Abstract
The application and promotion of water-saving irrigation technology are of great significance for maintaining food security and the sustainable development of agricultural water resources. Based on the characteristics of integrated agricultural technologies, a binary logistic model was used to analyze the impact of technology perception on farmers' adoption behavior regarding water-saving irrigation technologies using data from surveys of 775 wheat and maize farmers in the North China Plain. The results show the following: the perceived ease of use of technology significantly contributes to farmers' water-saving irrigation technology adoption behavior, but the effect of the perceived usefulness of technology is not significant. Government regulation plays a moderating role in the impact of the perceived ease of use of technology on the adoption of water-saving irrigation technologies by farmers. In addition to the perceived ease of use of technology and technology training, large-scale farmers are influenced by government advocacy and technology subsidies, while smallholders are mainly influenced by the perceived usefulness of technology. Therefore, the focus of future work should be on improving farmers' perceptions of the ease of use of water-saving irrigation technologies, expanding the scope of technical training and technical subsidies, and strengthening government advocacy and education.
HIGHLIGHT
According to the results of the logistic model, the improvement of technology perception can effectively promote the adoption of water-saving irrigation technology by farmers. Government regulation plays a moderating role in the impact of the perceived ease of use of technology on the adoption of water-saving irrigation technologies by farmers. Significant differences were found between farmers of different cropland sizes.
INTRODUCTION
Water scarcity is a key constraint to sustainable global agricultural development. With 9% of the world's arable land and 6% of the world's water resources, China feeds 20% of the world's population and provides 25% of the world's total food. The predominantly plantation-based North China Plain plays a pivotal role in ensuring China's food security. However, there is a serious imbalance in the regional distribution of water resources available for agriculture in China, with a serious water deficit in the North China Plain, and the mismatch between water resources and grain production is further exacerbated by the transportation of grain from the north to the south. In Hebei Province, China, in particular, long-term reliance on groundwater irrigation for agricultural production has created a serious underground leakage zone. Improving agricultural water use efficiency is an important path to saving water resources and ensuring food security. The Chinese government utilized the agricultural extension system to vigorously promote the application of agricultural water-saving irrigation technologies such as sprinkler and drip irrigation in the North China Plain. Nevertheless, the application area of water-saving irrigation technology in the North China Plain region only accounts for about 5% of the sown area of crops (China Rural Statistics Yearbook). At present, China's rural population is aging seriously, and the education level of farmers is generally low. Farmers generally use traditional large water diffuse irrigation, but the perceptions of water-saving irrigation technology are still at a low level, which may seriously impede the popularization and application of water-saving irrigation technology.
Farmers will thoroughly evaluate a technology based on their existing experience and knowledge before adopting it, and they will only adopt it if it is well accepted (Karahanna & Straub, 1999). The adoption of water-saving irrigation technologies by farmers is based on a combination of perceived technical–economic, ecological, and social benefits (Zhong et al., 2019; Espejo et al., 2021; Shahangian et al., 2021). Some scholars have analyzed the impact of farmers' perceptions of this technology's ease of use and usefulness on their technology adoption behavior from a technology perception perspective. Technology ease of use is perceived as the degree of acceptance by farmers of the ease of installation, maintenance, and other aspects of using a particular agricultural technology, and the perceived usefulness of a technology is the extent to which farmers recognize that the use of agricultural technology has led to reduced costs and increased outputs (Adrian et al., 2005; Folorunso & Ogunseye, 2008). The use of water-saving irrigation technology involves several aspects, such as installation, laying, maintenance, and recycling, which increases the difficulty and uncertainty for farmers in using it. Whether farmers end up using traditional irrigation methods or adopting water-saving irrigation technologies largely depends on how easy it is to learn to use water-saving equipment (Warner et al., 2020; Wang et al., 2023). The perceived ease of use and perceived usefulness of the technology have a significant positive impact on farmers' green production behavior (Aubert et al., 2012; Mfitumukiza et al., 2020; Venkatesh, 2000; Yue et al., 2023) and are important factors in a farmer's choice of precision agricultural technologies (Far & Rezaei-Moghaddam, 2015; Caffaro et al., 2020). In conclusion, scholars generally agree that increasing the level of technology perception among farmers can be effective in encouraging their adoption of agricultural technology.
Agricultural water conservation has strong positive externalities. With groundwater resources, as highly mobile public pond resources, it is difficult to internalize externalities by defining property rights and market-based means; therefore, the solution to externalities must rely on government intervention to change the external environmental constraints on farmers in adopting water-saving irrigation technology. Government promotion is an important channel for farmers to obtain information about water-saving irrigation technologies, and it has a significantly positive impact on farmers adopting them (Su et al., 2021; Si et al., 2022a, 2022b). Technology subsidies reduce the costs of initial investments, increase returns on investment, and have a significant positive impact on farmers' adoption of water-saving irrigation technologies (Dinar & Yaron, 1992). When farmers can only provide labor and limited capital, the main role of the government is to provide them with appropriate financial support (Zou et al., 2013; Serote et al., 2023). Developing agricultural water-saving technology in well irrigation areas is constrained by factors such as a lack of effective technical guidance in this process (Muenratch & Nguyen, 2023). Technical training reduces learning costs for farmers in well irrigation areas and can significantly promote their adoption of green agricultural production practices, such as water-saving irrigation technologies (Alauddin et al., 2020). At the same time, project demonstrations allow farmers to visually experience the operational use of agricultural technology and increase their confidence in using it, which, in turn, promotes their adoption of water-saving irrigation technology (Tanti et al., 2022).
The above scholars have empirically examined the impact of technology perception on farmers' adoption of water-saving irrigation technologies, and have examined the effects of technology perception and government regulations on the adoption behavior of water-saving irrigation technologies by farmers, respectively, and both technology perception and government regulations are important factors influencing the adoption of water-saving irrigation technologies by farmers. Previous studies have only analyzed the impact of government regulation on farmers' water-saving irrigation technology adoption behavior from the perspective of externalities, ignoring the role of government regulation in changing farmers' technology perception. Thus, this study develops a framework for analyzing farmers' adoption behavior based on the characteristics of water-saving irrigation technology integration among wheat and maize farmers in the North China Plain. The study uses data from surveys taken from 775 wheat–corn households to analyze this subject, explore the moderating role of government regulations, and further reveal behavioral differences across different business scales.
LITERATURE REVIEW
This section compiles studies related to the impact of technology perception and government regulations on farmers' water-saving irrigation adoption behavior. On this basis, the research perspective of this paper is further presented and a theoretical analysis framework is constructed.
Technology perception and farmers’ technology adoption behavior
The most widely used model for studying technology acceptance and use behavior is the technology acceptance model (TAM). This theoretical model assumes that individual decisions are determined by behavioral intentions, which are determined by a combination of desired attitudes and perceived usefulness. The attitude of wanting to use something is determined by a combination of perceived usefulness and ease of use: perceived usefulness is determined by a combination of perceived ease of use and external variables, and perceived ease of use is determined by external variables. The TAM identifies two dimensions of individual behavior from a psycho-cognitive perspective, namely, perceived ease of use and perceived usefulness, and external variables include individual characteristics, policy influence, organizational structure, etc. Technology awareness, as a general perception of the ease of use and usefulness of agricultural technology, is a fundamental unit that connects farmers to the diffusion of agricultural technology, and it can be seen as the core unit of the organic interface between agriculture and agricultural technology diffusion (Davis, 1989). Perceptions of economic and ecological value significantly and positively influence the adoption of green production technologies and agroecological conservation by farmers (Gefen & Straub, 2000; Davis & Venkatesh, 2004). Accordingly, this study proposes the following research hypotheses:
Hypothesis (H1). The perceived ease of use of technology positively influences farmers' water-saving irrigation technology adoption behavior.
Hypothesis (H2). The perceived usefulness of technology positively influences farmers' water-saving irrigation technology adoption behavior.
Government regulations and farmers’ technology adoption behavior
Government regulations such as technology subsidies, government advocacy, and penalties all significantly and positively influence farmers' technology adoption behavior (Gao et al., 2014; Shi et al., 2019; Thakur et al., 2022). Strategies to enhance water conservation in the agricultural sector have a significant positive impact on the sustainable use of water resources. Both strict water management and economic incentives will help farmers adopt water-saving irrigation technologies (Nair & Nair, 2019). The agricultural irrigation choices of farmers, especially those involved in large-field cultivation, are crucial to the sustainability of agricultural water use. The adoption of water-saving irrigation technologies has strong positive externalities. Government advocacy and project demonstrations can effectively widen farmers' access to information on water-saving irrigation technology and reduce uncertainty about its use. Technical training can significantly improve farmers' familiarity with and experience using water-saving irrigation technology. Technology subsidies directly reduce the cost of adopting water-saving irrigation technology for farmers. Accordingly, this study proposes a third research hypothesis:
Hypothesis (H3). Government regulation positively influences farmers' water-saving irrigation technology adoption behavior.
The moderating effect of government regulations
The TAM assumes that external variables directly influence the perceived technology ease of use and perceived technology usefulness, but the model itself does not provide a detailed explanation of the external variables, which indicates that the TAM is an open model. Some external characteristics influence the final decision-making behavior of farmers by affecting their perception of this technology's ease of use and technology usefulness (Jin et al., 2022; Yan et al., 2022; Xiang & Gao, 2023). As an important external environmental constraint, government regulation profoundly influences irrigation behavior with positive externalities. Formal institutions such as incentive regulation, guidance regulation, and constraint regulation work together with informal institutions such as technology perceptions and other value orientations to facilitate the adoption of green production technologies by farmers (Susanto & Aljoza, 2015; Jia et al., 2022). Some scholars have placed technology perceptions, government regulations, and farmers' adoption of water-saving irrigation technologies in the same analytical framework. They found that farmers' perceptions of the ease of use and usefulness of water-saving technologies, as well as related subsidies, significantly contributed to their adoption of these technologies (Tesfaye et al., 2021; Blanke et al., 2007). However, no one has further explored the possible interplay between technology perception and government regulation. Government regulation has a moderating effect on the transformation of farmers' green perceptions into green production behavior (Pan et al., 2020; Si et al., 2022a, 2022b). As a hard constraint on the external environment, government regulation influences farmers' perceptions of the ease of use and usefulness of water-saving irrigation technologies and, thus, encourages or discourages them to adopt these technologies.
Farmers with different business sizes allocate their resource factors in agricultural production in varying ways, and their business objectives can significantly differ. The size of the arable land and the degree of its fragmentation are important factors influencing the adoption of technology by farmers (Mi et al., 2021). There are significant differences in the impacts that policy incentives and technology perceptions have on the adoption behavior of farmers of different arable land operation sizes in terms of water-saving technologies (Larson et al., 2016). Large-scale farmers and smallholders greatly differ in their resource endowments for agricultural production, and their technical perceptions and actual agricultural technology needs also significantly differ. Large-scale farmers have wider social networks, more sources of information on agricultural technology, and a better understanding of water-saving irrigation technology (Shi, 2021; Xie & Huang, 2021). Accordingly, this study proposes a fourth and fifth research hypothesis:
Hypothesis (H4). Government regulation plays a moderating role in the influence of technology perceptions on farmers' water-saving irrigation technology adoption behavior.
Hypothesis (H5). There are significant differences in the adoption behavior of water-saving irrigation technologies between farmers with different business sizes.
Logical analysis framework.
METHODS
This section gives an account of the sample farmers in the study area who are engaged in agricultural production and the situation related to water saving and irrigation, and further gives an account of the research methodology of the empirical analysis.
Data sources
Equation (1) is the sample size calculation formula where n is the minimum sample size, t is the critical value corresponding to the confidence level, and e is the sampling error. To ensure that the sample size is large enough, the minimum sample size corresponding to a 95% confidence level (t = 1.96) and a permissible sampling error e of 5% was set to be approximately 385 in this study. The research is conducted using multi-stage sampling and the minium sample size needs to be further expanded, usually by multiplying it by the design validity deft, which typically takes the value of 2. Therefore, the minimum sample size for this study is theoretically 770.
Baoding City, Shijiazhuang City, Xinji City, Cangzhou City, Hengshui City, Xingtai City, and Handan City were selected as the research areas which are large wheat–corn growing areas, underground water leakage zones and highly dependent on groundwater irrigation in the North China Plain. A random stratified sampling method was used. Additionally, one to two townships were randomly selected from each county, and one to two villages were randomly selected from each township. In each sample village, the investigators asked the village leaders to provide a list of names for all the household heads. The first household was selected randomly, and the other households were selected according to the calculated interval distance (total number of households/number of households to be selected). Approximately 10–15 farmers were randomly selected from each village. The survey obtained a total of 850 farmers' questionnaires, and after excluding those with missing key information and abnormal data, finally, 775 valid questionnaires were obtained, with a sample validity rate of about 91%.
Variable selection
Dependent variable
Farmers' water-saving irrigation technology adoption behavior: In this paper, the adoption or otherwise of water-saving irrigation technologies by farmers was the dependent variable in the model. Specifically, the explanatory variables were measured using the following question: ‘Have you adopted water-saving irrigation technologies?’ A value of 1 was assigned if the farmer had adopted water-saving irrigation technology, and a value of 0 was assigned if the farmer had never adopted water-saving irrigation technology.
Key independent variable
Technology perception: the core definition of technology awareness is a perceived benefit versus cost tradeoff. This study divided technological cognition into perceived benefits and perceived payoffs, using both perceived ease of use and perceived usefulness at the level of perceived benefits to measure farmers' perceptions of water-saving irrigation technologies. The perceived ease of use of the technology included a measure of farmers' perceptions of five relevant aspects: operation, installation, maintenance, recovery, and field management. Each dimension was measured on a five-point Likert scale. The final score for perceived ease of use of technology was the arithmetic mean of the above five dimensions. The perceived usefulness of the technology included a measure of the perceived level of impact of water-saving irrigation technology on five areas: fertilizer, irrigation water use, labor inputs, irrigation electricity costs, and agricultural output. Each dimension was measured on a five-point Likert scale. The final score for the perceived usefulness of the technology was the arithmetic mean of the above five dimensions. The technology perception score was the arithmetic mean of two indicators: perceived ease of use of technology and perceived usefulness of technology.
Reliability and validity tests were conducted on the technology perception variables in SPSS 16.0. The reliability of the questionnaire was high. The analysis of the 775 responses revealed that the Cronbach's α value for the latent variable reliability of farmers' perceptions of water-saving irrigation technology was 0.839, the Cronbach's α value for the latent variable of farmers' perceived ease of use of water-saving irrigation technology was 0.818, and the latent variable of farmers' perceived usefulness of water-saving irrigation technology had a Cronbach's α value of 0.799. The combined latent variable reliability of farmers' perception of water-saving irrigation technology was 0.919, the combined latent variable reliability of farmers' perceived ease of use of water-saving irrigation technology was 0.873, and the combined latent variable reliability of the perceived usefulness of water-saving irrigation technology was 0.858. All three values were greater than 0.8, indicating good internal consistency for the latent variables.
The results of the validity analysis are shown in Table 1. Validity analysis using Kaiser–Meyer–Olkin (KMO) and Bartlett's spherical test showed that the value of the overall KMO indicator statistic was 0.825, which is greater than the criterion of 0.7, and the value of Bartlett's spherical test was 3,186.388, reaching a significance level at a degree of freedom of 45. The values of the KMO statistics for the latent variables of perceived ease of use and perceived usefulness were 0.821 and 0.737, respectively, and the values of Bartlett's spherical test for each latent variable reached the significance level, indicating that the correlation coefficient matrices of each indicator were significantly different and suitable for factor analysis. The results of the analysis of the mean variance extracted for each latent variable showed that all were greater than the criterion of 0.5, indicating good convergent validity.
Results of the reliability and validity tests for the latent variables.
| Latent Variables | Cronbach's α | CR | AVE | KMO | AC | p |
|---|---|---|---|---|---|---|
| Perceived ease of use of technology | 0.818 | 0.873 | 0.580 | 0.821 | 1,285.253 | 0.000 |
| Perceived usefulness of technology | 0.799 | 0.858 | 0.559 | 0.737 | 1,637.637 | 0.000 |
| Technology perception | 0.839 | 0.919 | 0.538 | 0.825 | 3,186.388 | 0.000 |
| Latent Variables | Cronbach's α | CR | AVE | KMO | AC | p |
|---|---|---|---|---|---|---|
| Perceived ease of use of technology | 0.818 | 0.873 | 0.580 | 0.821 | 1,285.253 | 0.000 |
| Perceived usefulness of technology | 0.799 | 0.858 | 0.559 | 0.737 | 1,637.637 | 0.000 |
| Technology perception | 0.839 | 0.919 | 0.538 | 0.825 | 3,186.388 | 0.000 |
Note: Cronbach's α is the reliability coefficient of the latent variable; CR is the combined confidence level; AVE is the average variance extracted; KMO is the Kaiser–Meyer–Olkin measure of the latent variable; AC is the approximate cardinality; AC and p are the results of Bartlett's sphericity test for latent variables; p represents the level of significance.
Moderator
Government regulation: Government regulation multidimensionally reflects the external environment of farmers' water-saving irrigation technology adoption behavior. The relevant policies implemented by the government were the key factors influencing the adoption of agricultural technology by farmers. In this study, four indicators reflected government regulation: government awareness, technology training, technology subsidies, and project demonstration.
Control variables and instrumental variables
Numerous studies have confirmed that the adoption of agricultural technology by farmers is influenced by several factors. Farmers' demands for green production technologies are influenced by a combination of technological characteristics, natural environmental factors, economic factors, and policy factors. Farmers' experiences, psychological characteristics, and geographical location all have an impact on adopting water-saving irrigation technologies in agriculture (Yamaguchi et al., 2019; Rodriguez-Sanchez & Sarabia-Sanchez, 2020). To control other factors that may influence farmers' water-saving irrigation technology adoption behavior, this paper used farmers' personal characteristics, household business characteristics, external environmental characteristics, and village characteristics as control variables. The ordinal variables, such as ‘Education level’, are not the focus of this paper and do not affect the analysis of the impact of the core variable of this paper. Although they limit the information the dataset can provide VS dummies, we retain the form of the ordinal variable (Dean et al., 2021; Shi, 2021). In addition, considering the possible endogeneity of the model, this study used several channels to obtain information on water-saving irrigation technologies and the depth of irrigation wells as instrumental variables, considering field research and drawing on the study by Xu et al. (2018). These two variables affected farmers' perceptions of water-saving irrigation technologies to some extent, but they did not have an impact on farmers' adoption behavior, which was only influenced by the perceived value of water-saving irrigation technologies.
Specific variable descriptions and descriptive statistics are shown in Table 2. In total, 157 farmers out of the 775 interviewed adopted water-saving irrigation technologies, representing about 20%. Of the farmers interviewed, 549 were men and 226 were women, with women accounting for approximately 29% of the total sample. The average age was about 56 years old, and the education level was mostly at the junior high school level. In addition, the vast majority of respondents had a political profile, with 85 farmers being party members, accounting for about 11%. The number of farmers with more than 30 years of farming experience was 509, accounting for 66% of the sample, indicating that farmers who currently engage in wheat and maize farming have more experience. The sample of farmers had a high level of technical awareness of water-saving irrigation technologies. The arithmetic mean of farmers' perception of water-saving irrigation technologies was 2.595. The levels of perceived ease of use and perceived usefulness were 2.726 and 2.464, respectively. With the average depth of irrigation wells reaching 142 m, the long-term dependence on groundwater resources for agricultural irrigation has created a dilemma regarding agricultural irrigation and sustainable water use. The average number of information channels for farmers to access water-saving irrigation technologies was 1.249. In total, 27% of farmers interviewed have seen government advocacy about water-saving irrigation technology, 15% have received training in agricultural technology, 18% have received subsidies for water-saving irrigation technologies, and 45% have water-saving irrigation technique project demonstrations held near them.
Variable definitions and descriptive statistics.
| Variable | Definition | Mean value | Standard deviation |
|---|---|---|---|
| Dependent variable | |||
| Water-saving irrigation technology adoption behavior | Whether to adopt water-saving irrigation technology: Yes = 1; No = 0. | 0.203 | 0.402 |
| Key independent variable | |||
| Technology perception | Farmers’ overall perception of water-saving irrigation technology includes both the ease of use of the technology and the usefulness of the technology. | 2.595 | 0.532 |
| Perceived ease of use of technology | Farmers’ perceptions of the difficulty of using water-saving irrigation technology include five aspects: operation, installation, maintenance, recovery, and field management. | 2.726 | 0.622 |
| Do you find water-saving irrigation technologies easy to use? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 3.147 | 0.889 | |
| Do you find water-saving irrigation technology complicated to install? Very complex = 1; Complex = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.626 | 0.889 | |
| Do you find water-saving irrigation technology easy to maintain and repair? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.655 | 0.856 | |
| Do you find it easy to recycle materials and equipment for water-saving irrigation technology? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.634 | 0.823 | |
| Do you feel that using water-saving irrigation technology affects your regular field management (e.g., sowing, weeding, pesticides, harvesting, etc.)? Very large influence = 1; A relatively large influence = 2; Appropriate influence = 3; A little influence = 4; Little influence = 5. | 2.568 | 0.595 | |
| Perceived usefulness of technology | Farmers’ perceptions of the effectiveness of using water-saving irrigation technologies include five aspects: fertilizer, irrigation water consumption, labor inputs, irrigation electricity costs, and agricultural output. | 2.464 | 0.632 |
| Are water-saving irrigation technologies conducive to reducing fertilizer applications? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 1.512 | 0.672 | |
| Does water-saving irrigation technology contribute to reducing the amount of water irrigated? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 2.795 | 0.943 | |
| Are water-saving irrigation technologies conducive to reducing labor inputs? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 3.317 | 0.928 | |
| Does water-saving irrigation technology help reduce irrigation electricity costs? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 2.765 | 0.903 | |
| Are water-saving irrigation technologies conducive to higher yields? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 1.941 | 0.772 | |
| Moderator | Government regulations | ||
| Do you know anything about water-saving irrigation technology through government advocacy? Yes = 1; No = 0. | 0.267 | 0.443 | |
| Do you attend the water-saving irrigation technology training? Yes = 1; No = 0. | 0.161 | 0.382 | |
| Are there subsidies for the use of water-saving irrigation technology? Yes = 1; No = 0. | 0.183 | 0.387 | |
| Is there a water-saving irrigation demonstration site near your village? Yes = 1; No = 0. | 0.454 | 0.498 | |
| Control variables | |||
| Personal characteristics of the householder | Age: years | 56.272 | 9.923 |
| Gender: Man = 1; Woman = 0. | 0.708 | 0.455 | |
| Education level: No degree = 1; Primary school = 2; junior middle school = 3; Senior middle school = 4; University = 5. | 2.855 | 0.993 | |
| Party member: Yes = 1; No = 0. | 0.110 | 0.313 | |
| Farming time: Years. | 31.299 | 14.112 | |
| Household business characteristics | Size of household labor force: people | 2.875 | 1.010 |
| Farmland size: mu (1 mu = 1/15 hm2). | 104.585 | 275.742 | |
| Quantity of farmland: plots. | 3.663 | 4.264 | |
| Income from farming accounts for the proportion of total household income: %. | 0.313 | 0.332 | |
| Are there agricultural water-saving irrigation services? Yes = 1; No = 0. | 0.360 | 0.480 | |
| Applied for an agricultural loan: Yes = 1; No = 0. | 0.048 | 0.213 | |
| Bought agricultural insurance: Yes = 1; No = 0. | 0.846 | 0.378 | |
| Technical exchange: No = 1; Little = 2; Ordinary = 3; Frequent = 4; Very frequent = 5. | 2.225 | 1.021 | |
| Participates in agricultural cooperative organizations: Yes = 1; No = 0. | 0.185 | 0.388 | |
| Village characteristics | The distance of your village from the town center: kilometers. | 4.762 | 3.248 |
| Instrumental variables | |||
| Number of information channels for obtaining water-saving irrigation technology: number. | 1.249 | 1.339 | |
| irrigation well depth: meters. | 142.603 | 101.303 | |
| Variable | Definition | Mean value | Standard deviation |
|---|---|---|---|
| Dependent variable | |||
| Water-saving irrigation technology adoption behavior | Whether to adopt water-saving irrigation technology: Yes = 1; No = 0. | 0.203 | 0.402 |
| Key independent variable | |||
| Technology perception | Farmers’ overall perception of water-saving irrigation technology includes both the ease of use of the technology and the usefulness of the technology. | 2.595 | 0.532 |
| Perceived ease of use of technology | Farmers’ perceptions of the difficulty of using water-saving irrigation technology include five aspects: operation, installation, maintenance, recovery, and field management. | 2.726 | 0.622 |
| Do you find water-saving irrigation technologies easy to use? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 3.147 | 0.889 | |
| Do you find water-saving irrigation technology complicated to install? Very complex = 1; Complex = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.626 | 0.889 | |
| Do you find water-saving irrigation technology easy to maintain and repair? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.655 | 0.856 | |
| Do you find it easy to recycle materials and equipment for water-saving irrigation technology? Very difficult = 1; Difficult = 2; Appropriate = 3; Easy = 4; Very easy = 5. | 2.634 | 0.823 | |
| Do you feel that using water-saving irrigation technology affects your regular field management (e.g., sowing, weeding, pesticides, harvesting, etc.)? Very large influence = 1; A relatively large influence = 2; Appropriate influence = 3; A little influence = 4; Little influence = 5. | 2.568 | 0.595 | |
| Perceived usefulness of technology | Farmers’ perceptions of the effectiveness of using water-saving irrigation technologies include five aspects: fertilizer, irrigation water consumption, labor inputs, irrigation electricity costs, and agricultural output. | 2.464 | 0.632 |
| Are water-saving irrigation technologies conducive to reducing fertilizer applications? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 1.512 | 0.672 | |
| Does water-saving irrigation technology contribute to reducing the amount of water irrigated? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 2.795 | 0.943 | |
| Are water-saving irrigation technologies conducive to reducing labor inputs? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 3.317 | 0.928 | |
| Does water-saving irrigation technology help reduce irrigation electricity costs? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 2.765 | 0.903 | |
| Are water-saving irrigation technologies conducive to higher yields? No = 1; A little = 2; Some = 3; Much = 4; Very much = 5. | 1.941 | 0.772 | |
| Moderator | Government regulations | ||
| Do you know anything about water-saving irrigation technology through government advocacy? Yes = 1; No = 0. | 0.267 | 0.443 | |
| Do you attend the water-saving irrigation technology training? Yes = 1; No = 0. | 0.161 | 0.382 | |
| Are there subsidies for the use of water-saving irrigation technology? Yes = 1; No = 0. | 0.183 | 0.387 | |
| Is there a water-saving irrigation demonstration site near your village? Yes = 1; No = 0. | 0.454 | 0.498 | |
| Control variables | |||
| Personal characteristics of the householder | Age: years | 56.272 | 9.923 |
| Gender: Man = 1; Woman = 0. | 0.708 | 0.455 | |
| Education level: No degree = 1; Primary school = 2; junior middle school = 3; Senior middle school = 4; University = 5. | 2.855 | 0.993 | |
| Party member: Yes = 1; No = 0. | 0.110 | 0.313 | |
| Farming time: Years. | 31.299 | 14.112 | |
| Household business characteristics | Size of household labor force: people | 2.875 | 1.010 |
| Farmland size: mu (1 mu = 1/15 hm2). | 104.585 | 275.742 | |
| Quantity of farmland: plots. | 3.663 | 4.264 | |
| Income from farming accounts for the proportion of total household income: %. | 0.313 | 0.332 | |
| Are there agricultural water-saving irrigation services? Yes = 1; No = 0. | 0.360 | 0.480 | |
| Applied for an agricultural loan: Yes = 1; No = 0. | 0.048 | 0.213 | |
| Bought agricultural insurance: Yes = 1; No = 0. | 0.846 | 0.378 | |
| Technical exchange: No = 1; Little = 2; Ordinary = 3; Frequent = 4; Very frequent = 5. | 2.225 | 1.021 | |
| Participates in agricultural cooperative organizations: Yes = 1; No = 0. | 0.185 | 0.388 | |
| Village characteristics | The distance of your village from the town center: kilometers. | 4.762 | 3.248 |
| Instrumental variables | |||
| Number of information channels for obtaining water-saving irrigation technology: number. | 1.249 | 1.339 | |
| irrigation well depth: meters. | 142.603 | 101.303 | |
Logistic model
is the probability that a farmer will adopt a water-saving irrigation technology or not. Y indicates the adoption of water-saving irrigation technologies by farmers. Y = 1 means that the farmer adopted the water-saving irrigation technology, while Y = 0 means that the farmer did not adopt the water-saving irrigation technology. Y is a linear combination of R, G, and C. The expression of the model is as follows:
In Equation (3), R is the farmer's technical perception variable of water-saving irrigation technology, including perceived ease of use and perceived usefulness. G is the government regulation variable, which includes government advocacy, technology training, technology subsidies, and project demonstration. RG is an interactive term for technology perception and government regulation. C is the control variable, including the personal characteristics of the householder, household business characteristics, and village characteristics. 
represents a constant term.
is the random error term. It is worth noting that the signs of the regression coefficients of the independent variables, such as λ and θ, only provide the direction of the effect of the respective variables on the adoption of water-saving irrigation technologies by farmers and not the specific marginal effects. Therefore, to address the inconsistent scaling of regression coefficients for categorical and continuous variables in the mediated effects model, marginal effects need to be applied to the regression coefficients in the logistic model.RESULTS
This section presents the empirical results and the mechanism of action of the influence of technology perception and government regulations on farmers' water-saving irrigation technology adoption behavior based on the logistic model and further analyzes the differences between large-scale farmers and smallholder farmers.
Benchmark model results and analysis
Before estimating the model, the independent variables were first tested for multicollinearity. The Variance Inflation Factor (VIF) values in this section, analyzed using Stata 14.0, ranged from 1.05 to 3.80, indicating that there were no serious covariance issues between the variables.
Table 3 shows that there is a significant positive effect of technology perception variables on farmers' water-saving irrigation technology adoption behavior in the logistic model. To further confirm the reliability of the empirical results, the empirical analysis was conducted again using the probit model. The technology perception variables exhibited a relatively consistent significance in the model's estimation results in both the logistic and probit models, indicating the strong robustness of these estimation results. This study analyzed and interpreted its results by estimating the logistic model.
Benchmark regression model.
| Logistic | Probit | |||
|---|---|---|---|---|
| Coefficient | Marginal Effects (%) | Coefficient | Marginal Effects (%) | |
| Perceived ease of use of technology | 1.956*** (0.468) | 5.622*** (0.012) | 1.001*** (0.221) | 5.315*** (0.011) |
| Perceived usefulness of technology | 0.049 (0.472) | 0.140 (0.014) | 0.052 (0.229) | 0.279 (0.012) |
| Government advocacy | 1.807*** (0.527) | 5.192*** (0.015) | 0.922*** (0.244) | 4.895*** (0.013) |
| Technical training | 1.114** (0.513) | 3.202** (0.014) | 0.619** (0.265) | 3.287*** (0.014) |
| Technical subsidies | 2.365*** (0.494) | 6.797*** (0.012) | 1.317*** (0.253) | 6.993*** (0.012) |
| Project demonstration | −0.453 (0.521) | −1.302 (0.015) | −0.256 (0.242) | −1.361 (0.013) |
| Age | 0.044 (0.041) | 0.126 (0.001) | 0.025 (0.019) | 0.134 (0.001) |
| Gender | 0.572 (0.613) | 1.644 (0.018) | 0.232 (0.288) | 1.230 (0.015) |
| Education level | −0.348 (0.270) | −0.999 (0.008) | −0.219* (0.131) | −1.164 (0.007) |
| Political status | −0.754 (0.659) | −2.167 (0.019) | −0.295 (0.328) | −1.564 (0.017) |
| Farming time | −0.070*** (0.026) | −0.202*** (0.001) | −0.038*** (0.012) | −0.201*** (0.001) |
| Number of laborers | 0.216 (0.293) | 0.622 (0.008) | 0.108 (0.128) | 0.574 (0.007) |
| Scale of arable land operations | 0.002 (0.001) | 0.007 (< 0.001) | 0.001** (0.001) | 0.006** (< 0.001) |
| Number of plots | 0.001 (0.037) | 0.002 (0.001) | −0.001 (0.020) | −0.004 (0.001) |
| The proportion of income from farming | 2.686*** (0.980) | 7.717*** (0.028) | 1.544*** (0.444) | 8.200*** (0.024) |
| Agricultural productive services | 1.446*** (0.454) | 4.156*** (0.012) | 0.734*** (0.220) | 3.899*** (0.011) |
| Applied for an agricultural loan | −1.299 (0.990) | −3.732 (0.028) | −0.738 (0.477) | −3.920 (0.025) |
| Bought agricultural insurance | 1.349*** (0.496) | 3.877*** (0.015) | 0.741*** (0.257) | 3.936** (0.014) |
| Technical exchange | 0.328 (0.265) | 0.944 (0.008) | 0.169 (0.130) | 0.899 (0.007) |
| Participates in a cooperative organization | 2.042*** (0.488) | 5.867** (0.014) | 1.080*** (0.241) | 5.736*** (0.013) |
| Residential-to-township distance | −0.097 (0.074) | −0.280 (0.002) | −0.052 (0.037) | −0.278 (0.002) |
| Constant term | −13.949*** (3.172) | – | −7.334*** (1.490) | – |
| Pseudo R2 | 0.8010 | – | 0.8040 | – |
| Logistic | Probit | |||
|---|---|---|---|---|
| Coefficient | Marginal Effects (%) | Coefficient | Marginal Effects (%) | |
| Perceived ease of use of technology | 1.956*** (0.468) | 5.622*** (0.012) | 1.001*** (0.221) | 5.315*** (0.011) |
| Perceived usefulness of technology | 0.049 (0.472) | 0.140 (0.014) | 0.052 (0.229) | 0.279 (0.012) |
| Government advocacy | 1.807*** (0.527) | 5.192*** (0.015) | 0.922*** (0.244) | 4.895*** (0.013) |
| Technical training | 1.114** (0.513) | 3.202** (0.014) | 0.619** (0.265) | 3.287*** (0.014) |
| Technical subsidies | 2.365*** (0.494) | 6.797*** (0.012) | 1.317*** (0.253) | 6.993*** (0.012) |
| Project demonstration | −0.453 (0.521) | −1.302 (0.015) | −0.256 (0.242) | −1.361 (0.013) |
| Age | 0.044 (0.041) | 0.126 (0.001) | 0.025 (0.019) | 0.134 (0.001) |
| Gender | 0.572 (0.613) | 1.644 (0.018) | 0.232 (0.288) | 1.230 (0.015) |
| Education level | −0.348 (0.270) | −0.999 (0.008) | −0.219* (0.131) | −1.164 (0.007) |
| Political status | −0.754 (0.659) | −2.167 (0.019) | −0.295 (0.328) | −1.564 (0.017) |
| Farming time | −0.070*** (0.026) | −0.202*** (0.001) | −0.038*** (0.012) | −0.201*** (0.001) |
| Number of laborers | 0.216 (0.293) | 0.622 (0.008) | 0.108 (0.128) | 0.574 (0.007) |
| Scale of arable land operations | 0.002 (0.001) | 0.007 (< 0.001) | 0.001** (0.001) | 0.006** (< 0.001) |
| Number of plots | 0.001 (0.037) | 0.002 (0.001) | −0.001 (0.020) | −0.004 (0.001) |
| The proportion of income from farming | 2.686*** (0.980) | 7.717*** (0.028) | 1.544*** (0.444) | 8.200*** (0.024) |
| Agricultural productive services | 1.446*** (0.454) | 4.156*** (0.012) | 0.734*** (0.220) | 3.899*** (0.011) |
| Applied for an agricultural loan | −1.299 (0.990) | −3.732 (0.028) | −0.738 (0.477) | −3.920 (0.025) |
| Bought agricultural insurance | 1.349*** (0.496) | 3.877*** (0.015) | 0.741*** (0.257) | 3.936** (0.014) |
| Technical exchange | 0.328 (0.265) | 0.944 (0.008) | 0.169 (0.130) | 0.899 (0.007) |
| Participates in a cooperative organization | 2.042*** (0.488) | 5.867** (0.014) | 1.080*** (0.241) | 5.736*** (0.013) |
| Residential-to-township distance | −0.097 (0.074) | −0.280 (0.002) | −0.052 (0.037) | −0.278 (0.002) |
| Constant term | −13.949*** (3.172) | – | −7.334*** (1.490) | – |
| Pseudo R2 | 0.8010 | – | 0.8040 | – |
Note: ***, **, and * indicate significance at the levels of 1, 5, and 10%, respectively; the robust standard errors are in parentheses.
The perceived ease of use of technology had a significant positive effect on farmers' adoption of water-saving irrigation technology: those with a high perception of ease of use were 5.622% more likely to adopt this technology than those with a low perception. The operational applicability of water-saving irrigation technology can significantly contribute to farmers' acceptance and adoption of it. The effect of perceived technology usefulness on the adoption of these technologies was not significant. Farmers' perceptions of the ease of use and the suitability of the technology to their existing agricultural production conditions were key factors influencing their adoption of these technologies.
Government advocacy, technical training, and technical subsidies significantly and positively influenced farmers' water-saving irrigation technology adoption behavior, with technical subsidies playing the largest role, followed by government advocacy and technical training. Farmers who received government advocacy regarding water-saving irrigation technologies were 5.192% more likely to adopt these technologies than those who had not received such advocacy. Government advocacy can make older and less educated farmers better aware of water-saving irrigation technologies. Farmers who had attended training on these technologies were 3.202% more likely to adopt them than those who had not attended training. Technical training can deepen a farmer's knowledge of water-saving irrigation technologies in a more systematic and intuitive way. The policy subsidies increased the probability of farmers adopting these technologies by a corresponding 6.979% compared with no technology subsidies. The use of water-saving irrigation technology is a large investment in agricultural production. Government subsidies can alleviate the financial pressure on farmers, thus encouraging them to adopt water-saving irrigation technologies. Project demonstrations did not have a significant impact in this study because most farmers did not know much about these sites.
Regarding farmers' personal characteristics, only farming time had a significant negative effect on adopting water-saving irrigation technology. Age, gender, education level, and political status did not have a significant effect on this issue. The proportion of income from farming, participation in cooperative organizations, purchasing agricultural insurance, and agricultural productive services for water-saving irrigation in the household business characteristic variables all significantly contributed to farmers' adoption of water-saving irrigation technologies: the order of importance was the proportion of income from farming, participation in cooperative organizations, agricultural productive services, and purchasing agricultural insurance. The distance from the village to the township center characteristic did not have a significant effect on farmers' adoption of water-saving irrigation technologies.
Endogenous test of technology perception
The endogenous nature of technology perception is an issue that cannot be ignored. The endogeneity problem arises for two reasons: Firstly, there is an interaction between farmers' perceptions of water-saving irrigation technologies and their technology adoption behavior. Secondly, in discussing the impact of farmers' perceptions of water-saving irrigation technologies on their technology adoption, this study cannot exclude individual unobservable heterogeneity. These unobservable factors can lead to omitted variables, creating an endogeneity problem, which, in turn, leads to regression bias. To address these endogeneity issues, a number of channels are used to access information on water-saving irrigation technologies in this paper. According to the empirical results in Table 3 irrigation well depth does not have a significant effect on farmers' water-saving irrigation technology. Therefore the issue of endogenous test of irrigation well depth is not under discussion. A two-stage regression was conducted using the IV probit model.
Table 4 shows the results of the Wald test for the original hypothesis on the exogeneity of the perceived ease of use. The p-value is 0.0001, so the perceived ease of use can be considered as endogenous variables at the 1% level. The results of the first-stage regression showed that the number of channels to access information on water-saving irrigation technologies had strong explanatory power for the perceived ease of use of this technology. The results of the second-stage regression showed that the estimated coefficient of 8.001 for the perceived ease of use was much larger than the estimated coefficient of 1.001 from the original probit model. It was significant at the 1% level. The above analysis suggests that the positive impact of technology perceptions on the adoption of water-saving irrigation technologies by farmers, particularly the perceived ease of use, is underestimated because of endogeneity.
Endogeneity test of technology perception.
| Stage 1 | ||
| Dependent variable | Independent variable | Coefficient |
| Perceived ease of use of technology | Number of information channels | 0.083*** (0.022) |
| Control variables | Yes | |
| Stage 2 | ||
| Water-saving irrigation technology adoption behavior | Perceived ease of use of technology | 8.001*** (2.579) |
| Control variables | Yes | |
| Wald test of exogeneity: χ2(2) = 15.62 | Prob > χ2 = 0.0001 |
| Stage 1 | ||
| Dependent variable | Independent variable | Coefficient |
| Perceived ease of use of technology | Number of information channels | 0.083*** (0.022) |
| Control variables | Yes | |
| Stage 2 | ||
| Water-saving irrigation technology adoption behavior | Perceived ease of use of technology | 8.001*** (2.579) |
| Control variables | Yes | |
| Wald test of exogeneity: χ2(2) = 15.62 | Prob > χ2 = 0.0001 |
Note: ***, **, and * indicate significance at the levels of 1, 5, and 10%, respectively; the robust standard errors are in parentheses.
Testing the moderating effects of government regulations
In the regression results in Table 3, only the perceived ease of use had a significant effect on farmers' water-saving irrigation technology adoption behavior among the technology perceptions. Therefore, only an interaction term between the perceived ease of use and government regulation was constructed in this section and added to the model for testing.
Table 5 shows that none of the interaction terms constructed for project demonstration and perceived ease of use passed the significance test; however, the interaction terms for technology training, perceived ease of use, technology subsidy, and perceived ease of use of technology had a significant positive effect on farmers' adoption of water-saving irrigation technologies. Both technical training and technical subsidies increased farmers' perceived ease of use of technology and encouraged them to adopt water-saving irrigation technologies. The interaction between government advocacy and perceived ease of use had a significant negative effect on farmers' adopting water-saving irrigation technologies, with government advocacy weakening their ease of use perceptions. This is because the current advocacy for water-saving irrigation technology is too specialized and does not provide farmers with a complete picture of its installation, maintenance, and use, creating a misconception about the ease of use.
The moderating effect results of government regulations.
| Logistic | Probit | |||
|---|---|---|---|---|
| Coefficient | Marginal Effects (%) | Coefficient | Marginal Effects (%) | |
| Perceived ease of use of technology | 1.672** (0.841) | 4.374** (0.022) | 0.846** (0.399) | 4.150** (0.020) |
| Government advocacy × perceived ease of use of technology | −1.705* (0.879) | −4.461* (0.023) | −0.914** (0.406) | −4.483** (0.020) |
| Technical training × perceived ease of use of technology | 1.712* (0.980) | 4.478* (0.024) | 0.896** (0.451) | 4.396** (0.021) |
| Technical subsidies × perceived ease of use of technology | 2.075* (1.177) | 5.428* (0.029) | 1.014** (0.517) | 4.976** (0.025) |
| Project demonstration × perceived ease of use of technology | 0.319 (0.884) | 0.836 (0.023) | 0.251 (0.390) | 1.232 (0.019) |
| Control variables | Yes | Yes | Yes | Yes |
| Pseudo R2 | 0.8182 | – | 0.8200 | – |
| Logistic | Probit | |||
|---|---|---|---|---|
| Coefficient | Marginal Effects (%) | Coefficient | Marginal Effects (%) | |
| Perceived ease of use of technology | 1.672** (0.841) | 4.374** (0.022) | 0.846** (0.399) | 4.150** (0.020) |
| Government advocacy × perceived ease of use of technology | −1.705* (0.879) | −4.461* (0.023) | −0.914** (0.406) | −4.483** (0.020) |
| Technical training × perceived ease of use of technology | 1.712* (0.980) | 4.478* (0.024) | 0.896** (0.451) | 4.396** (0.021) |
| Technical subsidies × perceived ease of use of technology | 2.075* (1.177) | 5.428* (0.029) | 1.014** (0.517) | 4.976** (0.025) |
| Project demonstration × perceived ease of use of technology | 0.319 (0.884) | 0.836 (0.023) | 0.251 (0.390) | 1.232 (0.019) |
| Control variables | Yes | Yes | Yes | Yes |
| Pseudo R2 | 0.8182 | – | 0.8200 | – |
Note: ***, **, and * indicate significance at the levels of 1, 5, and 10%, respectively; the robust standard errors are in parentheses; the regression results for the other variables in the model fit are generally consistent with Table 3, and the estimates are omitted here.
Heterogeneity analysis
Based on the results of the Third China Agricultural Census, farmers with agricultural operations of 50 mu or more are considered large-scale farmers, while those with less than 50 mu are considered smallholder farmers.
Table 6 shows that the impact of technology perception and government regulation considerably varied on the adoption of water-saving irrigation technologies by farmers with different arable land operation sizes. Perceived ease of use significantly contributed to the adoption of water-saving irrigation technologies by both large-scale and smallholder farmers, with a greater effect on large-scale farmers. Perceived usefulness only significantly encouraged smallholder farmers to adopt water-saving irrigation technologies. Government advocacy, technical training, and technical subsidies all had a significant positive impact on the adoption of water-saving irrigation technologies by large-scale farmers, with technical subsidies having the greatest effect, followed by government advocacy and technical training, and project demonstration did not have a significant impact. Only technical training significantly encouraged smallholder farmers to adopt water-saving irrigation technologies, while government advocacy, technical subsidies, and project demonstrations had no significant impact.
Differences in water-saving irrigation technology adoption behavior between farmers with different operation sizes.
| Large-scale farmers | Smallholder farmers | |||
|---|---|---|---|---|
| Variables | Logistic | Logistic | ||
| Coefficient | Marginal effects (%) | Coefficient | Marginal effects (%) | |
| Perceived ease of use of technology | 2.160*** (0.702) | 16.043*** (0.042) | 2.969*** (1.114) | 2.081*** (0.008) |
| Perceived usefulness of technology | −0.561 (0.597) | −4.163 (0.044) | 4.002*** (1.426) | 2.806** (0.013) |
| Government advocacy | 1.734*** (0.639) | 12.874*** (0.039) | 3.477 (2.392) | 2.437 (0.017) |
| Technical training | 0.953* (0.574) | 7.074* (0.041) | 4.249*** (1.097) | 2.978*** (0.009) |
| Technical subsidies | 2.336*** (0.638) | 17.345*** (0.041) | 4.841 (5.755) | 3.394 (0.039) |
| Project demonstration | −0.774 (0.802) | −5.747 (0.058) | 0.275 (1.232) | 0.193 (0.008) |
| Government advocacy | Yes | Yes | Yes | Yes |
| Pseudo R2 | 0.5745 | – | 0.8122 | – |
| Large-scale farmers | Smallholder farmers | |||
|---|---|---|---|---|
| Variables | Logistic | Logistic | ||
| Coefficient | Marginal effects (%) | Coefficient | Marginal effects (%) | |
| Perceived ease of use of technology | 2.160*** (0.702) | 16.043*** (0.042) | 2.969*** (1.114) | 2.081*** (0.008) |
| Perceived usefulness of technology | −0.561 (0.597) | −4.163 (0.044) | 4.002*** (1.426) | 2.806** (0.013) |
| Government advocacy | 1.734*** (0.639) | 12.874*** (0.039) | 3.477 (2.392) | 2.437 (0.017) |
| Technical training | 0.953* (0.574) | 7.074* (0.041) | 4.249*** (1.097) | 2.978*** (0.009) |
| Technical subsidies | 2.336*** (0.638) | 17.345*** (0.041) | 4.841 (5.755) | 3.394 (0.039) |
| Project demonstration | −0.774 (0.802) | −5.747 (0.058) | 0.275 (1.232) | 0.193 (0.008) |
| Government advocacy | Yes | Yes | Yes | Yes |
| Pseudo R2 | 0.5745 | – | 0.8122 | – |
Note: ***, **, and * indicate significance at the levels of 1, 5, and 10%, respectively; the robust standard errors are in parentheses; the regression results for the other variables in the model fit are generally consistent with Table 3, and the estimates are omitted here.
CONCLUSIONS
The following are the study's primary conclusions.
Technology perception and government regulations significantly encouraged farmers to adopt water-saving irrigation technologies. A one-level increase in farmers' perceived ease of use of water-saving irrigation technologies increases their probability of adopting the technologies by about 5%. The presence of government advocacy, technical training, and technical subsidies increased the probability of farmers' adoption of water-saving irrigation technologies by about 5, 3, and 6%, respectively.
Government regulations played a moderating role in the influence of the perceived ease of use of technology on farmers' adoption behavior. Technology training and subsidies increased farmers' perceptions of ease of use and reinforced its positive impact on their adoption of water-saving irrigation technologies. Government advocacy weakened the positive impact of the perceived ease of use on farmers' adoption behavior.
There were significant differences in the water-saving irrigation technology adoption behavior between farmers with different business sizes. Technology perception and government regulations can significantly increase the probability of adoption of water-saving irrigation technologies by large-scale farmers compared to smallholder farmers.
Based on the above conclusions, the following policy recommendations can be drawn. First, grassroots village committees, family farms, cooperatives, and other agricultural business entities can become important channels of policy advocacy, which can provide a variety of field guidance technologies and other forms of technical training. Strengthen project demonstrations in major food-producing regions should be strengthened in order to improve farmers' perceived ease of use of technology. Second, large-scale farmers are the focus of government regulations. Resources such as technical subsidies, government advocacy, and technical training are prioritized for distribution to large-scale farmers. Third, encourage farmers, especially smallholder farmers, to participate in agricultural cooperative organizations. Support village collective economic organizations and agricultural cooperatives to absorb farmers in order to increase their level of organization, develop socialized agricultural service system, and reduce the costs of water-saving irrigation technology through professional divisions of labor.
Although this study analyzes the effects of technology perception and government regulations on the adoption behavior of water-saving irrigation technologies by farmers, it does not further reveal the changes in the dynamics of water-saving irrigation technology adoption by farmers, such as how the ratio of the area applying water-saving irrigation technologies to the total area of cultivated land varies, and how the technology perceptions and government regulations affect the changes in the ratio of the area applying water-saving irrigation technologies. Government regulations are simply categorized and specific policy implementation is not refined. This study does not further explore the impact of the implementation mode and strength of government regulations on the adoption of water-saving irrigation technology by farmers, which weakens the explanatory power of government regulations on the adoption behavior of water-saving irrigation technology by farmers. Therefore, in the future, we will continue to track the promotion and application of water-saving irrigation technologies in the North China Plain. Based on more detailed data, we will further explore the influence of technology perception and the way and strength of government regulations on the continuous adoption behavior of water-saving irrigation technologies by farmers.
AUTHOR CONTRIBUTIONS
Yongqiang Wang: Conceptualization, Writing-original draft, Data curation, Formal analysis, Investigation, Methodology, Software. Zhe Wang: Investigation, Project administration, Funding acquisition. Mingyue Zhao: Investigation, Data curation. Bingrong Li: Writing-review ＆ edition.
FUNDING
This research was funded by the National Social Science Foundation of China ‘Empirical Measurement, Generation Mechanism and Policy Guidance of Agricultural Water Rebound Effect in Beijing-Tianjin-Hebei Region’ (17BJY093) and the Major Research Projects in the Humanities and Social Sciences of the Hebei Provincial Department of Education in 2021 ‘Research on Agricultural Water Conservation in Hebei Province under the Coordination of Economy and Ecology’ (ZD202115).
INFORMED CONSENT STATEMENT
Informed consent was obtained from all subjects involved in the study.
ACKNOWLEDGEMENT
We thank the Hengshui Municipal Bureau of Agriculture and Rural Affairs and Malan Farm for their assistance with data collection.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.
%20cropped.png?versionId=7148)