Abstract
Introduction:
Exploring the influence of social capital on farmers’ adoption of plant protection drone technology is crucial for guiding mechanized plant protection practices and achieving high-quality agricultural development.
Methods:
Based on survey data from farmers in Henan Province in 2024, this study employs an ordered probit model to empirically investigate the impact of social capital on farmers’ adoption of plant protection drone technology and its underlying mechanisms.
Results and discussion:
Baseline regression results indicate that social capital positively influences farmers’ adoption of plant protection drone technology, with cognitive social capital playing a more significant role than structural social capital. Mechanism analysis reveals that social capital affects adoption behavior by improving farmers’ access to credit and non-farm employment opportunities. Heterogeneity analysis shows that social capital’s influence is stronger among households with joint male–female decision-making, smallholder farmers, and those engaged in single-grain cropping, compared to male-dominated decision-making households, large-scale farmers, and those practicing diversified cropping. The study concludes with recommendations to enhance the promotion of plant protection drone technology and to recognize the role of social capital in shaping farmers’ technology adoption decisions.
1 Introduction
Agricultural production faces the challenge of meeting the demand for raw materials and food supply of a growing world population. At the same time, resource efficiency and production sustainability concerns are considered increasingly important aspects of agricultural production (; ). Precision agriculture (PA) is a management strategy which is based on the use of data from multiple sources to improve farmers’ decision making (). The main goal is to tailor management practices to the need of the crop by considering spatial and temporal information concerning the crop, soil and environment (; ). Bramley et al. (2017) noted that drones being regarded as one of the latest tools in PA, so drone adoption in agriculture remains limited (; ). In recent years, developing countries have been actively promoting agricultural mechanization to address challenges such as an aging agricultural workforce, low utilization efficiency of agricultural inputs, suboptimal agricultural output efficiency, and farmland system pollution. These efforts have achieved significant mechanization in key processes such as tillage, sowing, and harvesting (). However, the mechanization of crop protection—a critical component of agricultural production—has lagged behind, with manual crop protection practices still dominating (). Manual crop protection methods are characterized by inefficiencies, high labor intensity, and low utilization rates of pesticides and fertilizers, posing threats to ecological sustainability, food safety, and the physical health of operators (). These issues significantly hinder the progress of sustainable agricultural development, highlighting the urgent need to advance the mechanization of crop protection. Crop protection drones, which operate as unmanned aerial vehicles designed for high-altitude spraying, have gained increasing attention in recent years due to their advantages, including higher input efficiency, operational safety, efficiency, and ease of adoption (). Promoting farmers’ adoption of crop protection drone technology is critical for enhancing resource utilization efficiency and meeting the growing global demand for food ().
Existing research on crop protection drone technology has primarily focused on product development and industry analysis (). A smaller body of studies has examined the adoption of crop protection drones from the perspective of individual farmers, identifying factors such as land size, perceived value, and technological awareness as significant determinants of adoption decisions (; ; ). However, these studies predominantly emphasize individual characteristics while neglecting the embeddedness of individual decision-making within the broader social environment. As the integration of economic theory and agricultural production practices deepens, scholars increasingly recognize the diversity and complexity of farmers’ behavior. This shift has reframed the understanding of farmers from “homo economicus” to “homo socialis,” prompting calls for research to move beyond individual-level analyses and consider the influence of the social environment on farmers’ decision-making processes. Furthermore, constrained by limited awareness and risk aversion, farmers often exhibit significant behavioral delays between passively receiving information about new technologies and actively applying them (). Although developing countries are actively promoting the adoption of crop protection drones, many farmers remain hesitant, leading to low levels of mechanization in crop protection. According to individual decision-making theory, social resources—such as information, assistance, and norms—acquired from the social environment can help farmers update their perceptions of new technologies, enhance trust, reduce perceived risks, and ultimately encourage the adoption of crop protection drones (). The total amount of social resources accessible to an individual is referred to as social capital (). Does social capital influence farmers’ adoption of crop protection drone technology? If so, what are the mechanisms through which this influence occurs, and are there heterogeneities in these effects? These questions have received limited attention in the existing literature and warrant further exploration.
Some studies have explored the influence of social capital on farmers’ adoption of precision agricultural technologies, but these investigations remain incomplete and exhibit three main limitations. First, most existing research that incorporates social capital into analyses of farmers’ adoption behavior tends to focus on single dimensions of social capital, such as trust or norms, without conducting a comprehensive analysis from a systematic perspective (; ; ). Second, evaluations of farmers’ social capital levels have not adequately accounted for the specific social contexts in which they operate. Due to the lack of consensus in academia regarding the concept and definition of social capital, its meaning can vary significantly across different research contexts. Consequently, assessing individual social capital requires contextualization (Claudia et al., 2017). Rural China is undergoing profound social structural transformations driven by urbanization and modernization efforts. These changes have fundamentally altered traditional rural environments, shifting interpersonal relationships from the hierarchy of kinship-based networks to group-based structures. Interactions are increasingly guided by rational principles of cooperation and reciprocity rather than kinship proximity. As a result, the social capital of Chinese farmers today comprises both networks formed through reciprocal individual activities and influences rooted in kinship-based village society. While some studies have considered the multidimensional characteristics of farmers’ social capital in China, they predominantly focus on individual-level social capital, neglecting the role of village-level social capital (). Given the kinship-oriented nature of rural Chinese society, village-level social capital has a subtle yet profound influence on farmers’ decision-making, effectively guiding and constraining their adoption of new technologies. Therefore, current evaluations of social capital among Chinese farmers are somewhat biased. Third, existing research rarely investigates the pathways through which social capital influences farmers’ adoption of precision agricultural technologies from the perspective of input factors.
In summary, there is a notable gap in research on the impact of farmers’ social capital on the promotion of precision agricultural technologies. Taking the adoption of crop protection drones as an example, this study aims to address this gap by investigating how social capital influences farmers’ adoption behavior. It seeks to enrich the existing literature in three key ways: By considering the context of rural social transformation in China, this study draws on previous research (Claudia et al., 2017) to define farmers’ social capital from two dimensions: structural social capital at the individual level and cognitive social capital at the village level. A comprehensive evaluation model for farmers’ social capital is constructed accordingly. Using survey data collected from farmers in Henan Province, China, in 2024, this study provides micro-level empirical evidence on the effects of social capital on the adoption of crop protection drones. It also develops a theoretical framework for understanding how social capital facilitates adoption through two transmission pathways: access to productive credit and levels of off-farm employment. This study conducts an in-depth analysis of the heterogeneous effects of social capital on the adoption behavior of crop protection drones across different types of production decisions, farm sizes, and cropping patterns. The goal is to refine the applicable scenarios for crop protection drone adoption and provide new theoretical insights and practical tools for promoting precision agricultural technologies and crop protection drones.
The paper is organized as follows. An analytical framework and research hypotheses are proposed in Section 2. Study area, data collection and methods are introduced in Section 3. Results are presented in Section 4. And conclusions are summarized in Section 5.
2 Theoretical analysis and research hypotheses
2.1 Definition and measurement of farmers’ social capital
The concept of social capital was first introduced by Lyda Hanifan in 1916, who described it as the shared values and social relationships—such as trust, goodwill, and empathy—that individuals derive from community groups (). Since then, scholars have extensively debated the definition and research boundaries of social capital in relation to various study contexts. Although no consensus has been reached, two primary perspectives have emerged: First, from the perspective of economic development, social capital is defined at the individual level, emphasizing its relationship with human capital. This view regards social capital as the sum of actual or potential resources that individuals can access through their social networks (; ; ). Second, from a sociological standpoint, focusing on policy, institutions, and governance, social capital is characterized as the shared norms, trust, and cultural elements within a community that constitute cognitive resources (). Building on these perspectives, differentiated social capital into structural social capital and cognitive social capital. Structural social capital, also referred to as bonding social capital, is centered on the structural characteristics of individual social networks, such as network size, density, and diversity (). Cognitive social capital, on the other hand, pertains to the norms, trust, and cultural values intrinsic to social organizations. These cognitive resources are internalized by group members through social interactions and communication, fostering a shared consensus within the group ().
In the context of individual decision-making, scholars such as and , building on systematic approach to analyzing social issues, advocate for integrating individual-level and organizational-level factors to explore decision-making processes. They argue that individual social structures are shaped by organizational cognitive factors, while individual social structures, in turn, provide a platform for the dissemination of these cognitive factors. Through social interaction and social learning, these elements collectively guide individual decision-making. Building on this foundation, Claudia et al. (2017) proposed that when constructing a social capital framework to analyze individual decision-making, it is essential to incorporate both structural social capital at the individual level and cognitive social capital at the community level. The adoption of crop protection drone technology by farmers falls within the scope of individual decision-making. Therefore, to analyze farmers’ decision-making regarding the adoption of crop protection drones, this study defines farmers’ social capital as the sum of social resources obtained from individual-level social networks and village-level community environments. Specifically, farmers’ social capital is composed of two dimensions: resources derived from the structural characteristics of their social networks and cognitive resources such as shared norms, trust, and cultural values within their village community.
Based on the definition outlined earlier and drawing on existing literature (), this study examines farmers’ structural social capital through four dimensions: network size, network strength, network heterogeneity, and network reachability. Specifically: Network size refers to the total number of individuals with whom a farmer discusses agricultural production issues. Network strength reflects the closeness of a farmer’s connections with other stakeholders. Network heterogeneity captures the diversity of members’ roles and statuses within the farmer’s social network. Network reachability indicates the access farmers have to high-level resources within their network. Additionally, informed by prior studies (; ), this study evaluates farmers’ cognitive social capital through three dimensions: social norms, social trust, and social culture. Specifically: Social norms represent the agricultural production rules and guidelines endorsed by the majority of villagers. Social trust reflects the mutual trust commonly shared among members of the village community. Social culture pertains to the emphasis placed by the majority of villagers on preserving and passing down traditional farming practices.
2.2 Research hypotheses
2.2.1 The direct impact of social capital on farmers’ adoption of plant protection drone technology
Structural social capital plays a critical role in influencing farmers’ decisions to adopt plant protection drone technology. Farmers embedded in different social network structures have access to varying types of information, which leads to differences in their decision-making behavior (
Deepak et al., 2022). Specifically:
Social Network Size: Farmers with larger social networks are exposed to a broader range of information about the economic and ecological benefits of adopting plant protection drones. They also have more opportunities to observe the outcomes of adoption by others, which reduces their concerns about the risks associated with new technology and encourages its adoption ().
Social Network Strength: Farmers with stronger ties within their social networks are more likely to develop cooperative agreements with network members. This collaboration, such as through collective bargaining, can reduce the costs of acquiring technical services, thereby promoting the adoption of plant protection drones ().
Social Network Heterogeneity: Unlike homogeneous networks, which often circulate redundant and repetitive information, heterogeneous networks provide farmers with diverse and non-redundant information. This enables farmers to gain a comprehensive understanding of the technical benefits of plant protection drones, reducing information asymmetry and increasing the likelihood of adoption. Additionally, farmers in heterogeneous networks are more likely to establish close connections with technical service providers, such as cooperatives and agricultural extension agents, facilitating easier access to technical support and after-sales services. These factors lower the perceived risks of adoption and enhance adoption rates (; Deepak et al., 2022).
Social Network Reachability: Farmers with high network reachability often maintain closer ties with government officials. This increases their awareness of government policies supporting plant protection drone technology and enhances their perception of the economic benefits due to the credibility of government endorsements, thereby boosting the likelihood of adoption.
In summary, structural social capital significantly promotes farmers’ adoption of plant protection drone technology by providing access to valuable resources, reducing risk perceptions, and fostering trust and collaboration.
Cognitive social capital significantly influences farmers’ adoption of plant protection drone technology by shaping the norms, attitudes, and motivations underpinning their behavioral decisions, as suggested by the Theory of Planned Behavior. Specifically:
Behavioral Constraining Effect of Social Norms: Social norms create coercive pressure on farmers to align their agricultural production decisions with the majority of their village counterparts (). When a social norm emerges within a village that emphasizes the perceived importance of adopting plant protection drone technology, farmers who deviate from this norm risk condemnation and ostracism from other villagers. Such consequences, including barriers to agricultural cooperation and damage to social reputation, motivate farmers to conform to the norm. By adhering to collective expectations, farmers are more likely to adopt agricultural technologies, thereby increasing the probability of their adoption of plant protection drones ().
Behavioral Conversion Effect of Social Trust: Social trust facilitates mutual assistance and information exchange among farmers, acting as a catalyst for translating the intention to adopt plant protection drones into action (; ). In cooperative production environments, early adopters of plant protection drones often disseminate information about the technology’s advantages, such as improved fertilizer and pesticide efficiency, reduced labor costs, enhanced farmland ecology, and lower risks of poisoning incidents among agricultural workers. Meanwhile, in trust-based social environments, non-adopting farmers are more inclined to believe in the reliability of these benefits and actively seek advice from early adopters, amplifying the demonstration effect and encouraging wider adoption.
Behavioral Guiding Effect of Social Culture: Social culture subtly shapes individual behavioral orientations (). As an integral part of rural social culture, farming culture encapsulates agricultural practices and lifestyle norms, profoundly influencing farmers’ decision-making (). Farming culture emphasizes the preservation of the environment and ecological balance, guiding farmers to minimize damage to natural resources and adopt sustainable agricultural practices. As a modern green digital technology, plant protection drones execute precision operations for pesticide application, irrigation, and fertilization, thereby reducing non-point source pollution and enhancing farmland sustainability. Consequently, farmers in regions that prioritize traditional farming culture are more likely to adopt such technologies. Additionally, for farmers in areas where farming culture and economic livelihoods are tightly interconnected, the pursuit of cost reduction and efficiency becomes a key driver. Plant protection drones, by improving fertilizer and pesticide utilization and reducing labor input, meet farmers’ economic goals, further increasing adoption likelihood.
In conclusion, cognitive social capital facilitates the adoption of plant protection drone technology by influencing social norms, trust, and cultural values. Based on this, the following hypotheses are proposed:
H1: Social capital promotes farmers’ adoption of plant protection drone technology.
H1a: Structural social capital has a positive effect on farmers’ adoption of plant protection drone technology.
H1b: Cognitive social capital has a positive effect on farmers’ adoption of plant protection drone technology.
2.2.2 The indirect impact of social capital on farmers’ adoption of plant protection drone technology
Based on investment theory, technology functions as a production factor, and farmers’ adoption of plant protection drone technology can essentially be considered an investment decision. Generally, in agricultural production, whether farmers invest in technology is influenced by their endowment of other production factors, such as financial resources and labor. Social capital indirectly affects farmers’ adoption of plant protection drone technology by influencing their access to productive credit and off-farm employment opportunities. Specifically:
Social capital enhances farmers’ access to productive credit, thereby facilitating their adoption of plant protection drone technology. Farmers with high levels of structural social capital maintain close connections with other network members, enabling these members to gain a better understanding of their financial conditions and personal creditworthiness through daily interactions. This reduces information asymmetry between lenders and borrowers, increasing farmers’ likelihood of obtaining productive credit from friends and relatives (). Additionally, farmers with higher structural social capital often maintain stronger ties with village committee members and government officials, allowing them to stay informed about agricultural credit loan and guarantee policies. This increases their awareness of relevant policies, mitigates their perceptions of difficulty or resistance toward formal productive credit, and enhances their likelihood of accessing such loans (). Furthermore, farmers with higher cognitive social capital tend to live in villages with strong traditions of cultural preservation. In such villages, social norms and customs—such as sanctions or ostracism—are established to monitor and constrain loan default behavior among farmers. This increases the likelihood of other villagers providing credit and enhances farmers’ access to productive loans (). Historically, constraints on productive funding have been a key factor limiting farmers’ decisions to adopt new technologies (). Access to productive credit alleviates financial pressures during agricultural production and operations, enabling farmers to transition to modern production methods and increasing the likelihood of adopting plant protection drone technology.
H2: Social capital facilitates farmers’ adoption of plant protection drone technology by enhancing their access to productive credit.
Social capital influences farmers’ adoption of plant protection drone technology by enhancing their non-farm employment levels. According to farmer behavior theory, transaction costs are a critical factor affecting farmers’ decisions regarding non-farm employment (). Farmers with higher levels of structural social capital possess extensive and diversified social networks, which reduce the cost of searching for non-farm employment information and increase access to high-quality and stable non-farm job opportunities. Additionally, these farmers maintain close ties with their network members, which lowers the likelihood of fraud or losses during their pursuit of non-farm employment, further reducing transaction costs and improving the probability of securing such opportunities. Moreover, farmers with higher levels of cognitive social capital demonstrate higher trust in other villagers and are more likely to follow their peers in pursuing non-farm employment. This collective behavior fosters a mutual aid effect among farmers from the same village, reducing living costs during migration for work and increasing the likelihood of non-farm employment (). As a critical production factor, non-farm employment significantly influences farmers’ decisions regarding technology adoption (). When farmers have low levels of non-farm employment, agricultural income constitutes a major portion of their total household income. In such cases, farmers tend to adopt agricultural technologies to ensure the long-term stability of their agricultural income (). Plant protection drone technology, representing agricultural technological advancements, improves the efficiency of input utilization, such as fertilizers, reduces production costs, enhances soil quality, and ensures stable and orderly agricultural production. Therefore, non-farm employment can encourage farmers to adopt plant protection drone technology.
Conversely, when farmers’ non-farm employment levels are high, non-farm income accounts for a larger proportion of their household income, reducing their reliance on and attention to agricultural production. Additionally, in such scenarios, the higher out-migration of household labor increases the monitoring costs associated with implementing new technologies, thereby lowering the motivation to adopt agricultural production technologies for ensuring agricultural income (). As a result, non-farm employment may decrease farmers’ likelihood of adopting plant protection drone technology.
H3: Social capital influences farmers’ adoption of plant protection drone technology by enhancing their non-farm employment levels.
3 Materials and methods
3.1 Data sources
The data for this study were sourced from a farmer survey conducted by the research team in July 2024 in Henan Province, one of China’s major grain-producing regions. The survey took place in Huaxian County of Anyang City, and Dancheng and Taikang Counties of Zhoukou City. As a key agricultural province, Henan is one of China’s 13 major grain-producing areas and has advanced agricultural technology. Henan was one of the earliest regions in China to promote plant protection drone technology, and local farmers have a relatively high level of awareness and adoption of this technology. Therefore, the study area serves as a representative sample for investigating the adoption of plant protection drone technology among farmers.
Based on a combination of stratified and random sampling methods, the research team considered factors such as agricultural production conditions, the development level of digital agriculture, and population density. A total of 10–14 towns were selected from each county, 2–3 villages were chosen from each town, and 6–10 households were randomly surveyed from each village. The questionnaire covered topics including basic household information, production and living conditions, digital agriculture practices, and village characteristics.
In total, 900 questionnaires were completed during the survey. After excluding incomplete responses relevant to the study, 857 valid questionnaires were obtained, resulting in an effective response rate of 95.22%.
3.2 Variable selection
3.2.1 Dependent variable
Farmers’ Adoption of Plant Protection Drone Technology. Based on existing research and the current situation of the surveyed area (Cheryl et al., 2021), this study defines the adoption of plant protection drone technology by farmers as their purchase of agricultural drone services for tasks such as fertilization or pesticide application. Furthermore, considering that the adoption behavior of farmers involves the entire process from cognition, evaluation, to decision-making (), in order to deeply understand the adoption behavior of farmers regarding plant protection drones and to promote the development of targeted technology dissemination measures, this study references existing research () and uses the Trans-theoretical Model (TTM) to decompose and quantify the different stages and behavioral change processes of farmers’ adoption of plant protection drone technology. The different stages of adoption are assigned values ranging from 1 to 4.
3.2.2 Independent variable
Social Capital. Based on existing literature (Claudia et al., 2017), this study constructs an evaluation index system for farmers’ social capital from two dimensions: cognitive social capital and structural social capital. The social capital of farmers is then assessed using a combined subjective and objective entropy weight method. The subjective, objective, and relative weights between them are determined through two rounds of anonymous scoring by 45 experts. We invited 45 experts to participate in the Delphi scoring process. The selection criteria were: (a) having a doctoral degree or senior professional title in relevant fields such as agricultural economics, rural sociology, or gender studies; (b) having published at least three peer-reviewed papers on topics related to rural development, agricultural technology adoption, or gender issues; and (c) having more than five years of research experience in their respective fields. The Delphi process consisted of two rounds. In the first round, we distributed questionnaires to all 45 experts, asking them to assign importance weights to each indicator on a 5-point Likert scale. After collecting the first-round responses, we calculated the mean scores and standard deviations for each indicator, and summarized the results. In the second round, we provided each expert with the summary statistics from the first round, along with their own previous responses, and invited them to revise their scores if they wished. This process continued until consensus was reached (defined as a reduction in standard deviation to below 0.5 for each indicator). Ultimately, all 10 experts completed both rounds, and consensus was achieved.
3.2.3 Mechanism variables
Drawing on existing studies (; ) and theoretical analysis, this study selects credit accessibility and non-agricultural employment level as the mechanism variables.
3.2.3.1 Control variables
Based on existing literature (), control variables are chosen from four dimensions: characteristics of agricultural decision-makers, family characteristics, agricultural production characteristics, and village characteristics. Additionally, a county-level dummy variable is included to control for regional differences.
3.2.4 Instrumental variables
To overcome potential endogeneity issues such as reverse causality and omitted variables, this study uses an instrumental variable approach to test the empirical results. The instrumental variable is the average level of social capital of other respondents residing in the same village, excluding the respondent themselves. This is because social capital within the same village tends to be similar due to village environment and cultural norms, and individual social capital is influenced by the social capital levels of others in the same village, meeting the relevance requirement. Moreover, the adoption behavior of plant protection drone technology by the surveyed farmers is not directly related to the social capital of other farmers, fulfilling the exogeneity requirement (Tables 1, 2). The definitions of each variable and their descriptive statistics are shown in Table 3.
Table 1
| Stage | Theoretical concept | Variable meaning | Dependent variable code | Sample size |
|---|---|---|---|---|
| Pre-contemplation | Individual does not intend to change behavior | Farmers are unwilling to use plant protection drone technology | 1 | 359 |
| Contemplation | Individual begins to consider changing behavior | Farmers are willing to use plant protection drone technology | 2 | 128 |
| Preparation | Individual has a plan to change behavior | Farmers have a plan to use plant protection drone technology | 3 | 49 |
| Action | Individual’s behavior has changed | Farmers have already adopted plant protection drone technology | 4 | 321 |
Trans-theoretical model for farmers’ adoption of plant protection drone technology (Total sample size = 857).
Table 2
| Dimension | Classification | Indicator | Meaning | Objective weight | Subjective weight | Comprehensive weight |
|---|---|---|---|---|---|---|
| Structural social capital | Social network size | Total number of people in your agricultural production exchange network | Actual number | 0.1432 | 0.1428 | 0.1430 |
| Social network strength | Frequency of communication with other entities about agricultural production issues | Very low = 1; Low = 2; Average = 3; High = 4; Very high = 5 | 0.1428 | 0.1429 | 0.1429 | |
| Social network heterogeneity | Ii refers to percentage distribution of various members within the network | 0.1428 | 0.1429 | 0.1429 | ||
| Social network accessibility | Highest level of government personnel in your agricultural production communication network | Unfamiliar = 0; Township level = 1; County level = 2; City level = 3; Provincial level = 4; National level = 5 | 0.1432 | 0.1432 | 0.1432 | |
| Cognitive social capital | Social norms | Majority of farmers in the village consider plant protection technology important | Strongly disagree = 1; Disagree = 2; Neutral = 3; Agree = 4; Strongly agree = 5 | 0.1426 | 0.1427 | 0.1426 |
| Social trust | Majority of farmers in the village have a mutual aid consciousness | Strongly disagree = 1; Disagree = 2; Neutral = 3; Agree = 4; Strongly agree = 5 | 0.1427 | 0.1428 | 0.1427 | |
| Social culture | Majority of farmers in the village emphasize the inheritance of agricultural culture | Strongly disagree = 1; Disagree = 2; Neutral = 3; Agree = 4; Strongly agree = 5 | 0.1427 | 0.1427 | 0.1427 |
Evaluation index system for farmers’ social capital.
Comprehensive weight = Objective weight × 0.455 + Subjective weight × 0.545, where 0.455 and 0.545 are determined by expert scoring methods.
Table 3
| Type | Variable name | Definition | Mean | Standard deviation |
|---|---|---|---|---|
| Dependent variable | Farmer adoption of plant protection drone technology | Refer to Table 1 | 2.3873 | 1.3514 |
| Independent variables | Structural social capital | Refer to Table 2 | 0.1438 | 0.0699 |
| Cognitive social capital | Refer to Table 2 | 0.3265 | 0.0614 | |
| Social capital | Refer to Table 2 | 0.0614 | 0.0990 | |
| Mechanism variables | Credit accessibility | Very difficult = 1; Difficult = 2; Average = 3; Easy = 4; Very easy = 5 | 1.8646 | 1.2744 |
| Non-agricultural employment level | Non-agricultural employment number/total labor force | 0.2597 | 0.2586 | |
| Control variables | Agricultural decision-maker’s social experience | Farmer = 1; Village leader = 2; Cooperative member = 3 | 0.4130 | 0.7360 |
| Agricultural decision-maker’s education | 0 = No education; 1 = Primary school; 2 = Middle school; 3 = High school; 4 = Vocational college; 5 = Junior college; 6 = University; 7 = Postgraduate | 1.9054 | 1.1577 | |
| Digital agricultural technology understanding | Very unfamiliar = 1; Unfamiliar = 2; Average = 3; Somewhat familiar = 4; Very familiar = 5 | 2.4457 | 1.2312 | |
| Plan to retire with farmland | No = 0; Yes = 1 | 0.6102 | 0.4879 | |
| Total family income | Actual value in logarithm | 11.4741 | 1.1492 | |
| Family life cycle | No dependents = 1; Child-rearing family = 2; Elderly-care family = 3; Child-rearing and elderly-care family = 4 | 1.8109 | 0.9517 | |
| Agricultural production decision type | Male household head independently decides = 1; Both spouses jointly decide = 2 | 1.6732 | 0.4692 | |
| Agricultural production area | Actual value in logarithm | 2.7168 | 1.1452 | |
| Agricultural production crop structure | Area of food crops/total agricultural production land area | 0.9399 | 0.1713 | |
| Agricultural production experience with disasters | No = 0; Yes = 1 | 0.7339 | 0.4421 | |
| Village leaders’ governance satisfaction | Very dissatisfied = 1; Dissatisfied = 2; Neutral = 3; Satisfied = 4; Very satisfied = 5 | 3.7631 | 1.0063 | |
| Number of village technicians | Actual value | 0.7047 | 2.5913 | |
| Distance from village to township | Actual value | 5.0908 | 4.3326 | |
| Instrumental variable | Social capital of other samples in same village | Actual value | 0.4648 | 0.0742 |
Descriptive statistics of variables.
“No dependents family” refers to households with no children under 16 and no elderly members over 65. “Child-rearing family” refers to households with children under 16 and no elderly members over 65. “Elderly-care family” refers to households with no children under 16 and elderly members over 65. “Child-rearing and elderly-care family” refers to households with both children under 16 and elderly members over 65.
3.3 Model setting
Since the dependent variable is ordinal, this study applies the ordered Probit model for the baseline regression. The model is specified as follows:
Equation 1, where, represents the adoption behavior of the farmer regarding plant protection drone technology, with four possible categories: unwilling to adopt, willing to adopt, planning to adopt, and already adopted. represents the social capital level of the farmer. represents the control variables, is the error term. , , are the coefficients to be estimated.
4 Results
4.1 Benchmark regression
Table 4 reports the estimated results of the impact of social capital on farmers’ adoption of plant protection drone technology. The results in column (1) show that both structural social capital and cognitive social capital influence farmers’ adoption behavior, reducing their unwillingness to adopt and increasing their willingness and actual adoption of the technology. Moreover, compared to structural social capital, cognitive social capital has a greater impact on farmers’ adoption behavior of plant protection drone technology. This suggests that although rural China is gradually transitioning from a relationship-based society (rooted in kinship and locality) to a contract-based society driven by market forces, for farmers deeply rooted in rural society, village-level social norms, trust, and culture still have a strong influence on their behavior. Additionally, this result highlights the significant role of village-level social capital in farmers’ production decision-making, further validating the inclusion of village-level social capital in the farmer social capital assessment framework. The results in column (2) indicate that social capital promotes farmers’ adoption of plant protection drone technology, providing preliminary support for Hypothesis H1.
Table 4
| Variable | Farmers’ plant protection drone adoption behavior (1) | Plant protection drone adoption behavior (2) | |||
|---|---|---|---|---|---|
| Dependent variable = 1 | Dependent variable = 2 | Dependent variable = 3 | Dependent variable = 4 | ||
| Structural social capital | −0.4378** (0.2229) | 0.0023 (0.0062) | 0.0173* (0.0093) | 0.4181** (0.2131) | |
| Cognitive social capital | −0.5202** (0.2591) | 0.0027 (0.0074) | 0.0205* (0.0108) | 0.4968** (0.2476) | |
| Social capital | 0.9955** (0.4908) | ||||
| Agricultural decision-maker’s social experience | −0.0151 (0.0216) | 0.0080 (0.0248) | 0.0060 (0.0086) | 0.0145 (0.0206) | 0.0448 (0.0635) |
| Agricultural decision-maker’s education | −0.0253* (0.0131) | 0.0013 (0.0035) | 0.0100* (0.0054) | 0.0242* (0.0126) | 0.0747* (0.0387) |
| Digital agricultural technology understanding | −0.1162*** (0.0111) | 0.0061 (0.0163) | 0.0459*** (0.0096) | 0.1110*** (0.0105) | 0.3407*** (0.0367) |
| Plan to retire with farmland | 0.0273 (0.0296) | −0.0014 (0.0041) | −0.0010 (0.0011) | −0.0260 (0.0282) | −0.0898 (0.0865) |
| Total family income | −0.0385** (0.0163) | 0.0020 (0.0054) | 0.0152** (0.0070) | 0.0367** (0.0156) | 0.1165** (0.0482) |
| Family life cycle | −0.0010 (0.0151) | 0.0056 (0.0821) | 0.0042 (0.0601) | 0.0010 (0.0145) | 0.0082 (0.4419) |
| Agricultural production decision type | 0.0083 (0.0155) | −0.0044 (0.0141) | −0.0032 (0.0061) | −0.0079 (0.0148) | −0.0205 (0.0451) |
| Agricultural production area | −0.0148 (0.0180) | 0.0078 (0.0235) | 0.0058 (0.0072) | 0.0142 (0.0172) | 0.0470 (0.0530) |
| Agricultural production crop structure | −0.0034 (0.8286) | 0.0018 (0.4391) | 0.0013 (0.3277) | 0.0033 (0.7914) | 0.0013 (0.2434) |
| Agricultural production experience with disasters | 0.0214 (0.0324) | −0.0011 (0.0034) | −0.0084 (0.0129) | −0.0204 (0.0310) | −0.0611 (0.0942) |
| Village leaders’ governance satisfaction | −0.0075 (0.0167) | 0.0039 (0.0136) | 0.0029 (0.0066) | 0.0071 (0.0160) | 0.0372 (0.0489) |
| Number of village technicians | −0.0117** (0.0056) | 0.0062 (0.0169) | 0.0046* (0.0024) | 0.0112** (0.0053) | 0.0350** (0.0166) |
| Distance from village to township | −0.0060* (0.0032) | 0.0031 (0.0086) | 0.0023* (0.0013) | 0.0057* (0.0030) | 0.0169* (0.0095) |
| County-level dummy variables | Controlled | Controlled | |||
| Pseudo R2 | 0.0901 | 0.0879 | |||
Baseline regression results.
Column (1) reports marginal effects, and Column (2) reports impact coefficients.
Among the control variables, the education level of agricultural decision-makers and the understanding of digital agricultural technology have a positive impact on farmers’ adoption decisions. This may be because farmers with higher education levels have a better understanding and acceptance of new technologies like plant protection drones. As their understanding of the technology increases, their awareness of the economic benefits post-adoption also rises, thereby increasing the likelihood of adoption. Family income also has a positive effect on farmers’ adoption decisions, likely because as household income increases, farmers have more financial resources to invest in agricultural technology services, which makes adoption more feasible. The number of village-level technical personnel positively affects farmers’ adoption decisions, possibly because a higher number of technical personnel in the village facilitates more frequent and effective promotion of plant protection drone technology to farmers, increasing their awareness of the technology and its benefits. The distance from the village to the township also has a positive impact on adoption decisions, as farmers living farther away from townships may face higher costs for external employment. As their livelihoods are more reliant on agriculture, these farmers are more likely to prioritize agricultural income and are thus more inclined to adopt plant protection drone technology to improve input efficiency and prevent pest and disease outbreaks.
4.2 Endogeneity test
Farmers adopting plant protection drone technology may expand their social networks and strengthen social ties with other farmers through demonstration effects. However, this introduces potential reverse causality issues that could bias the empirical estimation. To address this, an instrumental variable (IV) approach is employed to test the robustness of the baseline regression results. Table 5 presents the regression outcomes from both the two-stage least squares (2SLS) and the conditional mixed process (CMP) models.
Table 5
| Variable | 2SLS model | CMP model | ||
|---|---|---|---|---|
| Social capital | Adoption of plant protection drones | Social capital | Adoption of plant protection drones | |
| Instrumental variable | 0.4793*** (0.0571) | 0.4793*** (0.0396) | ||
| Social capital | 2.9951** (1.2462) | 3.5439*** (1.2130) | ||
| Control variables | Controlled | Controlled | ||
| Weak instrument test | 70.4049 | |||
| DWH test | 3.8032* | |||
| atanhrho | −0.2339** | |||
Instrumental variable analysis.
First, the results of the 2SLS model indicate that the Durbin–Wu–Hausman (DWH) test is significant, confirming the presence of endogeneity. Additionally, the F-test indicates no weak instrument problem, and the instrumental variable exhibits a significant effect on farmers’ social capital, satisfying the relevance condition. The regression results reveal that social capital has a statistically significant and positive effect on farmers’ adoption of plant protection drone technology at the 1% level.
Second, the CMP model results show that the atanhrho statistic significantly rejects the null hypothesis of social capital being an exogenous variable, further confirming the presence of endogeneity. The regression results also demonstrate that social capital positively and significantly influences the adoption of plant protection drone technology at the 1% level.
The findings from the instrumental variable analysis suggest that, after accounting for potential endogeneity, social capital has a significant positive impact on farmers’ adoption of plant protection drone technology, providing strong support for research hypothesis H1.
The earlier analysis employed the 2SLS model and the CMP approach to address endogeneity issues as comprehensively as possible. However, due to limitations in data and variables, the impact of social capital on farmers’ adoption of plant protection drone technology may still be affected by self-selection bias, potentially leading to selective estimation errors. To address this, the study divides farmers into two groups—low social capital and high social capital—based on whether their social capital level exceeds the sample mean. An endogenous switching regression (ESR) model is then applied to construct a counterfactual framework to analyze the effect of social capital on farmers’ adoption of plant protection drone technology, mitigating potential selection bias in the model. In the simultaneous estimation of the selection and outcome equations, the Wald test is significant at the 1% level, rejecting the null hypothesis that the selection equation and outcome equation are independent. Additionally, the significance of the correlation coefficients at the 1% level confirms the presence of self-selection bias, validating the use of the ESR model (). On this basis, the study further calculates the Average Treatment Effect (ATE) to estimate the impact of social capital on farmers’ adoption of plant protection drone technology. Table 6 reports the ATE estimation results. For farmers with high social capital, the probability of adopting plant protection drone technology is 2.9586. Under the counterfactual scenario where these farmers are assumed to have low social capital, their adoption probability decreases significantly to 2.6039. Similarly, for farmers with low social capital, the probability of adoption is 2.1943. Under the counterfactual scenario where these farmers are assumed to have high social capital, their adoption probability increases significantly to 2.9586. These results demonstrate that social capital positively and significantly influences farmers’ adoption of plant protection drone technology, consistent with the baseline regression results. This confirms the robustness of the baseline findings.
Table 6
| Farmer category | High social capital | Low social capital | ATT | ATU |
|---|---|---|---|---|
| Low social capital | 2.9586 (0.0270) | 2.6039 (0.0288) | 0.3547*** (0.0395) | |
| Low social capital | 2.9586 (0.0270) | 2.1943 (0.0269) | 0.7642*** (0.0382) |
Average treatment effect (ATE) estimation.
Drawing on existing research (), this study evaluates the impact of unobserved variables on the model results by comparing the proportional differences in the OLS coefficients of the social capital variable under restricted and unrestricted sets of control variables. County-level fixed effects are included in all restricted and unrestricted models. The effect of omitted variables is assessed based on observable factors, using the formula:
Where and represent the coefficients of the social capital variable in restricted and unrestricted models, respectively. Three groups of restricted control variables are considered in this analysis: (1) only the characteristics of agricultural decision-makers, (2) characteristics of agricultural decision-makers and household characteristics, and (3) characteristics of agricultural decision-makers, household characteristics, and agricultural production characteristics. The unrestricted model includes all control variables.
Table 7 presents the proportional differences calculated based on the regression results for the three groups of restricted control variables. The results indicate that the proportional differences range between 1.8183 and 5.4182, with an average greater than 1. This implies that for unobserved omitted variables to have a disruptive effect on the baseline regression results, there would need to be an implausibly large number of omitted variables compared to those already included. Thus, the likelihood of omitted variables substantially biasing the results is minimal. These findings suggest that even in the presence of omitted variables, the baseline regression results remain robust.
Table 7
| Control variables | Coefficient (restricted) | Coefficient (unrestricted) | Proportional difference |
|---|---|---|---|
| Agricultural decision-maker characteristics | 1.2891*** (0.4345) | 0.8317* (0.4979) | 1.8183 |
| Agricultural decision-maker and household characteristics | 0.9945** (0.4429) | 0.8317* (0.4979) | 5.1087 |
| Agricultural decision-maker, household, and agricultural production characteristics | 0.9852** (0.4466) | 0.8317* (0.4979) | 5.4182 |
Omitted variable bias test.
4.3 Robustness test
Further robustness checks were conducted on the baseline regression results by altering the dependent variable, changing the sample, and removing outliers, as shown in Table 8.
Table 8
| Variable | Altering dependent variable | Changing sample | Changing sample |
|---|---|---|---|
| Adoption of plant protection drones | Adoption of plant protection drones | Adoption of plant protection drones | |
| Social capital | 0.9757* (0.4908) | 1.3442** (0.6718) | 0. 9341* (0.4972) |
| Control variables | Controlled | Controlled | Controlled |
| Pseudo R2 | 0.1306 | 0.0866 | 0.0899 |
| Observations | 857 | 684 | 857 |
Robustness checks.
First, altering the dependent variable: The original four-category variable for farmers’ adoption of plant protection drone technology was replaced with a binary variable, where farmers who had adopted drones for fertilization or pesticide application were assigned a value of 1, and others a value of 0. A Probit model was used for the empirical analysis, and the results indicate that social capital has a significant positive effect on farmers’ adoption of plant protection drone technology.
Second, changing the sample: A new sample was created by randomly selecting 80% of the original sample for regression analysis. The results confirm that social capital significantly and positively affects farmers’ adoption of plant protection drone technology.
Finally, removing outliers: All variables were winsorized at the 1st and 99th percentiles to eliminate extreme values, and the regression was repeated. The results again demonstrate a significant positive effect of social capital on farmers’ adoption of plant protection drone technology.
Overall, the robustness check results are consistent with the baseline regression findings, confirming the robustness of the baseline results.
4.4 Pathway analysis
According to theoretical analysis, access to productive credit and the level of non-farm employment may serve as the mechanisms through which social capital influences farmers’ doption of plant protection drone technology. A two-step method was employed to examine these pathways, with the results presented in Table 9.
Table 9
| Variable | Access to credit | Access to credit | Non-farm employment | Non-farm employment |
|---|---|---|---|---|
| Structural social capital | 3.6625*** (0.6699) | 0.2579** (0.1206) | ||
| Cognitive social capital | 2.8580*** (0.8184) | 0.1206** (0.1403) | ||
| Social capital | 3.4366*** (0.5280) | 0.2609*** (0.0908) | ||
| Control variables | Controlled | Controlled | Controlled | Controlled |
| Pseudo R2 | 0.0878 | 0.0875 | ||
| R-squared | 0.2856 | 0.2829 |
Pathway mechanism analysis.
First, the pathway mechanism of access to productive credit was tested. The empirical results show that social capital enhances farmers’ access to credit. Farmers with higher social capital maintain close ties with other network members, who are more familiar with their credibility and financial conditions. This reduces information asymmetry between borrowers and lenders, increasing the likelihood of farmers obtaining productive credit. Enhanced access to productive credit enables farmers to transform their agricultural production practices, thereby increasing their likelihood of adopting plant protection drone technology. These findings support Hypothesis H2, indicating that social capital promotes farmers’ adoption of drone technology by improving access to credit.
Second, the pathway mechanism of non-farm employment levels was examined. The results indicate that social capital enhances farmers’ non-farm employment levels. Farmers with high structural social capital possess extensive and diverse social networks, which lower the cost of searching for employment information and facilitate access to high-quality and stable non-farm job opportunities. As non-farm employment levels increase, farmers become less dependent on agricultural income, reducing their motivation to adopt plant protection drone technology services. These findings support Hypothesis H3, demonstrating that social capital influences farmers’ adoption behavior by shaping non-farm employment opportunities.
4.5 Heterogeneity analysis
The analysis above demonstrates that social capital promotes farmers’ adoption of plant protection drone technology. However, prior research has highlighted that farmers’ production characteristics significantly influence both their level of social capital and their technology adoption behavior (). This raises the question: Does the impact of social capital on the adoption of plant protection drones vary across different types of agricultural decision-making, scales of operation, and cropping structures? To address this, a heterogeneity analysis was conducted, with the results presented in Table 10.
Table 10
| Variable | Decision-making type analysis | Operational scale analysis | Cropping structure analysis | |||
|---|---|---|---|---|---|---|
| Male decision | Joint decision | Smallholder farmers | Large-scale farmers | Single-grain crop | Diversified cropping | |
| Social capital | 1.3441 (0.8941) | 1.3319** (0.6259) | 1.1497** (0.5565) | 0.5242 (1.0940) | 0.9347* (0.5247) | 1.6086 (1.4674) |
| Control variables | Controlled | Controlled | Controlled | Controlled | Controlled | Controlled |
| Pseudo R2 | 0.0893 | 0.1079 | 0.0898 | 0.0714 | 0.0869 | 0.1193 |
| Observations | 280 | 577 | 638 | 219 | 724 | 133 |
Heterogeneity analysis.
Farmers were grouped by operational scale using a threshold of 20,000 m2. Farmers operating less than 20,000 m2 were categorized as smallholders (assigned a value of 0), while those operating 20,000 m2 or more were categorized as large-scale farmers (assigned a value of 1). Farmers were grouped by cropping structure. Mono-cropping grain farmers were assigned a value of 0, while those engaged in both grain and cash crops were classified as diversified cropping farmers (assigned a value of 1).
First, heterogeneity by agricultural decision-making type. In traditional agrarian societies, agricultural decisions were predominantly made by male household heads. Consequently, much of the existing research on technology adoption relies heavily on male decision-makers’ data. However, with the significant outflow of male agricultural laborers, female spouses have increasingly participated in agricultural decision-making. Studies focusing solely on male decision-makers risk producing biased estimates (). This study analyzes heterogeneity by household agricultural decision-making types to explore whether these differences influence the relationship between social capital and technology adoption. The results indicate that compared to households where males make decisions independently, social capital has a stronger positive effect on the adoption of plant protection drone technology in households where decisions are made jointly by both male and female members. Joint decision-making implies shared risk, with both parties leveraging their respective social networks to gather comprehensive information on the benefits and risks of new technologies. This reduces biases or limitations associated with single-gender decision-making and enhances the likelihood of adoption. These findings underscore the importance of considering gender differences and decision-making types when promoting the adoption of agricultural innovations.
Second, heterogeneity by scale of operation. The results reveal that compared to large-scale farmers, social capital has a stronger impact on smallholder farmers’ adoption of plant protection drone technology. This is likely because social capital facilitates access to technical information and enhances farmers’ access to productive credit, thereby promoting adoption decisions. Large-scale farmers typically possess abundant material and cultural capital, allowing them to independently acquire information and evaluate the risks of adopting drone technology without relying heavily on social capital. Smallholders, in contrast, often lack sufficient material and cultural resources and have limited capacity to withstand risks. They frequently learn about drone services through the demonstration effects of large-scale farmers. Consequently, social capital plays a more significant role in supporting smallholders’ adoption of plant protection drone technology. These findings suggest that social capital can substitute for material and human capital, alleviating resource constraints and enhancing smallholders’ ability to adopt new technologies.
Finally, heterogeneity by cropping structure. The results show that compared to farmers engaged in diversified cropping, social capital has a stronger positive effect on the adoption of plant protection drone technology among farmers engaged in mono-cropping of grains. This is likely because the study area, Henan Province, is a major grain production region where most farmers specialize in grain mono-cropping. Compared to farmers with diversified cropping systems, grain mono-cropping farmers participate in larger agricultural production groups, possess greater bargaining power for drone service costs, and face lower adoption costs. Additionally, regional food security policies prioritize demonstration projects for drone technology in grain production, increasing the likelihood of adoption among grain mono-cropping farmers.
5 Conclusions and policy implications
Based on survey data from 857 households in Henan Province, China, this study empirically examines the impact of social capital on farmers’ adoption of plant protection drone technology and its underlying mechanisms. The following conclusions are drawn:
Social capital significantly and positively influences farmers’ adoption of plant protection drone technology, with the effect of cognitive social capital exceeding that of structural social capital.
Social capital influences adoption behavior by improving farmers’ access to credit and non-farm employment opportunities.
Heterogeneity analysis reveals: (i) compared to households where males dominate agricultural decision-making, social capital has a stronger impact on drone adoption among households with joint male–female decision-making; (ii) compared to large-scale farmers, social capital more significantly promotes adoption among smallholder farmers; and (iii) compared to farmers engaged in diversified cropping, social capital has a stronger positive effect on adoption among those practicing monocropping of grain.
Based on these findings, the following policy recommendations are proposed:
First, enhance the promotion of plant protection drone technology. Include plant protection drones in agricultural machinery purchase subsidy programs and increase subsidy levels. Additionally, integrate drone technology into agricultural social services to improve farmers’ accessibility to the technology. Provide training to improve farmers’ digital agricultural technology literacy. Various initiatives should focus on cultivating innovative farmers and helping them recognize the economic benefits of adopting drone technology.
Second, recognize the role of social capital in driving farmers’ technology adoption decisions. On one hand, expand farmers’ structural social capital. Organize regular production experience-sharing events among new agricultural entities, encourage agricultural extension officers to conduct field visits, and strengthen connections between farmers, extension personnel, and agricultural organizations. This will broaden farmers’ social networks, enhance network diversity, and facilitate access to information about new technologies. On the other hand, enhance cognitive social capital within villages. Develop clear village rules and regulations to protect farmland quality, promote agricultural production, and preserve farming culture. Include appropriate reward and penalty mechanisms to ensure compliance and foster trust among villagers.
Third, develop tailored digital agricultural technology promotion strategies for farmers with different production and operational characteristics. Based on heterogeneous characteristics such as production decision types, farm sizes, and cropping patterns, targeted promotion strategies should be implemented to improve the relevance and effectiveness of technology adoption policies.
Due to the availability of micro-level survey data, the discussion of potential endogeneity issues related to social capital in this study requires further improvement. Furthermore, the current study focuses on the agricultural context of a developing country. The generalizability of our findings to agricultural systems in developed countries requires further validation. Future research could incorporate samples from multiple developed countries to systematically compare the heterogeneous effects of social capital on farmers’ adoption of precision agricultural technologies across different levels of economic development, agricultural operation systems, and technology extension models. Such efforts would help further elucidate the external validity of our findings.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving human participants were reviewed and approved by the Ethics Committee of the College of Economics and Management, Northwest A&F University. Written informed consent to participate in this study was provided by the participants.
Author contributions
LX: Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. ZY: Methodology, Supervision, Writing – review & editing. ZK: Formal analysis, Funding acquisition, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This research was supported by the project of the Ministry of Agriculture and Rural Affairs of China (Grant No. 10200071), and the Innovation Capability Support Program of Shaanxi Province in 2022 (Grant No. 2022KRW20). These funding sources had no involvement in study design, data collection, analysis, interpretation, or the decision to submit the article for publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
Aguilar-GallegosN.Muñoz-RodríguezM.Santoyo-CortésH. (2015). Information networks that generate economic value: a study on clusters of adopters of new or improved technologies and practices among oil palm growers in Mexico. Agric. Syst.135, 122–132. doi: 10.1016/j.agsy.2015.01.003
2
AltonjiJ. G.ElderT. E.TaberC. R. (2005). Selection on observed and unobserved variables: assessing the effectiveness of Catholic schools. J. Polit. Econ.113, 151–184. doi: 10.1086/426036
3
BourdieuP. (1989). Social space and symbolic power. Sociol Theory7, 14–25. doi: 10.2307/202060
4
BramleyR. G. V.OuzmanJ.GobbettD. L. (2017). Yield mapping at different scales to improve fertilizer decision making in the Australian sugar industry. Adv. Anim. Biosci.8, 630–634. doi: 10.1017/S2040470017000607
5
BurchanA. (2019). Public acceptance of drones: knowledge, attitudes, and practice. Technol. Soc.59:101180. doi: 10.1016/j.techsoc.2019.101180
6
CandiagoS.RemondinoF.De GiglioM.DubbiniM.GattelliM. (2015). Evaluating multispectral images and vegetation indices for precision farming applications from UAV images. Remote Sens.7, 4026–4047. doi: 10.3390/rs70404026
7
CaoH.ZhaoK. (2019). Off-farm employment, farmland protection policy awareness, and the adoption of environmentally friendly agricultural technologies: evidence from 1,422 survey responses in major grain-producing counties. J. Agricul. Tech. Econom.5, 52–65. doi: 10.13246/j.cnki.jae.2019.05.005
8
CherylW.FanL. F.ShiZ. (2021). Adoption of unmanned aerial vehicles for pesticide application: Role of social network, resource endowment, and perceptions. Technol. Soc., 64, 101470. doi: 10.1016/j.techsoc.2020.101470
9
CialdiniR. B.GoldsteinN. J. (2004). Social influence: compliance and conformity. Annu. Rev. Psychol.55, 591–621. doi: 10.1146/annurev.psych.55.090902.142015
10
ClaudiaH.AlejandraE.RobertoJ. R. (2017). Understanding the role of social capital in adoption decisions: An application to irrigation technology. Agric. Syst.153, 221–231. doi: 10.1016/j.agsy.2017.02.003
11
ColemanJ. S. (1988). Social capital in the creation of human capital. Am. J. Sociol.94, S95–S120. doi: 10.2307/2780243
12
DeepakV.AshokK.MishraP. K. (2022). Social networks, heterogeneity, and adoption of technologies: Evidence from India. Food Policy, 112, 102360. doi: 10.1016/j.foodpol.2022.102360
13
EU SCAR (2012). Agricultural knowledge and innovation Systems in Transition – a reflection paper. Brussels Belgium. doi: 10.2777/34991
14
European Commission (2018). Drones in Agriculture. Brussels: Belgium.
15
FederG.JustR.ZilbermanD. (1985). Adoption of agricultural innovations in developing countries: a survey. Econ. Dev. Cult. Chang.33, 255–298. doi: 10.2307/1153228
16
FengX. (2024). How does digital agricultural technology influence agricultural governance transformation?J. China Agricul. Univ.41, 56–77. doi: 10.13240/j.cnki.caujsse.20240416.004
17
FisherR. (2013). A gentleman's handshake: the role of social capital and trust in transforming information into usable knowledge. J. Rural. Stud.31, 13–22. doi: 10.1016/j.jrurstud.2013.02.006
18
FukuyamaF. (2001). Social capital, civil society and development. Third World Q.22, 20–40. doi: 10.2307/3993342
19
GranovetterM. (1985). Economic action and social structure: the problem of embeddedness. Am. J. Sociol.91, 481–510. doi: 10.2307/2780199
20
GuoC.WeiF. (2015). The impact of social capital on farmers' technology adoption behavior. J. Manag. Stud.28, 30–33. doi: 10.19714/j.cnki.1671-7465.2025.0031
21
HanX.LeiY.ZhenT. (2022). Analysis of factors influencing farmers' continuous usage intention of plant protection drones based on TAM. J. Southwest Univ. National48, 332–339. doi: 10.16815/j.cnki.11-5436/s.2022.36.024
22
HanJ.ZhuW.ZhangB. (2022). Research on the promotion of modern smart agriculture development through equipment and information coordination. China Engineering Science24, 55–63. doi: 10.16178/j.issn.0528-9017.20212219
23
LiY.ChenW. (2023). The impact of online social networks on the business performance of rural e-commerce: mechanism and evidence. China Rural Economy9, 165–184. doi: 10.20077/j.cnki.11-1262/f.2023.09.008
24
LiB.XuX. (2017). Social networks, information flow, and farmers' adoption of new technologies: a re-examination of Granovetter's "weak ties hypothesis". J. Agricul. Tech. Econom.12, 98–109. doi: 10.13246/j.cnki.jae.2017.12.009
25
LiJ.ZhangY. (2024). The historical logic and practical path of rural construction promoted by the "ten thousand villages project". J. Nanjing Agricul. Univ.24, 16–26. doi: 10.19714/j.cnki.1671-7465.2024.0015
26
LiC.ZhangQ.ZhouH. (2021). Social norms, economic incentives, and farmers' behavior in pesticide packaging waste recycling. J. Nanjing Agricul. Univ.21, 133–142. doi: 10.19714/j.cnki.1671-7465.2021.0013
27
LokshinM.SajaiaZ. (2004). Maximum likelihood estimation of endogenous switching regression models. Stata J.4, 282–289. doi: 10.1177/1536867X0400400306
28
LolloE. (2012). “Toward a theory of social capital definition: its dimensions and resulting social capital types,” in 14th World Congress of Social Economics, (Glasgow: United Kingdom).
29
LuJ.GuoR. (2023). Rural labor aging under the rural revitalization strategy: development trends, mechanism analysis, and coping strategies. J. China Agricul. Univ.40, 5–21. doi: 10.13240/j.cnki.caujsse.20230918.011
30
LvP.FengS.WangB. (2022). Farmers' land and labor resource allocation decisions and their influencing factors. Resour. Sci.44, 1577–1588. doi: 10.13950/j.cnki.jlu.2023.06.013
31
MaS.HeG.GuoJ. (2022). Welfare effects of digital agriculture: a deconstruction from the perspective of value re-creation and redistribution. Agricultural Economic Issues5, 10–26. doi: 10.13246/j.cnki.iae.2022.05.006
32
MariaL.MalabayabasA. K.MishraV. O. (2023). Joint decision-making, technology adoption and food security: evidence from rice varieties in eastern India. World Dev.171:106367. doi: 10.1016/j.worlddev.2023.106367,
33
MariusM.Cord-FriedrichH.OliverM. (2020). A trans-theoretical model for the adoption of drones by large-scale German farmers. J. Rural. Stud.75, 80–88. doi: 10.1016/j.jrurstud.2020.01.005
34
Mesas-CarrascosaF. J.SantanoD. V.MeroñoJ. E.de La OrdenM. S.García-FerrerA. (2015). Open source hardware to monitor environmental parameters in precision agriculture. Biosyst. Eng.137, 73–83. doi: 10.1016/j.biosystemseng.2015.07.005
35
MoskvitchK. (2015). Take off: are drones the future of farming?Eng. Technol.10, 62–66. doi: 10.1049/et.2015.0721
36
NarayanD.CassidyM. F. (2001). A dimensional approach to measuring social capital: development and validation of a social capital inventory. Curr. Sociol.49, 59–102. doi: 10.1177/0011392101049002006
37
OstromE. (1993). Design principles of long-enduring irrigation institutions. Water Resour. Res.29, 1907–1912. doi: 10.1029/92WR02991
38
PanH.LaduS.VulturiusM. (2022). Social networks and renewable energy technology adoption: empirical evidence from biogas adoption in China. Energy Econ.106:105789. doi: 10.1016/j.eneco.2021.105789
39
PannellD. J.MarshallG. R.BarrN. (2006). Understanding and promoting adoption of conservation practices by rural landholders. Aust. J. Exp. Agric.46, 1407–1424. doi: 10.1071/EA05037
40
PuigE.McFadyenA.GonzalezF., (2018). Advances in unmanned aerial systems and payload technologies for precision agriculture. In: ChenG. (Ed.), Advances in Agricultural Machinery and Technologies. CRC Press, Boca Raton, Florida. 227, 109601
41
PutnamR. D. (1993). The prosperous community. Social capital and public life. American Prospect13, 35–42.
42
RenZ.GuoY. (2023). The impact of environmental regulation and social capital on farmers' adoption of low-carbon agricultural technologies. J. Nat. Resour.38, 2872–2888. doi: 10.19629/j.cnki.34-1014/f.250123013
43
ValinH.SandsR. D.van der MensbruggheD.NelsonG. C.AhammadH.BlancE.et al. (2014). The future of food demand: understanding differences in global economic models. Agric. Econ.45, 51–67. doi: 10.1111/agec.12089
44
Van RijnF.BulteE.AdekunaleA. (2012). Social capital and agricultural innovation in sub-Saharan Africa. Agric. Syst.108, 112–122. doi: 10.1016/j.agsy.2011.12.003
45
WeiJ.GaoM. (2023). The impact of agricultural labor aging on the technology adoption of grain farmers: a case study of conservation tillage and high-quality seeds. China Soft Science12, 49–58. doi: 10.13240/j.cnki.caujsse.20250320.009
46
WenX.NieY. (2023). The impact of labor mobility on farmers' participation in internet finance: an empirical analysis from the perspective of social capital. Macroeconomic Research2, 60–75. doi: 10.16304/j.cnki.11-3952/f.2023.02.006
47
XieH.ShiJ.LengK. (2023). Differences in farmers' abandonment behavior of cultivated land and its influencing factors from the perspective of the family life cycle: a case study of the hilly areas in Jiangxi Province. Resour. Sci.45, 2170–2182. doi: 10.13448/j.cnki.jalre.2024.168
48
YangJ.ZhangK. (2024). Social capital, crop differences, and the adoption of alternative agricultural chemical technologies: evidence from a survey of 942 farmers in Shandong. J. Agricul. Tech. Econom.6, 90–106. doi: 10.13246/j.cnki.jae.2024.06.004
49
ZhangW.LuoY.ZhaoM. (2024). How does risk preference influence farmers' participation in "grain-to-feed" behavior? The mediating effect of credit and the moderating effect of risk perception. J. Northwest A&F University24, 99–109. doi: 10.13968/j.cnki.1009-9107.2024.01.11
50
ZhengS.ChenX.WangZ. (2018). Land scale, cooperative membership, and the recognition and adoption of plant protection drone technology: a case study of Jilin Province. J. Agricul. Tech. Econom.6, 92–105. doi: 10.13246/j.cnki.jae.2018.06.008
51
ZhouX.MaL. (2008). Corporate social capital, cultural orientation, and turnover intention: an empirical study from a localized psychological perspective. Manage. World6, 109–117. doi: 10.19744/j.cnki.11-1235/f.2008.06.012
52
ZhuangT.LiG.LuoJ. (2024). Does social capital alleviate financing constraints for new agricultural business entities? A case study of agricultural credit guarantee financing. Research Financial Issues2, 105–119. doi: 10.19654/j.cnki.cjwtyj.2024.02.009
Summary
Keywords
cognitive social capital, farmer and economical, plant protection drone technology adoption, structural social capital, trans-theoretical model
Citation
Xinyi L, Yutian Z and Kai Z (2026) Cognitive and structural social capital in farmers’ adoption of plant protection drone technology: evidence from Henan Province, China. Front. Sustain. Food Syst. 10:1839929. doi: 10.3389/fsufs.2026.1839929
Received
26 March 2026
Revised
24 April 2026
Accepted
30 April 2026
Published
22 May 2026
Volume
10 - 2026
Edited by
Vijay Singh Meena, ICAR - Mahatma Gandhi Integrated Farming Research Institute, India
Reviewed by
Hafeez Noor, Shanxi Agricultural University, China
Sabinus Beni, Institut Shanti Bhuana, Indonesia
Updates
Copyright
© 2026 Xinyi, Yutian and Kai.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Zhao Kai, 18729511655@163.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.