Empirical model of public participation intention in biomass project based on integrated TAM-TPB model

Abstract

Effective implementation of sustainable biomass projects requires active participation and public acceptance to supply and transport biomass feedstock. Therefore, this study aims to identify and measure the factors encouraging the public to participate actively in biomass projects and suggest strategies to increase their participation. This study started with a systematic literature review and conceptual model development by integrating the Technology Acceptance Model (TAM) and Theory of Planned Behavior (TPB) models. In addition, this study considers other relevant and significant variables such as education level, environmental concern, experience, information, and self-transcendence values. This study employed purposive sampling to target corn farmers and wood artisans in Banyuwangi. Then, data analysis was conducted on 75 collected questionnaires (75% response rate) using Partial Least Square-Structural Equation Modelling (PLS-SEM) with SmartPLS 4.0. The result of hypothesis testing showed that 13 of the 18 proposed hypotheses are significantly supported; one hypothesis is significant with a different direction from the study proposed, and four are not significantly supported. According to the PLS-SEM calculation, the most influential factor for Intention (INT) is Perceived Behavioral Control (PBC). The study indicates that people will most likely participate in a biomass project by supplying or transporting biomass feedstock if they can devote their time or resources. This study’s results offer crucial insights for the power generation industry and policymakers regarding implementing sustainable biomass projects, equipping them with actionable insights to increase public participation.

Introduction

As Indonesia’s economy grows, the nation’s energy supply in 2023 is projected to increase by 1.55% from the previous year, reaching 1,853 million barrels of oil equivalent (BOE), the highest in six years (refer to Fig. 1). Although there was a slight decrease in the supply of fossil fuels such as crude oil, its derivatives, and coal compared to the previous year, natural gas and renewable energy (NRE) sources grew by 3% and 13.8%, respectively in 2023. The energy mix is still primarily composed of coal at 39.69%, followed by petroleum at 29.91%, natural gas at 17.11%, and NRE at 13.29%. According to the General National Energy Plan, the government is committed to increasing the share of NRE by 23% in the primary energy mix in 2025 and 31% in 2050. Compared to target of achievement in 2025, there is still a shortage about 11,71%. In the electricity sector, the target has only produced 65.23 terawatt-hours (TWh) in 2023, accounting for 18.60% of the total national production (about 350.6 TWh) (see Fig. 2).

Fig. 1

The Indonesia’s Energy Supply (Source: Anditya et al.¹).

Fig. 2

Electricity Generation (TWh) (Source: Anditya et al.¹).

Indonesia’s State Electricity Company has launched an initiative to meet NRE targets in the electricity sector by using biomass as a co-firing fuel in coal-fired power plants². Biomass co-firing involves substituting a portion of coal with biomass while ensuring the plants’ fuel quality and operational feasibility^3,4. Biomass consists of renewable organic materials from plants and animals, such as wood pellets, oil palm kernels, corncob, or sawdust, which can be burned directly for heat or processed into liquid or gaseous fuels. There are 114 coal-fired power plants in Indonesia that could potentially adopt biomass co-firing^5,6. The goal of this program is to utilize 10.2 million tons of biomass in 52 coal-fired power plants by 2025, which would contribute 3.5% to the new and renewable energy mix, reducing carbon and greenhouse gas emissions by 11 million tons annually. By the end of 2020, the completed biomass co-firing trials at 29 coal-fired power plants, ranging from Kalimantan to Java. Among the 29 plants, six are already in commercial operation: Paiton, Pacitan, and Surabaya in Java; Jeranjang in West Nusa Tenggara; and Ketapang and Sanggau in West Kalimantan⁶. As of May 2022, 32 coal-fired power plants had implemented co-firing technology, produced 487 MWh of electricity and reduced CO2 emissions by 184,000 tons as of April 2022⁷. However, around 20 coal-fired power plants must implement biomass co-firing by 2025.

Reaching the target may be more complex than it sounds; it could be for several reasons, such as technical difficulties related to integration with existing infrastructure⁸, the high operational cost of the feedstock supply chain process⁹and farmers’ willingness to supply biomass is contingent upon market demand and profitability¹⁰. Although the tropical country of Indonesia is entitled to enormous forests, agricultural land, and various biomass sources, such as wood plantations, rice husks, wood scraps, and leftover corn (of the 74.4 million hectares of production forest, over 1.3 million hectares are reserved for plants related to the energy industry)¹¹, the supply chain of biomass raw materials has not been well developed⁴. Regardless of technical factors, getting public support to develop a supply chain of biomass raw materials is essential since many biomass sources come from community plantations, agriculture, or farmers. The intention and ability of farmers or producers to supply biomass or alternative biofuel feedstocks and their participation in other biomass projects still need to be thoroughly studied. Then, related to the intention of farmers to participate, some previous studies have highlighted the willingness of farmers to supply biomass, such as Altman et al.¹², Caldas et al.¹³, Convery et al.¹⁴, and Thompson & Tyner¹³. Altman et al.¹² found that price significantly determines farmers willingness to supply biomass. Higher prices generally increase the likelihood that farmers will participate in supplying biomass. Besides price, factors such as farm size, type of crops grown, and previous experiences with biomass markets also play critical roles in influencing farmers’ decisions. Caldas et al.¹⁰ found that economic factors, such as expected profitability and market access for biofuel feedstocks, significantly influence farmers’ willingness to grow alternative biofuel feedstocks across Kansas. Farmers are more likely to consider planting alternative feedstocks if they perceive a strong market demand and financial viability. Farmers’ environmental attitudes also played a crucial role; those who valued sustainability and environmental benefits were more inclined to grow biofuel crops. In line with Caldas et al.¹⁰, Convery et al.¹⁴ also found that, besides economic incentives, farmers’ awareness of bioenergy technologies and their understanding of the benefits associated with biomass production played a crucial role. Environmental concerns also influenced farmers’ decisions, and supportive policies and programs from government entities and agricultural institutions positively impacted farmers’ willingness to engage. Thompson and Tyner¹³ found that market prices significantly influence farmers’ willingness to supply corn stover. Higher prices for corn stover lead to increased supply from farmers, highlighting the price elasticity of supply in this context.

So, giving the point that developing a biomass supply chain relies heavily on public support, mainly when biomass sources come from community plantations or agriculture or farmers, and previous studies have consistently shown that multiple factors influence farmers’ willingness to supply biomass (such as price, farmers’ awareness and education, and supportive policies and programs), researching the intentions and capabilities of farmers within theoretical models like Theory of Planned Behavior (TPB) (first developed by Ajzen¹⁵) and the Technology Acceptance Model (TAM) (first developed by Davis¹⁶) becomes imperative for several reasons:

• 1. These models are theoretical constructs and practical tools for understanding and predicting behavioral intentions. They provide a framework for elucidating how psychological factors influence farmers’ intentions to adopt new technologies or participate in novel activities such as biomass supply.

• 2. The models could be used to identify the critical success factors. By examining factors such as perceived usefulness, ease of use, subjective norms, attitude toward behavior, and control beliefs within these models, researchers can pinpoint critical success factors necessary for enhancing farmer participation.

• 3. The insights from applying the models can guide intervention strategies to improve public perception and farmer engagement in biomass projects. This practical application of the research will not only enhance the understanding of farmer behavior but also provide actionable strategies for improving the effectiveness of biomass projects, thereby engaging the audience in the potential real-world impact of the research.

Ultimately, thorough research into farmer intentions using established theoretical frameworks like TPB or TAM will provide invaluable insights for crafting effective policies and interventions to optimize biomass supply in Indonesia. This move aligns perfectly with achieving renewable energy targets, a noble and urgent goal. The TAM complements the TPB by focusing on how perceived ease of use and perceived usefulness of technology influence acceptance. In the context of biomass projects, understanding how farmers perceive the technologies involved in biomass production—such as harvesting equipment or processing methods—can significantly impact their intention to engage. If farmers believe that adopting these technologies will enhance their productivity and profitability, their willingness to participate will likely increase, contributing to the larger mission of sustainable energy.

Hence, this study established three objectives in light of the above discussions: (i) identify the factors that encourage the public to participate actively in biomass projects, (ii) measure the influence of those factors on the public’s willingness to participate actively in biomass projects, and (iii) provide recommendations for encouraging public participation in biomass projects and improving their sustainability.

The structure of this paper is as follows. This manuscript meticulously executed a systematic literature approach in Section “Literature Review”. Section “Research Methodology” presents a comprehensive discussion of the hypothesis and the conceptual model, the research sample, the measurement tool, and the data analysis tool related to the intention to participate in the biomass project. The characteristics of the respondents, the outcomes of the measurement test, the assessment of the structural model, and the results of the hypothesis testing are all covered and explained in Section “Results”. Section “Conclusion and Recommendation for Further Research” presents the research’s conclusion, limitations, and recommendations.

Literature review

A review on biomass for electricity generation in Indonesia

Natural gas and renewable energy sources (NRE) are regenerative or have an endless capacity to produce energy. “Renewable” refers to energy derived from natural resources such as solar, wind, rain, tides, geothermal, hydropower, biomass, and biofuel¹⁷. Renewable energy sources have a minimal environmental impact and are considered sustainable as they are naturally replenished and do not emit contaminants like carbon dioxide¹⁸. Among these, biomass is a significant energy source with several advantages over traditional fossil fuels and other renewable sources. It is a widely available energy source that can be gained from agricultural residues, forest products, and urban waste, reducing dependence on imported fuels and enhancing energy security. Unlike solar and wind energy, which are intermittent, biomass stands out for its ability to provide a continuous and reliable energy supply, making it a crucial part of the renewable energy landscape¹⁹. The depletion of conventional fossil fuels has led to a rising need for renewable energy, with a particular focus on biomass energy consumption²⁰.

The primary potential source of biomass projects is wood, crops, and their waste, waste from animal and food processing, algae, and aquatic plants²¹. Waste from forestry and agricultural activities (including rice husks, corn stalks, and palm oil waste) represents an underutilized potential biomass resource in Indonesia. With over 86 million hectares of forest and 8 million hectares of agricultural land, Indonesia is well-positioned to harness these natural resources for biomass-based electricity generation^22,23. The diversity of biomass sources offers a strong foundation for sustainable electricity generation, supporting both rural electrification and urban energy demands. In addition, a review of the mapping analysis of biomass potential on Java highlighted the importance of biomass residue in contributing to the country’s energy development trajectory and reducing environmental impact²⁴.

According to Primadita et al.²⁰, Indonesia has employed several technologies for biomass electricity generation, including direct combustion, co-firing with coal, gasification, refuse-derived fuel (RDF), sanitary landfills, and incineration. Out of these, direct combustion is the most widely used method (82%) due to its simplicity and lower cost, where biomass is burned either alone (firing) or mixed with coal (co-firing) to generate heat for steam turbines. However, the reliance on cheaper fossil fuels like coal and oil has slowed the development and adoption of more advanced biomass technologies²⁵. The high costs associated with biomass processing technologies and the need for technological advancements have hindered the transition to bio-mass as a primary energy source.

To address the obstacles in biomass development, such as cost and technological barriers, the Indonesian government has enacted a series of regulations to support biomass development. Key regulations include Ministerial Regulation of Energy and Mineral Resources #21/2016, which facilitates the purchase of electricity from biomass and biogas power plants, and Ministerial Regulation of Energy and Mineral Resources #39/2017, which regulates the physical implementation of renewable energy projects. Further, Presidential Regulations #18/2016 and #35/2018 focus on accelerating the development of environmentally friendly waste-to-energy power plants, essential for biomass electricity supply in urban areas. These policies demonstrate the government’s commitment to expanding the biomass sector and are crucial for attracting investment in bioenergy projects²⁶.

The potential for biomass electricity generation in Indonesia is promising; it’s a beacon of hope for a sustainable future. With several new power plants and waste-to-energy facilities in the pipeline, focusing on palm oil products and urban waste using direct combustion technologies, the future looks bright. The expansion of these projects, alongside rural electrification initiatives, is set to drive the development of biomass power plants in off-grid areas. As stated by Lantasi et al.²⁷, the continued development of biomass power generation is a critical component of Indonesia’s strategy to reduce its reliance on fossil fuels and meet its renewable energy targets by 2025 and beyond.

To effectively implement biomass for electricity generation, Indonesia requires improved policies and strategies prioritizing biomass utilization. This includes fostering investment in technology and infrastructure to support biomass energy projects and developing regulations that incentivize the use of biomass and address the environmental impacts associated with biomass production. Enhanced policies could facilitate the adoption of biomass technologies and improve the overall energy mix, moving towards a more sustainable energy future. According to Susanto Sadirsan et al.²⁸, the development of renewable energy, including biomass, is also influenced by factors such as the balancing between supply and demand, the readiness of the electricity system to tap renewable energy power generation at competitive costs, and the availability of alternative potential industrial biomass raw materials in local areas. Therefore, public participation is more than just meaningful; ensuring a steady supply of raw biomass materials is crucial..

Previous research related to biomass, technology acceptance model, and theory of planned behavior

This study used a systematic approach to the literature review to identify and analyze the previous research materials related to the objectives of the current study and also to find other factors that may be important and have a significant relationship with the variables used in TAM and TPB. Restrictions on keywords are utilized to find relevant articles. In titles, abstracts, and keywords, the terms “biomass” AND "technology acceptance model" OR "theory of planned behavior" is utilized as search queries, following the aim of this study, which discusses the integrated TAM and TPB for modeling the public intention to participate in biomass project. The search found 24 articles between 2008 and 2023. Based on documents published each year, the peak year was 2018, followed by 2023. However, several articles are excluded because of the non-user acceptance model, which is inaccessible, and they discuss willingness to pay instead. In the end, this research found 14 relevant articles to be used as reference, as shown in Table 1.

Table 1 The previous research.

As indicated in Table 1, prior research suggests that more thorough research still needs to be done on the models explaining the factors influencing public intention to participate in biomass projects. Research discussing the acceptance model of biomass in general is also uncommon. Therefore, this study can contribute to the present research by integrating the TAM and TPB models to examine the issue of public participation in biomass projects. More detailed, based on the previous research listed in Table 1, the study novelty can be explained as follows.

• The integration of theoretical models: The study combines two well-established behavioral models—the Technology Acceptance Model (TAM) and the Theory of Planned Behavior (TPB)—to create an integrated framework for understanding public participation in biomass projects. This integration allows for a more comprehensive analysis of the factors influencing public intention to participate.

• Focus on biomass projects: While previous studies have examined public perceptions of renewable energy in general, this study explicitly targets mass projects. This is significant because biomass has unique characteristics and challenges compared to other renewable energy sources.

• Empirical approach: The study aims to develop an empirical model based on real-world data rather than relying solely on theoretical constructs. This approach can provide more practical insights for policymakers and project developers.

• Dual participation perspective: The research considers public participation in providing biomass feedstock and other forms of active involvement.

• Regional context: The study focuses on Java Island, providing insights into public participation intentions in a specific geographical and cultural context. This localized approach can help tailor strategies for enhancing public engagement in biomass projects in the region.

• Application to power generation industry: The research addresses a critical sector for sustainable energy transition by targeting the power generation industry. This focus can provide valuable insights for industry stakeholders and policymakers.

Research methodology

Development of research models and hypotheses

According to Yazdanpanah and Forouzani⁴³, the Theory of Planned Behavior is an extensible psychology theorem developed by Ajzen¹⁵ that predicts people’s intention and actual behavior by using the variable of attitude, perceived behavioral control, and subjective norm. On the other hand, the Technology Acceptance Model (TAM) was first presented by Davis in 1985 as the expansion of the Theory of Reasoned Action (TRA), in particular, and social psychology theory⁴⁴. TRA implies that intention influences behavior, which begins with the effect of belief towards attitude. Accordingly, the four original TAM variables established by Davis^16,45 are perceived usefulness (PU), perceived ease of use (PEU), behavioral intention, and attitudes.

Hsiao and Tang⁴⁶ stated that TAM and TPB have an explanatory power of up to 64% and 61%. TPB and TAM expanded on the ideas of Ajzen and Fishbein’s Theory of Reasoned Action (TRA). As a result, the two models can be integrated¹⁵. When the TPB model and TAM are integrated, it will be possible to analyze internal and external elements that affect a person’s intention to use or participate in something. Combining the theory of planned behavior (TPB) with the technology acceptance theory (TAM), Taylor and Todd⁴⁷ presented an expanded model that aligned with their study. Chang et al.⁴⁸ used the C-TPB-TAM to examine the telehealth system’s purpose. Then, using the C-TPB-TAM, Chen and Chao⁴⁹ investigated the purpose of public transportation transfer and green smart mobility. Then, TAM or TPB or integrated or TAM and TPB are often extended with some variables to encompass broader psychological and sociological aspects, making them more comprehensive frameworks for studying human behavior in various technological contexts^{50,50,51,52,54}. Abdullah et al.⁵⁰ incorporate self-efficacy, enjoyment, computer anxiety, and experience as precursor background variables of the TAM to use the e-portfolio system for learning. Lee et al.⁵¹ incorporates environmental variables as precursor background variables of the TPB to predict quitting-related intentions. Wang et al.⁵⁴ and Prakash et al.⁵² have taken the practical applications of TAM and TPB a step further by integrating them with external variables. Wang et al.⁵⁴ focused on patient-centered factors, while Prakash et al.⁵² explored the level of openness and dominance in small enterprises’ intention to use social networking sites (SNS) in their business activity. More recently, Tang and Jiang⁵³ used trust in the integrated model of TAM and TPB to explain the behavioral intention in the sharing economy.

Based on previous studies, this study uses the integrated model of TPB and TAM withs some additional experience and information as external variables and level of education as a moderating variable. The detailed explanation of the hypotheses proposed in this study can be seen as follows. As seen in Table 1,

Attitude (ATT), perceived behavioral control (PBC), and subjective norm (SN) towards intention to participate (INT)

Intention is the main TPB driving force behind behavior, indicating the willingness to attempt something new and the amount of work someone intends to put into it. The intention of entrepreneurs to develop a renewable energy portfolio by transforming horse manure into bioenergy is presented in a previous study by Lupoae et al.²⁹. In this case, Lupoae et al.²⁹ found that attitude, perceived behavioral control, and subjective norms are important determinants of entrepreneurial intention used in the TPB. Research conducted by Zhu⁵⁵ has also shown that attitude, perceived behavioral control, and subjective norms are significant predictors of the public’s smog-reducing behavior. These results align with earlier studies that utilized the TPB model⁵⁶. De Jong et al.⁵⁷ state that attitude is the outcome of an individual’s assessment after considering the benefits and drawbacks of a specific behavior. Wang & Tsai⁵⁸ implied that participation levels are positively impacted by attitude. One’s willingness to collaborate may be increased by having a positive attitude toward substation initiatives^59,60. In this instance, people will be more willing to participate in a biomass project if their attitudes toward the project are favorable⁶¹.

Moreover, a person’s assessment of how easy or difficult it is to manage a specific action is known as perceived behavioral control¹⁵. In this case, a person will have a solid intention to carry out the behavior when their perceived behavioral control over it is sufficiently strong⁶². Several prior studies have demonstrated this relationship, such as Kharuhayothin et al.⁶³, Suntornsan et al.⁶⁴, and Qalati et al.⁶⁵. From their research, Kharuhayothin et al.⁶³ proved that individuals will effectively carry out an action or engage in something if they can devote their time or resources to it. Then, according to research conducted by Suntornsan et al.⁶⁴, although perceived behavioral control did not directly affect behaviors, it had a significant effect on the intentions of energy-saving behaviors of students with physical impairments. Qalati et al.⁶⁵ found a positive and significant effect of perceived behavior control on households’ intention to save energy. So, extrapolating the findings from Suntornsan et al.⁶⁴ and Qalati et al.⁶⁵ to biomass projects, this study can infer that individuals who perceive they have control over their ability to participate in such projects are more likely to form intentions to do so. Referring to the finding from Kharuhayothin et al.⁶³, in the context of biomass projects, this suggests that people who feel they have the necessary resources (time, knowledge, financial means) to participate are more likely to form intentions to do so. The consistent positive relationship between perceived behavioral control and intentions across different energy-related contexts (as shown in previous studies) indicates that feeling empowered and capable is crucial. For biomass projects, enhancing individuals’ sense of self-efficacy regarding their ability to contribute meaningfully could significantly boost their intention to participate.

The third primary variable of TPB is the subjective norm. The more favorable the subjective norms are toward an act, the more likely an individual is to participate in or carry out that action¹⁵. Referring to Fu et al.⁶⁶, subjective norms is an indication of the social pressure that people who are close to someone give off, which might influence someone’s choice as to whether to engage in a particular action or not. Subjective norms can positively influence people’s behavioral decisions. Before participating or doing something, people may experience social pressure from friends, family, and the government⁶⁷. In their research, Suntornsan et al.⁶⁴ found that subjective norms were the most significant predictor of behavioral intentions and that intentions significantly impacted energy-saving behaviors. In line with Suntornsan et al.⁶⁴, Qalati et al.⁶⁵ also reveal the positive and significant effect of subjective norms on households’ intention to save energy. Given the strong theoretical and empirical foundation, this study can expect that individuals who perceive positive subjective norms (i.e., social approval and encouragement from their social circles and community) regarding biomass projects will be more likely to intend to participate in such projects. This hypothesis, if confirmed, would underscore the significant role of community attitudes in promoting participation in biomass initiatives, demonstrating the power of collective action. Therefore, referring to the previous studies on the impact of TPB variables on public participation, this study proposed the following hypotheses.

• Hypothesis 1 (H1): Attitude (ATT) has a significant positive effect on intention to participate in biomass project (INT).

• Hypothesis 2 (H2): Perceived Behavioral Control (PBC) has a significant positive effect on intention to participate in biomass project (INT).

• Hypothesis 3 (H3): Subjective norm (SN) has a significant positive effect on intention to participate in biomass project (INT).

Perceived usefulness (PU), Perceived ease of use (PEU), Attitude (ATT) and intention (INT)

Perceived usefulness has a significant impact on how people utilize and adapt technology. If people think technology will benefit them, they will be more likely to employ it. Perceived usefulness is the degree to which one believes utilizing a specific technology would be beneficial and promote the accomplishment of essential objectives^16,68. Besides perceive usefulness, the other factor influenced the person attitude is perceived easy to use which can be defined as the level at which a person anticipates a behavior or technology to be simple to use and involve little to no effort¹⁶. Perceived easy to use is a crucial factor influencing people’s decision to adopt or participate in new activities. It means, easy-to-use technology will be more readily adopted, according to the basic principles of TAM¹⁶.

Pikkarainen et al.⁶⁹ found that perceived usefulness was the main factor influencing online banking acceptance among private banking customers in Finland. Kim et al.⁷⁰ found that perceived usefulness and perceived ease of use were key elements influencing the traveler’s attitudes and intentions to use mobile devices in the trip decision-making process. Min et al.⁷¹found that perceived usefulness and perceived ease of use have a significant influence on attitudes in the context of the Uber mobile application. The findings by Ejigu and Yeshitela⁷² also approved that both perceived usefulness and perceived ease of use positively affect the attitude of farmers toward the adaptation of excreta-based organic fertilizers. More recently, Toros et al.⁷³ confirmed that perceived usefulness influences the attitudes of refreshment students toward technology usage. When users perceive a technology as useful, they are more inclined to consider it a tool that can improve their daily tasks and work efficiency.

Given the solid theoretical and empirical foundation about the relationship between perceived usefulness and attitude, this study can expect that the degree to which people believe participating in or supporting biomass initiatives will provide benefits (such as environmental benefits, potential for job creation, local economic development, energy cost savings, and energy security reduced dependence on fossil fuels and imported energy sources) will influence their attitude toward biomass project. Moreover, this study can expect the people with perceived ease in accessibility of biomass resources (availability of feedstock and ease of collection/transportation), used technology (user-friendly biomass conversion technologies and integration with existing energy systems), and regulatory ease (streamlined permitting processes and clear policy frameworks for biomass projects) will have a good attitude toward biomass project.

• Hypothesis 4 (H4): Perceived usefulness (PU) has a significant positive effect on attitude (ATT).

• Hypothesis 5 (H5): Perceived ease to use (PEU) has a significant positive effect on attitude (ATT).

Perceived ease of use also influences the perceived usefulness. In this case, the relationship between perceived ease of use and perceived usefulness is grounded in the TAM developed by Davis¹⁶. This model posits that perceived ease of use can influence perceived usefulness, affecting attitudes and behavioral intentions. Once someone learns how to participate in a new project or use a new technology, they can quickly familiarize themselves with the process and understand how the new thing can benefit them¹⁶. The effect of perceived ease of use on perceived usefulness has been proved by Ayeh et al.⁷⁴. Based on their research, perceived ease of use significantly affected the perceived usefulness of using consumer-generated media (CGM). Kim and Chiu⁷⁵, in their study on sports and fitness wearable devices, also found the positive influence of perceived ease of use on perceived usefulness. More recently, a study by Hua and Wang⁷⁶ indicated that perceived ease of use had a significant impact on perceived usefulness; moreover, it positively influenced consumers’ attitudes. So, given the solid theoretical and empirical foundation, this study can expect enhancing the perceived ease of farmer or other stakeholder to participate in biomass projects could significantly improve their perceived usefulness, potentially leading to higher participation rates and more successful implementation of biomass projects.

• Hypothesis 6 (H6): Perceived ease to use (PEU) has a significant positive effect on intention to participate in biomass project (INT).

• Hypothesis 7 (H7) : Perceived ease of use (PEU) has a significant positive effect on perceived usefulness (PU).

Experience

Experience is a person’s observations, interpretations of sensations, and emotions generated while participating or interacting with a system⁷⁷. Harryanto et al.⁷⁸ build the TAM model by including the experience variable because it assumes experience is also a determinant of behavior for system users. However, the prior studies demonstrate mixed findings regarding the relationship between experience and perceived usefulness, with some studies showing a positive effect and others finding no significant relationship. Among the previous researchers that support the significant positive effect of experience on perceived usefulness are Abramson et al.⁷⁹and Arndt and Peterson⁸⁰. Andika et al.⁸¹ found that experience had no significant effect on perceived usefulness. The mixed findings regarding perceived usefulness might be due to the more complex nature of usefulness perceptions, which various factors beyond experience could influence.

Unlike mixed findings on the relationship between experience and perceived usefulness, prior research related to the relationship between experience and perceived ease of use can prove the positive relationship between the two variables. Andika et al.⁸¹found that the perception of ease of use is influenced by experience. This research is supported by study conducted by Arndt and Peterson⁸⁰and Danurdoro & Wulandari⁸². In this case, the perceived ease of use is heavily influenced by prior experience. Users with more experience with a system typically perceive it as more straightforward.

Given that experience appears to have a more consistent and more substantial influence on perceived ease of use compared to perceived usefulness, this study can expect to enhance the perceived usefulness and ease of use of farmers or other stakeholders to participate in biomass projects when they have good experience. Therefore, this study proposes the following hypotheses.

• Hypothesis 8 (H8): Experience (EXP) has a significant positive effect on perceived usefulness (PU).

• Hypothesis 9 (H9): Experience (EXP) has a significant positive effect on perceived ease of use (PEU).

Information

Information refers to the extent to which people have encountered a particular message or set of messages/media content regarding environmental protection⁸³. Lin et al.⁸⁴ and Shah et al.⁸⁵ illustrated in their study that media exposure behaviors can positively or negatively affect the audience. The study results from Shah et al.⁸⁵indicated that exposure to media significantly influenced individuals’ altruistic values. The results are consistent with the studies of Tsay-Vogel and Krakowiak⁸⁶and Hogan⁸⁷. Media stimulates individuals’ feelings of promotion, provokes perceptions, and leads them to altruism⁸⁶. They have become sources of people’s interconnectivity and raise awareness of global, social, and environmental issues⁸⁸. Media enables and compels individuals to shift from single-minded and selfish thinking to collective human purpose on global and cosmic levels⁸⁷. Then, related to the impact of information on biospheric values, a study conducted by Lee⁸⁹ has proved that the more often people (specifically adolescents) are exposed to media on environmental issues, the more positive biospheric value they will hold.

Given that the empirical foundation indicates a significant and largely positive relationship between information exposure through media and the development of both altruistic and biospheric values, this study can expect to enhance the of farmers or other stakeholders altruistic and biospheric values to participate in biomass projects. Therefore, this study proposes the following hypotheses.

• Hypothesis 10 (H10): Information (INF) has a significant positive effect on altruistic values (ALT).

• Hypothesis 11 (H11): Information (INF) has a significant positive effect on biospheric values (BIO).

Self-transcendence values

In environmental research, there are generally two types of self-transcendence values: Altruistic and biospheric⁶¹. Altruistic values represent a concern for fellow human beings’ well-being and fair treatment. Environmentally conscious behaviors and beliefs are often associated with positive human outcomes, such as preserving our planet for future generations, improving people’s health, or participating in a biomass project to reduce waste⁶¹. Therefore, altruistic values are usually positively related to environmental care beliefs and behaviors if these behaviors also benefit others.

On the other hand, biospheric values represent a concern for the environment itself, with no firm link to humans⁹⁰. Therefore, acting with environmental care will directly support this value. Biospheric values are the most strongly and consistently associated with environmental care beliefs and behaviors compared to other values. A person who strongly supports biospheric and altruistic values usually acts with environmental concern and has stronger environmental concern beliefs, as they are generally positively related to environmental concern beliefs and behaviors unless they contradict each other⁶¹. According to that, the following hypotheses are proposed in this study.

• Hypothesis 12 (H12): Altruistic values (ALT) have a significant positive effect on environmental concern (EC).

• Hypothesis 13 (H13): Biospheric values (BIO) have a significant positive effect on environmental concern (EC).

Environmental concern

Environmental concerns are associated with a person’s knowledge of and concern for climate and environmental issues⁹¹. Environmental concerns are a significant motivation of environmental attitudes⁹²since environmental concerns have been cited as an influential determinant for individuals to change their current behavior to a more environmentally friendly one⁹³. Yadav and Pathak⁹² showed that environmental concerns significantly influence the attitude toward green products among young consumers in India. Maichum et al.⁹⁴ presented that environmental concerns directly impact attitudes towards green products among Thai consumers. According to Yang et al.⁹⁵, environmental concern is positively and significantly associated with consumers’ attitude toward sustainable consumption China. Tamar et al.⁹⁶ found that knowledge about the environment has an effect on attitude to pro-environmental behavior.

Various studies have shown that a person’s knowledge of the environment can influence more than just their attitude toward green behavior^97,98. These studies consistently found that personal knowledge of environmental issues has a positive effect on perceived behavioral control^96,99,99,101and subjective norms, a concept from the Theory of Planned Behavior that refers to an individual’s perception of the social pressure to perform or not perform a behavior^96,100. For instance, Kim et al.⁹⁹ found that a person’s knowledge of the environment positively influenced their perceived behavioral control toward participation in UNICEF’s Change for Good (CFG) and voluntary carbon offsetting (VCO) program. More recently, Galván-Mendoza et al.¹⁰¹ found that a person’s knowledge of the environment predicts perceived behavioral control, which predicts employee green behavior. Then, related to both perceived behavioral control and subjective norms, Tamar et al.⁹⁶ found that a person’s knowledge of the environment affects perceived behavioral control and subjective norms, although it had the slightest effect compared to other variables. In line with Tamar et al.⁹⁶, Sari et al.¹⁰⁰ also revealed that environmental knowledge affects consumers’ intention to participate in e-waste collection programs through perceived behavioral control variables and subjective norms.

Given the solid theoretical and empirical foundation, this study can expect that the degree to which farmers’ or other stakeholders’ knowledge about the environment will predict their perceived behavioral control and subjective norms to participate in biomass projects. Therefore, this study proposes the following hypotheses.

• Hypothesis 14 (H14): Environmental concern (EC) has a significant positive effect on attitude (ATT).

• Hypothesis 15 (H15): Environmental concern (EC) has a significant positive effect on perceived behavioral control (PBC).

• Hypothesis 16 (H16): Environmental concern (EC) has a significant positive effect on subjective norm (SN).

Level of education

In TAM, the level of education is one of the socio-demographic variables that can influence an individual’s intention to perform a behavior¹⁶. In line with that, the previous study by Zhang et al.¹⁰²found that respondents who believed they had enough time and resources to participate in environmental care activities and had higher education were more willing to contribute to green projects. Encouraging public participation in urban environment management requires timely information disclosure and education¹⁰². The study on green consumption intention by Liang et al.¹⁰³ implies that a higher level of education contributes to customers’ increased awareness of environmental and health issues, which may impact their consumption decisions. Moreover, the study by Zhao et al.¹⁰⁴ stated that age and education were positively correlated with intention to use. People with higher education may be more willing to use single-use plastic food container substitutes (SUPS) to reduce plastic waste. According to by Zhao et al.¹⁰⁴, highly educated people tend to embrace environmental concepts and exhibit a preference for environmentally conscious behaviors, which affects the influence of societal norms on people’s willingness. This underlines the role of education in enhancing environmental awareness, making individuals more enlightened and informed about their choices and actions.

By synthesizing the findings of previous studies in the context of biomass projects, it becomes evident that education plays a crucial role. Highly educated individuals are likely to have a better grasp of the benefits and rationale behind participating in biomass projects. Their heightened awareness of the social and environmental impacts of their actions may make them more susceptible to positive social cues and norms. As a result, the relationship between subjective norms and intention could be more pronounced among individuals with higher education. This study anticipates that education level moderates the effect of subjective norms on intention, making the influence of subjective norms on the intention to participate in a biomass project more significant as the level of education increases.

• Hypothesis 17 (H17): The level of education moderates the effect of subjective norms (SN) on intention to participate in biomass project (INT) positively and significantly.

Intention to participate

The extent to which someone intends to participate in or actively participate in a specific behavior is known as their intention to participate¹⁵. TPB is a well-established theory that predicts behavior through intention. The intention to carry out a specific behavior is the immediate antecedent of behavior in TPB. The greater the degree of intention, the higher the probability of the behavior occurring¹⁰⁵. Favorable psychological factors in the public will encourage individuals’ positive intentions to make behavioral choices, leading to favorable behavioral outcomes¹⁰⁶. In TAM, user behavioral intentions influence actual behavior⁶⁸. Many studies demonstrate that intentions are reliable predictors of conduct for a range of individuals, even though intentions do not always result in activity²⁹. For instance, research on children’s exercise intentions and behaviors over time also supports that intentions are the immediate determinants of exercise behaviors¹⁰⁷. Huang et al.¹⁰⁸ also proved that citizens’ intention positively impacts their participatory behavior in the urban green space government -the planning and management of urban green open spaces, such as parks, grasslands, and forests in the urban landscape.

With the solid theoretical and empirical base of previous studies, this study proposes a high likelihood of participation occurs when individuals genuinely intend to contribute to a biomass project. Accordingly, this study puts forth the following hypothesis.

• 18 (H18): Intention to participate (INT) has a significant positive effect on actual behavior (BEH).

Based on hypothesis 1 until 18, this study proposed the conceptual model as seen in Fig. 3.

Fig. 3

Proposed conceptual model.

Research philosophy and research design

A research philosophy consists of fundamental beliefs that shape how a research study is designed and carried out, with different philosophies offering distinct approaches to understanding scientific research. These philosophies reflect a worldview, outlining one’s perspective on the "world," their place, and their possible relationships¹⁰⁹. This study adopts a "post-positivist" research philosophy, which assumes an external reality independent of the observer and emphasizes objective observation and measurement. Post-positivism is a common stance in sport and exercise psychology research. In contrast, research grounded in a constructivist philosophy seeks to understand how individuals create meaning from their experiences. This study focuses on exploring participants’ intentions in a biomass project, aiming to understand the meanings farmers and other stakeholders assign to their experiences with the project.

Related to research design, this study utilized both qualitative and quantitative methods. The qualitative approach was applied to understand the studied topic deeply. Its core principle is identifying the problem or issue and researching to collect relevant data¹¹⁰. This study used the qualitative method to develop the research model and hypotheses, including defining measurement items through a thorough and systematic literature review to ensure data validity. Then, the research utilized a quantitative method to quantify constructs, using the widely accepted Likert-type scales, empirical item measurement, and model validation through Partial Least Square-Structural Equation Modelling (PLS-SEM) with Smart PLS 4.0 software. In detail, the explanation of the primary data collecting technique (such as the number of samples, location, and sampling strategy to collect the data) can be seen in sub-section "Sample of research". Then, detailed measurement items and data processing techniques can be seen in sub-sections "Sample of research" and "Measurement items". Figure 4 depicts a thorough step-by-step process employed in this research.

Fig. 4

Flowchart Describing the Steps in Conducting this Research.

Sample of research

Researchers are advised by Hair and Alamer¹¹¹to refer to guidelines for recommended sample sizes, such as those provided by Kock and Hadaya¹¹². In their research, Kock and Hadaya¹¹² advised PLS-SEM users to estimate the minimum sample size using the inverse square root method. This method, in addition to being user-friendly, is also reasonably accurate. A path coefficient of 0.3 plays a crucial role in establishing a meaningful relationship between variables in PLS-SEM, guiding the interpretation of the results. The minimum sample size required for the significance level of 5% can be calculated as follows.

$$\widehat{N}>{\left(\frac{2.486}{{\left|\beta \right|}_{min}}\right)}^{2}$$

$$\widehat{N}>{\left(\frac{2.486}{0.3}\right)}^{2}$$

$$\widehat{N}>68.66$$

Therefore, 69 is the minimum sample size requirement for this study.

Then, Banyuwangi was selected for this study because it has enormous corn and wood production and its potential for biomass cofiring project materials. This study used a convenience sampling method to select the respondents. In this case, only respondents who agree to participate will be involved as a source for answering the questionnaire. The questionnaire was delivered to farmers and business owners who produce sawdust or corn stalk waste in the hopes that the respondents would be knowledgeable about their respective wastes and able to provide a comprehensive understanding of their waste production and usage. The respondents’ knowledge is used to ensure the relevance and quality of data collection. A presentation outlining the study’s goal and the potential contributions of the respondents was given before the respondents filled out the questionnaire. At the end of data collection, 75 completed and valid questionnaires indicate that this study has attained the minimum needed sample size.

The method carried out in this study is following our institution’s guidelines and regulations. All statements and experiment protocols were approved by the Institute for Research and Community Services, Diponegoro University, responsible for ensuring the research’s ethical and legal conduct. This study believes in the importance of transparency in the research process, and all steps are clearly outlined. Then, informed consent was obtained from all subjects and their legal guardians. This study includes a brief overview of the study on a level of understanding of the participants in the questionnaire of intention to participate: "This study aims to identify and measure the factors encouraging the public to participate actively in biomass projects, a crucial step in guaranteeing the effectiveness of the implementation of sustainable biomass project." Besides, the questionnaires also describe what participation in the study entails. In this case, for the validation questionnaire: ‘" When you enter into the questionnaire, you will be asked to complete 24 questionnaires with four alternative answers (1 = strongly disagree, 2 = disagree, 3 = agree and 4 = strongly agree).

The four-point Likert scale, often called a ‘forced choice’ scale, eliminates a neutral option, empowering respondents to express a clear opinion. This design fosters more decisive data, encouraging participants to engage thoughtfully with the questions. A study comparing the reliability and validity of four-point versus six-point Likert scales found that while both formats have their strengths, the four-point scale effectively reduces method variance and enhances the clarity of trait measurement¹¹³. More recently, according to South et al.¹¹⁴, researchers often use a Likert scale with four levels to mitigate the issue of social desirability bias. To ensure confidentiality, participants are informed about the level of identity protection of any personal information collected for the study: ‘All information taken from the study will be coded to protect each subject’s name. No names, or other identifying information will be used when discussing or reporting data’. Participation is voluntary, and there is no coercion in participating in the questionnaire survey. Participants have the right to withdraw from the survey at any time without consequence. Lastly, if participants have any questions or concerns or would like further information about this study, they are provided with a contact number.

Measurement items

This study utilized 12 variables and 25 indicators to examine the relationship between the conceptual model variables and public participation in the biomass project. Twenty-three items measured the independent variable; the remaining two measured the dependent variable, public participation in biomass projects. The variable information was measured using three items, and the remaining variables were measured using two items each. The TPB variables (perceived behavioral control, attitude, and subjective norms) were measured using items modified from Baker et al.¹¹⁵. Items measuring TAM variables (perceived usefulness and perceived ease of use) were adopted from Kardooni et al.¹¹⁶ and Borrero et al.¹¹⁷. Biospheric and altruistic values were measured with items modified from Bouman et al.⁹⁰. The items to measure environmental concern were adopted from Sari et al.¹⁰⁰. Furthermore, the items measuring experience and information were modified from Huang et al.¹¹⁸ and Kardooni et al.¹¹⁶. Thus, Sulaiman et al.¹¹⁹ provided the items measuring actual behavior, and Mboya et al.¹²⁰ provided the items measuring intention to participate with some modifications. All of the items is measured using 4 point Likert-Scale.

Data processing technique

The PLS-SEM method was applied to the questionnaire results using SmartPLS 4.0 software. PLS-SEM is a popular method among researchers because it enables the estimation of complex models that include numerous constructs, indicator variables, and structural paths (such as the conceptual model in this study) without requiring distributional assumptions. Unlike other methods, PLS is applicable without requiring distributional assumptions¹²¹. It can handle non-normally distributed data and works well with categorical, ordinal (quasi-metric) data and single-item measurements¹²². This versatility makes it worthwhile in fields like behavioral sciences¹²³. Additionally, PLS is suitable for small datasets¹²⁴. Crucially, PLS-SEM is a causal-predictive approach to SEM, prioritizing prediction when estimating statistical models intended to offer causal explanations^125,126. Figure 5shows the process followed in this research. Then, there were two steps in data processing with PLS-SEM¹¹¹.

Fig. 5

Result of hypothesis testing.

First, assess the reliability and validity of the item measures by evaluating the outer model.

As Chin¹²⁷suggests, the initial step in outer model evaluation involves assessing the reliability and validity of the measurements representing each construct, such as convergent and discriminant validity and composite reliability. Convergent validity assesses how well groups of items align in measuring the construct they were designed for, checking if the loadings are sufficiently high and similar. Subsequently, discriminant validity ensures that each item is correctly associated with its intended construct and not more strongly linked to another. If an item correlates more with a different construct, it indicates a discriminant validity issue, meaning it may not effectively differentiate between constructs. In addition, composite reliability for each set of indicators is typically included in assessing internal consistency^111,127.

Second, evaluate the predictive capacity of the inner model.

Once the reliability and validity of the measures have been confirmed, the next step involves evaluating the predictive power of the inner model¹²⁷. The Variance Inflation Factor (VIF) is critical to ensure that constructs are not highly correlated, as this would lead to issues in both methodology and interpretation¹¹¹. Then, the model’s predictive strength is generally measured using the R² values for the endogenous constructs. Along with R², the Q²predictive can also be used to further gauge predictive relevance. Model fit can be checked using indices like the Standardized Root Mean Square Residual (SRMR), Chi-square divided by degree of freedom, and Normed Fit Index (NFI), which indicate whether the proposed model aligns well with the data¹²⁸.

Results

Characteristic of respondents

Almost all respondents were male (99%), and only one female respondent (1%). The majority of respondents were aged 41–50 years (39%), followed by 31–40 years (27%), 21–30 years (19%), and more than 50 years (5%). Furthermore, most respondents have a high school or equivalent education level (33%), followed by elementary school or equivalent (28%), junior high school or equivalent (16%), bachelor’s degree (11%), and diploma (7%). Over half of the respondents (57%) produce corn stalk waste. Other respondents produced sawdust or waste from the wood production process (35%), and six respondents (8%) had no waste. Most respondents who generated waste disposed of the waste without further treatment (70%). Other respondents sent the waste to other places (one of which was a tofu factory) (23%), both combination of self-use and sending to other places (3%), both combination of disposal and sending to other places (1%), self-use (1%), and combination of self-use and disposal (1%).

Result of data processing with PLS- result of measurement test (outer model)

This study uses convergent validity tests, composite reliability tests, and discriminant validity tests to conduct outer model tests or measurements. This study employs the Average Variance Extracted (AVE) value for each construct and the Loading Factor of each indicator for the convergent validity test. An indicator is considered convergence if its loading factor and AVE value are at least 0.5. According to Fornell & Larcker¹²⁹, a good measure of convergent validity means that more than half of the item variance may be explained by the latent variable on average. One indicator, SN2, has a loading factor value of less than 0.5 in the first test, as Table 2 depicts. The indicator states that the neighborhood around the house has initiated initiatives to supply biomass for energy sources (SN2; loading factor = 0,053). This leads to the conclusion that indication SN2 is deemed invalid as part of the subjective norm variable. Hence, the indicator SN2 was removed from SN variable in the second test. In the second test, all indicators showed loading factor values larger than 0.5 and AVE values greater than 0.5. This result indicates a good measure of convergent validity.

Table 2 Result of the validity and reliability test.

Additionally, this study employs each construct’s Composite Reliability (CR) value in the reliability test (composite reliability test). According to Chen & Chao⁴⁹, good reliability is indicated by a CR score of 0.7 or higher. Except for the subjective norm (SN) construct, all constructs in the first test had CR values larger than 0.7, as seen in Table 2. The CR value of the subjective norm construct is 0.477. In contrast, every construct in the second test has a value higher than 0.7. This shows that each construct has enough dependability or reliability.

Then, this study used the Heterotrait-Monotrait ratio (HTMT), Fornell Larcker criterion, and cross-loading criteria to evaluate discriminant validity. A Heterotrait-Monotrait ratio (HTMT) value smaller than 0.9 implies that all pairs of variables in the model have a good level of discrimination, which indicates that the constructs measured are quite different from each other and not too similar¹³⁰. As shown in Table 3, there is one pair of construct, INT and BEH, which has a value of 0.91. This indicates that the relationship between the constructs is high, thus the model needs to be modified. By analyzing the correlations between indicators, it is found that indicator BEH2 has the highest average correlations (0.445) with INT1 and INT2. Hence, the indicator BEH2 was removed from BEH construct.

Table 3 Result of the first discriminant validity check based on HTMT ratio.

To ensure good model measurement, the convergent validity and construct reliability need to be tested again after removing the indicator BEH2 in the third test. Table 2 shows that all indicators in the third test have loading factor values larger than 0.5 and AVE values greater than 0.5, which indicates a good measure of convergent validity. Additionally, each construct’s Composite Reliability (CR) value in the reliability test has a value higher than 0.7. This result shows that each construct has enough dependability or reliability.

Moving on to discriminant validity test, after removing indicator BEH2, all pairs of constructs have HTMT values below 0.9 as shown in Table 4. This indicates that the constructs in this research model can be considered statistically different from each other.

Table 4 Result of the second discriminant validity check based on HTMT ratio.

The primary value of AVE should be greater than its correlation with other variables¹²⁹. It is evident from the information in Table 5 that every variable satisfies the conditions needed to have a good Fornell-Larcker criterion value. Given that each latent variable in this study model has sufficient discriminant validity, it can be concluded that each variable has a greater correlation with its own indicators than it does with other variables.

Table 5 Result of the discriminant validity check based on Fornell-Larcker criterion.

Chin (1981) states that the loading factor of any item must be greater than each of its cross-loadings. Table 6 displays the findings of the final discriminant validity evaluation. All of the variables exhibit a definite display of discriminant validity since the value of each loading factor exceeds all of the cross-loadings.

Table 6 Result of the discriminant validity check based on cross-loading value.

Result of descriptive statistics

The descriptive statistics for each construct and items used in the research can be seen in Table 7.

Table 7 Result of the descriptive statistics.

Six indicators have an average value below 3, namely the role when participating in biomass projects (EXP1), the length of experience in biomass projects (EXP2), seeing or hearing information about energy shortages (INF1), seeing or hearing about participation in biomass projects (INF2), seeing or hearing information about the contribution of biomass as a source of electricity (INF3), and neighboring community have participated in activities to provide biomass for energy sources (SN2). This is because most of the respondents have never participated in a biomass project; if they have, the duration of participation is less than one year of participation. In this case, most of the respondents’ forms of participation were as passive waste providers, not delivering the waste to a specific destination or collecting the waste from others. Respondents acknowledged that waste collection is carried out periodically by parties in need. Furthermore, for various information related to energy and biomass, most respondents felt they had never heard of it.

Result of data processing with pls- result of structural model assessment (Inner model test)

First, it’s critical to make sure that there is no strong correlation between the constructs, as these could lead to problems with the interpretation of the route coefficient and is indicated by a high VIF value. According to Hair and Alamer¹¹¹, a VIF value of 5 or more is thought to indicate significant collinearity problems across the predictor factors. Table 8 showed that each relationship between latent variables has satisfied the requirements for a satisfactory collinearity statistics (VIF) value, which is smaller than 5.

Table 8 Result of the VIF value.

The inner model or structural model assessment is conducted using some measures, such as the coefficient of determination (R²), model fit (standardized root means square (SRMR), normed fit index or (NFI), and chi square divided by degree of freedom (χ²/df). The result can be seen in the Table 9.

Table 9 Result of the inner model test.

In addition to that, the value of Q² is also calculated to assess the PLS path model’s predictive accurracy. A cross-validated redundancy Q²above 0 is generally considered indicative of a predictive model¹²⁷. The calculation is shown as below.

$${\text{Q}}^{2}=1-\left(1-{\text{R}}_{1}^{2}\right)\left(1-{\text{R}}_{2}^{2}\right)\left(1-{\text{R}}_{3}^{2}\right)\left(1-{\text{R}}_{4}^{2}\right)$$

$${\text{Q}}^{2}=1-\left(1-0.144\right)\left(1-0.408\right)\left(1-0.214\right)\left(1-0.235\right)$$

$${\text{Q}}^{2}=1-\left(0.856\right)\left(0.592\right)\left(0.786\right)(0.765)$$

$${\text{Q}}^{2}=0.695$$

This study obtained a Q² of 0.695, which suggests a predictive model. Therefore, the structural model fits the data and has good predictive ability.

Result of data processing with PLS- result of hypothesis testing.

Since there must be more number of bootstrapping iterations than number of samples, 500 iterations of the bootstrapping approach are used for this hypothesis testing¹³⁷. The bootstrapping method was carried out by SmartPLS software, which yields a number of values: The p-value, which indicates the level of statistical significance used to decide whether to accept or reject H0 (significant effect), the t-statistic value, which indicates the significance of the construct, and the original sample value, which indicates the direction of the relationship between the two constructs. If the p-value is less than 0.1, 0.05, or 0.01, the link is deemed significant¹³⁸. It is possible to conclude that 13 out of the 18 hypotheses are supported based on the findings of the hypothesis testing as shown in Table 10. In this case, there are four hypotheses that do not receive significant support, namely the effect of biospheric value (BIO) on environmental concern (EC), environmental concern (EC) on attitude (ATT), experience (EXP) on perceived ease of use (PEU), and perceived ease of use (PEU) on intention to participate (INT). However, experience (EXP) significantly influences perceived usefulness (PU), though the effect is negative. Furthermore, it also can be seen that the most influential factor for Intention (INT) is Perceived Behavioral Control (PBC) with the highest loading factor (0.317) compared to Attitude (ATT), Perceived Ease to Use (PEU), and Subjective Norms (SN). Intention then influences actual behavior strongly with the loading factor 0.463.

Table 10 Result of the hypothesis testing.

The empirical model which is the result of hypothesis testing based on the proposed model can be seen in the Fig. 5. The value below the line shows the t-statistic value, while the value above the line shows the p-value.

Some insight from the result of this research can be explained as seen in Table 11.

Table 11 Practical implications.

Conclusion and recommendation for further research

Based on a systematic literature review, this research proposes a conceptual model that integrates TAM and TPB. The integration of the two models makes it possible to explore the technical and non-technical aspects that can influence the level of community participation in biomass projects. This TAM-TPB integration research model has 12 variables with 25 indicators. In addition, one moderating variable, education level, examines the influence of one demographic aspect on public participation. Based on the convergent validity test results, the SN2 indicator is considered invalid as part of the subjective norms construct and was excluded from the calculation. Since the HTMT ratio between INT and BEH exceeds 0.9, the BEH2 indicator was also removed. The result of hypothesis testing showed that 13 of the 18 proposed hypotheses are significantly supported; one hypothesis is significant with a different direction from the study proposed, and four are not significantly supported. The unsupported hypotheses include the effect of biospheric value on environmental concern, environmental concern on attitude, experience on perceived ease of use, and perceived ease of use on intention to participate. Meanwhile, experience significantly influences perceived usefulness with a negative effect.

According to PLS-SEM calculation and hypothesis testing, the most influential factor for Intention (INT) is Perceived Behavioral Control (PBC), with the highest loading factor (0.317) compared to Attitude (ATT), Perceived Ease to Use (PEU) and Subjective Norms (SN). The intention then influences actual behavior strongly, with a loading factor of 0.463. This study indicates that people will most likely participate in biomass projects by supplying or transporting biomass feedstock if they can devote their time or resources. The result aligns with the prior research by Zhang et al.¹⁴⁰, which found that perceived behavioral control was the most significant latent variable affecting the private sector’s willingness to participate in Public–Private-Partnership (PPP) projects.

Both sets of variables from the TAM and TPB models are combined to build a fully integrated model as a conceptual framework to satisfy the need in both the theoretical structure and empirical analysis of the public intention toward engaging in biomass projects. The conceptual framework improves upon the two distinct models’ explanatory shortcomings and provides further clarity regarding the objective and subjective aspects impacting participation intention in the biomass project. Hence, this study provides insightful implications for practitioners and academics. As for managerial implications, the provider or the power-generating industry should focus on the significant aspects to encourage public participation. This could be accomplished by informing the public about the significance of the biomass project, the capability given by the resources they have, or providing them with an opportunity to experience it first-hand. These methods are consistent with the empirical model, which states that experience significantly impacts how helpful something is seen and generates favourable attitudes that become intentions. Policymakers may encourage the public to engage in biomass projects through government socialization initiatives addressing issues like climate change and energy scarcity, which are promoted through public service announcements. This is important because information is crucial in raising values regarding environmental concerns and influencing public intention.

However, this study has limitations since the study’s sample related to factors influencing people to participate actively in biomass projects was limited to 75 corn farmers and wood craftsmen in Banyuwangi. Although this number qualifies for testing using the PLS method, increasing the sample size in future studies may improve the representativeness of the population of corn farmers and wood craftsmen. Furthermore, increasing the sampling area (not only in Banyuwangi) allows for comparing the factors influencing people to participate actively in biomass projects in two or more regions.