Assessing the Effectiveness of Climate-Smart Irrigation Practices in...

[Full title: Assessing the Effectiveness of Climate-Smart Irrigation Practices in Improving Household Income Among Smallholder Maize Farmers in Botswana]

Citation: Jagadeesh, D.; Christian, M.; Letsoalo, S. Assessing the Effectiveness of Climate-Smart Irrigation Practices in Improving Household Income Among Smallholder Maize Farmers in Botswana Sustainability 2024 , 16 , 9693 https://doi.org/10.3390/su 16229693 Academic Editor: Jan Hopmans Received: 24 September 2024 Revised: 31 October 2024 Accepted: 4 November 2024 Published: 7 November 2024 Copyright: © 2024 by the authors Licensee MDPI, Basel, Switzerland This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/) Article Assessing the Effectiveness of Climate-Smart Irrigation Practices in Improving Household Income Among Smallholder Maize Farmers in Botswana Dhanya Jagadeesh 1, * , Mzuyanda Christian 2 and Simon Letsoalo 1 1 Department of Agricultural Economics and Extension, North-West University, Private Bag X 2046, Mmabatho 2735, South Africa; simon.letsoalo@nwu.ac.za 2 School of Agricultural Sciences, University of Mpumalanga, Mbombela 1200, South Africa; mzuyanda.christian@ump.ac.za * Correspondence: dhanyajagadeesh 2022@gmail.com Abstract: The growing impacts of climate change have adversely affected smallholder farmers across the world, leading to low output, decreased incomes, and high levels of food insecurity As a result, farmers have been advised to find alternative ways of dealing with this phenomenon. The low adoption of climate-smart irrigation technology in Botswana warrants an investigation into the factors and the impact of adoption. This study used a semi-structured questionnaire to collect data from 271 smallholder maize farmers, who were selected through a multi-stage sampling approach. Descriptive statistics, probit regression, and propensity score matching technique (PSM) were employed to analyze the data. The results revealed that the majority of the respondents (55%) were male and 62% of farmers were above 50 years. The majority (62%) of the participants had a farm size of less than 5 ha and were heavily reliant on family labour for farm operations. Despite high (66%) awareness of climate-smart irrigation technology, many (52%) farmers did not adopt smart irrigation in Botswana. Age, gender, and access to credit had a statistical and negative influence on adoption. However, level of education and farming experience had a positive influence on adoption. The result of the propensity score matching model indicated that farmers using climate-smart irrigation techniques experienced positive and significant improvement in crop yield compared to dryland farmers. The study recommends that relevant institutions in Botswana should design a strategy that will be tailored to addressing issues of access to credit, facilitate training and education on advanced irrigation methods, and encourage more young farmers to engage in farming activities Keywords: adoption; climate-smart agriculture; climate-smart irrigation; PSM model; ATT; Botswana 1. Introduction Climate change is emerging as an unprecedented threat to global agriculture. In Africa and other developing countries, where agriculture is the main livelihood of most of the population, the impacts are dire because agriculture in these regions heavily relies on rain-fed systems [ 1 – 3 ]. Consequently, shifts in climate patterns pose significant risk to both food production and security [ 4 ]. Studies highlight the detrimental impacts of climate change on agriculture, including reduced crop yields, increased vulnerability to diseases and more frequent extreme weather events [ 4 ]. Addressing these challenges is critical in ensuring food security and agricultural sustainability To mitigate the impacts of climate change, there has been growing emphasis on adopting climate-smart agriculture (CSA) strategies [ 5 ]. Among these, climate-smart irrigation (CSIT) has garnered significant attention worldwide. Climate-smart irrigation encompasses sustainable and adaptive approaches to agricultural water management designed to improve water-use efficiency, improve production and minimize the impact of climate change [ 6 ]. Consequently, significant advancements have been made in irrigation systems, Sustainability 2024 , 16 , 9693. https://doi.org/10.3390/su 16229693 https://www.mdpi.com/journal/sustainability

Sustainability 2024 , 16 , 9693 2 of 14 which researchers are continuously exploring. Climate-smart irrigation strategies empower smallholder farmers to enhance, automate, and optimize traditional agricultural practices, leading to improved agricultural productivity and a more streamlined farming system [ 7 ]. The implementation of smart irrigation systems such as drip irrigation can significantly reduce the total amount of water required to produce field crops. Empirical evidence suggests that automated drip irrigation system can save 26% more water and enhance high crop productivity compared to traditional irrigation methods [ 8 ]. Additionally, Yang et al. [ 9 ] reported that during water shortage, drip irrigation not only saves water but it also maintains crop yields better compared to other types of irrigation methods Despite the potential benefits of climate-smart irrigation technology, its adoption remains low among smallholders in many developing countries [ 10 – 12 ]. Botswana serves as notable example of this trend. In this country, the majority of economically disadvantaged rural communities are directly or indirectly involved in crop production, which is crucial for poverty alleviation, employment, and income generation. However, crop production in Botswana is highly vulnerable to climate-related shocks, with rising temperatures and erratic rainfall patterns threatening crop productivity, household income and food security [ 13 ]. Therefore, policies that boost and improve irrigation management and encourage the adoption of climate-smart irrigation technology should be considered [ 14 ]. Considering this, since independence the government has implemented several policies and support programmes to guide, boost, and improve the production and productivity of various agricultural sub-sectors [ 15 ], which include arable agricultural programmes such as the Integrated Support Programme for Arable Agriculture Development (ISPAAD) and the Arable Land Development Programme (ALDEP). However, these policies did not improve the nation’s food security situation, so the Ministry of Agriculture (MoA) of the Government of Botswana created a new initiative called the Temo Letlotlo Programme and Horticulture Impact Accelerator Subsidy (IAS) fund. Despite the increasing recognition of climate change’s impact on agriculture in Botswana, all these programmes fail to encourage smallholder crop farmers to adopt climate-smart irrigation strategies directly. Hence, this paper investigates the drivers of climate-smart irrigation adoption and its impacts in this context 2. Conceptual Framework on the Adoption of Climate-Smart Irrigation Technology The conceptual framework of this study, as illustrated in Figure 1 below, has four main components. The first components is climate change and its vulnerability. Second, there is the adaptation process to mitigate climate change and vulnerability, which is CSIT; third is the determinant of the CSIT. The last component is the outcome of the adoption decision in terms of crop yield and the income of the farmers Climate change leads to climate vulnerabilities, such as droughts, dry spells, and irregular rainfall, leading to shortage of agricultural water that adversely affects crop yield Without interventions, many rural farmers are at a risk of losing their livelihoods due to the devastating effects of climate change [ 16 ]. Climate-smart irrigation technologies (CSIT) enhance smallholder farmers’ resilience to climate change [ 17 ] as a sustainable and adaptive solution for agricultural water management. The main objective of CSIT is to increase crop yield and enhance farms’ income [ 18 ] by mitigating climate risks and ensuring the availability of sufficient water for agricultural production. Therefore, the farmers who decided to adopt CSIT improved their crop productivity and farm income. In contrast, if farmers prefer either not to adopt CSIT, or to adopt a traditional way of watering, then CSIT may increase the risks and expose their crops to extreme weather, which may have adverse impacts on farm income through reducing per-crop yields or crop loss. Figure 1 shows the relationship between the variables discussed.

Sustainability 2024 , 16 , 9693 3 of 14 Sustainability 2024 , 16 , x FOR PEER REVIEW 3 of 14 Figure 1. Conceptual framework of climate-smart technology adoption by smallholder farmers. Source: Adapted and modi fi ed from Abegunde, Melusi and Obi, 2019 [2]. 3. Methodology 3.1. Description of Study Area Botswana is a landlocked country primarily characterized by the Kalahari Desert, which stretches west of the eastern hardveld and covers 84% of the country. The country is situated between longitudes 20 and 30 degrees east of Greenwich, and latitudes approximately 18 and 27 degrees south of the Equator [19]. Botswana is one of the smallest countries on the African continent and is home to 2.7 million people. It shares borders with South Africa to the South and East, Namibia to the West, and Zimbabwe to the East. According to World Bank [20]. Botswana is one of the most unequal countries in the world. Botswana is divided into 10 districts, each with its own unique characteristics and cultural heritage. The study was conducted only in the fi ve districts in Botswana, namely Central, Kgatleng, South-East, Southern and Kweneng, and these areas were randomly chosen. The central district at 24.0480° S, 26.7747° E latitude and longitude is sandveld, with a mean annual rainfall of 350 mm and high temperatures exceeding 35 °C [21]. The southeast district experiences a blend of average winter temperatures, rainfall, and summer temperatures; with a mean annual rainfall of 450 mm and high temperatures of 32 °C [22]. The Southern district at 25.0559° S, 26.0121° E latitude and longitude experiences highly variable rainfall pa tt erns, averaging about 450 mm annually [23]. During the summer season, the district experiences average daily high temperatures of 32 °C. Kweneng and Kgatleng districts in Botswana have contrasting climates. Deep Kalahari sandy soils primarily cover Kweneng, while Kgatleng has loamy clay soil. The semi-arid climate has annual rainfall ranging from 350 to 600 mm, with mean daily temperatures ranging from 25 °C to 32.6 °C. Kgatleng experiences annual rainfall between 450 and 550 mm, with winter temperatures ranging from 6 °C to 20 °C [24]. As a result of these climate conditions, crops grown in Botswana include maize, wheat, sun fl ower, and groundnuts. In addition, the country is also known for citrus, co tt on, and dairy products. Figure 1. Conceptual framework of climate-smart technology adoption by smallholder farmers Source: Adapted and modified from Abegunde, Melusi and Obi, 2019 [ 2 ]. 3. Methodology 3.1. Description of Study Area Botswana is a landlocked country primarily characterized by the Kalahari Desert, which stretches west of the eastern hardveld and covers 84% of the country. The country is situated between longitudes 20 and 30 degrees east of Greenwich, and latitudes approximately 18 and 27 degrees south of the Equator [ 19 ]. Botswana is one of the smallest countries on the African continent and is home to 2.7 million people. It shares borders with South Africa to the South and East, Namibia to the West, and Zimbabwe to the East. According to World Bank [ 20 ]. Botswana is one of the most unequal countries in the world Botswana is divided into 10 districts, each with its own unique characteristics and cultural heritage. The study was conducted only in the five districts in Botswana, namely Central, Kgatleng, South-East, Southern and Kweneng, and these areas were randomly chosen The central district at 24.0480 ◦ S, 26.7747 ◦ E latitude and longitude is sandveld, with a mean annual rainfall of 350 mm and high temperatures exceeding 35 ◦ C [ 21 ]. The southeast district experiences a blend of average winter temperatures, rainfall, and summer temperatures; with a mean annual rainfall of 450 mm and high temperatures of 32 ◦ C [ 22 ]. The Southern district at 25.0559 ◦ S, 26.0121 ◦ E latitude and longitude experiences highly variable rainfall patterns, averaging about 450 mm annually [ 23 ]. During the summer season, the district experiences average daily high temperatures of 32 ◦ C. Kweneng and Kgatleng districts in Botswana have contrasting climates. Deep Kalahari sandy soils primarily cover Kweneng, while Kgatleng has loamy clay soil. The semi-arid climate has annual rainfall ranging from 350 to 600 mm, with mean daily temperatures ranging from 25 ◦ C to 32.6 ◦ C. Kgatleng experiences annual rainfall between 450 and 550 mm, with winter temperatures ranging from 6 ◦ C to 20 ◦ C [ 24 ]. As a result of these climate conditions, crops grown in Botswana include maize, wheat, sunflower, and groundnuts. In addition, the country is also known for citrus, cotton, and dairy products.

Sustainability 2024 , 16 , 9693 4 of 14 3.2. Sampling Procedure, Sample Size, and Data Collection The study utilized cross-sectional survey data from smallholder maize farmers in Botswana, who were selected using a multi-stage sampling approach. The study used primary and secondary data. Primary data from 271 smallholder maize farmers from five districts in Botswana were collected using a structured questionnaire from December 2023 to January 2024. The survey questionnaire was developed in English and Setswana, and included a broad array of information, encompassing demographic, social-economic, farm, and institutional characteristics that affect the adoption of new technology. Trained research assistants, who were knowledgeable about the rural farming system and proficient about in the local language, administered the questionnaire to prevent misinterpretations or misunderstanding of words or questions. The survey collection tool was administered to 10 respondents prior to the actual data collection to assess its validity and reliability. The unit of analysis for this study was farmers. Additionally, secondary data were obtained from government documents, peer-reviewed journals, and books 3.3. Data Analysis Following the collection of primary data from smallholder farmers, the information was coded, cleaned, and organized into a Microsoft Excel spreadsheet. Subsequently, the coded data were imported from an MS Excel (365 MSO version 2409) spreadsheet to STATA version 13 for analysis. This study employed descriptive statistics, probit regression, and propensity score matching to analyze the data. The details of how the analytical tools were applied is provided below 3.3.1. Descriptive Statistics The study employed descriptive statistics to characterize the demographic, socioeconomic and institutional attributes of smallholder maize farmers in the study area. Tools for descriptive statistics included frequency tables for categorical variables and calculation of average mean, maximum, minimum, and standard deviation for continuous variables 3.3.2. Propensity Score Matching Model (PSM) Propensity score matching (PSM) is a non-experimental method for causal inference that aims to equate the treatment groups by balancing them on confounding factors to make them comparable [ 25 ]. The model is designed to reduce self-selection bias caused by observable features, by pairing a group that participates in adoption activities with a group that does not share similar observable traits [ 26 ]. As a result, it helps to minimize the potential bias that can come from differences in individual characteristics between the treated and untreated groups. Propensity score matching involves linking individuals in the treatment group with individuals in the control group based on their estimated likelihood of being in the treatment group determined by their observable features. As the name suggests, propensity score matching computes the propensity scores for each observation through a first-stage regression analysis, employing either a probit or a logistic regression model. These scores reflect each farmer’s likelihood of adopting CSIT. The propensity score (p-score) generated from the probit in the first-stage regression ranges from 0 to 1. A score closer to 1 indicates a higher likelihood of adopting CSIT practices, while a score closer to 0 suggests a lower likelihood. This estimated propensity score on the treated (ATT), matched treated, and non-treated farming households were calculated using probit regression in the first stage and was used to estimate the average treatment effect in the third stage of PSM. In the second stage, two balanced groups were created based on their estimated propensity score. This study employed various matching methods, including kernel matching, nearest neighbour matching, and radius matching, to achieve balance between the groups.

Sustainability 2024 , 16 , 9693 5 of 14 3.3.3. Probit Regression Model Smallholder households in Botswana often operate under uncertainty, impacting their capacity to produce agricultural products effectively. As a result, the probit model assumes that smallholder farmers make adoption decisions with a goal of maximizing their utility [ 27 ]. The theoretical framework of the probit model is grounded in random utility theory (RUT). In this context, let U 0 represent the expected benefits from nonadoption, and Uj denote the expected benefits from adopting CSIT. According to the RUT, a farmer i chooses to adopt CSIT, if the expected benefit from adoption surpass those from non-adoption [ 28 , 29 ]. The utility functions for adoption and non-adoption are defined as follows: U i 1 = β 1 X i + ε i 1 (for Adoption) (1) U i 0 = β 0 X i + ε i 0 (for Non-Adoption) (2) The decision-making framework can be expressed as follows: y i = 1 i f u i ≥ 0 0 i f u i < 0 (3) The probability of a smallholder farmer i’s decision to adopt CSIT is determined by the utility of that alternative compared to the utility of the current alternative (U i 1 > U i 0) Therefore, for the smallholder farmer i, the probability of adoption is indicated by: P ( y 1 ) = P (U i 1 > U i 0 ) (4) P ( y 1 ) = P ( β 1 X i + ε i 1 > β 0 X i + ε i 0 ) (5) P ( y 1 ) = P ( ε i 1 − ε i 0 < β 1 X i − β 0 X i ) (6) P ( y 1 ) = P ( ε i < β X i ) (7) P (1) = F ( β X i ) (8) where, U i 1 = the utility derived from the technology by a farmer from the CSIT; U i 0 = the absence of utility obtained from CSIT; P (1) = probability of adopting CSIT technology; β 0 to β 1 = estimated parameters; X i = independent variables; F = is the cumulative distribution function of the standard normal distribution ε i = disturbance term 3.3.4. Average Treatment Effect on the Treated (ATT) In the third stage of the propensity score matching model, the average outcomes for the two groups were estimated. The estimated impacts of CSIT interventions are determined by calculating the difference in average outcomes between the group that adopted CSIT practices and the group that did not implement CSIT interventions. This difference is referred to as the PSM estimator of average treatment effect on the treated households (ATT), represented as follows: ATT = E{Y 1 i − Y 0 i |A i = 1} (9) = E E{Y 1 i − Y 0 i |A i = 1, p(Xi)} (10) = E [E{Y 1 i |A i = 1,p(Xi)} − E{Y 0 i|Ai = 0, p(Xi)}|Ai = 1 (11) where y 1 and y 0 represent the outcomes for households that have adopted CSIT practices and the controlled group, respectively. A i = 1 indicates adoption of CSIT, and A i = 0 refers to comparison group does not. Table 1 presents a summary of the explanatory variables used in this study, specifying the measurement type and outlining the theoretical relationship posited between these variables and the dependent variable studied. Data collection for this study was conducted from December 2023 to January 2024.

Sustainability 2024 , 16 , 9693 6 of 14 Table 1. Explanatory variables used in the probit model and their expected outcome Variable Name Type of Measurement Prior Expectations Adoption (Yes/No) Dependent variable Gender of HH Farmer’s sex (female = 0; male = 1) (Dummy) + Marital status Marital status of the farmer (single = 0;married, divorced, widowed = 1) (Dummy) +/ − Age of a farmer Actual number in years (Continuous) + Education Level of education (Categorial) + Labour Availability of labour (Continuous) +/ − Household size Total number of individuals living in a unit (Continuous) + Farming Experience Number of years in farming (Continuous) +/ − Land tenure The type of land ownership (Categorical) + Land size Number of hectares that each household owns (Continuous) +/ − Access to credit If a farmer has access or not (Yes = 1; No = 0) (Dummy) + Frequency of extension visits Measures how often farmers receive visits from extension officers (Continuous) + Main occupation Measured the main occupation of the farmers (Categorical) − Access to electricity If a farmer has access or not (Yes = 1; No = 0) (Dummy) + Districts Different districts in which the farmers come from (Categorical) +/ − +/ − represents the direction of influence (either positive or negative) Source: Author, 2024 4. Results and Discussion 4.1. Demographic and Socio-Economic Profile of the Smallholder Maize Farmers in Botswana The demographic and socio-economic characteristics of farmers were assessed using a structured interview process. Information was gathered from the farming household heads, who provided the required information on behalf of their families. Key characteristics assessed included age, gender, level of education, marital status, household size, and involvement in off-farm activities, among others. As presented in Table 2 , the sample consisted of both those who adopted and those who did not adopt climate-smart irrigation technology. Based on the results from the survey, 48% of the respondents are those who adopted improved irrigation methods as a mechanism to cope with the changes in climatic conditions, while 52% selected for study relied on rainfed systems for farming. More than half of the respondents’ reliance on rainfed agriculture highlights a potential vulnerability. Furthermore, these farmers face increased risks of crop failure due to insufficient rainfall Gender is a very crucial aspect in a household for decision-making, especially in rural areas [ 30 ]. Table 2 indicates that the study participants were dominated by males (55%) rather than females (45%). This finding is a clear indication that agriculture in Botswana is still a male-dominated activity, with females participating in non-agricultural activities The findings of this study are consistent with those of Gayo [ 31 ]. This indicates that farming households are predominantly male due to the physically demanding nature of farming Additionally, in many African cultures, men are typically responsible for farming, while women often handle household chores.

Sustainability 2024 , 16 , 9693 7 of 14 Table 2. Socio-economic and demographic profile of sampled farmers in Botswana Variable Frequency ( n ) Percentage (%) Gender Male 150 55.35 Female 121 44.65 Marital status Single 150 55.35 Married 121 44.65 Adoption status Adopters 131 48.34 Non-adopters 140 51.66 Land tenure Own 182 67.16 Lease 89 32.84 Level of education No education 47 17.34 Primary education 62 22.88 Secondary education 80 29.52 Tertiary education 43 15.87 Others 39 14.39 Access to credit Yes 24 8.86 No 247 91.14 Mean Max Min SD Age 50.981 74 23 12.394 Household size 4.40 10 1 2.31 Farm size 1.811 9 1 1.436 Farming experience 3.394 6 1 1.294 Income 577 35,325 157 4723 Abbreviations: Max, Min, and SD; maximum, minimum, and standard deviation Source: Survey data (2024) Descriptive statistics reveal that only 45% of farmers were married, and 55% were single (Table 2 ). The result also indicated that 31.37% of the farmers were above 60 years, and 30.26% were in the age range from 51 to 60 years; meaning that only 20% of the population were in younger age groups (below 40 years), and around 62% of the respondents were aged above 50 years. This means that many farmers in Botswana are well experienced in farming (see Table 2 ), although older farmers will negatively impact on the supply of family labour force and agricultural production. In addition, older farmers with many years of experience are not always willing to abandon old tactics in favour of new ones, while young farmers are more flexible. Therefore, there is a need to encourage youth participation in agriculture to improve the adoption of climate-smart irrigation practices. This argument is supported by many scholars, such as Ruiz Salvago et al. [ 32 ] and Alrawashdeh et al. [ 33 ] to mention a few. Furthermore, the study demonstrated that a greater proportion (36.53%) of the respondents have 31 to 40 years of farming experience, and only 8.12% have fewer than 5 years of experience. These statistics imply that the respondents have many years of farming experience.

Sustainability 2024 , 16 , 9693 8 of 14 The farmers were also queried about their educational levels; classified as formal versus no formal education. The results presented in Table 2 show that a total of 17.34% of farmers had no formal education, while 22.88% had completed primary education. Interestingly, 29.52% of farmers had secondary education, and 15.87% had acquired tertiary qualifications. The statistics confirm that a greater proportion of the respondents (69.74%) possess a lower level of education. Level of education has a direct influence on the adoption of sustainable irrigation practices, leading to crop productivity. This was confirmed by Das, AB and Sahoo, D [ 34 ]. The findings show that the highest percentage (72%) of farmers belongs to a family with a maximum of five members, and 28% of the respondents had 6 to 10 members in the family. An increase in the number of households may lead to the adoption of more intensive and sustainable agricultural irrigation practices due to the increase in demand for food. Contrary to that, other studies have found that larger farm households are more likely to use labour-intensive agricultural practices due to insufficient labour supply [ 35 ]. However, the reality is that with smart irrigation, no or less human involvement is required [ 36 ]. Off-farm activities are “dummy variable”, showing whether the household head participates in off-farm activities or not. It is expected to impact adoption positively as engaging in off-farm activities can solve liquidity problems [ 37 ]. Approximately 46.86% of the respondents reported participating in off-farm activities, primarily due to low agricultural income and the seasonal nature of jobs in the agricultural sector [ 38 ]. Farmers were engaged in different income-generating activities such as engaging in small businesses selling vegetables, and other types off-farm activities such as “Ipelegeng” which is a short-term employment support scheme in Botswana by the government. The majority of the respondents (83.39%) considered farming as the main source of income, while only 16.61% reported their main source of income as coming from business and other types of employment. Access to and ownership of land is an important factor in farming A total of 67.16% of farmers reported that they cultivated their own farm, while the rest (32.84%) were either renting or leasing the land 4.2. Determinants of Climate-Smart Irrigation Technology To investigate the research question regarding the adoption of climate-smart irrigation technology, a probit regression model was applied, with the results summarized in Table 3 . The probit model indicate that various socio-economic and institutional factors influenced the adoption of climate-smart irrigation technology at different levels. Among the fourteen variables that were included in the model, seven were found to significantly influence the adoption of climate smart-irrigation technology. The model’s performance is evidenced by the value of Pseudo R 2 of 67%, the log-likelihood of − 59.79 and an LR Chi 2 that is significant at 5% level, demonstrating a good fit to the data. Additionally, the model showed no multicollinearity issues because it had a low average variance inflation factor (VIF) of 1.54 The probit model regression results reveal that age was significant at the 1% level and negatively influenced farmers’ decision to adopt irrigation technology. Specifically, an increase of one year in farmers’ age is associated with a 9% decrease in their likelihood of adopting climate-smart irrigation technology. This negative relationship is corroborated by studies from Van der Berg [ 39 ], Lebeta [ 40 ]. Kurgat et al. [ 41 ], and Aryal et al. [ 42 ]. This indicates that older farmers are generally reluctant to adopt new innovations. Additionally, household size negatively impacted the decision to adopt climate-smart irrigation technology, with a significant level of 10%. This suggests that, as a family size increases, the likelihood of adopting climate-smart irrigation technology increases by 20%. Farming experience was found to have negative and significant effect on adopting climate-smart irrigation technology with a significance level of 1%. This finding is supported by research from Serote et al. [ 43 ] and Tanti [ 44 ]. The negative relationship suggests that, as farming experience increases, the likelihood of adopting climate-smart irrigation technology decreases. This trend may be attributed to older farmers, who typically possess more

Sustainability 2024 , 16 , 9693 9 of 14 experience but are less willing to take risks associated with investing in new irrigation technologies due to their shorter planning horizon [ 45 , 46 ]. Land Tenure (land tenure = − 0.726) negatively and significantly impacts the adoption at 1%. The negative coefficient suggests that the probability of adoption decreases by 10% for those who use their own property This may be because those who use leased property are more serious about maximizing productivity within a limited timeframe compared to those who own land. On contrary to that Ma í ra [ 47 ], Wondmu [ 48 ] and Schuck, et al. [ 49 ] reported that an enterprises with the most owned land are most likely to invest in more efficient irrigation systems during severe droughts. Gender significantly influences farmers’ decision to adopt climate-smart irrigation technology at the 5% level, with the probability of males adopting this technology increasing by 44% compared to females. This finding is consistent with that of Abdulkareem and Azahinli [ 50 ], reflecting cultural biases that grant men exclusive rights to make farm decisions regarding short-term and long-term adjustments [ 43 ]. Level of education (Education = 0.075) is significant at 1% and directly impacts farmers’ awareness and understanding of improved irrigation techniques. Those with higher education levels are generally more informed about the benefits and operations of advanced irrigation systems. Conversely, lower educational attainment can lead to misconceptions about these technologies or a lack of confidence in using them effectively. Supporting this view, Schultz [ 51 ], Ali and Byerlee [ 52 ], and Hojo [ 53 ] reinforced the notion that education enhances a farmer’s capacity to respond swiftly and efficiently to technological changes. The availability of labour (Labour = 0.2633) is regarded as an important factor that influences technology adoption among smallholder farmers; this is confirmed by Adeoti [ 54 ], Feder et al. [ 55 ], Quan and Doluschitz [ 56 ], and the availability of labour directly impacts a farmer’s ability to implement and maintain an irrigation system. Access to credit positively influences the likelihood of farmers adopting climate-smart irrigation technology as anticipated. This finding is consistent with Yohannes’ [ 57 ] study, which demonstrated that credit access enhances farmers’ financial capacity to invest in modern farming technologies Table 3. Estimating the determinants of CSIT adoption by maize farmers using probit regression Adoption Coef. Std.Err Z p > |Z| Maritalstatus 0.0535 0.19099 0.28 0.779 Gender 0.4493 ** 0.19538 2.30 0.021 Age − 0.2851 *** 0.09660 − 2.95 0.003 Education 0.0752 *** 0.02053 3.67 0.000 Labour 0.2633 *** 0.07674 3.43 0.001 HHsize − 0.3355 * 0.20117 − 1.67 0.095 Farm_Exp − 0.3105 *** 0.07649 − 4.06 0.000 Land tenure − 0.7267 *** 0.21526 − 3.38 0.001 Access_credit 0.4126 0.34399 1.20 0.230 -cons 1.262 0.57478 2.20 0.028 Number of obs = 269 LR chi 2(9) = 134.21 Prob > chi 2= 0.0000 Pseudo R 2 = 0.3603 Log Likelihood = − 119.12591 * Significant at 10%, ** significant at 5%,and *** significant at 1%. Source: Survey data (2024) 4.3. Impact of Climate-Smart Irrigation Technology (CSIT) The impact of climate-smart irrigation technology (CSIT) was assessed by comparing the differences in “Gross Production” and in “Gross Value” between those who have adopted the CSIT and those who have not. This evaluation was performed using the propensity score matching model and the average treatment effect on the treated (ATT) analysis.

Sustainability 2024 , 16 , 9693 10 of 14 4.3.1. Estimation of Propensity Score The impact of adopting climate-smart irrigation technology on “Gross Production” and “Gross Value” was estimated using the propensity score matching (PSM) method, implemented through the Pscore and PSMatch 2 commands in STATA version 15.1. The propensity score indicates the likelihood of receiving a treatment based on observed factors Researchers can build comparable treatment and control groups by matching people who have similar propensity scores 4.3.2. Matching Effect Analysis The effectiveness of propensity score matching approach relies on two requirements; namely, (i) the balancing test and (ii) the common support test. Rosenbaum [ 58 ] examined the balance of the covariate between the treated and the controlled groups both before and after matching using the standardized bias as a measure. The success of the balance is determined by the percentage reduction in covariates bias and the changes in t-statistical significance levels before and after matching [ 59 ]. The balancing test requires that after matching, there are no systematic differences in covariates between the treated and nontreated groups. As shown in Table 3 , after matching, the standardized bias of all covariates were fewer below 5%, and most control variables were significant before matching. Additionally, the p -values of the t-statistics of all control variables were greater than 5% after matching, indicating that these variables are negligible after matching. This suggest that the matching process effectively reduces disparities in the distribution of explanatory variables between the control group and the treatment groups. Furthermore, the pseudo R 2 decreased after matching, indicating that the differences in controlled variables between the treated and untreated groups have been minimized. A lower psuedo R 2 value suggests that the treatment and control groups became more comparable after matching for the variables and reduced the risk of confounding bias (see Table 4 ). Overall, these results affirm the quality of the matching Table 4. Balance test results of propensity score matching (PSM) Variable Unmatched Matched Mean Treated Control %bas %Reduced |bias| t -Test t P > |t| Marital Status U M 0.4651 0.4214 8.8 7.0 20.2 0.72 0.473 0.4651 0.4302 0.56 0.575 Gender U M 0.6589 0.4642 39.9 1.9 95.2 3.26 0.001 0.6589 0.6495 0.16 0.874 Age U M 2.2636 3.1357 − 81.7 − 0.7 99.2 − 6.70 0.000 2.2636 2.2709 − 0.05 0.959 Education U M 12.07 7.0429 102.0 3.1 97.0 8.31 0.000 12.07 11.917 0.30 0.265 Labour U M 1.845 1.0786 61.4 − 25.0 59.3 5.05 0.000 1.845 2.157 − 1.61 0.109 Hhsize U M 1.2636 1.3429 − 16.6 − 24.8 − 48.5 − 1.36 0.174 1.2636 1.3813 − 2.02 0.044 Farm_Exp U M 2.938 3.8143 − 71.7 − 33.4 53.4 − 5.89 0.000 2.938 3.346 − 2.48 0.014 Landtenure U M 0.51938 0.82143 − 67.6 12.5 81.4 − 5.57 0.000 0.51938 0.46335 0.90 0.370 Access_Credit U M 0.14729 0.03571 39.3 − 6.6 83.3 3.26 0.001 0.4729 0.16594 − 0.41 0.682

Sustainability 2024 , 16 , 9693 11 of 14 Table 4. Cont Variable Unmatched Matched Mean Treated Control %bas %Reduced |bias| t -Test t P > |t| Sample Ps R 2 LR chi 2 p > chi 2 MeanBias MedBias B R %var Unmatched Matched 0.360 0.046 134.21 16.54 0.000 0.056 54.3 12.8 61.4 7.0 165.5 * 51.3 * 0.89 1.40 20 20 Source: Survey data (2024) 4.3.3. An Average Treatment Effect on the Treated (ATT) Analysis Once the propensity scores have been estimated, the next step is to conduct an average treatment effect on the treated (ATT) analysis. This analysis focuses specifically on evaluating the impact of the treatment, which is the adoption of CSIT, on those smallholder maize farmers who actually participated in it. We estimated the impact of CSIT on gross production and gross value using three comparison techniques: Nearest neighbour (NN) matching, radius matching, and kernel matching and Bootstrap standard errors are used to make estimations. The estimations from all matching methods showed that CSIT had a favourable and significant influence on maize production in the treated groups As shown in Table 5 , in the case of gross production, the average treatment effect on treated (ATT) ranges from 650 kg to 2489 kg per hectare of land, whereas the average treatment effect of CSIT on gross value ranges between BWP 1186 to 2944. Therefore, the study found that the adoption of CSIT has a significant relationship with gross production and gross value, which determines the farm income of the farmers Table 5. Average treatment effect of CSIT on gross production and gross value NN Matching Kernel Matching Radius Matching ATT 2489 *** 650 *** 938 *** SE 551 178 198 ATT 1186 *** 2039 *** 2944 *** SE 711 409 623 Source: Survey data (2024). *** significant at 1% 5. Conclusions and Recommendations Climate-smart irrigation technology has attracted substantial interest due to the increase in demand for enhanced water usage efficiency. Smart irrigation may conserve irrigation water and boost production at the farm level, thus contributing to better food security for the needy and household income for smallholder farmers This study mainly investigated the adoption and effectiveness of climate-smart irrigation technology among smallholder farmers in Botswana. The findings from descriptive statistics highlighted the ageing farming population, male domination and educational barriers. Additionally, the findings show that adoption of climate-smart irrigation technology is still low in Botswana. The results of the probit model indicated that age, household size, farm experience, and land tenure negatively and significantly impacted on the adoption decision, meaning that all these factors lower the probability of adopting CSIT among the smallholder maize farmers in Botswana, as expected. In contrast, gender, education, labour force, and access to credit lead to an increase in the likelihood of adoption among the farmers. Other variables such as marriage status, land tenure, frequency of extension visits, access to credit and region were insignificant in the adoption of climate-smart irrigation technology. While these factors were not significant drivers of adoption, they may be crucial in understanding the broader context of Botswana’s farming sector. Further studies may reveal new insights regarding these issues. using propensity score matching model (PSM) and average treatment effect on the treated (ATT), the results show that the adoption of

Sustainability 2024 , 16 , 9693 12 of 14 CSIT is economically and financially viable, pro-poor, and can mitigate the adverse effects of climate change on maize production in Botswana. Given the findings above, the study recommends the following regarding the adoption of climate-smart irrigation technology: - The concerned institutions in Botswana should work to improve access to credit for farmers; - Support should be strengthened to facilitate training and improve the educational attainment of farmers for farmers, to understand better the potential the benefits of adopting climate-smart irrigation technology; - Encourage more young farmers to engage in farming activities and ensure frequent extension services to encourage irrigation farming; - Gender-sensitive programmes that address unequal access to resources such as land should be prioritized Author Contributions: D.J. and M.C. conceived the project idea, co-wrote the paper, investigated, and did document collection and analysis; S.L. and D.J. revised the conceptual framework and analysis; D.J. aligned all referencing; M.C. revised the paper. All authors have read and agreed to the published version of the manuscript Funding: This research work received no external funding Institutional Review Board Statement: The study obtained an ethical clearance from the North West University and the protocol reference number is NWU-01206-23-A 9 Informed Consent Statement: All individuals participating in the study gave informed consent Participants were informed of their right to ask questions about the research. Steps were taken to ensure confidentiality and privacy throughout the process Data Availability Statement: The datasets used or analysed in this study is available from the corresponding author on request Acknowledgments: The authors would like to thank all the farmers who participated in providing the information required to achieve the study objective Conflicts of Interest: The authors declare no conflict of interest References 1 Belloumi, M Investigating the Impact of Climate Change on Agricultural Production in Eastern and Southern African Countries ; AGRODEP Working Paper 0003; African Growth and Development Policy Modeling Consortium, International Food Policy Research Institute (IFPRI): Kigali, Rwanda, 2014 2 Christian, M.; Obi, A.; Agbugba, I.K. Adoption of Irrigation Technology to Combat Household Food Insecurity in the Resource- Constrained Farming Systems of the Eastern Cape Province, South Africa S. Afr. J. Agric. Ext 2019 , 47 , 94–104. [ CrossRef ] 3 Christian, M.; Obi, A. Impact of Irrigation Adoption on Rural Farmers Welfare in Eastern Cape Province of South Africa: A Propensity Score Matching Approach Afr. J. Agric. Res 2018 , 13 , 2641–2650. [ CrossRef ] 4 Ringer, C.; Zhu, T.; Cai, X.; Koo, J.; Wang, D. Climate Change Impacts on Food Security in Sub-Saharan Africa: Insights From Comprehensive Change Scenarios. In IFPRI Discussion Paper 01042 ; International Food Policy Research Institute: Washington, DC, USA, 2010 5 Amadu, F.O.; McNamara, P.E.; Miller, D.C. Understanding the Adoption of Climate-Smart Agriculture: A Farm Level Typology With Empirical Evidence From Southern Malawi World Dev 2020 , 126 , 104692. [ CrossRef ] 6 Sikka, A.K.; Islam, A.; Rao, K.V. Climate-Smart Land and Water Management for Sustainable Agriculture Irrig. Drain 2018 , 67 , 72–81. [ CrossRef ] 7 Manogaran, G.; Alazab, M.; Muhammad, K.; De Albuquerque, V.H.C. Smart Sensing Based Functional Control for Reducing Uncertainties in Agricultural Farm Data Analysis IEEE Sens. J 2021 , 21 , 17469–17478. [ CrossRef ] 8 Bwambale, E.; Abagale, F.; Anornu, G.K. Smart Irrigation for Climate Change Adaptation and Improved Food Security. In Irrigation and Drainage—Recent ; Sultan, M., Ahmad, F., Eds.; IntechOpen: London, UK, 2022. Available online: https://www. intechopen.com/chapters/83182 (accessed on 1 December 2018) 9 Yang, P.; Wu, L.; Cheng, M.; Fan, J.; Li, S.; Wang, H.; Qian, L. Review on Drip Irrigation: Impact on Crop Yield, Quality, and Water Productivity in China Water 2023 , 15 , 733. [ CrossRef ] 10 Brown, B.; Llewellyn, R.; Nuberg, I. Global learnings to inform the local adaptation of conservation agriculture in Eastern and Southern Africa Glob. Food Secur 2018 , 17 , 17213–17220. [ CrossRef ]

Sustainability 2024 , 16 , 9693 13 of 14 11 Sova, C.; Grosjean, G.; Baedeker, T.; Nguyen, T.; Wallner, M.; Nowak, A.; Corner-Dolloff, C.; Girvetz, E.; Laderach, P.; Lizarazo, M Bringing the Concept of Climate-Smart Agriculture to Life: Insights From CSIT Country Profiles Across Africa, Asia, and Latin America ; World Bank Publications—Reports 31064; The World Bank and International Centre for Tropical Agriculture: Washington, DC, USA, 2018; Available online: https://hdl.handle.net/10986/31064 (accessed on 1 December 2018) 12 Westermann, O.; Förch, W.; Thornton, P.; Koerner, J.; Cramer, L.; Campbell, B.M. Scaling Up Agricultural Interventions: Case Studies of Climate-Smart Agriculture Agric. Syst 2018 , 165 , 283–293. [ CrossRef ] 13 FAO Climate Change and Food Security: Risks and Responses ; Food and Agricultural Organization: Rome, Italy, 2015; ISBN 978-92-5- 108998-9. Available online: https://openknowledge.fao.org/server/api/core/bitstreams/a 4 fd 8 ac 5-4582-4 a 66-91 b 0-55 abf 642 a 400/content (accessed on 1 December 2015) 14 Babakholov, S.; Bobojonov, I.; Hasanov, S.; Glauben, T. An Empirical Assessment of the Interactive Impacts of Irrigation and Climate on Farm Productivity in Samarkand Region, Uzbekistan Environ. Chall 2022 , 7 , 100502. [ CrossRef ] 15 Seleka, T.B. The Performance Of Botswana’s Traditional Arable Agriculture: Growth Rates and the Impact of The Accelerated Rainfed Arable Programme (ARAP) Agric. Econ 1999 , 20 , 121–133. [ CrossRef ] 16 FAO; MoFA Investment Framework for Mobilisation of Resources into Climate-Smart Agriculture (Csa) in Ghana , 1 st ed.; FAO: Accra, Ghana, 2018; ISBN 978-92-5-130546-1 17 Serote, B.; Mokgehle, S.; Senyolo, G.; du Plooy, C.; Hlophe-Ginindza, S.; Mpandeli, S.; Nhamo, L.; Araya, H. Exploring the Barriers to the Adoption of Climate-Smart Irrigation Technologies for Sustainable Crop Productivity by Smallholder Farmers: Evidence from South Africa Agriculture 2023 , 13 , 246. [ CrossRef ] 18 FAO Climate-Smart Agriculture Sourcebook: Summary , 2 nd ed.; Food and Agriculture Organization: Rome, Italy, 2017 19 Botswana Tourism. Location. 2021. Available online: https://www.botswanatourism.co.bw/location (accessed on 5 November 2021) 20 WB. World Bank Overview: Development News, Research, Data. 2022. Available online: https://www.worldbank.org/en/ country/botswana/overview (accessed on 27 October 2024) 21 Mcleod, G. Environmental Change at Bobonong in the Central District, Eastern Botswana Botsw. Notes Rec 1992 , 24 , 87–121 Available online: http://www.jstor.org/stable/40979918 (accessed on 11 May 2024) 22 Scotch, K.K. The Settlement Nexus of the Southern Tswana on Hilltops and Valleys in Present Day South East Botswana in the 19 th Century. Master’s Thesis, University of Pretoria, Pretoria, South Africa, 2008 23 Tsheko, R. Rainfall Reliability, Drought and Flood Vulnerability in Botswana Water SA 2003 , 29 , 389–392. [ CrossRef ] 24 Bhalotra, Y.P.R Climate of Botswana, Part ii, Elements of Climate. Republic of Botswana ; Department of Meteorological Services: Gaborone, Botswana, 1987 25 Luvsandorj, Z. A Beginner’s Guide to Propensity Score Matching. Builtin.com. 2023. Available online: https://builtin.com/datascience/propensity-score-matching (accessed on 17 February 2024) 26 Jena, P.R.; Tanti, P.C.; Maharjan, K.L. Determinants of Adoption of Climate Resilient Practices and Their Impact on Yield and Household Income J. Agric. Food Res 2023 , 14 , 100659. [ CrossRef ] 27 Shaw, C.S. Agricultural Technology Adoption in West Africa. Master’s Thesis, Texas A&M University, College Station, TX, USA, 2014 28 Leonardi, G.; Roberto, T. Random Utility Demand Models and Service Location Reg. Sci. Urban Econ 1984 , 14 , 399–431 [ CrossRef ] 29 Leonardil, G The Use of Random-Utility Theory in Building Location-Allocation Models ; IIASA Working Paper; IIASA: Laxenburg, Austria, 1981 30 Matebeni, F. Measuring Rural Household Food Security in the Nkonkobe Local Municipality, Eastern Cape Province of South Africa. Ph.D. Thesis, Stellenbosch University, Stellenbosch, South Africa, 2018 31 Gayo, L. Status, Determinants And Challenges Of Tree Planting In Dodoma Distinct, Tanzania Urban For. Urban Green 2023 , 81 , 127862. [ CrossRef ] 32 Ruiz, S.M.; Phiboon, K.; Faysse, N.; Nguyen, T.P.L. Young People’s Willingness to Farm Under Present and Improved Conditions in Thailand Outlook Agric 2019 , 48 , 282–291. [ CrossRef ] 33 Alrawashdeh, G.S.; Lindgren, S.; Reyes, M.; Pisey, S. Engaging Youth at School to Advance Sustainable Agriculture and Inspire Future Farming: Evidence From Cambodia J. Agric. Educ. Ext 2023 , 29 , 539–556. [ CrossRef ] 34 Das, A.B.; Sahoo, D. Farmers’ Educational Level And Agriculture Productivity: A Study of Tribals of KBK Districts of Odisha Int J. Educ. Econ. Dev 2012 , 3 , 363–374. [ CrossRef ] 35 Ndiritu, S.W.; Kassie, M.; Shiferaw, B. Are There Systematic Gender Differences In The Adoption Of Sustainable Agricultural Intensification Practices? Evidence From Kenya Food Policy 2014 , 49 , 117–127. [ CrossRef ] 36 Obaideen, K.; Yousef, B.A.; AlMallahi, M.N.; Tan, Y.C.; Mahmoud, M.; Jaber, H.; Ramadan, M. An Overview of Smart Irrigation Systems Using IOT Energy Nexus 2022 , 7 , 100124. [ CrossRef ] 37 Berihun, K.; Bihon, K.; Kibrom, A. Adoption and Impact of Agricultural Technologies on Farm Income: Evidence From Southern Tigray, Northern Ethiopia Int. J. Food Agric. Econ 2014 , 2 , 91–106 38 Iqbal, M.A.; Rizwan, M.; Abbas, A.; Makhdum, M.S.A.; Kousar, R.; Nazam, M.; Samie, A.; Nadeem, N. A Quest for Livelihood Sustainability? Patterns, Motives and Determinants of Non-Farm Income Diversification among Agricultural Households in Punjab, Pakistan Sustainability 2021 , 13 , 9084. [ CrossRef ]

Sustainability 2024 , 16 , 9693 14 of 14 39 Van Den Berg, J. Socio-Economic Factors Affecting Adoption of Improved Agricultural Practices by Small-Scale Farmers in South Africa Afr. J. Agric. Res 2013 , 83 , 4490–4500 40 Lebeta, T.H. Participation in and Impact of Small-Scale Irrigation Practice on Household Income: The Case of Abay Chomen District of Oromia National, Regional State, Ethiopia. Master’s Thesis, Haramaya University, Oromia, South Africa, 2017 41 Kurgat, B.K.; Lamanna, C.; Kimaro, A.; Namoi, N.; Manda, L.; Rosenstock, T.S. Adoption of Climate-Smart Agriculture Technologies In Tanzania Front. Sustain. Food Syst 2020 , 4 , 55. [ CrossRef ] 42 Aryal, J.P.; Rahut, D.B.; Maharjan, S.; Erenstein, O. Factors Affecting The Adoption of Multiple Climate-Smart Agricultural Practices in the Indo-Gangetic Plains of India Nat. Resour. Forum 2018 , 42 , 141–158. [ CrossRef ] 43 Serote, B.; Mokgehle, S.; Du Plooy, C.; Mpandeli, S.; Nhamo, L.; Senyolo, G. Factors Influencing the Adoption of Climate-Smart Irrigation Technologies for Sustainable Crop Productivity by Smallholder Farmers in Arid Areas of South Africa Agriculture 2021 , 11 , 1222. [ CrossRef ] 44 Tanti, P.C.; Jena, P.R.; Prakash, J.; Aryal, D.B.R. Role Of Institutional Factors In Climate-Smart Technology Adoption In Agriculture: Evidence from an Eastern Indian state Environ. Chall 2022 , 7 , 100498. [ CrossRef ] 45 Dolisca, F.; Carter, R.; Mcdaniel, J.; Shannon, D.; Jolly, C. Factors Influencing Farmers’ Participation in Forestry Management Programs: A Case Study from Haiti For. Ecol. Manag 2006 , 236 , 324–331. [ CrossRef ] 46 Anley, Y.; Bogale, A.; Haile, G.A. Adoption Decision And Use Intensity of Soil and Water Conservation Measures by Smallholder Subsistence Farmers in Dedo District, Western Ethiopia Land Degrad. Dev 2007 , 18 , 289–302. [ CrossRef ] 47 E Silva, M.F.; Sophie, V.S.; Jan, C.; Steven, V.P. A systematic review identifying the drivers and barriers to the adoption of climate-smart agriculture by smallholder farmers in Africa Front. Environ. Econ 2024 , 3 , 1356335. [ CrossRef ] 48 Golla, W. Determinants of Drip Irrigation Technology Adoption in Degua Woreda, Tigray Region, Ethiopia: Household Level Analysis. Master’s Thesis, Mekelle University, Mekelle, Ethiopia, 2010 49 Schuck, E.; Frasier, W.; Webb, R.; Doctorman, L.; Umberger, W. Adoption of More Technically Efficient System as a Drought Response. International Journal of Water Resources Development Int. J. Water Resour. Dev 2005 , 21 , 651–662. [ CrossRef ] 50 Abdul-Kareem, M.M.; Åžahinli, M.A. Demographic And Socio-Economic Characteristics of Cassava Farmers Influencing Output Levels in the Savannah Zone of Northern Ghana Afr. J. Agric. Res 2018 , 13 , 189–195 51 Schultz, T.W. The Value of the Ability to Deal with Disequilibria J. Econ. Lit 1975 , 13 , 827–846 52 Ali, M.; Derek, B. Economic Efficiency of Small Farmers in a Changing World: A Survey of Recent Evidence J. Int. Dev 1991 , 4 , 1–27. [ CrossRef ] 53 Hojo, M Measuring Education Levels of Farmers: Evidence From Innovation Adoption in Bangladesh ; Discussion Papers in Economics and Business; Graduate School of Economics and Osaka School of International Public Policy (OSIPP), Osaka University: Osaka, Japan, 2024 54 Adeoti, A. Factors Influencing Irrigation Technology Adoption and Its Impact on Household Poverty in Ghana J. Agric. Rural Dev. Trop. Subtrop 2008 , 109 , 51–63 55 Feder, G.; Just, R.; Zilberman, D. Adoption of Agricultural Innovations in Developing Countries: A Survey Econ. Dev. Cult Change 1985 , 33 , 255–298. [ CrossRef ] 56 Quan, X.; Doluschitz, R. Factors Influencing the Adoption of Agricultural Machinery by Chinese Maize Farmers Agriculture 2021 , 11 , 1090. [ CrossRef ] 57 Yohannes, G. Credit Access And Agricultural Technology Adoption Nexus in Ethiopia: A Systematic Review and Meta-Analysis J. Agric. Food Res 2022 , 10 , 100362. [ CrossRef ] 58 Rosenbaum, P.R.; Rubin, D.B. Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score Am. Stat 1985 , 39 , 33–38. [ CrossRef ] 59 Caliendo, M.; Kopeinig, S. Some Practical Guidance for the Implementation of Propensity Score Matching J. Econ. Surv 2008 , 22 , 31–72. [ CrossRef ] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Other Environmental Sciences Concepts:

[back to top]

Discover the significance of concepts within the article: ‘Assessing the Effectiveness of Climate-Smart Irrigation Practices in...’. Further sources in the context of Environmental Sciences might help you critically compare this page with similair documents:

Att, Land tenure, Financial capacity, Adoption, Poverty alleviation, Irrigation system, Gender, Conceptual framework, Water management, Significant influence, Food production, Climate change, Agricultural production, Developing countries, Food Security, Sustainable agriculture, Semi-structured questionnaire, Study area, Crop production, Crop yield, Rural Communities, Drip irrigation, Level of education, Household income, Socio-economic characteristics, Propensity score matching, Propensity score, Household size, Multi-stage sampling approach, Agricultural income, Agricultural technology, Demographic characteristic, Household head, PSM model, Descriptive statistic, Unequal access, Botswana, Smallholder farmer, Government of Botswana, Probit Regression, Water use efficiency, Balancing test, Average Treatment Effect, Climate-Smart Agriculture, Maize production, Access to Credit, Extension service, Probit model, Efficient irrigation systems, Pseudo R 2, Farming experience, Irrigation farming, Water Usage Efficiency, Irrigation Technology, Global Agriculture, Older Farmer, Common support test, Farm experience, Farming population.