Factors Influencing Drip Irrigation Technology Uptake Among Small Scale Farmers in Kenya, Exams of Economics

Download Factors Influencing Drip Irrigation Technology Uptake Among Small Scale Farmers in Kenya and more Exams Economics in PDF only on Docsity! 1 SOCIAL ECONOMIC AND INSTITUTIONAL FACTORS INFLUENCING DRIP IRRIGATION TECHNOLOGY UPTAKE AMONG SMALL SCALE HORTICULTURAL FARMERS IN SUBUKIA SUB COUNTY NAKURU COUNTY KENYA.LATEST UPDATE. 1.1 Overview This chapter covers the background information of the study, problem statement, objectives of the study, research hypotheses, and significance of the study. The last part in this chapter presents the theoretical scope and application of the study. 1.2 Agriculture in Kenya In Kenya, as in many parts of the sub Saharan Africa, agriculture is the mainstay of the livelihood of her citizens. Over 75% of the population of Kenya relies heavily on subsistent farming and 52% of her entire workforce directly practices small-scale farming including pastoral activities (Maina & Maina, 2012). Small-scale/subsistence farming produce accounts for over 75% of the entire agricultural output and over 70% of the marketed agricultural produce in Kenya. Sixty-six (66) percent of the country's manufacturing sector is agro based. These statistics go to show the importance of small-scale farming to Kenya's economy. The same statistics underscore the importance of focusing on this sector with interventions geared towards achieving 2 success (Muthui, 2015). The country enjoys a variety of climates and soils but less than 20% of the land size is considered arable under rain fed condition. The remaining 80% is classified as arid and semi arid lands (ASALs) and experiences perennial water shortage which is a major constraint to agricultural production. Due to population pressure in the high and medium potential areas, people whose livelihoods traditionally depended on subsistence farming have since moved to the ASALs and intensively cultivated them. Cultivation in this fragile ecosystem has not been sustainable without external inputs such as water and nutrients, (Okumu, 2004). According to Okumu, between the mid- 1960s and the mid-1980s, parastatal irrigation agencies were established and 5 farming technologies in citrus orchards (Sevier and Lee, 2004). Lapar and Ehui (2004) identified that small producers who have higher levels of education, higher incomes and access to credit are more likely to adopt dual-purpose forages in Philippine. According to Ogadaet al., (2014) the joint adoption of inorganic and improved maize varieties in Kenya was influenced by the use of manure, access to credit, distance to input markets, secure tenure, education and gender of the household head, cultivated area, drainage of the plots, and expected yields. With respect to irrigation technologies, the literature distinguish mainly two stages of the adoption process. The first was related to the primary adoption in which the producer did not use previously any type of irrigation. The second was related with the change of an irrigation system for another; usually more efficient in the use of water. This second stage of adoption especially addressed in countries or regions with water resource scarcity problems and environmental degradation. Techniques of irrigation vary across crops; common methods included surface irrigation (furrow or flood), overhead sprinklers, trickle irrigation (drip or buried), micro-sprinklers, Moneymaker pumps, and direct can watering (Kinyua, 2009). Research indicates that the kind of irrigation system used depends greatly on the type of farmer, size of farm, and range of operation, as well as the drought tolerance of particular plant standings (Uddin, Bokelmann, & Entsminger, 2014). Adoption also varied according to initial investment costs and was sometimes related to the gender of the producer. Tumboet al., (2011) observed that men usually had more power to make adoption decisions that involved general changes in farm topography; women could not have this power because of lack access to and/or ownership of land. In any case, smallholder farmer support was vital in order to boost 6 adoption of new irrigation technologies. 7 1.2.3 Agricultural Technology Adoption It is estimated that 76% of the population in Kenya live in rural areas, mainly as small-scale farmers, among the many factors that contributed in the growth of agricultural productivity; technology is the most important (Kinyua, 2009). The rate of adoption of a new technology is subject to its profitability and the degree of risk and uncertainty associated with it and is highly influenced by the capital requirement, agricultural policies and socio-economic characteristics of the farmers, RoK (2015). The question of adoption and non adoption is important, however, intensity of adoption is actually the most important criterion in the adoption process. According to Rogers (1995), there are several factors affecting farmer‟s decision to adopt irrigation technologies. Extension creates awareness on existence of irrigation technology, the farmers assess whether the technologies are acceptable to them given their land sizes, crops grown, education, experience, labour availability or demand, expected improvement in fertility, availability of credit facilities, input cost and other factors. According to Singh (2020)the decision to determine whether it is feasible and profitable for farmers to adopt and implement the irrigation technology on their farms may be instantaneous, that is they can adopt immediately in the same year when the technology is introduced or it can take several years depending on socio-economic factors such as education, frequency of extension contact, technology input prices and literacy levels. 1.2.4 Agricultural Irrigation Technology Reducing vulnerability to rainfall failure shocks and variability of production is extremely important for subsistence farmers. Fewer or less severe shocks mean the household is able to maintain proper consumption levels and is less likely to deplete 10 technology which farmers can effectively operate and maintain (Carter, 1994). 11 Kenya has an estimated irrigation potential of 1.3 million ha and a drainage potential of 600,000 ha. Currently, 114,600 ha of irrigation and 30,000 ha of drainage have been developed. Of the available irrigation potential, 540,000 hectares can be developed with the available water resources, while the rest of the area will require water harvesting and storage. The developed irrigation potential can be categorized into the following three main types: smallholder schemes, 49,000 hectares, (43 per cent); public/national schemes, 20,600 hectares, (18 per cent); and, private schemes, 45,000 hectares (Randall, 2012). The remaining potential of over 424,400 hectares and 570,000 hectares of irrigation and drainage calls for increased focus to unleash this potential (ASDS, 2010-2020). 1.3 Problem Statement In the study area, Lari Wendani irrigation scheme farmers use irrigation methods which use lots of water leaving the river with low volume or dry downstream during the dry period. Igwamiti River is the main water source. They are not able to have enough water during this period. Downstream are pastoralists who do not get enough water for their livestock and domestic use due to over abstraction from the river during the period. If the farmers upstream would adopt water efficient irrigation method (drip irrigation) it goes a long way increasing their production per unit area and downstream people would get enough water during dry period. However, in spite of some indications of improvements on the ground, in the study area there are not sufficient studies under-taken assessing the adoption decision of farmers. Given that the main driver for the promotion of drip irrigation in Kenya has been the provision of financial subsidies from the government. The present study focuses primarily on Subukia Sub County the evidence drawn upon and the conclusions drawn from the study expectantly is expected to have general applicability for other regions of the 12 county as well. 15 to be learnt shared across similar initiatives in Kenya. 16 The findings are also crucial in informing current irrigation technology adoption decision making processes among small scale farmers within a particular social context, identification of constraints (socio-economic and institution) that hinder wider adoption of irrigation technology. It will also provide the basis to work on their solutions and improve technology adoption among small scale farmers. 1.7 Scope of the Study The study covered only Subukia Sub County. This was mainly due to limitation of resources in terms of time and funds required to undertake the study on a larger scale. The study targeted all small-scale farmers in the area, and sample size was 277 households. The key issues in this study were social, economic and institutional characteristics of smallholder drip irrigation farmers. Structured questionnaire was used to collect data. To deal with the problem of illiteracy of respondents, there was training and close supervision of enumerators so as to eliminate distortion of information and improve on the quality and reliability of data that were collected. 1.8 Theoretical Scope and Application of the Study The current study is diffusion research and has focused on five areas: (1) the characteristics of an innovation which may influence its adoption; (2) the decision- making process that occurs when individuals consider adopting a new idea, product or practice in the current study drip irrigation; (3) the characteristics of individuals that make them likely to adopt an innovation (drip irrigation); (4) the consequences for individuals and society for adopting an innovation; and (5) communication channels used in the adoption process. 17 CHAPTER TWO: LITERATURE REVIEW 2.1 Overview This literature review summarizes research findings related to the application of drip irrigation for smallholder farming. Focusing in particular on, review of existing knowledge on theories of adoption, adoption of agricultural technologies among smallholder farmers, theoretical framework empirical literature review of models, summary of literature review and the conceptual framework. 2.2 Theoretical Framework There are a number of theories that explain adoption of technologies, the "top-down" and "bottom-up" models of adoption/diffusion provide a directional perspective to the process .Dichotomy theory relates to the scale of innovation efforts by distinguishing between macro-level theories and micro-level theories. Citing Wahid (2007) in Taylor and Todd (1995), the problem of innovation diffusion can be approached from several levels. Some researchers have approached it from macro view or at country level and still other researchers and academic scholars have approached this issue by exploring the factors influencing adoption and usage by individuals. Macro-level theories focus on the institution and systemic change initiatives. Innovation typically involves broad aspects of curriculum and instruction might encompass a wide range of technologies and practices. Micro-level theories, on the other hand, focus on the individual adopters and a specific innovation or product rather than on large-scale change. The following are some of the theories that have been used in explaining technology adoption. 20 the decision stage where there is a drive to seek additional information and a decision is made. Fourthly, is the implementation stage as regular use is attempted more information is sought. The confirmation stage where continued use is justified or rejected based on the evidence of benefits. 2.1.3 Rate of Adoption The rate of adoption is defined as the relative speed with which members of a social system adopt an innovation. Rogers (1995) defines the rate of adoption as the relative speed with which an innovation is adopted by members of a social system. An innovation's rate of adoption in a system, usually measured as the number of members of the system that adopt the innovation in a given time period. It is usually measured by the length of time required for a certain percentage of the members of a social system to adopt an innovation, Sunding and Zilberman (2001). Within the rate of adoption there is a point at which an innovation reaches its critical mass. Critical mass is the time in the adoption curve when enough individuals have adopted an innovation so that the continued adoption of the innovation is self- sustaining. The adoption process is an individual phenomenon describing the series of stages an individual undergoes from first hearing about a product to finally adopting it (Shoemaker et al., 1972). On the other hand, the diffusion process signifies a group of phenomena, which suggests how an innovation spreads among consumers. Overall, the diffusion process essentially encompasses the adoption process of several individuals over time. 2.1.4 Perceived Attributes Rogers (2003) defines several intrinsic characteristics of innovations that influence an individual's decision to adopt or reject an innovation: Relative Advantage: How 21 improved an innovation is over the previous generation; Compatibility: The level of 22 compatibility that an innovation has to be assimilated into an individual's life. Complexity or Simplicity: If the innovation is perceived as complicated or difficult to use, an individual is unlikely to adopt it. Trial ability: How easily an innovation may be experimented. If a user is able to test an innovation, the individual will be more likely to adopt it. Observability is the extent that an innovation is visible to others. An innovation that is more visible will drive communication among the individual's peers and personal networks, and will in turn create more positive or negative reactions. 2.1.5 Diffusion of Innovations Diffusion of innovation is a theory profound by Everett Rogers that seeks to explain how, why, and at what rate new ideas and technology spread. Rogers argues that diffusion is the process by which an innovation is communicated over time among the participants in a social system. For Rogers (2003), adoption is a decision of full use of an innovation as the best course of action available and rejection is a decision not to adopt an innovation. Haider (2004) defines diffusion as the process in which an innovation is communicated thorough certain channels over time among the members of a social system. As expressed in this definition, innovation, communication channels, time, and social system are the four key components of the diffusion of innovations (Sahin, 2006). Therefore, this study was anchored on Diffusion of Innovations Theory. 2.3 Empirical Literature Review This section presents the empirical literature of the study. 2.3.1 Technology Adoption by Smallholder Farmers A farmer‟s decision to adopt or discard a particular technology (such as drip irrigation) is influenced by a complex set of socio-economic, farm-related, and 25 from a source and contained in a bucket or drum, small amounts of water can 26 nevertheless irrigate an enormous area. On the basis that the average plant water requirement is 5 mm/day for land areas with a mean daily temperature of at least 20 °C, Haile et al., (2001) found that one bucket of water could irrigate up to 100 square meters of land, and it would be feasible to expand that area by operating more buckets or drums. One problem associated with the use of buckets to feed water into drip irrigation systems is that the transport of water to elevated reservoirs is complicated and difficult, especially if performed manually; ideally, this is best accomplished using simple mechanical lifting devices that require no fuel or electricity. 2.3.2 Barriers to Adoption of Agricultural Technology According to (Feder et al., 1985:98; William, 2010), the potential barriers to the adoption of a technology such as irrigation are; Inadequate information, education and training. Further, He Cao et al., (2007), underscores lack of access to credit especially when a significant expenditure is required to purchase equipment, inadequate or unreliable supply of equipment, insufficient transportation or infrastructure, Uncertainty and risk associated with information about the technology as other major barriers to adoption of a new technology. Gareth et al., (2007) in a related study finding reinforced that micro parameters are crucially important to understanding agricultural technology adoption and can best be statistically assessed using micro- level data. The same study also supports the findings that heterogeneity of asset quality is critical in the general study of technology adoption. Hochman et al., (1978) in their theoretical research identified three broad classes of factors affecting irrigation technology choices; economic variables, environmental characteristics and institutional variables. One of the major contributions of the past studies of 27 agricultural technology adoption to the general adoption literature is that they emphasize the role of heterogeneity of asset quality in the adoption process. Heterogeneity is a crucial element of the threshold model of diffusion (Davies et al., 2010), but many of the early threshold models focus exclusively on variations in wealth or related factors such as farm size. The agricultural technology problem highlights the importance of differences in physical or geographical conditions in explaining adoption behavior and points out that geographic information must be combined with economic data to predict adoption patterns. Rahman & Hickey (2019) found that social and cultural interactions between members of households and other specialized groups in society also help in understanding local innovation. Complex social and cultural relationships and norms affect the use and ownership of resources, how farming operations are undertaken, how new ideas and technologies are perceived within the family; male-female interactions also influence innovation. At household level gender power relations effect decisions on adoption or failure to adopt, some technologies are easily promoted through women depending on the cost implications or even economic significance. Busingye (2011) in explaining variance in technology adoption in time and space critically analyses training and visit (T&V)and the rigid ranch models as some extension methodologies that share common features; all being top-down, centre outwards, control oriented and intended to standardize and regulate behavior. The study concluded that in practice none could fit or serve local complex, diverse, dynamic and unpredictable conditions. They concluded that farmers do not think of adoption or non-adoption as scientists do, but select elements from the technological 30 than income maximization, in which case, they will not be expected to adopt an 31 income-enhancing technology. As a matter of fact, it is expected that the old that do adopt a technology do so at a slow pace because of their tendency to adapt less swiftly to a new phenomenon (Christensen et al., 2018). Studies in some areas have shown that smallholder farmers do not adopt all components of “packaged” technologies (Nguluuet al., 2006). When exposed to innovations, smallholder farmers only take those components that they perceive as useful and economically within their reach (Nguluuet al., 2012). Those that require a substantial cash outlay are not taken up easily (Ockwellet al., 2010). There are also technologies that do not require high investment costs and still exhibit low adoption. Rukandema (2004) and Muhammad and Parton (2012) have described other socio- economic factors such as farmers‟ innovativeness, age, off-farm income, risk and uncertainty that may result in low technology uptake. Lack of awareness of improved practices is another reason, particularly in remote areas (Nguluu et al., 2014). Studies that have sought to establish the effect of education on adoption in most cases relate it to years of formal schooling (Christensen et al., 2018), Feder and Slade, 2008). Generally, education is thought to create a favorable mental attitude for the acceptance of new practices especially of information-intensive and management- intensive practices (Waller et al., 2008; Caswell et al., 2011). IPM is frequently stated to be a complex technology (Pimentel, 2010; Boahene, Snijders and Folmer, 2009). What is more, adoption literature (Rogers 2003) indicates that technology complexity has a negative effect on adoption. Education is thought to reduce the amount of complexity perceived in a technology thereby increasing a technology‟s adoption. According to Ehler and Bottrell (2000), one of the hindrances to widespread adoption of IPM as an alternative method to 32 chemical control is that it requires greater ecological understanding of the production 35 2.3.4 Availability of Agricultural Extension Services and the Adoption of Agricultural Technology Additional constraints inhibiting increased fertilizer use among smallholders include lack of knowledge and ability to differentiate between various nutrient sources; and lack of understanding of cost-effective methods of soil fertility management (Muzari, Gatsi & Muvhunzi, 2012)). It has also been found that income from off-farm sources is important in the financing of purchased farm inputs (e.g. seeds, fertilizers, labor) (Muzari, Gatsi & Muvhunzi, 2012)). In addition, cash proceeds from crop sales, and income obtained from the sale of livestock and livestock products, also provide cash for the purchase of inputs in crop farming (Muzari, Gatsi & Muvhunzi, 2012). Higher levels of income from each of the above sources will lead to higher rates of adoption of yield-raising technology. Labor bottlenecks, resulting from higher labor requirements that new technologies often introduce, and seasonal peaks that may overlap with other agricultural activities, are important constraints to technology adoption. Acquisition of information about a new technology demystifies it and makes it more available to farmers. Information reduces the uncertainty about a technology‟s performance hence may change individual‟s assessment from purely subjective to objective over time (Caswell et al., 2011). Exposure to information about new technologies as such significantly affects farmers‟ choices about it. Feder and Slade (2004) indicate how, provided a particular technology, increased information induces its adoption. However, in the case where experience within the general population about a specific technology is limited, more information induces negative attitudes towards its adoption, probably because more information exposes an even bigger 36 information vacuum hence increasing the risk associated with it. 37 A good example is the adoption of recombinant bovine Somatotropin Technology in dairy production (Mc Guirk, Preston and Jones, 1992; Klotz, Saha and Butler, 1995). Information is acquired through informal sources like the media, extension personnel, visits, meetings, and farm organizations and through formal education. It is important that this information be reliable, consistent and accurate. Thus, the right mix of information properties for a particular technology is needed for effectiveness in its impact on adoption. Good extension programs and contacts with producers are a key aspect in technology dissemination and adoption. A recent publication stated that “a new technology is only as good as the mechanism of its dissemination” to farmers (IFPRI, 2005). Most studies analyzing this variable in the context of agricultural technology show its strong positive influence on adoption. In fact, Yaronet al., (2012) show that its influence can counter balance the negative effect of lack of years of formal education in the overall decision to adopt some technologies. A wide range of economic, social, physical, technical and institutional aspects of farming influence the adoption of agricultural production technologies. In a review of adoption of agro forestry technologies, Pattanayaket al., (2002) established that there were five basic categories of determinants of adoption. These were farmer preferences, resource endowments, market incentives, biophysical factors and risk and uncertainty. Farmer preferences include risk tolerance, conservation attitude and intra-household homogeneity. But since these are difficult to model, proxies such as age, gender, education and social status are used instead. Resource endowments include assets which a household has such as land, labour, livestock and earnings. Several authors identified a positive impact of the educational level of the household head on irrigation adoption (Barseet al., 2010; Vaezi and Daran, 2012; Shahzadi, 2013; Singh et al., 2015). Barseet al., (2010) found that the high level of education of 40 between scientific orientation and adoption level. Gupta et al., (2010) revealed that there was significant improvement in yield, quality, water and fertilizer use 41 efficiencies of capsicum under drip irrigation and fertigation. However, the combined effect of drip irrigation and fertigation was found superior than their individual effects. Kumar (2012) found that drip method of irrigation is found to have a significant impact on resources saving, cost of cultivation, yield of crops and farm profitability. The adoption of drip irrigation is significantly influenced by experience, farm size, proportion of wider spaced crops and participation in non-farm income activities. The policies should focus on promotion of drip irrigation in those regions where scarcity of water and labour is severe and where shift towards wider-spaced crops is taking place. Bahuguna (1996) stated that by drip system of irrigation, 95 percent of the irrigation water can be used efficiently and the production may be increased by 30-50 percent. The above facts show the importance of drip irrigation technology. A comprehensive adoption study by Feder et al., (2005) and Feder and Umali (2003) showed that farm size, risk, human capital, and labour availability, access to credit and land tenure systems were important factors. However, studies by Besely and Case (2012b) and Foster and Rosenzweig (2005) using panel data showed that learning from own experience and neighbor‟s experiences were important factors in determining adoption. Adoption studies in Europe, Asia and Africa have identified farm and technology specified factors, institutional, policy variables and environmental factors to explain the patterns and level of adoption. For example, Oladele (2005) highlights that some studies have shown strong and positive correlation between farming size and adoption 42 while others have shown a positive and significant association between age, farming experience, training received, social-economic status, economic motivation, 45 adoption, there is no clear distinguishing feature between variables in each category. Categorization is done to suit the current technology being investigated, the location, and the researcher‟s preference, or even to suit client needs (Bonabana-Wabbi, 2002). For instance, the level of education of a farmer has been classified as a human capital by some researchers while others classifies it as a household specific factor. According to Just and Zilberman (1983), there are various factors that influence the adoption of any technology. Technology may require some costs that are associated with new equipment‟s and investments, learning time, locating and developing markets and training labour. This view is supported by Bonabana- Wabbi (2002) adding that for farmers to adopt a technology, they must see an advantage or expect to obtain greater utility in adopting it. From the study, it is argued that without a significant difference in outcomes between two options and in the returns from alternative and conventional practices, it is less likely that farmers, especially smallholder farmers will adopt a new practice. Since adoption of a practice is guided by the utility expected from it, the effort put into adopting is reflective of its anticipated utility. Moreover, there is no standard way of classifying factors influencing adoption and classification cannot be uniform (Bonabana-Wabbi, 2002). This is because the factors influencing adoption may be a complex set of interactions and factors like the institution (administration), the potential/targeted adopter (the farmer) or the general setting in which the technology is introduced act either as barriers or enhancers of adoption. Several factors have been found to influence adoption. A study by Bonabana-Wabbi (2002) used multivariate Logit analysis to identify factors and their relative importance in explaining adoption of eight 46 Integrated Pest Management (IPM) agricultural technologies in Kumi District, Eastern Uganda. The study results indicated that size of household labour force had negative influence on IPM adoption but positive influence on growing improved IPM. For the gender variable, the study indicated that males were more likely to adopt IPM than females while experience positively influenced timely planting of cowpeas. The study argued that, farmers with accumulated farming experience may have acquired encouraging returns from the practice and thus continue with it anticipating continued benefits. Farm size and level of education did not show any significance with IPM adoption. Although the study analyzed quite a number of factors, access to market, infrastructure and land tenure were left out in the study. Nchinda, Hadley, Villano& Morales, 2020) used Tobit regression method as the main analytical tool in a study of factors influencing adoption and intensity of yam seedling technology in Cameroon. Farm size was not a significant determinant of adoption in their study. However, hired labour and membership to farmers‟ organizations positively and significantly influenced the adoption and intensity of yam minisett technology (is a way to obtain healthy planting materials in commercial quantities) in areas covered. They also showed that age had significant influence with farmers less than forty-one years of age being found to positively influence yam adoption and its intensity. Another study by Adeogun et al., (2009), aimed at estimating and explaining the parameters of the adoption process of Hybrid Clarias“Heteroclarias” by fish farmers in Lagos State Nigeria, showed age, farming experience and farm size to be statistically significant in explaining hybrid catfish adoption. However, their Logit 47 model results showed that education, contact with extension agents, access to seed 50 incorporate more than one prediction variable unlike Logit models. For this reason, probit models are widely used in limited dependent variable models. Shekya and 51 Flinn (1958) have recommended probit for functional with limited dependent variables that are continuous between 0 and 1. The model was specified by Theil, 1979) and Maddala, 1983) as shown in equation 2.1; In(E[Y | X i ])    Xi   i...................................................................2.1 Where  are estimated coefficients and X i are independent variables such as farmer and farm‟s characteristics  i are stochastic error terms. The probit model uses a logistic curve to transform binary responses into probabilities within the 0-1 interval. This postulates that the probability of a farmer (P) adopting drip irrigation technology is a function of some characteristics X i . These characteristics may be social, economic or institutional. The model is used to examine relationship between adoption and determinants of adoption which involve a mixed set of qualitative and quantitative analysis. Qualitative models have been extensively used in adoption studies although they have been criticized for their inability to account for partial adoption (Feder et al., 1985). Alternative specifications of qualitative choice models include the linear probability models and logit models. These are the two most frequently used applications in explaining the socio-economic phenomena, especially for analyzing relationship between dependent discrete variables (adoption) and explanatory variables (Polson et al., 1992). Both the probit and logit models yield similar parameter estimates and it‟s difficult to distinguish them statistically. Of the two models, the bivariate probit model is easier to estimate and simpler to 52 interpret (Abebaw and Belay, 2001). Quite a large number of studies have investigated the influence of various socio-economic, cultural and political factors on 55 Independent Variables Dependent Variable Figure 2. 1: Conceptual Framework Source: Author’s own Conceptualization, 2017 Institutional Factors Access to Credit Access to Extension services Frequency of Extension Visits Land Tenure Source of Extension knowledge Economic Factors Farm income Farm size Off-Farm Income Drip irrigation uptake Social Factors Age Gender Education Level Farmer experience Family size 56 CHAPTER THREE: RESEARCH METHODOLOGY 3.1Overview The chapter presents the study area, research design, data type and sources, target population, sample size, sampling procedure, data collection instrument, data analysis and model specification. 3.2Study Area The study was done in Nakuru County that comprises of eleven Sub counties namely: Subukia, Rongai, Molo, Njoro, Bahati, Naivasha, Kuresoi North, Kuresoi South, Gilgil, Nakuru East and Nakuru West. However, for this study Subukia Sub County was purposively selected. This is because there have been tremendous efforts to promote drip irrigation to increase food security in this region. 3.2.1 Subukia Sub County Subukia Sub county was curved from the Sub counties of Rongai and Nakuru North (now Bahati) is one of the eleven Sub counties in the Nakuru county. It lies within the Great Rift Valley and borders three other Sub counties namely, Rongai to the west, Laikipia to the north, Nakuru North to the south and south west. The Sub County covers an area of 390.8 Km2. The Sub County has three wards namely Subukia, Kabazi, and Waseges Ward. It has a total of 31,600 ha of agricultural land and 23,900 Ha is cultivated. The Sub County has a projected population of about 120,000 persons. There are 23,600 households and 21,500 farm families, (MoALF, 2016). The Sub County receives a bimodal rainfall. The long rains normally start from mid-March to August; the short rains are received in the months of September –December. The annual rainfall ranges from 700mm- 1400mm.Main Agricultural Economic Activities are: 57 Farming which includes, Maize- beans intercrop, Vegetables, tea, coffee, Livestock 60 3.4 : Data Types and Sources 3.4.1 Primary Data and Sources Primary data was obtained from the households‟ head including information on age (in years), gender (male or female), education level, farmer experience (in years), family size, crop income (in Kshs), farm size, off-farm income (in KSh.), land tenure (either freehold, communal or leased) access to credit (in Kshs) and access to extension services and frequency of extension visits and source of extension knowledge in Subukia Sub County. 3.5.2 Secondary Data and Sources Secondary data was used where historical information was required. Secondary information was obtained from the Ministry of Agriculture, Research Institutions Kenya National Bureau of Statistics (KNBS) publications journals, theses, and other government institutions. Government publications such as national and County, Sub County development plans, and annual reports among others were also used. 61 3.6 Target Population The target populations for the study were the smallholder horticultural farmers in Lari Wendani irrigation scheme, Subukia Sub County, comprising of both adopters and non-adopters. The number of households was 277, (MoALF, 2016).Since the total numbers of farmers in the scheme were 277, a census study was used. Therefore, the total number of respondents in this study was 277 farmers. There are 7 schemes in Lari Wendani irrigation scheme. 3.7RespondentsDistribution per Scheme Census was used and as such, no sampling procedure was required. Identifying the 1st respondent used the farmers register complied by the scheme management. Table 3. 1: Distribution of Respondents Per Scheme scheme 1 scheme 2 scheme 3 scheme 4 scheme 5 scheme 6 scheme 7 Total 29 28 36 27 52 39 66 277 Source: Author’s Own Computation (2017) 3.8 Data Collection Instrument A structured questionnaire was used to collect data from the surveyed farmers. Quantitative data was collected from the study area; pretest of the data collection tool was done in Arash location to establish reliability of the research instrument. 3.9 Data Analysis Data was analyzed using a combination of descriptive and inferential statistics. 3.9.1 Descriptive Statistics Descriptive statistics concerns the summarization of data (Saunders, Milyavskaya, Etz, Randles, Inzlicht & Vazire, 2018). Descriptive statistics were used to summarize and describe data from the surveyed households. This usually entails calculating 62 averages, standard deviation, minimum and maximum values (Saunders et al., 2018). The technique was useful in analyzing all the quantitative data. In this case, cross tabulation, frequency tables and descriptive statistics such as mean, standard deviation of study variables were calculated. Minimum and maximum values of each variable were identified. Descriptive statistics are useful as they represent pictorial view of the data. 3.9.2 Inferential Statistics Inferential statistics does more than descriptive statistics. There is an inference associated with the data set, a conclusion drawn about the population from which the data originated (Saunders et al., 2018). Inferential statistics such as correlation and regression analysis were used as to ensure efficient inferences are made to the larger population. Inferential statistics was used to infer sample results to the general population. In this study logit model was used. 3.10 Choice of Econometric Model There is no articulated model that provides a conceptual framework to determine the factors that influence drip irrigation adoption decision. However, studies have been carried out to relate farmers' adoption of new technologies to various socio economic factors (Anderson & Feder, 2004). Based on these studies, a conceptual model was developed to explain the effects of socio- economic factors on the adoption of drip irrigation technology. In adoption studies, responses to a question such as whether farmers adopt a given technology could be yes or no, is a typical case of dichotomous variable. The model that is suggested for such binary dependent variable is the linear probability model. 65 Subtracting 1 yield: ezi  1 1 1  p / p ...................................................................3.2 i i ip i 66 However since ez 1/ ez then e zi  p /1  p  so that by taking the natural logarithm on both sides of the equation we obtain zi  log( pi /(`1 pi )) or from equation (1) presented above, we have: …………………………………………………….….. 3.3 Where = the log of the odds that a certain decision will be made. = the constant of the equation = the coefficient of the predictor variables 3.10 Description and Measurement of Variables Table 3. 2: Description, Measurement and Expected Signs of Variables Variable Description Measurement Expected Sign Y1 Adoption 0- Not adopted 1-Adopted X1 Age of household head Number of years -/+ X2 Gender Male or female 0-Female 1 -Male -/+ X3 Level of education Subdivisions of formal learning, typically covering early childhood education, primary education, secondary education and tertiary education + X4 Farmer experience Number of years in farming . + X5 Farm income Household income from drip irrigation in Kenyan shillings + X6 Farm size Total land size of the household in hectares + X7 Land tenure Type of land ownership 1-Owned; 2-Communal 3-Rented/leased + X8 Access to credit Access to credit financial services 1-Yes;0-No + X9 Extension services Access to extension service 1-Yes; 0-No + X10 Frequency of extension visits Number of times visited per year + X11 Family size Number of members in Household +/- X12 Off-farm income Household income from other sources + Source: Author’s own Computation, 2017 67 For this study description and measurement of the variables has been illustrated in the Table 3.2 together with the expected sign. 3.11 Diagnostic Tests Logit regression analysis was used to test for heteroscedasticity to ensure that there was constant variance. 3.11.1 Heteroscedasticity Test The variance of linear regression model should be constant for the linear regression model to hold. If the error terms do not have the constant variance, they are heteroscedastic. Breusch-Pagan and Cook-Weisberg test was used to test for heteroscedasticity. It has the null hypothesis : constant variance. The Lagrange Multiplier test yields the following test statistic; …………………………………………………. 3.4 3.11.2 Test for Multicollinearity Variance inflation factor (VIF) was applied to check for Multicollinearity in logit regression analysis. VIF measures how a variance has increased the estimate of the slope High VIFs reflects an increase in the variances of estimated regression coefficients due to collinearity among predictor variables. VIF test for Multicollinearity is denoted as; ……………………………………………………………………....… 3.5 70 innovations. Dey (1981) and Ongiyo (2019) noted that male farmers are likely to have 71 more access to inputs, capital and information through farmers‟ networks and contacts with extension agents than female farmers. Table 4. 2: Gender Distribution of the Surveyed households Gender Frequency Percent Cum. Per cent Female (Coded 0) 69 24.91 24.91 Male (Coded 1) 208 75.09 100 Total 277 100 100 Source: Author’s Survey Data, 2017 4.2.3 Household Sizes of the Surveyed households Table 4.3 gives the household sizes of the surveyed households. Table 4.3 shows that the average number of persons living in one household among the surveyed farmers was 7 with standard deviation of 2.5. The minimum number in the surveyed households was one (1) person while the maximum were 15 people. This was an indication that there was enough provision of labor for drip irrigation because most of the households in developing countries use family members as source of labour for farming activities. Table4. 3: Household Sizes of the Surveyed households Variable Obs. Mean Std. Dev Minimum Maximum Household Size 277 6.7 2.5 1 15 Source: Authors Survey Data(2017) 4.2.4 Education Level of the Surveyed Households The results of education level of the surveyed households are depicted in Table 4.4. Result showed that majority of the surveyed household had primary education (137 that represented 49.46 per cent. This was followed closely by secondary level of 72 education (121 farmers that represented 43.68 per cent. A paltry 3 (1.08%) farmers 75 4.3 Land Size of the Surveyed Households Table 4.5 presents the land size of the surveyed households. Result showed that the average land holding was 2.9 acres with standard deviation of 1.9. The minimum holding was 0.25 acres while the maximum was 23.50 acres. This was an indication that majority of the surveyed households were small-scale farmers. Misra (1990) and Kannan (2002) in their respective studies in India reported that majority of the respondents had medium size (2-4 ha) of land holdings. Similarly, Ongiyo (2019) reported that most of the respondents were small-scale farmers in his study on adoption of dairy technologies in North Rift Kenya. Table 4. 5: Land Size Distribution of the Surveyed Households Variable Obs. Mean Std. Dev Minimum Maximum Land Size 277 2.9752 1.936 0.25 23.50 Source: Authors Survey Data, 2017 4.4 Land Ownership Pattern of the Surveyed Households Table 4.6 presents the results of land ownership pattern of the surveyed households. Results indicate that majority of the sampled households owned their lands (265 farmers that represented 95.67 per cent). Twelve farmers (4.33 per cent) leased their lands while none owned land under communal system. Table 4. 6: Land Ownership Pattern of the Surveyed Households Land Tenure System Frequency Percent Cum. Per cent Owned 265 95.67 95.67 Leased 12 4.33 100 Communal 0 0 0 Total 277 100 100 76 Source: Author’s Survey Data, 2017 77 4.5 Adoption Level of Drip Irrigation Technology The study sought to establish the number of farmers who used drip irrigation and the results are presented in Table 4.7. Results showed that adoption level was low because 228 farmers (82.31 per cent) did not use drip irrigation. This was an indication that the technology was expensive or farmers were not aware of the benefits of using the innovation or attitude of the farmers towards the technology was negative. Table 4. 7: Adoption of Drip Irrigation by the Surveyed Households Level of adoption Frequency Percent Cum. Per cent Non-Adopters 228 82.31 82.31 Adopters 49 17.69 100 Total 277 100 100 Source: Author’s Survey Data, 2017 4.6 Type of Drip Irrigation Technology The study also sought to establish type of irrigation technology used by surveyed households. There were three technologies; furrow, sprinkler and others. The results are reported in Table 4.8. Results in Table 4.8 showed that majority of the surveyed households 134 (48.38 per cent) used sprinkler technology while 83 (29.96 per cent) used other techniques like basins. Few of them 60 (21.66 per cent) used furrow irrigation. Farmers cited that furrow technology was expensive and results in loss of water through evaporation. This was in line with Irrigation Show (2009) who stated that furrow irrigation has high precipitation and water loss. 80 Table 4. 12: Access to Extension Services by Surveyed Households Access to Extension Services Frequency Percent Cum. Per cent Accessed 255 92.06 92.06 Did not Access 22 7.94 100 Total 277 100 100 Source: Author’s Survey Data, 2017 4.1.13 Sources of Extension Knowledge by Surveyed Households The study sought to establish sources of the extension knowledge by the surveyed households. Results in Table 4.13 showed that majority of the surveyed households; 156 farmers (56.32 per cent) accessed extension services extension personnel (MoALF). Ninety-five farmers accessed extension knowledge from mass media while 25 of them accessed extension knowledge from other sources. This was an indication that extension personnel from MoALF were available in the study area. Table 4. 13: Sources of Extension Knowledge by the Surveyed Households Sources of Extension Knowledge Frequency Percent Cum. Per cent Mass Media 25 9.03 9.03 Extension Personnel –MoALF 156 56.32 65.34 Research Institutes 95 34.30 99.64 Others 1 0.36 100 Total 277 100 100 Source: Author’s Survey Data, 2017 4.1.14 Frequency of Extension Visits of Surveyed Households The study sought to establish the frequency of extension visits by surveyed households. The results are reported in Table 4.14. Results in Table 4.14 showed that majority (121) of the surveyed households did not receive extension services while 4 of them were visited for 10 times. This was an indication that extension personnel 81 from MoALF were available in the study area but their contact was very minimal with the farmers. This is because in Table 4.13 farmers accessed extension services but Table 4.14 show that they did not get extension (farming) knowledge. Table 4. 14: Frequency of Extension Visits of Surveyed Households Frequency of Extension Visits Frequency Percent Cum. Per cent Zero Times 121 43.68 43.68 One Time 56 20.21 63.89 Two Times 51 18.41 82.30 Three Times 16 5.77 88.07 Four Times 10 3.61 91.68 Five Times 4 1.44 93.12 Six Times 8 2.88 96.00 Seven Times 1 0.36 96.36 Eight Times 6 2.16 98.52 Ten Times 4 1.44 100.00 Total 277 100.00 100.00 Source: Author’s Survey, 2017 4.1.15 Access to Credit by the Surveyed Households The study was to establish whether the surveyed households accessed credit facilities. The results of access to credit are presented in Table 4.15. Results indicated that majority (222 farmers representing 80.14 per cent)of the surveyed households did not access credit facilities. This may imply that credit facilities through banks and other financial institutions were rare facilities in the study area. Also it implies that such farmers lack collaterals to banks as security to get loan facilities. Table 4. 15: Access to Credit Facilities the Surveyed Households Access to Credit Facilities Frequency Percent Cum. Per cent Accessed Credit 222 80.14 80.14 Did not Access Credit 55 19.86 100 82 Total 277 100 100 Source: Author’s Survey Data, 2017 85 Constant -0.2223 0.1244 -1.79 0.0740 Note: ** indicates the variables that were significant 5% level of significance Source: Author’s Survey Data, 2017 86 The logit regression results on use of drip irrigation reported in Table 4.17. With 0 - 4 iterations, Likelihood ratio  2 (13) was 192.13, Pseudo R2 value of 0.4180 and Log likelihood of -2203.4075. The results showed that the model fitted the data well (P – Value>  2 = 0.00< 0.05). Results indicated that the model was well specified and were fit for inferential statistics (Greene, 2012; Cameron and Trivedi, 2005; Cameron and Trivedi, 2005). 87 4.2 Test of Hypotheses Section 4.2.1 presents hypothesis tests on social factors; section 4.2.2 gives the hypothesis test on economic and finally section 4.2.3 present hypothesis test on institutional factors 4.2.1 Hypothesis Test on Social Factors To determine the effect of social factors (gender, education level, farm experience and family size) on adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County, Kenya. The study sought to establish if social factors such as age, gender, and education level and farmer experience were significantly influencing the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. To test this the first hypothesis which stated that social factors such as age, gender, and education level and farmer experience do not significantly influence the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. Results from logit regression showed that age had negative and significant effect on the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County (β= -0.1125, p = 0.0000). This implies that the older the farmer the less likely to adopt new innovations as they stick to their older methods of production. This is consistent with Quddus (2010) and Anderson & Feder (2004) who found out that age is negatively related to technology adoption. This is because households tend to be tied up to the old culture of doing things thus being rigid to new ideas. This can also be referred to as cultural lag in technology adoption. Farming experience in drip irrigation had also positive and significant effect on the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County (β = 0.5607, p = 0.0000). Based on these findings it was 90 4.2.3 Hypothesis Test on Institutional Factors To determine the effect of institutional factors such as access to credit, availability of extension services, frequency of extension visits and land tenure on adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. Results indicated that access to credit had positive and significant effect on adoption of drip irrigation technology (β = 0.0608, p = 0.0040). This is because farmers who accessed credit facilities were able to acquire drip irrigation facilities such as water reservoir, main line, drip line, drip tapes and other associated accessories. Muthui, (2015) stated that irrigation technology adoption requires reasonable capital investment and which is beyond means of most small scale farmers. This finding was consistent with Rombo (2013) Rao et al., (2009) REN21 (2005) and Gatahun, Mwangi, Verkuijil and Wondimu, (2000). These results support prior studies by Kabir, Yegberney and Bauer (2013) Mtisi and Makore (2010) that found institutional factors (access to credit and access to extension services) as being significant determinants technology adoption. Ray 2001 argued that extension communication was a necessity in diffusion of innovations. Access to extension services was also positive and significant (β = 0.0879, p = 0.000). This is because extension services extend and educate farmers on new methods of production. Extension personnel help farmers to understand economic benefits of new innovations such as drip irrigation. They are change agents. This means that extension activities were readily available in the region. This was reflected by the number of extension contacts either through farm visits made or training sessions received during the preceding one-year production season. Most studies analyzing this variable in the context of agricultural technology show its 91 strong positive influence on adoption (Bonabana-Wabbi, 2002). This study is 92 consistent with a study by Nkonyaet al., (1997) that found contacts and access to extension services had positive and significant influence on adoption and intensity of technology. Frequency of extension visits was found to have positive and significant effect on adoption of drip irrigation (β = 0.0291, p = 0.0000). This is because when the farmer is frequently visited the farmer will learn more on the new technology as opposed to where contacts are limited. Land tenure was also positive and significant determinant of adoption of drip irrigation (β = 0.0098, p = 0.0020). This is because farmers who owned their land were able to use their title deeds as security to obtain credit and invest in long term projects like drip irrigation technology which is a capital intensive undertaking. Further source of extension knowledge had positive and significant effect on adoption of drip irrigation technology (β = 2.5914, p = 0.000). Studies by Besely and Further, Case (2012b) and Foster and Rosenzweig (2005) using panel data showed that learning from own experience and neighbors‟ experiences were important factors in determining adoption. The results of institutional factors support prior comprehensive adoption study by Feder et al., (2005) and Feder and Umali (2003) which showed that farm size, risk, human capital, labour availability, access to credit and land tenure systems were important factors in technology adoption. These findings also support prior study by Ogada et al., (2014) who found that joint adoption of inorganic and improved maize varieties in Kenya is influenced by the use of manure, access to credit, distance to input markets, secure tenure, education and gender of the household head, cultivated area, drainage of the plots, and expected yields. It also agrees with Njabulo, Ntshangase, Muroyiwa 95 CHAPTER FIVE: SUMMARY OF FINDINGS, CONCLUSIONS AND RECOMMENDATION 5.1Overview This chapter presents summary of findings, conclusions and recommendations from the study. The main purpose of this study was to analyze social, economic and institutional factors influencing adoption of drip irrigation technology among smallholder horticultural farmers in Subukia Sub County, Nakuru County, Kenya 5.2 Summary of the Findings The study carried out descriptive analysis of the variables. The average age of the surveyed farmers was 50 years. Majority (208) of the surveyed households were male. Most of the surveyed households had attained primary level of education and the average family size was found to be seven people per household. The average farm income was KSh. 7,151 while off-farm income had a mean of KSh. 22,689. The average farm size was found to be 3 acres. The study established that the adoption level was low because majority of the surveyed households (228) did not use drip irrigation while a paltry (49 households) used drip irrigation. The adoption of drip irrigation technology among small-scale farmers is still low despite the proven economic and environmental benefits of the technology (Njabuloet al., 2018) the main type of irrigation was use of sprinkler. Most of the surveyed farmers (222) accessed credit facilities, while majority of them accessed extension services. The maximum frequency of extension visit was ten times. Majority of the surveyed households (265) owned their lands under leasehold system. None of the surveyed households owned land on communal system showing 96 that in the study area there was no communal ownership of land. The main source of extension knowledge was MoALF. 97 The study documented that social factors such as age, farm experience and family size affected adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County, Kenya. Economic factors such as farm income, farm size and off farm income influenced adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. The study further documented that institutional factors such as access to credit, access to extension services, and frequency of extension visits, land tenure and source of extension knowledge affected adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. 5.3 Conclusions from the Study Following hypotheses that were tested the following conclusions were drawn from the study. Social factors such as age, and farmer experience significantly influence the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. Therefore, it was concluded that some social factors significantly affect adoption of drip irrigation among small holder farmers in Subukia Sub County, Nakuru County. Economic factors such as farm income, farm size and land tenure significantly influence the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County. Therefore, it was concluded that economic factors significantly affect adoption of drip irrigation among small holder farmers in Subukia Sub County, Nakuru County. Institutional factors such as access to credit, availability of extension service and frequency of extension visits significantly influenced the adoption of drip irrigation technology among smallholder farmers in Subukia Sub County, Nakuru County.