1. Introduction
China is a major tea-producing country, and tea has a long history in China [
1,
2]. As an important cash crop, tea is the basis for many rural farmers to sustain their livelihoods. Therefore, the development of the tea industry is crucial for tea farmers. However, tea plants are susceptible to a variety of pests such as nematodes, termites, cockchafer grub, stem pests, stem borer, leaf pests, and all mite species. If these pests and diseases are not effectively controlled, they could cause 10% to 20% of yield losses [
3,
4]. Since tea farmers are risk-averse, they have to apply pesticides to rapidly control pests and diseases for reducing production losses [
5,
6]. However, excessive usage of pesticides for reducing losses has created a serious problem of agricultural pollution. Some of the pesticide residues enter the ecosystem through the atmosphere, soil, and groundwater, causing incalculable damage to the ecosystem [
7], and the pesticide residues on tea leaves can affect human health when brewed [
4].
To curb the excessive use of chemical pesticides and ensure food quality and safety and ecological sustainability, the Chinese government emphasizes the promotion and application of green control techniques. Similar to integrated pest management (IPM) [
8], green control techniques enable the effective control of crop pests and diseases by reducing chemical pesticide use and adopting techniques such as agricultural control, physical control, biological control, ecological regulation, and the scientific pesticide use to ensure the agricultural production safety, quality safety, and agroecological environment safety, as well as to improve farmers’ welfare. However, data from the Ministry of Agriculture and Rural Development showed that by the end of 2021, the adoption rate of green control techniques in China still had not reached 50%. The reasons for the low adoption rate can be attributed to several aspects. First, although the adoption of green control techniques can bring positive environmental effects, their effectiveness in controlling pests and diseases is not immediate compared to the traditional means of applying chemical pesticides [
6]. Second, the adoption of green control techniques requires farmers to learn the knowledge and operational skills and even to shoulder the high cost of technology transfer and usage risk [
9]. Finally, farmers have insufficient knowledge about pesticide application and weak awareness of ecological environment protection [
5].
Moreover, many studies have shown that risk perception is one of the main factors influencing farmers’ adoption of agricultural techniques. For example, Liu and Huang [
10] found that risk-averse farmers tended to apply more pesticides. Fernandez-Cornejo et al. [
11] pointed out that farmers who adopted the IPM tended to be less risk-averse compared to non-adopters. Li et al. [
12] found that risk perception positively mediated the relationship between farmer differences and soil and water conservation techniques. Pingali et al. [
13] argued that the greater the risk farmers perceived, the more they intend to adopt pest control. The high risk of pests and diseases could lead to crop yield reduction, thus affecting their choice of control techniques. To our knowledge, although studies have discussed the impact of risk perception on pesticide use and sustainable technique adoption in agriculture, there is little literature examining the impact of risk perception of pests and diseases on the usage behavior and behavioral intentions of green control techniques. In recent years, many countries have emphasized sustainable and green development in agriculture. The development of green agriculture implies the transformation of agricultural production methods, and the transformation of control methods is also a key to it. It is well known that the promotion and application of green control techniques have a positive effect on the improvement of the ecological environment and agricultural product quality. Therefore, it is important to explore the influence of risk perception of pests and diseases on the adoption of green control techniques by tea farmers for the development of policies related to the promotion of green agricultural techniques. As alluded above, this paper involves the risk perception of pests and diseases in the Unified Theory of Acceptance and Use of Technology (UTAUT) framework, using the micro-data of tea farmers from Sichuan province, China, attempting to explore the impact and mechanism of risk perception on tea farmers’ adoption of green control techniques. We found that risk perception of pests and diseases has a negative effect on the behavioral intention of green control techniques. Additionally, performance expectancy, effort expectancy, and social influence have positive effects on behavioral intention, and facilitating conditions could positively affect usage behavior. Furthermore, performance expectancy can mediate the influence of risk perception on behavioral intention, and risk perception can mediate the effect of effort expectancy on behavioral intention.
This paper has some contributions. First, although the UTAUT has been used to study farmers’ adoption of electronic commerce, communication technologies, Internet of Things technologies, and project payments [
14,
15,
16,
17,
18,
19,
20], they have been rarely applied in the context of green agricultural techniques. In this paper, we introduce risk perception of pests and diseases into the original UTAUT framework in an attempt to explore the effect of risk perception on farmers’ behavioral intention and usage behavior of green control techniques and contribute to the existing literature. Second, the data used in this paper are more representative because Sichuan is a major tea-producing province [
4], and sufficient samples can also improve the estimation results under the structural equation model (SEM). Third, we also identified mediating effects of risk perception and performance expectancy. On the one hand, risk perception plays a mediating role between effort expectancy and behavioral intention. On the other hand, performance expectancy also mediates the relationship between risk perception and behavioral intention. These results are new findings for relevant studies and help us to understand the impact of risk perception of pests and diseases on the adoption of green control techniques at a deeper level. Fourth, our findings can provide policy implications for the government in making policies for the promotion of green control techniques as well as green agriculture development.
The remainder of this paper is shown as follows:
Section 2 explains the theory and hypotheses;
Section 3 introduces the questionnaire and data source;
Section 4 and
Section 5 report and discuss the results, respectively;
Section 6 states the conclusions, limitations, and future research.
3. Methodology
3.1. Questionnaire Design
The questionnaire consists of three parts: the first part is the basic information on demographic characteristics, and their planting and technology adoption behavior will be investigated. The second part is the core of the questionnaire, and questions are set under each of the 7 items. The third part mainly enquires about the tea harvest and income as well as the feeling of using green control techniques. The investigation design of four constructions (performance expectancy, effort expectancy, social influence, facilitating conditions) of the UTAUT is designed by referring to the research of previous literature [
22,
64,
65]. Each item is designed with about 4 to 5 questions, using seven variables of a 5-point Likert scale (1 = strongly agree, 5 = strongly disagree). Due to commonly used in behavioral research, the Likert scale is chosen to organize the perception and subjective evaluation of technology adoption of respondents, and the structural equation model (SEM) is used in further statistical analysis.
Appendix A presents the details of the questionnaire.
3.2. Data Collection
During the period between May and October 2019, this paper used stratified random sampling to randomly select nine prefecture-level cities from the main tea-producing regions in Sichuan, including Chengdu, Yaan, Leshan, Mianyang, Yibin, Bazhong, Meishan, Leshan, and Guangyuan. Then, two or three counties with a high concentration of business entities are selected in each city, and one or two townships are chosen in each county. The survey initially targeted two groups of tea farmers: green control techniques adopters and non-adopters, and the two-stage stratified random sampling procedure was applied. In the first step, using the information offered by the agricultural departments of the townships, the villages in each sample township were divided into two categories: green prevention and control demonstration villages and non-demonstration villages. Based on the principle of stratified random sampling, we selected one or two green prevention and control demonstration villages in each sample township for sampling. To ensure strong comparability between the two types of villages, we also considered the economic conditions, transportation conditions, and tea production potential of each village. In the second step, we obtained the list of all tea growers in each sample village with the help of the village committee. We randomly selected ten tea farmers from the demonstration villages of green prevention and control and ten tea farmers from the non-demonstration villages. Professionally trained investigators interviewed tea farmers one-on-one and recorded questionnaire information. Finally, we received 800 questionnaires in total, and 747 valid questionnaires remained after removing invalid questionnaires.
3.3. Data Analysis
The Smart-PLS 3.0 application is used to model a structural equation in this study. PLS-SEM (partial lead squares structural equation modeling) offered a structural equation modeling method, which integrates the advantages of principal component analysis, linear regression analysis, and typical correlation analysis. It is suitable for small data and large correlations. It has a higher statistical function, which is more conducive to the exploration and development of theoretical models. In addition, it takes into account the sample size, measurement range, and residual distribution. We divide it into two parts: the first part measures the validity and reliability of the research model and determines the fit. The second part is to test the structural relationship between latent variables to verify our hypotheses.
5. Discussion
In China, pesticide overuse has become a major source of environmental pollution. As the main users of pesticides, farmers’ pesticide use behavior is directly related to environmental pollution. Generally speaking, most farmers are risk-averse and have little knowledge about the safe use of pesticides, which encourages farmers to use pesticides in excess [
6]. Data from the National Bureau of Statistics show that before 2015, pesticide use in China was on a trend of increasing year by year. In order to curb the excessive use of chemical pesticides and ensure the quality and safety of agricultural products and ecological sustainability, the government introduced the
Zero Growth Action Plan for Pesticide Use by 2020 in 2015, which clearly stated that it was necessary to accelerate the transformation of pest prevention and control methods and vigorously promote green control techniques. Although the policy of pesticide use reduction has generated significant benefits in the past five years, farmers have not been very active in adopting green control techniques. This may be because green control techniques are a new alternative technique to green agriculture, requiring farmers to have the relevant knowledge and operational skills. However, the diffusion of green control techniques is difficult. In the case of the tea industry, most tea production areas are in the mountains. Due to the closed information, farmers cannot keep informed of the outside world’s environmental protection concepts and new agricultural technologies. At the same time, most farmers have a low level of education and have great difficulties in learning new knowledge and mastering new techniques. Tea is one of the special agricultural products in China, with a high economic value. Tea plants are vulnerable to pests and diseases, causing huge production losses. However, excessive use of pesticides can lead to excessive pesticide residues in tea, which brings disadvantages to farmers’ income and ecological environment. Therefore, it is important to study the intention and behavior of tea farmers to adopt green control techniques for the development of the tea industry and green agriculture.
This paper examines Chinese tea farmers’ intention and behavior to use green control techniques in the extended UTAUT framework and obtains some useful findings. First, performance expectancy, effort expectancy, and social influence have a positive effect on farmers’ behavioral intention to use green control technologies, which is consistent with the prediction of the classical UTAUT [
10,
19,
42,
43,
44,
70]. That is, the usefulness of techniques, perceived ease, and the influence of surrounding people will affect their adoption intention. In addition, risk perception of pests and diseases can reduce farmers’ behavioral intention to adopt green control techniques. Most farmers are risk-averse, and if they perceive a high pest threat, they would take strong measures to control pests and diseases (e.g., applying large amounts of pesticides) rather than adopting new green control techniques with uncertainty [
34,
36]. In addition, facilitating conditions can promote farmers to use green control techniques, which is consistent with the findings of Venkatesh et al. [
22], Sykes et al. [
61], and Nistor et al. [
48]. Facilitating conditions including infrastructure, technical support, and financial subsidies can encourage farmers to use green control techniques [
62,
63]. Furthermore, this paper also tests two mediating mechanisms. On the one hand, the effect of risk perception on behavioral intention is mediated by performance expectancy. When the perceived risk is high (pests and diseases seriously threaten the quality and yield of tea), farmers tend to perceive green control techniques as being of little use [
46,
47]. Therefore, technicians need to focus on improving the performance of green control techniques to effectively prevent pests and diseases while protecting the environment. On the other hand, the effect of effort expectancy on behavioral intentions is mediated by risk perception. Easier use of the technique can lead tea farmers to perceive it as not effective enough for control, thus reducing their intention to adopt it.
Therefore, the government should strengthen policy incentives to promote the effective implementation of green control techniques. For example, appropriate subsidies and support should be given to farmers who use green control techniques, thereby enhancing their adoption intention of green agricultural techniques [
71,
72]. In addition, they should also pay attention to the introduction and promotion of green control techniques, expanding farmers’ scope of adopting new techniques, and improving farmers’ acceptance and recognition. Moreover, technology training should be conducted to reduce the usage difficulty of green control techniques. The most effective way to promote agricultural techniques is to carry out field schools [
73]. Furthermore, the social learning function and information channel function of the social network is very important for adoption behavior [
74]. Thus, it is also necessary to establish a good social learning network in the local area.
6. Conclusions
This paper explores the intention and behavior of tea farmers to use green control techniques in the UTAUT framework. The results showed that risk perception of pests and diseases can negatively impact the behavioral intention of green control techniques. Performance effect, effort expectancy, and social influence can enhance behavioral intention significantly, and facilitating conditions has a positive on usage behavior. In addition, performance expectancy plays a mediating role between risk perception and behavioral intention, while risk perception mediates the effect of effort expectancy on behavioral intention. Finally, some policy implications for promoting the diffusion and application of green control techniques are presented in the discussion section based on these findings.
Of course, the findings of our study may not be exactly generalizable, but they could at least provide some policy insights for promoting green agriculture in Asian countries with predominantly small-scale agriculture. Moreover, we have to acknowledge that this study has some limitations. First, this study is one of the few contributing to research on green control techniques adoption in tea planting. Some studies about green control techniques added factors such as environmental policy, environmental tax, carbon emissions, supply chain, and so on, and this paper mainly considers the risk perception from farmers’ perspectives. Therefore, we can also consider new factors to study farmers’ micro behavior in the future. Second, some of the population characteristics in this study are relatively dense. For example, most of the samples are middle-aged people and people who are not cadres, so future research can be more decentralized in selecting samples. Third, due to the complexity of China’s agricultural management system, the types of farmers are mainly small farmers, family farms, and agricultural cooperatives. Their intention toward the adoption of green control techniques would be different. As a result, different types of farmers could be studied in the future.