Individual willingness and behavior decision-making is a complex and systematic process influenced by multiple factors (Sahu et al., 2020). Depending on the specific research domain, scholars have adopted different theoretical models to explain public participation behavior (PH).

Recently, a few studies have examined Individual participation in knowledge-based crowdsourcing conceptually (Terwiesch and Xu, 2008) and empirically (Boons et al., 2015). In addition, most existing studies believe that economic reward, enjoyment, improvement of skills, trust, and other factors are essential motives that affect individual participation in crowdsourcing. For example, Shao et al. (2012) developed a model to explain individual participation in crowdsourcing and tested it using data from Chinese crowdsourcing websites. The results showed that higher rewards, more manageable tasks, longer task time, and lower competition intensity would attract more crowdsourcing participants. Based on the social exchange theory, Ye and Kankanhalli (2017) developed a model to explain how rewards, costs, and trust affect individual participation behavior in crowdsourcing. The study found that monetary rewards, skill enhancement, job autonomy, entertainment, and trust have a positive effect on solvers’ participation in crowdsourcing. In contrast, cognitive efforts hurt their participation. In addition, Li and Hu (2017) developed a model based on expectation theory to explore the impact of rewards and competition intensity on participant registration and submission behavior. The research found that the task reward was positively correlated with the number of registered and submitted tasks, while the competition intensity was negatively correlated with the solver’s submission. Based on the self-determination theory (SDT), Suen et al. (2022) advanced the theoretical understanding of the effects of employee motivation antecedents (the three basic psychological needs: autonomy, competence, and relatedness) on participation efforts in internal crowdsourcing activities.

At present, crowdsourcing delivery mode is still in its early stages of development. Previous studies have typically focused on the current development status of crowdsourcing delivery and analyzed the advantages and disadvantages of crowdsourcing delivery (Pourrahmani and Jaller, 2021). Research on the PW of crowdsourcing delivery was rare. Some scholars have noted that crowdsourcing delivery could reduce delivery costs while ensuring fast and reliable delivery (Devari et al., 2017).

Among the limited research on PW of crowdsourcing delivery, the majority of studies suggest that factors such as expected reward, effort, the environment of the crowdsourcing website, and cost of execution would affect individuals’ PW of crowdsourcing delivery. For instance, Bin et al. (2020) developed a model to explore the influencing factors of enterprises’ willingness to implement crowdsourcing logistics based on the technology–organization–environment (TOE) theoretical model. In a similar vein, Liang et al. (2017) developed a model of influencing factors for the continuous PW of crowdsourcing delivery and found that participation motivation, subjective norms, and satisfaction would affect individual continuous PW in crowdsourcing delivery. Liu Y. et al. (2024) explored the moderating effect of perceived risk on the correlations between satisfaction and continuous-use intention, perceived value and continuous-use intention, and trust and continuous-use intention. The findings revealed that trust, perceived value, and satisfaction positively contributed to continuous-use intention, where trust contributed the most.

According to our review, we found that due to the rapid development of crowdsourcing in China in recent years, there is a lack of existing theoretical foundations. Additionally, the majority of the existing research focused on the feasibility of crowdsourcing delivery and the advantages and disadvantages of crowdsourcing delivery. However, relatively few have examined individuals’ PW in crowdsourcing delivery. Furthermore, the majority of researchers believed that expected revenue, effort expectations, crowdsourcing website environment, execution cost, and other factors would affect the individuals’ PW in crowdsourcing delivery. However, there is very little empirical research examining the influence of risk and travel characteristics factors on individual participation in crowdsourcing delivery of fresh agricultural products. Therefore, this study selects crowdsourcing delivery as the research object, considering various factors including risk and travel characteristics from the perspective of social exchange theory, and analyzes the factors that affect individual participation in crowdsourcing delivery.

In this study, we investigate the PW in crowdsourcing delivery of fresh agricultural products, considering the influencing factors of individual trust, participation risks, travel characteristics, and constraints within the research model.

Expected reward (ER) refers to the prediction of the reward that crowdsourcing delivery participants can obtain, assuming no accidents occur. Except for monetary rewards, rewards include non-monetary rewards, such as enjoyment, job opportunities, and skill enhancement. Previous literature has suggested that expected rewards are important drivers for individuals to participate in crowdsourcing delivery. For example, Peng and Zhang (2010) found that learning knowledge and financial rewards are the primary reasons for attracting individuals to participate in crowdsourcing. Goncalves et al. (2014) analyzed some cases and found that crowdsourcing platforms can obtain better service quality and more individual participation through incentives. Following the discussion above, we propose hypothesis H1 that the expectation of reward will enhance individuals’ PW in crowdsourcing delivery of fresh agricultural products:

Expected costs (EC) refer to the various costs that individuals perceive when participating in crowdsourcing delivery of fresh agricultural products. Similar to expected rewards, expected costs encompass both monetary costs and non-monetary expenses. The non-monetary costs primarily refer to time, search, and travel costs (Patel et al., 2023). When individuals decide whether to participate in crowdsourcing delivery, they will consider not only their benefits but also the various costs they incur for participating in crowdsourcing delivery of fresh agricultural products. Such as the time required to participate in the various operational links of the task process, the energy required to participate in the delivery task, searching for suitable delivery tasks will take up time, etc. Therefore, this study proposes hypothesis H2a:

Additionally, we propose that the expected cost will affect individuals’ perceived participation risk. The higher the expected cost perceived by individuals, the more worried individuals are about their participation in crowdsourcing delivery, and it will increase their perceived risk or weaken their PW. That is, the higher the expected cost, the greater the perceived risk of participation by individuals. Thus, this study proposes hypothesis H2b:

During crowdsourcing delivery activities, an exchange occurs between the individual and crowdsourcing delivery platform, with trust (TR) playing a key role in the process. The crowdsourcing delivery of fresh agricultural products involves many aspects of trust-related concerns, such as whether the platform delivers services effectively, adheres to its commitments to the public, and objectively defines responsibilities while taking into account the interests of all participants. Therefore, we believe that these trust issues will affect the public’s PW in crowdsourcing delivery according to the above discussion. Thus, this study proposes hypothesis H3:

Participation risk in this study refers to various risks that the public perceives when participating in crowdsourcing delivery activities, such as time risk and security risk (Cui et al., 2022). We believe that the public may face a few risks when participating in crowdsourcing delivery. For example, time risk refers to the fact that customers cannot pick up on time, or participating in crowdsourcing delivery may take up your original travel time. Security risk refers to the possibility that the individual may encounter traffic accidents during the delivery. Since there are currently no relevant laws or regulations that clearly stipulate this, it may result in participants being at a disadvantage in crowdsourcing delivery, and weaken the public’s PW in crowdsourcing delivery of fresh agricultural products. Therefore, this study proposes hypothesis H4:

People have sociality, so they will inevitably be affected by people around them in their daily lives and interactions with others. In this research, social influence (SI) refers to the impact of the surrounding groups’ attitudes, thoughts, and behaviors on the public’s participation in crowdsourcing delivery when the public determines whether they are willing to participate in crowdsourcing delivery. Thus, this study proposes hypothesis H5a:

In addition, if people around the individuals have good evaluations of crowdsourcing delivery, the individuals will have more positive attitudes toward crowdsourcing delivery and it will increase their trust in the crowdsourcing delivery platform. Thus, this study proposes hypothesis H5b:

Unlike the knowledge-based crowdsourcing, crowdsourcing delivery accompanies individual travel. Thus, travel characteristics will affect individuals’ PW in crowdsourcing delivery. The indicators used to measure the characteristics of residents’ travel mainly include the frequency of travel, the purpose of travel, the mode of transportation, the travel distance, and the travel time. However, the purpose of travel and the transportation of travel cannot be quantified and cannot be researched in this study. Combined with the characteristics of crowdsourcing delivery, we primarily consider three aspects: residents’ travel time, the travel distance, and the frequency of residents’ travel. So, we propose hypothesis H6:

Promotion conditions (PCs) in our study refer to the help that the public can get from firms or crowdsourcing delivery platforms when they participate in crowdsourcing delivery. Economic incentives are important factors that can increase the public’s motivation to participate in an activity (Ali and Anwar, 2021). Drawing on the above discussion, we believe that promotion conditions can strengthen the individuals’ PW in crowdsourcing delivery. The firms or platforms can encourage the public to participate in crowdsourcing delivery through a series of incentive measures. Specifically, promotion conditions in our study include closer orders to travel destinations, the precise location of merchants and customers, an incentive system, updating the delivery tasks in time, a reasonable web design of a platform and a more convenient operation. Thus, this study proposes hypothesis H7:

Crowdsourcing delivery differs from traditional crowdsourcing activities, which typically involve completing tasks online. Crowdsourcing delivery of fresh agricultural products assumes that the packages were delivered on time. This leads to some constraints that prevent the public from participating in crowdsourcing delivery. For example, an individual may not want to participate in crowdsourcing delivery when there is a traffic jam or the weather is bad. Thus, this study proposes hypothesis H8:

Willingness is an individual’s view and thoughts on something, and personal subjective thinking directly affects the probability and intensity of the individual’s execution of something. In this study, participation willingness (PW) refers to the subjective thinking of an individual’s attempt to participate in crowdsourcing delivery, including searching for appropriate tasks and accomplishing the delivery.

The questionnaire in this study consisted of three parts. The first part provided a description of crowdsourcing delivery of fresh agricultural products, so that the interviewees could better fill out the questionnaire. The second part collected personal information from the interviewee, such as gender, age, education, and income. The third part was items of the research variables.

This survey involved nine variables they are expected reward, expected cost, trust, risk, social influence, promotion conditions, constraints, travel characteristics, and PW. To analyze all the measurements conveniently, items in our survey are adopted by Likert’s 5-level scale, with a 5-point format from 1 = strongly disagree to 5 = strongly agree. The items of all variables in the survey are listed in Table 2.

The respondents in this study were selected from potential participants, specifically individuals who had not previously engaged in crowdsourcing delivery. To test the research model, we collected questionnaires using both online and offline methods, resulting in a total of 354 questionnaires. Excluding questionnaires with very short response times and a highly consistent selection of items, we obtain 332 valid questionnaires.

To ensure the accuracy of the analysis results, we conducted a descriptive statistical analysis of the 332 valid samples collected. The Statistical Package for the Social Sciences (SPSS, the software developed by IBM Corporation) was used to analyze the interviewees’ sex, age, education level, and monthly income. The respondent demographics and background were summarized in Table 3.

According to Table 3, the age of the valid sample is mainly distributed in the 18–45 age range, accounting for 85.5% of the overall valid sample. In real life, individuals involved in delivery activities are primarily between 18 and 45 years old, young and middle-aged people. Therefore, the age structure of the sample in this study is reasonable. In addition, 97% of the respondents in this study have an education level of a bachelor’s degree or below. In real life, the majority of individuals involved in delivery activities have an education of a bachelor’s degree or below, so the education level of the sample in this study is in line with reality. Additionally, Table 3 lists that the majority of respondents have a monthly income of less than 5,000 renminbi (RMB), with a sample size of 74.7%. Crowdsourcing delivery is still predominantly part-time, so groups with low or no income are more likely to participate in crowdsourcing delivery of fresh agricultural products.

To validate our survey instrument, convergent validity tests were conducted. Convergent validity was assessed by examining composite reliability (CR), average variance extracted (AVE), and the results of the factor analysis. When the value of AVE > 0.5 and CR > 0.7, it indicates that the validity of the instrument is reliable (Cheung et al., 2024). Table 4 shows that the values of CR and AVE for each model construct satisfy the thresholds, except for travel characteristics. Therefore, in the subsequent analysis, this study excluded the items related to travel characteristics.

Individual participation in crowdsourcing delivery of fresh agricultural products is critical to platforms’ survival and profitability of firms and platforms. To better understand how to motivate individual participation, we developed and tested a model to identify the key factors influencing individuals’ willingness to engage in crowdsourcing delivery of fresh agricultural products. Our findings indicate that expected reward, trust, social influence, and promotion conditions positively influence individuals’ PW in crowdsourcing delivery, while expected cost is negatively related to individuals’ PW in crowdsourcing delivery of fresh agricultural products. From a practical standpoint, this study offers several recommendations for firms and crowdsourcing platform administrators seeking to attract and retain individual participants.

First, platforms should focus on enhancing various forms of rewards to incentivize participation. According to the results of this study, expected rewards (including monetary rewards, enjoyment, skill enhancement, and work autonomy) and incentive mechanisms have a significant effect on the PW in crowdsourcing delivery of fresh agricultural products. Therefore, additional revenue is the most effective incentive. To capitalize on this, crowdsourcing delivery platforms can establish an incentive mechanism to attract individuals to participate in crowdsourcing delivery of fresh agricultural products. For example, after the cumulative number of deliveries or miles per month reaches a certain threshold, the platform can provide participants with a specified additional reward.

Second, according to the results of this study, trust was positively related to individual participation in crowdsourcing delivery of fresh agricultural products. Therefore, platforms should focus on establishing effective trust-building mechanisms to enhance public confidence. Specifically, platforms could ensure that participants obtain appropriate rewards once their deliveries are accomplished. Additionally, the platform should uphold its commitment to participants and enhance its service level.

Third, the results of this study show that social influence has a significant effect on individuals’ PW in crowdsourcing delivery of fresh agricultural products. Therefore, from the perspective of social influence, platforms can set up an interactive community that relies on the app for crowdsourced delivery. Using interpersonal networks to influence the attitude of potential participants and increase their PW in crowdsourcing delivery of fresh agricultural products. An interactive community as a communication platform of participants in crowdsourcing delivery can focus on exchanging work experience, helping each other, and so on. Additionally, platforms can regularly organize some offline activities to create a relaxed and pleasant community environment. These activities can strengthen the bonds among participants by expanding their interpersonal networks, thereby enhancing individuals’ PW in crowdsourcing delivery of fresh agricultural products.

This study has several limitations. First, expected benefits, expected costs, trust, participation risk, key travel characteristics, facilitating conditions, and constraints were identified as the primary factors influencing individual participation in this study. Although the data analysis indicates that the proposed model demonstrates strong explanatory power, these seven factors may not encompass all the variables that influence individuals’ intentions to participate in crowdsourcing delivery of fresh agricultural products. Future research could further explore additional influencing factors and underlying mechanisms affecting individuals’ PW in crowdsourcing delivery of fresh agricultural products. These may include attributes of the fresh produce, supportive government policies, political environment, and broader economic conditions. In addition, as the number of participants in crowdsourcing delivery of fresh agricultural products is currently relatively limited, this study focuses on how to attract potential participants. However, with the continued development of the industry, the number of actual participants is expected to grow significantly. Accordingly, future research should shift its focus toward optimizing operational models and enhancing platform functionalities to sustain and strengthen individual engagement in fresh produce crowdsourcing delivery over the long term.