Method for analyzing influence factors of intercropping mode selected by farmers

Abstract

The invention discloses a method for analyzing influence factors of an intercropping and intercropping mode selected by a peasant household. The method comprises the following steps: step 1, selecting an investigation site according to an investigation purpose; 2, designing a semi-structured questionnaire of a farmer intercropping planting production mode based on a theoretical framework, performing field investigation and issuing the questionnaire for pre-investigation, and modifying the questionnaire according to a pre-investigation condition to realize localization of the questionnaire; 3, selecting a survey site and a farmer, issuing a survey questionnaire, and obtaining an initial data sample according to the content of the survey questionnaire filled by the farmer; step 4, performing data preprocessing on the recovered farmer intercropping planting production mode questionnaire; step 5, on the basis of multiple regression Tobit model analysis, constructing a farmer intercropping planting mode influence factor model; and step 6, importing the processed data into R language statistical software, and carrying out regression analysis on the processed sample data. According to the method, a basis can be provided for making a popularization policy of intercropping in a peasant household level.

Description

Technical Field

[0001] This invention relates to the field of agricultural technology and economics, specifically to an analytical method for analyzing the factors influencing farmers' choice of intercropping patterns. Background Technology

[0002] Intercropping, an ancient agricultural technique, has played a vital role in my country's agricultural production. It increases crop yield per unit area by planting two or more crops sequentially on the same plot, achieving higher and more stable yields with reduced inputs. Furthermore, intercropping helps suppress pests, diseases, and weeds, improving soil fertility. Especially when intercropped with legumes, the legumes' biological nitrogen fixation increases nitrogen input into the system, affecting the quantity and quality of crop residues and further enhancing soil fertility, thus increasing the overall productivity of the planting system. Because intercropping reduces environmental costs, improves soil function, and does not affect crop yield, some have proposed using intercropping as an alternative to traditional intensive planting systems.

[0003] However, with the evolution of modern agriculture towards large-scale and mechanized operations, traditional labor-intensive intercropping and relay cropping models are facing a strong challenge from the intensive production of single crops. Traditional intercropping and relay cropping, such as the corn / wheat intercropping model, can make full use of light, heat, water, and soil resources to increase total yield per unit area. However, its production process is complex, requires high levels of field management, is labor-intensive, and has poor compatibility with large-scale mechanized operations. In contrast, monoculture planting has a highly simplified production process, is easy to mechanize throughout the entire process, has high labor productivity, and obvious economies of scale; however, its system stability is poor, it is highly dependent on chemical fertilizers and pesticides, has low resource utilization efficiency, and faces greater production and environmental risks. The resource efficiency advantages of traditional intercropping and relay cropping contrast sharply with the labor-saving and cost-saving advantages of monoculture, leaving many farmers, especially farms in the stage of expansion, in a dilemma regarding planting system selection, unable to choose the optimal path based on their own resource endowments.

[0004] Intercropping and relay cropping can only contribute to sustainable food production in my country if they are widely adopted. Therefore, there is an urgent need for an analytical method to analyze the factors influencing farmers' choice of intercropping and relay cropping patterns in order to solve the aforementioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide an analytical method for identifying factors influencing farmers' choice of intercropping patterns, accurately identifying key driving and hindering factors. This provides a basis for formulating more targeted policies and promotion strategies, thereby effectively enhancing farmers' willingness and ability to adopt sustainable planting systems and promoting the widespread application and innovative development of intercropping as a more sustainable agronomic practice under modern agricultural conditions.

[0006] The objective of this invention is achieved as follows:

[0007] A method for analyzing the factors influencing farmers' choice of intercropping and relay cropping patterns includes the following steps:

[0008] Step 1: Select the research location according to the research objectives; within the selected county, select representative townships according to the research objectives (such as distance from the county center, level of economic development, etc. for stratification), and use stratified sampling to determine specific townships, villages and households to ensure that the sample can represent the population and contain the key variations required for the research.

[0009] Step 2: Design a semi-structured survey questionnaire on the intercropping production model among farmers based on the theoretical framework, conduct on-site investigations and distribute the survey questionnaires for preliminary surveys, revise the survey questionnaire based on the preliminary survey results, and complete the localization of the questionnaire content.

[0010] Step 3: Select the survey location and farmers, distribute the survey questionnaire designed in Step 2, and obtain the initial data sample according to the survey questionnaire filled out by the farmers;

[0011] Step 4: Using the questionnaire on intercropping planting patterns among farmers collected in Step 3 as the initial data sample, perform data preprocessing to obtain processed sample data.

[0012] Step 5: Based on the analysis of the multiple regression Tobit model, construct a model of the influencing factors of intercropping planting patterns among farmers;

[0013] Step 6: Import the processed sample data from Step 4 into the R language statistical software and perform regression analysis on the processed sample data to determine the socioeconomic factors that influence farmers' choice of intercropping patterns.

[0014] Step 2, the design and localization of the questionnaire, includes the following steps:

[0015] Step 2.1: Based on the comprehensive theoretical framework influencing farmers' choices, the main socioeconomic factors influencing farmers' choices of intercropping patterns are divided into three categories: household assets (including natural assets, physical assets, financial assets, human assets and social assets), household characteristics (including demographic characteristics and risk aversion), and other external factors (including non-agricultural employment opportunities, market prices, and current economic, agricultural, and environmental policies). This theoretical framework forms the basis for analyzing the factors influencing smallholder farmers' choices of intercropping patterns.

[0016] Step 2.2: Go to the designated survey area, select 3-5 farmers, and conduct in-depth interviews based on the survey questionnaire designed according to the theoretical framework. In the interviews, confirm whether the concepts and questions in the questionnaire are understood by the local farmers, and whether there are differences in the independent variables set according to the theoretical framework among the local farmers.

[0017] Step 2.3: Modify the survey questionnaire based on the feedback from the preliminary survey to complete the localization of the survey questionnaire.

[0018] In step 3, the questionnaire on factors influencing smallholder farmers' choice of intercropping patterns includes family characteristics, family assets, and intercropping planting patterns:

[0019] Family characteristics include the proportion of male labor force, dependency ratio (dependents / total family members), and risk awareness;

[0020] Household assets include: 1) Natural assets: the area and number of contracted land plots; 2) Physical assets: the types and quantities of agricultural machinery and the types and quantities of livestock; 3) Financial assets: the proportion of income from crop farming to total household income; 4) Human assets: the labor force / land ratio, the average education level of the labor force, the average age of the labor force, and experience in intercropping.

[0021] Intercropping and relay cropping patterns include crop combinations and planting area.

[0022] Step 4, data preprocessing includes the following steps:

[0023] Step 4.1: Eliminate samples with incomplete or missing data from the questionnaires. The questionnaires obtained after screening are the final valid questionnaires.

[0024] Step 4.2: For the initial data sample in the final valid questionnaire, map the discrete data to numeric type;

[0025] Step 4.3: Based on Step 4.2, for the multiple-choice questions on agricultural machinery ownership and intercropping experience among farmers, the final score for the question is obtained by summing up the scores of all the respondents' options.

[0026] In step 5, the analytical model for the factors influencing farmers' choice of intercropping patterns is constructed based on the Tobit regression equation. The Tobit model is a restricted dependent variable model with distribution truncation characteristics, used for the analysis of continuous selection behavior with censoring.

[0027] Because the area of ​​intercropping accounts for a small proportion of the total planting area in the village, there are a large number of zero values ​​in the dataset; the Tobit model is a suitable method for handling a large number of zeros in the dependent variable, therefore the Tobit regression method is used for analysis; the model is constructed as follows.

[0028]

[0029] IC i Intercropping indicators for farmer i;

[0030] AS jki Let j be the j-th indicator of farmer i's k-th asset;

[0031] HC li For farmer household i, the l-th household characteristic indicator;

[0032] VD mi This is a dummy variable; if farmer i lives in village m, then this variable takes the value 1.

[0033] c0 / c jk / c l / c m For unknown coefficients;

[0034] i This is a disturbance term with standard properties.

[0035] Step 6, the regression analysis process includes the following steps:

[0036] Step 6.1: The Tobit model regression analysis uses the total intercropping area and the planting area of ​​different intercropping patterns as dependent variables, and household assets, household characteristics and village dummy variables as independent variables. The maximum likelihood estimation method is used to estimate the model parameters.

[0037] Step 6.2: Determine the Tobit model fit based on the overall significance test of the Tobit model. If the significance of the likelihood ratio test is less than 0.05, it indicates that the overall specification of the Tobit model is reasonable and the fitting result is ideal.

[0038] Step 6.3: Analyze the direction and degree of influence of each factor on the intercropping area of ​​farmers by analyzing the significance and coefficient signs of the independent variables. *, **, and *** indicate that the independent variable has a significant impact on the potential intercropping area of ​​farmers (i.e., the joint decision from "not adopting" to "adopting" and deciding "how much to adopt") at the significance levels of 10%, 5%, and 1%, respectively.

[0039] Step 6.4: Perform a stability test on the Tobit model using a fractional probability model;

[0040] Step 6.5: Summarize the key factors that significantly influence farmers' intercropping and relay cropping decisions, and propose policy recommendations to promote the intercropping and relay cropping model.

[0041] The beneficial effects of this invention are: 1. This invention, through field investigation and the design and distribution of questionnaires, explores the reasons that influence farmers' choice of intercropping patterns from multiple aspects such as household assets and household characteristics. It also reflects the influencing factors with a scientific mathematical model, confirming that the same factor (such as livestock breeding and labor force age) may have completely opposite effects on technology adoption under different socio-economic and ecological backgrounds. This helps local governments and agricultural extension personnel, including experts and professors, to prescribe the right medicine for regional socio-economic conditions, providing empirical evidence for governments at all levels to formulate differentiated intercropping promotion models and policies, and promoting the widespread application of intercropping patterns.

[0042] 2. This invention uses the Tobit regression model as its basic method to analyze the factors influencing farmers' choice of intercropping patterns. Compared with other quantitative analysis methods, this invention can more effectively analyze the main factors affecting farmers' choice of intercropping patterns from a socio-economic perspective. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0045] This invention analyzes data on intercropping patterns in farmers' households in Gaotai County and Jingyuan County, Gansu Province. The analysis reveals that in Jingyuan County, livestock breeding and intercropping experience positively influence intercropping choices, while the labor-to-land ratio, average education level, and average age negatively affect farmers' choices. In Gaotai County, farmers' intercropping choices are primarily influenced by land size, number of plots, agricultural machinery, and intercropping experience, while livestock breeding negatively impacts these choices. In Jingyuan County, the focus should be on promoting the rice-soybean intercropping pattern among livestock farmers. In Gaotai County, intercropping patterns more suited to mechanized operations should be promoted to facilitate land transfer and large-scale management, creating conditions for the mechanization and large-scale application of intercropping technology. When promoting intercropping in both areas, farmers should be guided to recognize its yield potential and economic benefits through methods such as science and technology promotion, training, demonstrations, and introductions and observations of successful large-scale farmers.

[0046] Combination Figure 1 An analytical method for analyzing the factors influencing farmers' choice of intercropping patterns includes the following steps:

[0047] Step 1: Select the research location based on the research objectives. Within the selected county / city, select representative townships based on the research objectives (e.g., stratified by distance from the county center, level of economic development, etc.). Use stratified sampling to determine specific villages for research, ensuring that the sample represents the population and includes the key variations required for the research.

[0048] Step 2: Based on the theoretical framework, design a semi-structured survey questionnaire on intercropping production patterns among farmers, conduct field investigations, and distribute the questionnaires for preliminary research. Modify the questionnaire based on the preliminary research results. During the preliminary research, go to the designated research area and conduct in-depth interviews with 3-5 farmers. In the interviews, confirm whether the concepts and questions in the questionnaire are understood by local farmers, and whether there are differences in the independent variables set according to the theoretical framework among local farmers, thereby completing the localization of the questionnaire.

[0049] Step 3: Select survey households, distribute survey questionnaires, and obtain initial data samples based on the content of the survey questionnaires completed by the farmers. Among them, the survey questionnaire on factors influencing smallholder farmers' choice of intercropping and relay cropping patterns includes family characteristics, family assets, and intercropping and relay cropping patterns; the selection of survey households should follow the principle of random sampling.

[0050] Step 4 involves preprocessing the collected questionnaires on intercropping and relay cropping patterns among farmers. Specifically, this includes:

[0051] 1) Based on the reasonable range of data values, samples with incomplete or missing data in the questionnaires are removed. The questionnaires obtained after screening are the final valid questionnaires.

[0052] 2) Map the discrete data in the data sample to a numerical type. Table 1 defines and explains the variables involved in the questionnaire on the influencing factors of intercropping patterns among farmers. According to Table 1, intercropping patterns and planting area are the explained variables. Household assets include 1) Natural assets: contracted land area (mu), number of contracted land plots (plots); 2) Physical assets: number of agricultural machines owned (integrated seeding, fertilizing, and mulching machine = 1; rotary tiller = 1; micro tiller = 1; seeder = 1; mulching machine = 1; top dressing machine = 1; harvester = 1; hand tractor = 1; four-wheel tractor = 1; self-propelled tractor = 1; threshing machine = 1; none of the above = 0), livestock breeding status (no ruminants = 0; 10 or less = 1; 20 or less = 2; 30 or less = 3; 40 or less = 0). 4) Number of animals per head or less = 4; 50 animals per head or less = 5; more than 50 animals per head = 6); 3) The proportion of income from crop farming in total household income (10% or less = 1; 11%-30% = 2; 31%-50% = 3; 51%-70% = 4; 71% or more = 5); 4) Human assets: labor force / land ratio, average education level of the labor force (illiterate = 0; primary school = 1; junior high school or secondary vocational school = 2; high school = 3; technical school / higher vocational school = 4; undergraduate = 5; postgraduate = 6), average age of the labor force (0-10 years = 0; 1 = 10-20 years old = 1; 20-30 years old = 2; 30-40 years old = 3; 40-50 years old = 4; 50-60 years old = 5; 60-70 years old = 6; 70 years old and above = 7) and intercropping experience (never used intercropping = 0; planted within 5 years = 1 point; planted 5 to 10 years ago = 1 point; planted 10 years or more ago = 1 point); family characteristics include male labor force ratio (the proportion of men in the family labor force), dependency ratio (the proportion of dependent family members aged 0-14 and over 66 years old to the total number of family members) and risk awareness (are you the kind of person who would never be the first in the village to adopt new technologies? 1 = completely disagree, 2 = disagree, 3 = don't know, 4 = agree, 5 = completely agree).

[0053] 3) For multiple-choice questions regarding agricultural machinery ownership and intercropping experience, the final score for this question is obtained by summing the scores of all the respondents' options. For example, if a respondent selected both a seeder and a hand tractor, the final score is 2 points. A higher score indicates that the respondent owns more agricultural machinery. If the respondent selects both "used within the last 5 years" and "used 5 to 10 years ago," the final score is also 2 points. A higher score indicates that the respondent has more experience in intercropping.

[0054] Table 1. Variable Definitions and Explanations

[0055]

[0056]

[0057] Step 5: Based on the Tobit multiple regression model, construct a model of influencing factors of intercropping and relay cropping patterns among farmers. Since the area of ​​intercropping and relay cropping often accounts for a small percentage of the total planting area in a village, the dataset will contain a large number of zero values. The Tobit model is a suitable method for handling a large number of zero points in the dependent variable; therefore, the Tobit regression method is used for analysis. The model is constructed as follows:

[0058]

[0059] IC i For farmer i, the intercropping index; AS jki For farmer i, the j-th indicator of the k-th asset; HC li VD is the l-th household characteristic indicator for farmer i. mi This is a dummy variable; if farmer i lives in village m, then this variable takes the value 1; c0 / c jk / c l / c m For unknown coefficients; i For disturbance terms with standard properties

[0060] Step 6: Import the processed sample data into the R language statistical software and perform regression analysis on the processed sample data to obtain the influencing factors on whether farmers choose intercropping. Specifically:

[0061] 1) Using the intercropping area among farmers as the dependent variable and household assets, household characteristics, and village dummy variables as independent variables, the maximum likelihood estimation method was used to estimate the model parameters.

[0062] 2) Judge the model fit based on the overall significance test of the model. If the significance of the likelihood ratio test is less than 0.05, it indicates that the overall model specification is reasonable and the fit result is ideal.

[0063] 3) Analyze the direction and degree of influence of each factor on the intercropping area of ​​farmers by the significance and coefficient signs of the independent variables. *, **, and *** indicate that the independent variable has a significant impact on the intercropping area of ​​farmers (i.e., the joint decision from "not adopting" to "adopting" and deciding "how much to adopt") at the significance levels of 10%, 5%, and 1%, respectively.

[0064] 4) Calculate the marginal utility of the Tobit model, which represents the average change in actual observed values ​​(such as the total intercropping area) for every unit change in the explanatory variables;

[0065] 5) Perform a stability test on the Tibit model using a fractional probability model;

[0066] 6) Summarize the key factors that significantly influence farmers' intercropping and relay cropping decisions, and propose policy recommendations to promote intercropping and relay cropping models.

[0067] The following uses data from 24 different administrative villages in Gaotai County and Jingyuan County, Gansu Province, on intercropping and relay cropping patterns as an example to analyze the influencing factors of farmers' choice of intercropping and relay cropping patterns using R language.

[0068] 1. The research location was selected based on the research objectives. In order to understand the adoption of intercropping and relay cropping by smallholder farmers in Gansu Province and the influencing factors, this research selected typical townships in two counties (Jingyuan County of Baiyin City and Gaotai County of Zhangye City) in Gansu Province, which are located in two different agricultural ecological zones. From each township, stratified sampling was carried out according to the distance from the county center (far, medium, and near), and a total of 24 villages were selected for the research.

[0069] 2. Based on the theoretical framework, a semi-structured questionnaire for a survey on intercropping and relay cropping production patterns among farmers was designed. A field survey was conducted and the questionnaires were distributed for a preliminary survey. The questionnaire was then revised based on the preliminary survey results. The questionnaire content included family characteristics, family assets, and intercropping and relay cropping patterns.

[0070] 3. Select survey locations and farmers, distribute survey questionnaires, and obtain initial data samples based on the content of the survey questionnaires completed by the farmers. In each surveyed administrative village, 15 households were randomly selected to conduct the survey, resulting in a total of 360 questionnaires.

[0071] 4. Data preprocessing was performed on the 360 ​​questionnaires collected from farmers regarding intercropping and relay cropping production patterns.

[0072] 4.1 After removing questionnaires that were incomplete or had missing data, a total of 13 questionnaires were deleted, leaving 347 valid questionnaires, with a validity rate of 96.2%.

[0073] 4.2 For the final valid questionnaires, the discrete data in the initial data sample will be mapped to numeric type;

[0074] 4.3 Based on 4.2, the scores for the respondents' agricultural machinery ownership and intercropping experience were summed to obtain the final scores. The processed final data are shown in Tables 2 and 3. Ten sample data points from Gaotai County (Table 2) and ten sample data points from Jingyuan County (Table 3) are listed here as examples.

[0075] Table 2 Data on intercropping patterns among farmers in Gaotai County

[0076]

[0077] Table 3 Data on intercropping patterns among farmers in Jingyuan County

[0078]

[0079] 5. Based on the analysis of the multiple regression Tobit model, a model of influencing factors of intercropping and relay cropping patterns among farmers was constructed; the model construction is as follows:

[0080]

[0081] IC i For farmer i, the intercropping index; AS jki For farmer i, the j-th indicator of the k-th asset; HC li VD is the l-th household characteristic indicator for farmer i. mi This is a dummy variable; if farmer i lives in village m, then this variable takes the value 1; c0 / c jk / c l / c m For unknown coefficients; i This is a disturbance term with standard properties.

[0082] 6. Import the processed sample data from step 4 into the R language statistical software, and perform regression analysis on the processed sample data to determine the socioeconomic factors influencing farmers' choice of intercropping patterns. The model fit information and regression results are shown in Tables 3 and 4.

[0083] Step 6) Import the processed data into R statistical software and perform regression analysis on the processed sample data to obtain the influencing factors on whether farmers choose intercropping. Specifically:

[0084] 6.1 Using the intercropping area among farmers as the dependent variable and household assets, household characteristics, and village dummy variables as independent variables, the maximum likelihood estimation method was used to estimate the model parameters.

[0085] 6.2 As can be seen from the Tobit model fitting results in Table 4, the significance of the likelihood ratio test is less than 0.05, indicating that the overall Tobit model specification is reasonable and the fitting results are ideal.

[0086] Table 4 Tobit Model Fitting Information

[0087]

[0088] 6.3 As shown in the regression results in Table 5, in Gaotai County, the area of ​​contracted land has a positive impact on the area of ​​corn / cumin intercropping. For every 1 mu increase in contracted land area, the area of ​​corn / cumin intercropping increases by 0.17 mu. The number of contracted land plots has a negative impact on the total intercropping area and the corn / cumin intercropping area. For every additional plot, the total intercropping area decreases by 0.26 mu, and the corn / cumin intercropping area decreases by 0.20 mu. The raising of ruminant animals such as cattle and sheep has a negative impact on the total intercropping area. For every 10 additional head of cattle, the total intercropping area decreases by 0.64 mu. Agricultural machinery has a positive impact on the total intercropping area and the area of ​​the new intercropping method, cumin / lettuce. For every additional type of machinery, the total intercropping area increases by 1.14 mu, and the cumin / lettuce intercropping area increases by 0.32 mu. Intercropping experience has a positive impact on the adoption of all intercropping patterns. In general, for every 1 acre increase in farm area, the total intercropping area increases by 4.15 acres.

[0089] As shown in Table 6, the model regression results indicate that in Jingyuan County, the raising of ruminant animals such as cattle and sheep has a positive impact on the total intercropping area. For every 10 additional head of livestock, the total intercropping area increases by 0.41 mu, and the corn / pea intercropping area increases by 0.29 mu. The labor-to-land ratio, average education level, and average age have negative impacts on the total intercropping area and the traditional corn / pea intercropping area. The dependency ratio has a negative impact on the corn / lentil intercropping area. Intercropping experience has a positive impact on the adoption of all intercropping patterns. Overall, for every 1 mu increase in farm area, the total intercropping area increases by 1.74 mu.

[0090] 6.4 According to the regression analysis of the models in Tables 5 and 6, the factors influencing farmers' choice of intercropping patterns include contracted land area, number of plots, livestock, agricultural machinery, labor / land ratio, average education level, average age, dependency ratio, and intercropping experience. Therefore, in Gaotai County, the local government should integrate intercropping with large-scale operations and mechanized production, and remove obstacles to the modern application of intercropping technology through supporting policies such as standardizing land transfer platforms, providing subsidies for large-scale planting, and organizing mechanized production demonstrations. For large-scale producers or cooperatives with large land areas and good machinery, the focus should be on guiding and encouraging them to adhere to the advantageous "corn / cumin" model and boldly pilot new, more efficient models such as "cumin / lettuce," while promoting specialized agricultural machinery to fully leverage their scale and equipment advantages. In Jingyuan County, the local government should focus on strengthening the promotion of the concept of "integrated crop-livestock farming and ecological cycle." Through supporting measures such as science and technology promotion, door-to-door guidance, testimonials from successful farmers, and on-site demonstrations, farmers should fully understand the ecological and economic value of the mutually beneficial relationship between corn-soybean intercropping and family farming. Simultaneously, agricultural extension personnel need to implement precise policies: for farmers with abundant labor and higher education levels, the focus should be on promoting the long-term benefits of intercropping in reducing feed costs and improving soil fertility, along with the promotion of small-scale silage machinery to alleviate labor competition; for farmers with rich farming experience, they should be guided to view the intercropping system as a "family feed workshop," encouraging them to expand the scale of forage crop intercropping to form a closed loop of "cropping for livestock."

[0091] Table 5 Regression Analysis of Factors Influencing Smallholder Farmers' Choice of Intercropping Patterns in Gaotai County

[0092]

[0093]

[0094]

[0095] Table 6 Regression Analysis of Factors Influencing Smallholder Farmers' Choice of Intercropping Patterns in Jingyuan County

[0096]

Claims

1. A method for analyzing the influencing factors of farmers' choice of intercropping and relay cropping patterns, characterized in that, Includes the following steps: Step 1: Select the research location according to the research objectives; within the selected county, select representative townships according to the research objectives, and use stratified sampling to determine specific townships, villages and households to ensure that the sample can represent the population and include the key variations required for the research. Step 2: Design a semi-structured survey questionnaire on the intercropping production model among farmers based on the theoretical framework, conduct on-site investigations and distribute the survey questionnaires for preliminary surveys, revise the survey questionnaire based on the preliminary survey results, and complete the localization of the questionnaire content. Step 3: Select the survey location and farmers, distribute the survey questionnaire designed in Step 2, and obtain the initial data sample according to the survey questionnaire filled out by the farmers; Step 4: Using the questionnaire on intercropping planting patterns among farmers collected in Step 3 as the initial data sample, perform data preprocessing to obtain processed sample data. Step 5: Based on the analysis of the multiple regression Tobit model, construct a model of the influencing factors of intercropping planting patterns among farmers; Step 6: Import the processed sample data from Step 4 into the R language statistical software and perform regression analysis on the processed sample data to determine the socioeconomic factors that influence farmers' choice of intercropping patterns.

2. The method for analyzing the influencing factors of farmers' choice of intercropping patterns according to claim 1, characterized in that, Step 2, the design and localization of the questionnaire, includes the following steps: Step 2.1: Based on the comprehensive theoretical framework that influences farmers' choices, the main socioeconomic factors influencing farmers' choices of intercropping patterns are divided into three categories: household assets, household characteristics, and other external factors. This theoretical framework forms the basis for analyzing the factors influencing smallholder farmers' choices of intercropping patterns. Step 2.2: Go to the designated survey area, select 3-5 farmers, and conduct in-depth interviews based on the survey questionnaire designed according to the theoretical framework. In the interviews, confirm whether the concepts and questions in the questionnaire are understood by the local farmers, and whether there are differences in the independent variables set according to the theoretical framework among the local farmers. Step 2.3: Modify the survey questionnaire based on the feedback from the preliminary survey to complete the localization of the survey questionnaire.

3. The method for analyzing the influencing factors of farmers' choice of intercropping patterns according to claim 1, characterized in that, In step 3, the questionnaire on factors influencing smallholder farmers' choice of intercropping patterns includes family characteristics, family assets, and intercropping planting patterns: Family characteristics include the proportion of male labor force, dependency ratio, and risk awareness; Household assets include: 1) Natural assets: the area and number of contracted land plots; 2) Physical assets: the types and quantities of agricultural machinery and the types and quantities of livestock; 3) Financial assets: the proportion of income from crop farming to total household income; 4) Human assets: the labor force / land ratio, the average education level of the labor force, the average age of the labor force, and experience in intercropping. Intercropping and relay cropping patterns include crop combinations and planting area.

4. The method for analyzing the influencing factors of farmers' choice of intercropping patterns according to claim 1, characterized in that, Step 4, data preprocessing includes the following steps: Step 4.1: Eliminate samples with incomplete or missing data from the questionnaires. The questionnaires obtained after screening are the final valid questionnaires. Step 4.2: For the initial data sample in the final valid questionnaire, map the discrete data to numeric type; Step 4.3: Based on Step 4.2, for the multiple-choice questions on agricultural machinery ownership and intercropping experience among farmers, the final score for the question is obtained by summing up the scores of all the respondents' options.

5. The method for analyzing the influencing factors of farmers' choice of intercropping patterns according to claim 1, characterized in that, In step 5, the model of influencing factors of intercropping planting patterns among farmers is constructed based on the Tobit regression equation. The Tobit model is a restricted dependent variable model with distribution truncation characteristics, which is used for the analysis of continuous selection behavior with censoring. Because the area of ​​intercropping accounts for a small proportion of the total planting area in the village, there are a large number of zero values ​​in the dataset; the Tobit model is a suitable method for handling a large number of zeros in the dependent variable, therefore the Tobit regression method is used for analysis; the model is constructed as follows. 6.IC i Intercropping indicators for farmer i; AS jki Let j be the j-th indicator of farmer i's k-th asset; HC li For farmer household i, the l-th household characteristic indicator; VD mi This is a dummy variable; if farmer i lives in village m, then this variable takes the value 1. c0 / c jk / c l / c m For unknown coefficients; i This is a disturbance term with standard properties.

7. The method for analyzing the influencing factors of farmers' choice of intercropping patterns according to claim 1, characterized in that, Step 6, the regression analysis process includes the following steps: Step 6.1: The Tobit model regression analysis uses the total intercropping area and the planting area of ​​different intercropping patterns as dependent variables, and household assets, household characteristics and village dummy variables as independent variables. The maximum likelihood estimation method is used to estimate the model parameters. Step 6.2: Determine the Tobit model fit based on the overall significance test of the Tobit model. If the significance of the likelihood ratio test is less than 0.05, it indicates that the overall specification of the Tobit model is reasonable and the fitting result is ideal. Step 6.3: Analyze the direction and degree of influence of each factor on the intercropping area of ​​farmers by analyzing the significance and coefficient signs of the independent variables. *, **, and *** indicate that the independent variable has a significant impact on the potential intercropping area of ​​farmers at the significance levels of 10%, 5%, and 1%, respectively. Step 6.4: Perform a stability test on the Tobit model using a fractional probability model; Step 6.5: Summarize the key factors that significantly influence farmers' intercropping and relay cropping decisions, and propose policy recommendations to promote the intercropping and relay cropping model.

Images