustainable intensification was not financially viable.

Increased water use efficiency was made possible through precision agriculture technology, and crop rotation was encouraged through market feedback systems. Sales information helped in more accurate production, reducing post-production losses due to overproduction. Market transparency about sustainable activities through blockchain technology enticed environment-sensitive consumers, generating positive loops between sustainable investment and market benefits.

To examine these hypotheses, structural equation modeling was conducted on the proposed paths involving blockchain traceability and transparency, vocational training intensity, consumer trust, purchase intention, success of entrepreneurs, and growth of farmer income. Good model fit was confirmed for the final model by its χ²/df figures of 2.34, CFI of 0.946, TLI of 0.931, and RMSEA of 0.061, indicating that the proposed model was a very close representation of observed covariance matrices.

As illustrated in Figure 5, the structural equation model path analysis results strongly supported all three hypotheses. Before examining the structural paths, we assessed the measurement model to ensure construct validity. Measurement model diagnostics confirmed adequate convergent and discriminant validity. Factor loadings for all indicators exceeded 0.70 (range: 0.73–0.91), indicating that observed variables adequately represented their underlying latent constructs. Composite Reliability (CR) values ranged from 0.88 to 0.94 (all exceeding the 0.70 threshold), demonstrating internal consistency of the measurement scales. Average Variance Extracted (AVE) values ranged from 0.68 to 0.82 (all exceeding the 0.50 threshold), supporting convergent validity (Sathyanarayana and Mohanasundaram, 2024). Furthermore, the square root of AVE for each construct exceeded inter-construct correlations, establishing discriminant validity and confirming that constructs measured distinct theoretical concepts.

Path coefficients in the model strongly supported all hypotheses proposed. In hypothesis one, blockchain traceability and transparency positively impacted consumer purchase intention, having a total effect of β = 1.02 (p < 0.001), through its direct effects (β = 0.68) and indirect effects mediated through consumer trust (β = 0.34). These findings strongly supported theoretical expectations that as transparency increases, risk perception is decreased and more customer confidence is gained about products and services provided through them. In hypothesis two, vocational training intensity positively and significantly predicted success of entrepreneurs (β = 0.54, p < 0.001). Additional tests showed that practical experience and mentorship had great influential value on vocational trainings undertaken by them. In hypothesis three, digital empowerment composite index had a strongly positive effect on growth rate of farmer income (β = 0.72, p < 0.001). Blockchain, co-operation of students, and e-commerce and online marketing facilities significantly explained its effects on growth of farmer income. Moderated regression analysis showed that blockchain and purchase intention relationship was significantly observed only for people having high cultural interests (interaction effect β = 0.29, p < 0.01), and powerful mentorship had great effects on vocational trainings (interaction effect β = 0.21, p < 0.05). All hypotheses explored by this model explained large portions of variance, having a value of R² of 0.67, 0.51, and 0.74, for purchase intention, success of entrepreneurs, and growth of farmer income, respectively.

Government authorities should prioritize strategic investments in digital rural infrastructure to enable widespread adoption of blockchain-based agricultural traceability systems. Counties with significant cultural heritage resources and agricultural production should allocate 5–10 million RMB over three years for blockchain infrastructure deployment, including the establishment of “Digital Village Centers” that provide internet connectivity, computer facilities, and technical training resources accessible to rural communities. To reduce adoption barriers for smallholder farmers, government subsidies should cover approximately 80% of blockchain system implementation costs, which average 2,800 RMB per household for initial setup. This targeted subsidy approach ensures that resource-constrained farmers can participate in digital value chains without prohibitive upfront investment, thereby promoting inclusive rural development.

Regulatory frameworks must evolve to support the emergence of blockchain-verified cultural creative agricultural products as a distinct market category. Provincial authorities should develop comprehensive blockchain traceability standards for agricultural products by 2026, establishing technical specifications, data management protocols, and verification procedures that ensure system integrity while remaining accessible to small-scale producers. Complementary certification mechanisms should be created specifically for cultural creative agricultural products, recognizing their unique value proposition that combines food quality, cultural heritage, and sustainable production practices. Furthermore, dedicated funding programs with annual budgets of 50–100 million RMB should be established to support cultural heritage commercialization projects that integrate traditional knowledge with modern digital technologies, thereby creating economic opportunities while preserving intangible cultural assets for future generations.

Vocational education institutions must undertake comprehensive curriculum reforms to equip students with competencies relevant to the digital rural economy. E-commerce and entrepreneurship programs should integrate specialized modules on blockchain technology applications and cultural heritage creative design, with a minimum requirement of 120 credit hours dedicated to these emerging fields. This curriculum expansion should emphasize practical, project-based learning that connects students with real-world rural development challenges. Mandatory industry partnerships should require all students to complete at least 3 months of practical placement with agricultural cooperatives, cultural creative enterprises, or rural e-commerce ventures, ensuring that theoretical knowledge translates into applicable skills. Institutions should establish dedicated Cultural Creative Product Development Centers with minimum annual operating budgets of 500,000 RMB, providing physical infrastructure, equipment, and resources that enable student innovation teams to develop, prototype, and launch cultural creative products in collaboration with rural communities and heritage inheritors.

Faculty development initiatives are essential to support these curriculum transformations effectively. Vocational institutions should implement systematic training programs targeting 100 or more faculty members annually, focusing on digital technologies, blockchain applications, cultural heritage preservation and commercialization, and rural entrepreneurship ecosystem development. To bridge the gap between academic instruction and industry practice, institutions should recruit experienced practitioners from e-commerce platforms, blockchain enterprises, cultural creative industries, and successful rural ventures as adjunct instructors, with a target of at least 20% of teaching staff coming from industry backgrounds. This dual-faculty approach ensures that students receive both theoretical foundations and practical insights informed by current market realities, while also strengthening industry-education partnerships that facilitate student employment and entrepreneurial venture success.

Farmers and agricultural cooperatives should pursue collective approaches to blockchain technology adoption to achieve economies of scale and reduce per-household costs. The formation of farmer cooperatives with a minimum of 20–30 participating households enables shared investment in blockchain infrastructure, coordinated quality management, unified branding strategies, and enhanced bargaining power with buyers and platform providers. Individual farmers should actively seek partnerships with vocational student entrepreneurship teams, aiming to collaborate with 2–3 teams annually to access marketing expertise, branding support, digital content creation, and e-commerce operational assistance. Initial investments in basic digital infrastructure, including smartphones with adequate data capabilities and QR code printing equipment, typically require 3,000–5,000 RMB per household and represent essential enablers for participation in blockchain-verified supply chains and direct-to-consumer digital marketing.

Market development strategies should prioritize the creation of distinctive brand identities that authentically communicate cultural heritage narratives while signaling quality and traceability. Farmers and cooperatives should work with design professionals and cultural heritage experts to develop brand systems that visually and narratively express the unique cultural stories embedded in their products, differentiating them from generic commodity offerings. E-commerce presence should be established on a minimum of two major platforms such as Taobao and Douyin, recognizing that multi-platform strategies reach diverse consumer segments and reduce dependence on any single distribution channel. Collaborative marketing efforts, potentially coordinated through farmer cooperatives or industry associations, should target at least 50,000 potential consumers within the first year through strategic content marketing, influencer partnerships, live-streaming commerce, and participation in agricultural product festivals and cultural heritage promotion events. Active participation in industry associations and ongoing dialogue with policymakers enables farmers and businesses to advocate for supportive policies, share best practices, and contribute to the evolution of standards and regulations governing cultural creative agricultural products.

The integrated model presented in this research aligns with multiple United Nations Sustainable Development Goals, including SDG2 (Zero Hunger), SDG8 (Decent Work and Economic Growth), SDG9 (Industry, Innovation and Infrastructure), SDG12 (Responsible Consumption and Production), and SDG17 (Partnerships for the Goals). Our findings demonstrate how cultural heritage and traditional knowledge can be strategically leveraged as economic resources through the complementary application of digital technology, human capital development, and supportive institutional frameworks. For developing countries confronting the intersecting challenges of rural poverty, agricultural sustainability, cultural heritage loss, and youth unemployment, this research offers a comprehensive approach that addresses multiple development objectives simultaneously rather than treating them as competing priorities. The study challenges conventional assumptions about inevitable trade-offs in sustainable food systems by demonstrating that transparency, cultural meaning creation, and direct producer-consumer connections can simultaneously enhance producer livelihoods, consumer satisfaction, and environmental sustainability, offering a viable pathway for rural transformation in an era of climate change and evolving consumer values.