Farmers’ Acceptance of Water–Fertilizer Integration Technology: Theory and Evidence

Abstract

The traditional rough development method for irrigation and fertilization techniques has resulted in the waste of fertilizer and water sources and the degradation of black soil. The implementation of integrated water and fertilizer technology has the potential to address these issues. However, its success depends on farmers’ willingness to adopt it. This study aims to explore the incentives for farmers to adopt water and fertilizer integration technology through a practical investigation in China, revealing the driving mechanisms. The study constructed a technology adoption model and conducted a cross-sectional field study with farmers in Northeastern China. Financial consequences were incorporated into the integrated UTAUT-NAM to examine farmers’ acceptance. The validity and applicability of the model were evaluated through a partial least squares approach to structural equation modeling. The results showed that personal norms and financial consequences were the most critical factors influencing farmers’ willingness to adopt water–fertilizer integration technology. In addition, expected performance, facilitating conditions, and effort required were also significant predictors. The study further highlighted the pivotal role of awareness of consequences and responsibility in influencing farmers’ intentions to adopt the new technology, while social influence had no significant impact. The findings demonstrated that the established research model elucidated 69.1% of the observed variation in farmers’ intention to adopt water–fertilizer integration technology. The results of this study provide theoretical support for promoting water–fertilizer integration technology and inform practical strategies for its implementation. The study offers actionable insights for policymakers, agricultural advisors, and technology developers to promote resource-efficient irrigation and fertilization methods.

1. Introduction

In recent years, global agriculture has been facing increasingly severe resource pressures and environmental challenges, and how to improve agricultural production efficiency and reduce resource waste has become a focus of attention for all countries. The water–fertilizer integration technology, an intensive and precise agricultural management approach, has garnered widespread attention and promotion [1,2]. This technology can effectively improve the utilization efficiency of fertilizer and water resources through the organic combination of irrigation and fertilization processes [3,4]. This technology not only reduces the waste of water but can also accurately regulate the supply of nutrients needed by the crop, improve the efficiency of fertilizer use, and thus improve crop yield and quality [5,6]. Especially in areas with relatively scarce water resources, such as Northeastern China, water–fertilizer integration technology provides significant support for addressing agricultural resource waste and enhancing agricultural productivity.

Although water–fertilizer integration technology offers significant ecological and economic benefits, its successful adoption largely depends on farmers’ acceptance. As key decision-makers in agricultural production, farmers’ willingness and actions play a crucial role in the application of new technologies [7]. Their acceptance or rejection of the technology determines whether it can be effectively implemented in practical production and fully realize its potential benefits. Understanding farmers’ acceptance can help extension agencies and policymakers optimize extension strategies, making it easier to guide farmers toward adopting new technologies more effectively. Examining the motivations and barriers influencing farmers’ acceptance of this technology encourages the broader adoption of advanced irrigation technologies in agricultural production.

It is noteworthy that the extant literature contains a paucity of discussion on the behavioral mechanisms of adopting integrated water and fertilizer technology. The prevailing research trajectory is oriented toward technical efficiency, environmental benefits, and economic advantages [8,9]. Through field trials and model simulations, scholars have verified the significant effects of water–fertilizer integration technology in improving agricultural production efficiency and enhancing water and fertilizer use efficiency [1,10]. Tian et al. [11] demonstrated that water–fertilizer integration technology could effectively increase crop yield by 28% compared to traditional flood irrigation. This technology also reduced nitrogen oxide emissions by 4.23% and decreased water and fertilizer consumption by 5.78% and 14.03%, respectively. Additionally, the implementation of advanced irrigation technology has the potential to markedly reduce the input of water and fertilizer, as well as energy and economic costs, compared to conventional practices. These reductions have been observed to range from 12.18% to 28.57% [12]. However, these technology-oriented studies do not fully explain a key contradiction: why do farmers still have significantly different adoption rates when the technological benefits have been scientifically proven? This incongruity between theory and practice underscores the limitations of extant research, which predominantly focuses on technical and environmental aspects while paying comparatively less attention to the behavioral factors that influence farmers’ acceptance of the technology. In particular, extant literature has paid insufficient attention to the decision-making processes, motivations, and barriers of farmers, and there has been a paucity of systematic analysis of the dynamic links between technology adoption behavior and socio-economic conditions. Consequently, technology promotion strategies frequently diverge from the actual needs of farmers.

To address this issue, our research aims to examine the factors influencing farmers’ acceptance of integrated water and fertilizer technology. Initially, a conceptual model is developed based on the specific nature of technologies and farmers in the Northeast Black Soil region. Then, data were collected through an anonymous cross-sectional online survey to determine the magnitude of technology adoption by farmers in Northeast China. The questionnaire included an informed consent form, demographic information, and a structured measurement scale. Subsequently, we deployed PLS-SEM to validate the model. Furthermore, the outcomes of this research are examined in terms of their theoretical and practical implications for farmers’ uptake of advanced technology. Based on these findings, recommendations for future policy are presented.

The following contributions differentiate our research from existing research. First, this research addresses a novel issue of farmers’ acceptance of water–fertilizer integration technology in the black soil region. Our study presents a theoretical acceptance model of integrated water and fertilizer technology based on the extended Unified Theory of Acceptance and Use of Technology (UTAUT). We introduce the Norm Activation Model (NAM) to describe the prosocial characteristics of water–fertilizer integration technology and use financial consequences to measure farmers’ perceptions of the input–output ratio. Second, 69.1% of the behavioral intent variance has been explained by the proposed acceptance model, showing an excellent explanatory power of the model. Finally, research results show that personal norms and financial consequences are the most critical factors in farmers’ willingness to adopt water–fertilizer integration technology. The results offer insights into developing and implementing this advanced irrigation technology in the Northeast Black Soil region.

This paper is composed of six parts. The second part presents the model development of the water–fertilizer integration technology adoption model. Part 3 shows the methods and data. In the fourth part, we present the research results and evaluate them. Part 5 discusses the key findings, implications, and limitations. The sixth part summarizes the research.

2. Materials and Methods

2.1. Model Development

A number of studies have been conducted to elucidate the rationale behind the decision of numerous theoretical farmers to adopt new technologies [13]. This phenomenon has been the subject of considerable academic interest. The UTAUT model [14] represents an integration of the Technology Acceptance Model (TAM) with other technology-related models. The UTAUT posits four major determinants: performance expectancy, effort expectancy, social influence, and facilitation condition. The preliminary three constructs influence behavioral intention variables, which, in turn, influence use behavior. The fourth construct exerts its direct influence on use behavior. Moreover, the validity of the model in the agricultural field has been widely confirmed [15,16]. Therefore, this study uses UTAUT as the basic theoretical framework.

NAM is a psychological framework developed to explain prosocial behavior, particularly in situations where individuals are activated by a sense of moral obligation rather than personal gain. NAM attempts to predict how people are likely to behave according to personal norms or social obligations [17]. The model posits that individuals are more inclined to engage in prosocial behaviors when they are aware of the consequences of their actions, feel responsible for their actions, and are motivated by internal social norms. In a similar vein, numerous preceding studies have corroborated the explanatory prowess of the NAM in environmental contexts [18,19]. Consequently, in this study, the NAM framework was selected to model farmers’ acceptance of integrated water and fertilizer technology, given its capacity to encapsulate the prosocial characteristics inherent in this technology. Water–fertilizer integration generates broader environmental and societal benefits, including resource conservation and reduced environmental impact. The NAM framework allows us to incorporate these prosocial dimensions into our analysis, capturing how farmers’ awareness of the environmental and societal benefits, combined with their sense of responsibility toward sustainable agricultural practices, influences their adoption decisions.

Furthermore, a review of the existing literature and empirical studies suggests that, moreover [13], the economic feasibility of technology adoption is a key factor influencing farmers’ decision-making processes. In accordance with the theory of economic rationality [20], the heterogeneity of farmers’ resource endowments is significantly and positively correlated with their willingness to adopt technological innovations. This suggests that, while cost is the primary factor determining technology adoption, farmers undertake a multifaceted economic evaluation during the decision-making process. This evaluation encompasses not only the initial investment cost but also broader economic indicators, such as marginal benefits, risk premiums, and opportunity costs, that arise from the technology’s implementation. Consequently, it can be deduced that farmers’ decision-making process regarding technology adoption is fundamentally a comprehensive evaluation of the potential economic ramifications of technology implementation.

In summary, the theoretical model proposed in this study is a novel integration of UTAUT and NAM with the extension of financial consequences. The UTAUT model is reflected in its systematic explanation of the rational decision-making mechanism of technology adoption. The NAM is based on the endogenous moral drive of farmers to adopt technology. The study also incorporates the concept of economic drivers, which is based on the theory of economic drivers. This model explains the motives for adopting the water–fertilizer integration technology from a multidimensional perspective, solving the dilemma of traditional research that only explains the motivation for adopting agricultural technology in a single way. A visual representation of the model is shown in Figure 1.

Behavioral Intention

Behavioral intention is defined as the extent to which farmers are willing to accept and adopt new technologies. According to previous studies [21], perception factors shape farmers’ behavioral intentions by influencing their attitudes toward using technologies. In addition, the UTAUT model further emphasizes the core role of behavioral intention in the technology adoption process, pointing out that multiple factors can affect adoption behavior through behavioral intention. Therefore, we will measure farmers’ acceptance of adopting water–fertilizer integration technology through behavioral intention.

Personal norms

In the context of our research, personal norms represent that when farmers perceive the possible impact of a particular behavior on others or the environment, they will have the corresponding behavioral intention if they feel the drive of the personal norms [17]. According to the research by Zhang et al. [22], the positive impact of personal norms on users’ willingness was found in environmental behavior studies. Thus, we hypothesize the following:

H1: 

Personal norms positively affect the behavioral intention of using water and fertilizer integration technology among farmers.

Ascription of responsibility

In our research, the ascription of responsibility implies the cognition that farmers are responsible for the adverse consequences that may be caused by their non-implementation of a particular behavior. According to previous studies [17,22], if farmers clearly understand the results of a particular behavior and believe that they have specific responsibility for these consequences, their personal norms will be activated, thus affecting their behavioral intentions. For this reason, we put forward that:

H2: 

Ascription of responsibility positively affects the personal norms of using water and fertilizer integration technology among farmers.

Awareness of consequences

In our research, awareness of consequences represents farmers’ perception of the possible negative consequences of their actions. As evidenced by prior research [17], when farmers are aware of a new technology’s environmentally and socially beneficial implications, they may be more inclined to assign responsibility for environmental stewardship. Consequently, they may be more willing to establish personal norms that consider the use of this technology a moral obligation. To this end, we present the following hypotheses:

H3: 

Awareness of consequences positively affects the personal norms of using water and fertilizer integration technology among farmers.

H4: 

Awareness of consequences positively affects farmers’ ascription of responsibility for using integrated water and fertilizer technology.

Performance expectancy

In the context of our study, performance expectancy represents farmers’ belief that adopting new technologies can improve work performance. This construct is correlated with the job-related factors that comprise the UTAUT. According to previous studies [14,23,24], if farmers feel that the performance expectation is relatively high, they will increase their willingness to adopt this technology. Hence, the hypothesis of this study is as follows:

H5: 

Performance expectancy positively affects farmers’ willingness to use water and fertilizer integration technology among farmers.

Effort expectancy

In the context of our study, effort expectancy represents how difficult it is for farmers to adopt new technologies. It brings together the difficult-related factors in the models that make up UTAUT. As demonstrated by prior research [25], the anticipation of effort will influence the behavioral intention of farmers, subsequently impacting their inclination to adopt the technology. As a result, as technology is perceived as less difficult, adoption intentions increase. Other studies have shown that [21,26,27], if farmers feel that the new technology is easier to use, then their performance expectancy of this technology will be improved. As a result, our study hypotheses include the following:

H6: 

Effort expectancy positively affects farmers’ willingness to use water and fertilizer integration technology.

H7: 

Effort expectancy positively affects the performance expectancy of using water and fertilizer integration technology among farmers.

Social influence

In the context of our research, social influence represents how farmers’ decisions to adopt new technologies are affected by the views and attitudes of individuals or groups and is correlated with socially relevant factors in the models that make up UTAUT. According to previous studies [28], social influence will affect farmers’ behavioral intentions toward acquiring new technologies and further jeopardize their adoption behaviors. In addition, studies have shown [17] that when farmers are influenced by society, their personal norms are activated when they need to make decisions to adopt new technologies and feel responsible for taking action. Consequently, we formulate the following research hypotheses:

H8: 

Social influence positively affects farmers’ willingness to use water and fertilizer integration technology.

H9: 

Social influence positively affects the personal norms of using water and fertilizer integration technology among farmers.

Facilitating conditions

In the context of our research, facilitating conditions represent the extent to which farmers are confident that their resources, knowledge, and technical assistance can support the adoption of new technologies. It is related to the conditional resource factors that make up the UTAUT model. According to previous research [24], farmers are more likely to employ new technologies if they have the enabling conditions. Thus, we hypothesize that:

H10: 

Facilitation conditions positively affect the behavioral intention of using water and fertilizer integration technology among farmers.

Financial consequences

In the context of our research, financial consequences reflect the extent to which farmers perceive the costs and incomes of adopting water and fertilizer integration technology. Previous studies believe that financial consequences, including expected cost savings and income increases, will positively impact farmers’ behavioral intentions [13,20]. For this reason, we posit the hypotheses below:

H11: 

Financial consequences positively affect the behavioral intention of using water and fertilizer integration technology among farmers.

2.2. Measurements

The questionnaire comprised multiple sections. (1) Participants were presented with an informed consent form before beginning the online questionnaire. It was made clear that taking the survey was completely optional, that responses would be anonymous, and that the data would be kept confidential. This was performed to address any concerns and ensure the data collected were accurate and valid. Before starting the research project, participants had to read the declaration of consent and approve their willingness to participate in the survey. (2) In the second part of the questionnaire, participants provided information regarding their basic demographic characteristics and other basic information, including their age, gender, level of education, planting area, and crops. (3) The third section provided a concise overview of the integrated water and fertilizer technology to ensure that participants have a clear understanding of this technology, which was described in detail below: “Water and fertilizer integration technology represents a contemporary approach to agricultural management, whereby water and fertilizer are delivered directly to the root of the crop via an irrigation system, thereby facilitating precise irrigation and fertilization. The financial implications of implementing integrated water and fertilizer management strategies are contingent upon the quality of the water and the quantified amount of fertilizer utilized. The expenditure incurred is anticipated to range from 2000 to 10,000 yuan per mu. The deployment of this technology is expected to result in a notable reduction in the consumption of water resources and fertilizers, along with a simultaneous decline in production costs. Implementing integrated water and fertilizer technology will likely result in enhanced resource efficiency, mitigated environmental impact, and a notable improvement in crop yield and quality”. (4) The fourth part of the study examined participants’ cognitive processes and opinions regarding the water and fertilizer integration technologies to facilitate the final measurement of the predictors of farmers’ acceptance. The participants were invited to rate the projects on a 7-point Likert scale, with responses ranging from disagreement to agreement, according to their perceptions and feelings about water and fertilizer integration technologies. The design included a reverse question to ensure data accuracy and identify informal responses. The measurement construct items utilized in the investigation are presented in

Table 1. Construct measurement items.

Construct	Items	Origin
Performance Expectancy (PE)	PE1: Water–fertilizer integration technology can improve the efficiency of water resources and fertilizer use.	[14]
	PE2: Employing water–fertilizer integration technology helps me complete agricultural production tasks faster.
	PE3: I believe water–fertilizer integration technology is highly beneficial for agricultural production.
Effort Expectancy (EE)	EE1: It was easy for me to learn how to use water and fertilizer integration technology.	[21]
	EE2: Employing water–fertilizer integration technology is straightforward to understand.
	EE3: I can quickly become proficient in employing water–fertilizer integration technology.
	EE4 *: It is challenging for me to learn how to employ water–fertilizer integration technology, and it requires a lot of effort.
Social Influence (SI)	SI1: People vital to me (such as family and friends) believe I should use water and fertilizer integration technology.	[28]
	SI2: People who influence my behavior think I should use water and fertilizer integration technology.
	SI3: People whose opinions I value think I should use water and fertilizer integration technology.
Facilitating Conditions (FC)	FC1: I have the resources needed to use integrated water and fertilizer technology (such as space and money).	[24]
	FC2: I have the necessary knowledge to use integrated water and fertilizer technology.
	FC3: When I encounter difficulties employing water–fertilizer integration technology, I can receive the necessary technical guidance and services.
Financial Consequences (FCF)	FCF1: Employing this technology can reduce financial waste caused by separate irrigation and fertilization.	[29]
	FCF2: Employing this technology can save water and fertilizer, thereby lowering my production costs.
	FCF3: The initial investment may be higher, but employing this technology can increase my net income overall.
Awareness of Consequences (AC)	AC1: I think employing this technology can conserve water resources.	[17]
	AC2: I think employing this technology can reduce fertilizer runoff.
	AC3: I think employing this technology can promote the sustainable development of the environment.
Ascription of Responsibility (AR)	AR1: Saving water resources is my responsibility.	[22]
	AR2: Reducing fertilizer runoff is my responsibility.
	AR3: Promoting the sustainable development of the environment is my responsibility.
Personal Norms (PN)	PN1: I feel that conserving water resources, reducing fertilizer runoff, and promoting environmental sustainability is a moral obligation.	[17]
	PN2: I think it is generally important for farmers to save water, reduce fertilizer loss, and promote environmental sustainability.
	PN3: I feel obliged to use integrated water and fertilizer management because of my own values/principles.
Behavioral Intention (BI)	BI1: I am interested in using water and fertilizer integration technology.	[14]
	BI2: I am willing to try using integrated water and fertilizer technology.
	BI3: In summary, I intend to use integrated water and fertilizer technology.

* Note that EE4 was a reverse problem that facilitates the identification of invalid responses and guarantees data quality.

.

2.3. Data Collection

The Chinese government has demonstrated a robust commitment to the advancement of water–fertilizer integration technology since 2016. The Northeast area of China, which comprises black soil, is the primary grain-producing region of the country. However, the loss of black soil has become increasingly severe in recent years. Furthermore, the Chinese government implemented many policies with the objective of preserving black soil. Accordingly, the survey was conducted in Heilongjiang Province, China, from June to July 2024. The survey used an offline field research method to determine the propensity of farmers and agricultural practitioners to adopt integrated water and fertilizer technology. Data were collected online using a document collection method and mobile questionnaires. The research questionnaire was developed based on a technology adoption theoretical framework. The measurement dimensions were initially determined through literature analysis, and the content validity of the questionnaire was tested using the Delphi expert consultation method. An expert panel consisting of five agricultural technology extension experts, agricultural engineering professors, and policy researchers was invited to conduct two rounds of consultation to optimize the questionnaire structure and item wording. Before the formal survey, the research team conducted a pre-survey in the surrounding counties and districts of Harbin and tested the questionnaire on 30 agricultural operators. Based on the feedback, the wording and logical structure of the questionnaire were further adjusted. In addition, this research project has been reviewed by the Academic Ethics Committee of the College of Water Resources and Civil Engineering of Northeast Agricultural University. The ‘Research Group on the Application Analysis and Promotion Strategy of Water and Fertilizer Integration Technology for Black Soil Protection’ is responsible for data collection to ensure that the research process complies with academic norms.

Participants were selected using a cross-sectional survey method in Heilongjiang Province. Stratified random sampling was used to ensure the representativeness of the sample. The sampling process followed a multi-stage principle: first, 2 counties were randomly selected from each of the 13 prefecture-level cities in Heilongjiang Province (26 counties in total); second, 1 township was randomly selected from each sample county (26 townships in total); and finally, 2 administrative villages were randomly selected from each sample township (52 villages in total). Among the village-level sample units, 36 agricultural operators (including ordinary farmers, family farms, cooperatives, and agricultural enterprises) were selected as survey subjects using systematic sampling. The survey was carried out in cooperation with the local agricultural department. A cross-sectional survey method was used to ensure that the sample was diverse, including individuals from different geographical areas and backgrounds, including farmers, agricultural practitioners, and professionals in the field of agricultural consulting [30]. The random sampling method minimizes the possibility of selection bias, thereby improving the representativeness of the sample.

The final sample consisted of 1931 questionnaires completed by farmers and agricultural practitioners from Northeast China. As not all the surveyed farmers completed the questionnaire meticulously, we excluded those questionnaires that provided responses deemed insufficiently serious in accordance with the responses provided to the opposite questions. In addition, questionnaires were classified as invalid samples if the total response time was less than one minute and the informed consent page was completed with disagreement. After the screening process, 606 of the 1931 questionnaires were determined to be eligible for analysis, yielding a response rate of 31.4%.

Figure 2 illustrates the distribution of farmers’ responses across various measurement items related to their acceptance of water–fertilizer integration technology. The chart categorizes responses into three groups: low response (scores 1–3, represented by green triangles), medium response (score 4, represented by blue circles), and high response (scores 5–7, represented by red squares). Most items received a high level of response, with the red line generally extending farther from the center, indicating strong agreement or positive perception among farmers.

Table 2. Demographic characteristics of the valid sample.

Characteristics	n (%)
Gender
Male	319 (52.6%)
Female	287 (47.3%)
Education level
Elementary school or lower	125 (20.6%)
Middle school	269 (44.4%)
High school/secondary school	87 (14.4%)
College and above	125 (20.6%)
Age
≤25	98 (16.2%)
26–35 years	57 (9.4%)
36–45 years	92 (15.2%)
46–55 years	190 (31.4%)
56–65 years	119 (19.6%)
≥66	50 (8.2%)
Land area
50 acres or lower	330 (54.5%)
50–100 acres	107 (17.7%)
100–150 acres	53 (8.7%)
150–200 acres	54 (8.9%)
200 acres or more	62 (10.2%)
Agricultural income per year
10,000 Yuan or lower	175 (28.9%)
10,000–50,000 Yuan	255 (42.1%)
50,000–100,000 Yuan	103 (17.0%)
100,000 Yuan or more	73 (12.0%)
Plant variety
Maize	62 (10.2%)
Corn	406 (67.0%)
Rice	105 (17.3%)
Else	33 (5.4%)
Heard of integrated water and fertilizer technology or not
Yes	264 (43.6%)
No	342 (56.4%)
Used integrated water and fertilizer technology or not
Yes	194 (32.0%)
No	412 (68.0%)

provides an overview of demographic data, agricultural planting area, rural income, crop varieties, and knowledge and usage of water and fertilizer integration management, as derived from the 606 valid samples.

2.4. Data Analysis

This study used partial least squares structural equation modeling (PLS-SEM) techniques to identify the main factors influencing end users’ adoption of integrated water and fertilizer technology. We believe that this particular method choice is most appropriate for the specific research area under study. The sample size for PLS-SEM is usually determined by the ten-fold rule, which was originally proposed by Hair et al. [31,32]. This rule states that the required sample size should be no less than ten times the maximum number of observed variables in the path model. The actual size of this sample is 606. Therefore, it can be concluded that the sample size required for this study meets this requirement. Data analysis was carried out using the two-stage procedure in Smart PLS 3.0 [33]. The reliability and validity of the model were assessed using confirmatory factor analysis, and the structural model was subsequently assessed using the robust 5000-subsample bootstrap method. This method is used to assess the ability of the model to predict results and to determine the nature of the relationships between the model structures.

3. Results

3.1. Measurement Model Evaluation

Reliability is an indicator that assesses the consistency and stability of sample data, thereby reflecting the credibility of the study results. We employed Cronbach’s alpha and composite reliability (CR) to evaluate the internal consistency and overall reliability of the model. As demonstrated in

Table 3. Reliability and validity.

Constructs	Items	Factor Loadings	Mean	Cronbach’s Alpha	CR	AVE
AC	AC1	0.922	4.904	0.899	0.937	0.833
	AC2	0.861	4.746
	AC3	0.952	4.929
AR	AR1	0.926	5.215	0.892	0.933	0.824
	AR2	0.847	5.036
	AR3	0.947	5.267
BI	BI1	0.931	4.906	0.897	0.936	0.831
	BI2	0.862	4.715
	BI3	0.939	4.929
EE	EE1	0.842	4.561	0.836	0.891	0.674
	EE2	0.819	4.540
	EE3	0.900	4.620
	EE4	0.712	4.087
FC	FC1	0.876	4.512	0.841	0.904	0.759
	FC2	0.843	4.398
	FC3	0.893	4.746
FCF	FCF1	0.914	4.901	0.883	0.928	0.811
	FCF2	0.862	4.715
	FCF3	0.925	4.861
PE	PE1	0.899	4.998	0.862	0.916	0.785
	PE2	0.838	4.818
	PE3	0.919	5.010
PN	PN1	0.927	5.210	0.869	0.920	0.794
	PN2	0.849	4.997
	PN3	0.895	5.059
SI	SI1	0.907	4.706	0.875	0.924	0.802
	SI2	0.834	4.525
	SI3	0.943	4.748

, both Cronbach’s alpha and CR values surpassed 0.70, signifying that the measurement model exhibits an acceptable level of reliability. Validity is the degree to which measurement variables accurately reflect underlying constructs. The concept of convergent validity is extended in this case to encompass discriminant validity as well. The model’s convergent construct validity was assessed through an analysis of two key indicators: the average variance extracted (AVE) and item reliability for each construct. As illustrated in Table 3, all AVE values surpassed 0.50, while item loadings achieved a value of 0.70 or above, thereby confirming the model’s satisfactory convergent validity. The assessment of predictive validity was conducted in accordance with the criteria set forth by Fornell–Larcker. This entailed an examination of item cross-loading.

Table 4. Factor loading and the shared variance between the different structures.

	FC	AR	BI	AC	EE	FCF	PE	PN	SI
FC1	0.876	0.320	0.510	0.514	0.532	0.580	0.498	0.393	0.661
FC2	0.843	0.341	0.571	0.512	0.569	0.560	0.511	0.440	0.565
FC3	0.893	0.432	0.571	0.578	0.542	0.594	0.506	0.471	0.561
AR1	0.404	0.926	0.599	0.661	0.407	0.605	0.574	0.787	0.445
AR2	0.285	0.847	0.511	0.544	0.328	0.483	0.503	0.725	0.326
AR3	0.444	0.947	0.617	0.651	0.432	0.611	0.598	0.799	0.473
BI1	0.604	0.608	0.931	0.679	0.541	0.695	0.668	0.685	0.613
BI2	0.541	0.572	0.862	0.688	0.503	0.650	0.636	0.679	0.548
BI3	0.581	0.560	0.939	0.655	0.541	0.672	0.660	0.656	0.561
AC1	0.574	0.600	0.667	0.922	0.515	0.719	0.700	0.647	0.654
AC2	0.515	0.638	0.678	0.861	0.504	0.73	0.687	0.696	0.580
AC3	0.59	0.636	0.679	0.952	0.528	0.732	0.709	0.656	0.636
EE1	0.585	0.430	0.531	0.534	0.842	0.515	0.558	0.486	0.556
EE2	0.546	0.381	0.515	0.503	0.819	0.509	0.523	0.428	0.532
EE3	0.612	0.425	0.569	0.579	0.900	0.573	0.601	0.498	0.627
EE4	0.287	0.147	0.255	0.197	0.712	0.186	0.181	0.180	0.223
FCF1	0.595	0.573	0.679	0.751	0.495	0.914	0.697	0.647	0.665
FCF2	0.557	0.571	0.639	0.703	0.448	0.862	0.649	0.622	0.598
FCF3	0.640	0.550	0.674	0.697	0.556	0.925	0.677	0.605	0.648
PE1	0.509	0.591	0.652	0.686	0.529	0.659	0.899	0.644	0.601
PE2	0.503	0.537	0.607	0.662	0.493	0.661	0.838	0.603	0.534
PE3	0.529	0.511	0.649	0.686	0.523	0.671	0.919	0.588	0.629
PN1	0.456	0.820	0.653	0.668	0.467	0.617	0.615	0.927	0.516
PN2	0.363	0.762	0.575	0.586	0.358	0.567	0.577	0.849	0.396
PN3	0.511	0.688	0.742	0.691	0.495	0.667	0.652	0.895	0.577
SI1	0.626	0.423	0.566	0.628	0.559	0.671	0.603	0.516	0.907
SI2	0.586	0.386	0.544	0.561	0.477	0.581	0.581	0.477	0.834
SI3	0.626	0.425	0.583	0.646	0.580	0.646	0.604	0.509	0.943

demonstrates that the loads for each construct exceed their cross-loadings in measurement compared to those of any other construct. Furthermore, as illustrated in

Table 5. Test results for discriminant validity.

	AR	AC	BI	EE	FC	FCF	PE	PN	SI
AR	0.908
AC	0.683	0.913
BI	0.636	0.738	0.911
EE	0.430	0.565	0.580	0.821
FC	0.419	0.614	0.632	0.629	0.871
FCF	0.626	0.796	0.738	0.556	0.664	0.901
PE	0.616	0.765	0.718	0.582	0.580	0.749	0.886
PN	0.849	0.729	0.738	0.496	0.499	0.693	0.690	0.891
SI	0.460	0.684	0.630	0.603	0.684	0.707	0.665	0.559	0.896

, the square root of each construct’s AVE is greater than its correlation coefficient with other latent variables. Consequently, we conclude that discriminant validity is acceptable.

3.2. Structural Model Evaluation

The PLE-SEM calculation results show that the SRMR value is equal to 0.065, which is less than 0.08, indicating that the standardized residuals of the model covariance matrix and the observed data covariance matrix are at an excellent level [34]. The NFI value is 0.845, which is close to 1, indicating that the goodness of fit of the model is good [31]. Therefore, the model of farmers’ acceptance of water and fertilizer integration technology demonstrated a good model fit [35,36,37]. As illustrated in Figure 3, the results of the structural model evaluation are presented. The findings of the hypothesis testing and path coefficient analysis are presented in

Table 6. Hypothesis test results and path coefficients.

Hypothesis	Path Coefficients (β)	p-Value
H1: PN→BI	0.347	0.000
H2: AR→PN	0.192	0.000
H3: AC→PN	0.661	0.000
H4: AC→AR	0.683	0.000
H5: PE→BI	0.175	0.000
H6: EE→BI	0.083	0.029
H7: EE→PE	0.582	0.000
H8: SI→BI	0.018	0.683
H9: SI→PN	0.124	0.000
H10: FC→BI	0.158	0.001
H11: FCF→BI	0.202	0.002

. The results demonstrate that the accuracy of the proposed model is 69.1% in predicting farmers’ acceptance of water fertilization technology. Conversely, Chin asserts that a threshold value exceeding 67% signifies substantial explanatory power [38]. Consequently, this model demonstrates adequate explanatory power with regard to farmers’ adoption of integrated water and fertilizer technology.

3.3. Hypothesis Test

The relationship between variables is shown by the path coefficients and their associated p-values, where the p-value indicates the probability of the current data being true if the null hypothesis is true. The significance level is typically set to 0.001, 0.01, or 0.05, with a lower value indicating higher statistical significance [39]. The proposed hypothesis is displayed in Table 6, which corroborates the p-value of H1, H5, and H10 being less than 0.001, which is considered to be highly significant. The p-value for H11 is less than 0.001, indicating a reliably significant result, while the p-value for H6 is lower than 0.05, suggesting a horizontally significant outcome. Additionally, the level of significance for H8 is greater than 0.05, indicating that social influence is not a notable predictor of behavioral intention within the specified context. Consequently, H8 is rejected. Thus, behavioral intention is positively influenced by personal norms, performance expectations, enabling conditions, financial consequences, and effort expectations. In addition, the corresponding p-values for H1, H2, and H9 are not greater than 0.001, implying that these hypotheses are reliably significant at this level. Therefore, it can be concluded that personal norms exert the strongest influence on behavioral intentions and that these intentions are positively influenced by the meaning of consequences, sense of responsibility, and social influence. Furthermore, the p-values for hypotheses H4 and H7 are below 0.001, suggesting a positive effect of AC→AR and EE→PE.

Table 7. Effect of drivers on the willingness to adopt.

	Indirect Effect	Direct Effect	Total Effect
PN→BI	_	0.347 ***	0.347 ***
AR→BI	0.229 ***	_	0.229 ***
AC→BI	0.224 ***	_	0.224 ***
FCF→BI	_	0.202 **	0.202 **
EE→BI	0.102 ***	0.083 *	0.185 ***
PE→BI	_	0.175 ***	0.175 ***
FC→BI	_	0.158 ***	0.158 ***
SI→BI	0.043 ***	0.018	0.061

* p < 0.05, ** p < 0.01, *** p < 0.001.

illustrates the direct, indirect, and total effects of predictor variables on farmers’ acceptance of water and fertility integration management technology. The overall results demonstrate that the most significant influence on behavioral intention is exerted by personal norms, followed by the sense of consequence, the sense of responsibility, the financial consequences, the expectation of effort, the expectation of performance, and the enabling conditions. In addition, the relationships AC→PN→BI, AR→PN→BI, SI→PN→BI, and EE→PE→BI were identified as significant. This suggests that awareness of consequences, responsibility, social influence, and effort expectancy play a moderating role in farmers’ intentions to adopt.

4. Discussion

4.1. Theoretical Influences

The study demonstrated that personal norms were the most significant factor influencing farmers’ adoption of integrated water and fertilizer technologies. This finding is consistent with the core mechanism of the NAM [40], which states that farmers’ moral drives have a positive effect on technology adoption. One potential explanation for this phenomenon is that farmers exhibit a strong propensity to act when they perceive the adoption of water- and fertilizer-saving technologies as a moral obligation. This intrinsic motivation can supersede the constraints imposed by external factors, including economic disincentives [41]. This may be related to the growing sense of social responsibility and environmental awareness among farmers. In modern agriculture, integrated water and fertilizer technologies not only improve water and fertilizer utilization efficiency but also help reduce environmental pollution and enhance soil sustainability. Therefore, farmers may view the adoption of such technologies not only as an economic decision but also as a demonstration of responsibility toward the environment and society.

The study also confirmed that consequence awareness and responsibility attribution, as two ways of activating norms, together shape personal norms. When farmers are aware of the environmental benefits of technology application (AC) and believe that they have a responsibility to implement it (AR), a strong sense of moral obligation (PN) is formed, which directly promotes behavioral intention. This finding aligns with the results of Zhang (2018) [22], who determined that personal norms are the primary factor driving citizens to file environmental complaints. The study also found that awareness (AC) has a positive impact on responsibility (AR). One potential explanation for this phenomenon is that if farmers perceive the positive impact of using integrated water and fertilizer technology, they will attribute it to responsibility, which, in turn, will have a positive effect on their personal norms and behavioral intentions, as also confirmed by Dong et al. [41].

This study found that the adoption of new agricultural technologies by farmers is not directly influenced by social influence but indirectly influenced through the mediating mechanism of personal norms. This finding contradicts the predictions of the traditional UTAUT model and may be attributed to the unique sociocultural context of rural China. When farmers are granted complete autonomy in land management, their motivation to adopt technology is more likely to be based on an internal sense of responsibility rather than external social pressure [42]. Furthermore, the findings of Adnan et al. [28] demonstrate that farmers’ decision-making is influenced by a variety of factors, including psychological factors, communication channels, innovation attributes, and environmental factors. Consequently, the conclusion that social influence exerts no direct impact on the willingness to adopt is well explained. It is noteworthy that Schwartz proposed that social influence activates personal norms [17], thereby motivating individuals to adopt technology. It is consistent with our research findings.

Financial consequences have a reliably significant relationship to the path of behavioral intention, which means that farmers often weigh the economic benefits of a new technology when considering whether to adopt it. Farmers are more likely to embrace integrated water and fertilizer technology if they believe it can bring cost savings and increased income. A possible explanatory reason is the inherent risk aversion among farmers. Adopting new technologies involves initial costs, such as purchasing equipment or altering established practices. For farmers, particularly smallholders who operate under tight financial constraints, the decision to invest in water–fertilizer integration technology hinges on whether the perceived financial benefits outweigh the risks. If farmers are confident that the technology will reduce input costs and increase profitability, they are more likely to overcome the barriers associated with adoption. Previous studies confirmed this result [13,43], showing that farmers’ perceptions of financial gain significantly drive their willingness to adopt sustainable agricultural practices.

Effort expectancy has a reliable and significant relationship to the path of behavioral intention, which is consistent with previous findings [44], which means that farmers’ perceptions of the ease of using water and fertilizer integration technology directly affect their willingness to adopt the technology. If farmers think the technology is easy to use and does not require much effort, they are more likely to be interested in adopting it. One key factor is that easy-to-use technology can help reduce the cognitive load associated with learning and implementing new systems. For farmers already managing the daily demands of agricultural work, minimizing the effort required to adopt a new system is essential. Effort expectancy reduces the uncertainty and resistance associated with the learning curve, making it more likely that farmers will explore and eventually adopt the new technology. As agricultural technologies often need adaptation, a lower effort expectancy helps alleviate fears of technology failure or time investment, leading to greater adoption rates.

Performance expectancy is found to exhibit a significant and reliable relationship with the path of behavioral intention, which means that farmers have positive expectations about the ability of water and fertilizer integration technology to improve agricultural production efficiency, which directly promotes their willingness to adopt this technology. This finding is consistent with the extended unified theory model proposed by Venkatesh et al. [14,45]. This model emphasizes the importance of users’ perception of technology usefulness to their adoption intentions. According to prior studies [23,29], farmers tend to adopt advanced agricultural technologies that they believe will improve productivity and crop yield, which aligns with our study’s results. In addition, performance expectancy has a moderating effect on EE→BI, suggesting that farmers’ perception of ease of using technology is related to their adoption of water and fertilizer integration technology, a view supported by several previous studies [26,27].

Facilitating conditions have a reliably significant relationship with the path of behavioral intention, indicating that farmers’ perceptions of their availability of the resources and support required to use the integrated water and fertilizer technology, such as finance, technology, and knowledge, significantly influence their willingness to adopt the technology. This finding is in accordance with the conclusions of Venkatesh et al. [14], who highlight the pivotal role of material circumstances and social support in the technology adoption process. Our research explains that the availability of facilitating conditions is especially crucial in rural settings, where access to resources and services can be limited. For farmers, adopting new technology requires financial resources for initial investments and ongoing technical support to ensure proper usage and maintenance. Without access to adequate technical assistance, farmers may feel uncertain about their ability to successfully implement water–fertilizer integration systems, which can create barriers to adoption. Concurrently, prior research indicates that the accessibility of technical service organizations and technical assistance significantly influences the willingness and behavior of agricultural producers with regard to green production practices [13]. This result aligns with the conclusions drawn from our research.

4.2. Practical Application Significance

This study explores in depth the willingness of farmers to adopt water and fertilizer integration technology and its influencing factors, providing valuable insights for manufacturers and policymakers. Moreover, this study elucidates several crucial factors affecting farmers’ acceptance and utilization of water and fertilizer integration technology. These include personal norms, awareness of consequences, responsibility, social impact, performance expectations, effort expectations, enabling conditions, and financial consequences.

First, government-led agricultural extension programs should implement targeted educational campaigns to raise farmers’ awareness of the environmental benefits of water–fertilizer integration technology, particularly its contribution to sustainable farming practices and soil conservation. These programs could collaborate with local cooperatives to disseminate knowledge through workshops or field demonstrations, aiming to align farmers’ values with sustainable practices. These initiatives have the potential to influence farmers’ personal norms, perceived consequences, sense of responsibility, and social influence, which, in turn, can affect their propensity to adopt integrated water and fertilizer technologies, either directly or indirectly. By fostering a stronger connection between farmers’ sense of environmental stewardship and their daily practices, these efforts can lead to a higher rate of technology adoption.

Furthermore, the findings highlight the importance of effort expectancy in the adoption process, indicating that farmers are more likely to adopt the technology if they perceive it as easy to use. Therefore, technology providers should prioritize user-centered design, with manufacturers focusing on simplifying the design and user experience of the technology to ensure that it can be easily integrated into existing agricultural workflows with minimal disruption. Government subsidies for technical training could further support these efforts. In addition, creating user-friendly interfaces, step-by-step guides, and clear operational instructions will be crucial in reducing the learning curve associated with the technology. Agricultural cooperatives can implement technical guidance programs to strengthen the link between farmers’ awareness of environmental management and their daily practices, thereby increasing the rate of technology adoption. This increased awareness could potentially eliminate any technology concerns that potential users may have in the early stages of use, and lower effort expectations would raise performance expectations, which, in turn, would increase long-term adoption rates.

Given that financial consequences play an important role in shaping farmers’ adoption decisions, the study suggests that local agricultural bureaus and financial institutions increase financial incentives, such as subsidies for equipment procurement or low-interest loans, to reduce the economic burden of adopting water–fertilizer integration technology. These incentives would make the initial investment in the technology more feasible, especially for smallholder farmers or those operating on tight profit margins. Farmers will be more inclined to invest in the technology when they gain financial support, knowing that the potential financial risks are minimized. It is important to acknowledge that farmers also have expectations regarding the benefits of applied technologies, which are influenced by performance expectations. Consequently, agribusinesses and technology providers should prioritize not only the cost-effectiveness of the technology but also publish case studies demonstrating return on investment for early adopters, thereby demonstrating its economic feasibility and profitability. This will serve to enhance farmers’ propensity to adopt the technology. Concurrently, manufacturers should proactively communicate the substantial cost savings enabled by integrated water and fertilizer technology, which are achieved by reducing water and fertilizer usage, enhancing resource utilization, and augmenting crop yields. They should also highlight how this technology can improve both the quality and quantity of agricultural production, making it attractive to farmers focused on maximizing their output while minimizing input costs.

Lastly, access to strong technical support and resources is essential to ensure that the adoption of water–fertilizer integration technology is sustainable. Policymakers and industry stakeholders should focus on establishing multi-party collaboration mechanisms, such as promoting partnerships between agricultural cooperatives, agricultural technology companies, and farmers, to integrate resources and lower the threshold for technology adoption. At the same time, they should focus on establishing a strong network that provides continuous technical assistance, maintenance services, and troubleshooting resources. Ensuring that farmers have access to these resources will enhance their confidence in using the technology and address potential operational challenges that could arise, leading to sustained and widespread adoption.

5. Conclusions

Our research provides a complete theoretical framework for exploring the driving mechanism of farmers’ adoption of water and fertilizer integration technology. This study extends the current research field by combining UTAUT, NAM, and financial consequences to assess the constructs influencing behavioral intention. It provides strong support for a better understanding of the mechanisms driving farmers’ adoption of water and fertilizer integration technology. The research results reveal that individual norms are the most critical factors affecting the behavioral intention of farmers to adopt water and fertilizer integration technology. Concurrently, performance expectations, facilitating conditions, financial consequences, and effort expectations are also significant factors influencing farmers’ behavioral intention to adopt water and fertilizer integration technology. Furthermore, our findings indicate that performance expectancy serves as a moderating factor between effort expectancy and behavioral intention. Additionally, we discovered that personal norms are moderating factors between awareness of consequences, responsibility, social influence, and behavioral intention. The theoretical model outlined in this study is an appropriate framework for interpreting the research data. The outcomes of our investigation have tangible applications for encouraging and formulating policies designed to facilitate the implementation of this advanced irrigation technology among agricultural producers, thereby supporting soil quality preservation in Northeast China.

However, it should be noted that the response rate of the effective sample in this study was 31.4%, and this result may be affected by many factors. First, respondents with lower education levels or older people may have cognitive difficulties in understanding and using the 7-point Likert scale. Second, the lack of cooperation of some farmers with the questionnaire survey may also affect the effective response rate of the sample. To improve the quality of the research, in the future, consideration can be given to using a more concise 3-point or 5-point scale for specific populations, supplemented by appropriate incentives to increase participation. In addition, since the data for this study mainly come from the Northeast region, the generalizability of its conclusions may be constrained by regional characteristics. Consequently, subsequent studies will augment the sample scope to encompass additional agroecological regions within China and international cases or undertake comparative analysis of diverse farmer demographics (e.g., large and small producers) to substantiate the universality of research conclusions and enhance theoretical comprehension. Concomitantly, a longitudinal research design will be employed to further assess the timeliness and practical value of the research outcomes.