Technology adoption and extension services for inclusive rural development : assessing the livelihoods of smallholder farmers in the Hai and Moshi districts (Tanzania)

(1)

Technology Adoption and Extension Services for Inclusive Rural

Development: Assessing the Livelihoods of Smallholder Farmers in the

Hai and Moshi Districts (Tanzania)

MSc International Development Studies 2018-2019 Graduate School of Social Sciences

June 2019 Word count: 26,646

by Jet de Kort

12060534

Supervisor: Drs. Josh Maiyo Second reader: Dhr. Dr. Fred Zaal

(2)

Abstract

In Africa, most people who live in rural areas are dependent on agriculture for their

livelihood. Their productivity is relatively low, given their limited land size and vulnerability to external factors such as droughts and diseases. As a result, farmers in the rural parts of Africa are confronted by high levels of poverty. One of major challenges in rural development therefore is to improve agricultural production of smallholder farmers. Extension programs have been installed to improve the efficiency of agricultural production to achieve pro-poor development. They promote new agricultural technologies to improve farmers livelihoods, however, to what extent do these interventions contribute to inclusive development? Despite the increasing adoption of improved seeds and other new technologies among smallholders, there is little empirical evidence about the developmental impact of extension services in developing countries, including the unintended effects of interventions in rural areas. Therefore, this study investigates how new farming technologies and extension services influence smallholder farmers livelihoods in Northern Tanzania and analyses how these contribute to inclusive development patterns. The research adopts a mixed-method

approach, an uses the SEVIA extension project as its case study, and Hai and Moshi districts as research location. Semi-structured interviews, surveys, focus groups and document analyses have served as the major sources of data collected. The study finds that most farmers who have participated in the project experience positive changes in their livelihoods, however, some farmers struggle to sustain the use of modern technologies. The project has not

anticipated to the preferences and needs of economically and socially disadvantaged farmers. As a result, several subsistence farmers have been excluded from development processes which creates greater inequalities in rural areas. The thesis proposes that future extension programs should pay more attention to the different needs of farmers. In this case, the financing and marketization of crops has been proved to be essential in order to take full advantage of the impacts of technology adoption.

Key words: Sustainable livelihoods, technology adoption, extension services, smallholder farmers, pro-poor development, rural areas, Tanzania

(3)

Acknowledgements

First, I would like to extend my upmost gratitude to my supervisor Drs. Josh Maiyo for his valuable guidance throughout the research process. Second, I would like to thank SEVIA and its staff members for providing me with this research opportunity and their warm welcome in Tanzania. Furthermore, I would like to thank Lawrence, SEVIAs extension officer and also my research assistant, for helping me get in contact with crucial respondents. I would also like to thank my other research assistant Mseti, for translating the interviews. Most of all, I want to express my thanks to all respondents from interviews and surveys for sparing their time to participate in this study. The research would not have been possible without the personal stories the research participants shared with me.

(4)

List of images, tables and figures

Image 1 Front cover photo from SEVIAs ''Quarterly Report 1 October - 31 December 2018'' (SEVIA, 2018)

Image 2 SEVIA complex Image 3 SEVIA locations

Image 4 Districts within the Kilimanjaro Region Image 5 Wired trellising

Image 6 Stick trellising

Image 7 Farmer engaging in greenhouse production Image 8 Farmer engaging in open field production Image 9 Farmer engaging in open field production Table 1 Coding scheme

Table 2 Characteristics of improved varieties Table 3 Technology characteristics

Table 4 Rate of adoption of seed varieties among different farmer groups Table 5 Types of adopters within different farmer groups

Table 6 Size of farm land among different farmer groups Table 7 Use of different seed varieties across different gender Table 8 Use of different seed varieties across different age groups Table 9 Age groups and education levels of farmers in group 1 Table 10 Age groups and education levels of farmers in group 2 Table 11 Years of farming experience within different farmer groups

Table 12 Length of participation in SEVIA project across different types of adopters Table 13 Functional typology for household categorization in the Hai and Moshi districts Figure 1 The sustainable livelihood framework (DFID, 1999)

(7)

Acronyms

AISE Adverse Incorporation and Social Exclusion

FG Focus Group

OPV Open-Pollinated Variety

SEVIA Seeds of Expertise for the Vegetable Sector of Africa SLA Sustainable Livelihood Approach

TSH Tanzanian Shilling USD United States Dollars WB World Bank

(8)

Chapter 1. Introduction

In Africa, most people who live in rural areas are dependent on agriculture for their livelihood. Much of the rural land belongs to smallholder family farms. These smallholder farmers often have access to less than one to three acre of land and depend on labor supplied by household members and a few other inputs such as plows and tractors. Their productivity is relatively low, given their limited land size and vulnerability to external factors such as droughts and diseases. As a result, farmers in the rural parts of Africa are confronted by high levels of poverty (Leyaro and Morrissey, 2013). Dixon et al. (2001) believe that poverty

alleviation and providing better livelihoods is possible by promoting agricultural development. Growth within the agricultural sector can be ensured by increasing its productivity.

Development programs emphasize the potential of technology to increase agricultural production. In this research, the role of technologies in relation to smallholder farmers livelihoods in Tanzania will be assessed.

Leyaro and Morrissey (2013) argue that the majority of small farms in Tanzania are family owned and grow their crops in the open field which are rain-fed. Much of the land in Tanzania is relatively fertile but is not used optimally. Technologies such as improved seed varieties and fertilizers are seen as key drivers to ensure higher yields. Improvements in agricultural

productivity contribute to the subsequent increase in income levels and improved livelihoods within rural communities. However, to make use of these technologies, several conditions apply. First, improved seeds and fertilizers and other technologies are relatively expensive. As a result, smallholder farmers with few financial resources are not able to use these advanced technologies on their lands. Secondly, knowledge of the potential of these technologies is required in order to integrate new farming innovations successfully.

Anderson and Feder (2007) state that extension programs aim to introduce smallholder farmers to new technologies that can improve their productivity. These programs are mainly directed towards technical information on improved production and to train farmers in efficient and effective crop management. However, the expansion of extension programs is being perceived with a degree of skepticism within development studies. Issues of inclusivity are immanently present within the development arena; therefore, the evaluation of

extension services requires a bottom-up perspective. Aker (2011) stresses that it is important to look beyond farmer training and facilitation of technologies. In this research, I seek to address how and whether extension programs promoting the use of improved seeds and other technologies play into inclusive development patterns.

1.1 Problem statement and research aim

Despite the increasing adoption of improved seeds and other new technologies among smallholders, there is little empirical evidence about the developmental impact of extension services in developing countries, including the unintended effects of interventions in rural areas. Reference to extension program impacts is usually made on the rate of adoption and processes of economic growth, without taking into account the impacts on smallholders’ livelihoods, and especially vulnerable groups. Therefore, this research aims to find out technologies and extension programs influence the livelihoods of smallholder farmers. The sustainable livelihood approach (SLA) is used as an analytical framework to assess why some

(9)

smallholders benefit more from technologies and development interventions than others. In addition, inclusive development theories are explored to gain an understanding of the inclusion of vulnerable farmer groups in the debate around extension services. To paint a picture of the adverse effects of technology adoption and extension services, it analyzes how economically, and socially disadvantaged farmers participate in these development

processes. 1.2 Relevance

This research is socially relevant because it discusses the contextual dynamics that lie behind the different livelihood outcomes of smallholder farmers that participate in extension

programs. By studying the livelihoods of smallholder farmers, one can assess how extension programs affect the capabilities, assets, and activities required for a means of living. It is relevant to study the rate and scope of inclusivity of extension services to smallholder farmers as it seeks to understand who the winners and losers of these interventions are. By examining the factors affecting the adoption of new seeds and technologies, one can determine several assets that are required to make use of them. Moreover, it can also identify which farmers are left out of the extensions programs and understand why. In this way, future extension workers can adjust their programs to become more inclusive, and work towards the joint goal of improving smallholder farmers lives through the adoption of improved seed varieties and other agricultural technologies. The research is therefore designed to examine the effects of improved seeds and agriculture extension services on smallholder farmers’ livelihoods in the Hai and Moshi districts in Northern Tanzania. The research will be conducted by way of a case study of the Seeds of Expertise for the Vegetable Sector of Africa (SEVIA), an extension

project in Tanzania.

1.3 Research questions and thesis outline

The central research question that guides this research is: How do improved seed varieties and extension services influence smallholder farmers’ livelihoods and how do these play into inclusive development patterns in the Hai and Moshi districts? To answer this question, the following sub-questions are addressed:

1 - What factors influence farmers adoption of improved seed varieties and new farming technologies?

2 - How has the implementation of new farming technologies and participation in SEVIAs extension program affected farmers’ livelihoods?

3 – In what ways has the implementation of the SEVIA project contributed to inclusive development in the Hai and Moshi districts?

First, chapter 2 consists of the theoretical framework, in which the main concepts of this thesis are examined. These concepts are factors of technology adoption, sustainable livelihoods and inclusive development. This is followed by chapter 3, which discusses the methodology of this thesis. Chapter 4 gives an overview of the empirical context of this research. The findings are discussed in chapter 5 and 6. Chapter 5 aims to gain an

(10)

understanding of the factors that determine technology adoption. Chapter 6 analyzes the livelihood outcomes after technology adoption by examining farmers livelihood assets. Chapter 7, the discussion, addresses to what extent the SEVIA project has contributed to inclusive development patterns. Lastly, chapter 8 offers a conclusion and suggestions for future research.

(11)

Chapter 2: Theoretical framework

The theoretical framework studies the theories and arguments related to the concepts used in this research. The first sections explore the role of technology in agriculture in developing countries, theories on decision-making, and the factors that are determinant for technology adoption. Secondly, the chapter examines the theory of rural livelihoods and assesses how extension programs intervene in these. These sections form the theoretical foundation to understand how smallholder farmers experience new farming innovations and how these influences their livelihoods. Then, the chapter analyses inclusive development; it examines the role of extension programs and assesses how adverse incorporation can occur. Finally, the chapter concludes with the conceptual framework to illustrate how extension services

contribute to poverty reduction in rural areas. There exists a scholarly gap in the literature, as many authors focus on growth alone as components of inclusive development. There lacks a relational analysis of the dynamics of sustainable livelihoods and the exclusion of subsistence farmers. Therefore, this study uses theory to support the conceptualization of inclusive development patterns in the context of extension programs.

2.1 Technology adoption in developing countries

Innovation and technologies are a prerequisite for the development of equitable and sustainable agricultural systems in Africa (Glover et al., 2016). Technology adoption is especially relevant to smallholder farmers in developing countries; they mostly grow their crops under rain-fed conditions, and their crops are commonly affected by pests and

diseases. Improved seed varieties and management regimes such as soil fertility management and irrigation management can help to overcome farming challenges (Mwangi and Kariuki, 2015). Improved farming techniques and practices seem to offer an opportunity to increase agricultural production and income substantially. Policy makers, developing partners, and private investors emphasize how agrarian technology development is essential in achieving wider socio-economic development and addressing poverty issues in rural areas. They are keen to look for evidence that upholds these claims. However, new agricultural innovations are often adopted slowly despite their great potential (Glover et al., 2016). This study discusses the preferences of smallholders and actual adoption of farming innovations.

Theories of decision-making and the evaluation of factors that influence technology adoption can give insights into adoption processes and are discussed in the next sections.

2.2 Theories of decision-making

Technology adoption usually proceeds by a time: ''It is a mental process an individual pass from first hearing about an innovation to final utilization of it (Mwangi and Kariuki, 2015, p. 209)''. Rogers (2003) describes how technologies are adopted over time and assesses how the diffusion of technologies occurs and conceptualizes this in five steps; (1) knowledge, (2) persuasion, (3) decision, (4) implementation, and (5) confirmation. In the innovation-decision process, knowledge forms the basis for adoption. Information about the existence of the technology, its application, the benefits, and costs shape the individual's perceptions and attitudes towards it. Knowledge refers to the understanding of the functioning of technology and its potential. Perceptions are closely related to the knowledge individuals have about an innovation; however, they are shaped by prior experiences about an innovation which do not

(12)

per se adjust to reality. Knowledge and perceptions together form the attitude towards technologies. The decision stage leads to two outcomes; if the individual develops a favorable attitude towards the technology, it leads to adoption, and unfavorable attitudes lead to rejection (Meijer et al., (2015).

Models help to explain the decision-making process and integrate multiple variables and theories to provide a complete picture of the adoption process. Scholars have applied many different models consisting of numerous variables to understand individuals' attitudes towards innovations. The innovation-diffusion model results from Rogers (2003) work on innovation-diffusion processes. The classical model considers the role of the adopter of the innovation as a relatively passive accepter. It believes that technology is appropriate for use unless the lack of effective communication (Negatu and Parikh, 1999, p.208). However, more recent strands of literature argue that the adoption of innovations is also affected by receiver variables. Meijer et al. (2015) and Mwangi and Kariuki (2015) stress how extrinsic variables such as the characteristics of the farmer, the external environment, and the technology variables affect technology adoption. In their studies, individuals are considered as active end-users of technology; their capacities and abilities to use innovations are found to be critical. Traditionally, most models emphasize the importance of capital endowments in adoption processes. More recently, social and technological variables are included in models that determine the factors of adoption. These different variables help to analyze the uptake of new farming innovations. Inspired by previous studies, this theoretical framework now operationalizes a set of variables and assesses how these can affect technology adoption. 2.3 Operationalizing the factors of technology adoption

It is crucial to understand how different variables influence or hinder adoption. No apparent distinguishing features can differentiate between the variables; therefore, in this study, categorization is done to suit the research context and my preferences. This study on factors of adoption is far from complete; nevertheless, it is comprehensive of the rural settings in which smallholders pursue their livelihood strategies. The next sections operationalize the variables that are determinant for the adoption of agricultural technologies and organize them into different factors, respectively, perceptions of technology, economic and social factors. Studying the factors of adoption allows for a deeper understanding of the choices made by farmers to engage with modern farming practices or not.

2.3.1 Perceptions of technology

The characteristics of innovations play a critical role in the decision-making process. Following Rogers (2003) diffusion-model, perceptions form the basis of technology adoption.

The expected contributions to farmers livelihoods primarily drive the decision to adopt improved farming innovations. Perceptions include the prior experiences about innovation; if farmers perceive the technology consistent with their needs and compatible in terms of their environment, they are more likely to adopt (Meijer et al., 2015). Mwangi and Kariuki (2015) indicate that the perception of farmers towards improved seed varieties significantly influenced its uptake. The perceived preconceptions of technologies are primarily based on the potential of the technology, of which increased productivity, output stability through risk reduction, and improved income are typically expected among smallholder farmers.

(13)

Perceptions can be changed by knowledge-sharing mechanisms such as extension services. When an innovation sparks enough interest, for example, its promising returns, farmers want to get more information. Therefore, access to extension services has found to be critical in the promotion of farming innovations. Extensions programs can enhance technology

adoption by informing smallholders about the benefits and application of innovations. Meijer

et al. (2015) and Mwangi and Kariuki (2015) have reported a positive correlation between

extension services and the uptake of technologies. Extension services are mechanisms that can change farmers perceptions in a persuasive way as they form favorable or unfavorable attitudes towards innovations. Extension agents influence farmers' attitudes and reduce uncertainties about a technology's performance, which generally results in higher likelihoods of adoption. However, studies have also reported on the adverse effects of extension

programs. Theories on inclusive development that will be later discussed in this framework are pointing these out.

2.3.2 Economic factors

This factor is inspired by the economic constraint model, which contends that the distribution of different factor endowments determines technology adoption. This study uses capital and land as variables and operationalizes the costs of adoption (1), sources of income (2), access to credit facilities (3) and farm size (4) as indicators of technology adoption. The net gain to the farmer from adoption is perceived as a key determinant of technology adoption. The costs of technologies are generally considered as a hindrance to adoption (Obayelu et al., 2017). ''Risk and uncertainty play an important role in the adoption of new agricultural technologies (Meijer et al., 2015, 42). The fear of losing capital is imminent; therefore, farmers are hesitant to adopt new technologies. Secondly, access to credit facilities is expected to increase the probability of adoption. Most smallholder farmers in developing countries cannot afford basic production innovations such as fertilizers because they have limited access to credit. Increased linkages to formal credit can help farm households to invest in new technologies (Mohamed and Temu, 2008). Furthermore, differentiation between sources of income is believed to enhance production. Obayelu et al. (2017) argue that income derived from agricultural production is generally insufficient to sustain the use of new technologies. Off-farm income can help to overcome economic constraints faced by farmers in developing countries as it provides them with extra capital resources that can be used to purchase farm inputs such as improved seeds. Lastly, farm size has mixed effects on technology adoption; its impact could be positive, negative, or neutral. Some studies have found a positive correlation between farm size and technology adoption: ''Farmers with large farm size are likely to adopt a new technology as they can afford to devote part of their land to try new technology unlike those with less farm size (Mwangi and Kariuku, 2015, p.210).'' On the contrary, some studies have found a negative correlation. For example, farmers with small lands may adopt more efficient technologies such as improved seeds to achieve higher yields per land unit.

2.3.3 Social factors

The social factors operationalized in this study are the age of farmers (1), formal education (2), and gender (3). Age is assumed to affect technology adoption, both negatively and positively. Muzari et al. (2012) stress that older farmers are better able to evaluate technical

(14)

information because, over time, they have gained more experience than younger farmers. On the contrary, younger farmers are also more eager to adopt new technologies as they are more open to innovations. Conversely, older farmers often stick to their traditional ways of farming and are less willing to try new technologies. Enrollment in education among farmers has been assumed to influence the uptake of innovations positively. ''Education is favorable for farmers' ability to obtain, process and use information that is relevant in the technology adoption process (Obayelu et al., 2017, p.3). Education influences individuals' attitudes and makes them more rational and able to consider the advantages of innovations. Studies on gender issues have found mixed evidence that explains the different roles of men and women regarding technology adoption. Doss and Morris (2000) argue that differences in gender do not affect the probability to adopt. However, in most farming communities, men are the head of the household and are therefore considered as the primary decision makers. In several socio-cultural context's vegetable production is regarded as a women's activity. Men are in control over the household assets such as capital and decide whether to adopt a technology or not. If they do not participate in crop production activities, the likeability of technology adoption is generally low (Theis et al., 2018).

2.4 Livelihoods approaches and rural development in the context of smallholder farmers

2.4.1 Sustainable livelihood approach (SLA)

In the past decades, development agencies and practitioners have been trying to understand what comprises rural livelihoods and what can help to transform these. The SLA has been essential in the examination of rural development. Scoones (2009) argues that a ''a livelihood comprises the capabilities, assets (including both material and social resources) and activities required for a means of living. A livelihood is sustainable when it can cope with and recover from stresses and shocks, maintain or enhance its capabilities and assets, while not

undermining the natural resource base (p.5)''. The sustainable livelihood framework displayed below conceptualizes livelihoods and is commonly used to measure the impact of

development programs. Livelihood approaches look holistically at households and assess all the assets that are used to provide the necessities of its members, which are financial, human, natural, physical, and social capital. Greater access to assets reduces households' vulnerabilities, which can result in more secure and resilient livelihoods. Based on those assets, individuals undertake a range of livelihood strategies to achieve their livelihood objectives.

(15)

2.4.2 Livelihood strategies of smallholder farmers

Livelihood assets play a dominant role in the construction and transformation of livelihood strategies. The SLA links inputs (also referred to as assets or capitals) to outputs (livelihood strategies) and connects this to livelihood outcomes. It assumes that higher levels of asset endowments correlate to more advanced livelihood strategies. Farmers engage in several activities to sustain their livelihoods. According to Scoones (2009), the possible strategy in rural areas is either agricultural production, diversification with other activities or migration. Diversification of livelihood strategies refers to the attempt undertaken by individuals and household to find ways to maximize their livelihood outcomes. Ellis (1998) argues that farmers are rational agents that find new ways to raise income and reduce environmental risks. Examination of the different livelihood strategies that farming households attain is crucial for development agencies as it gives insight into the smallholders preferred outcomes and strategies. Morse and McNamara (2013) critique the approach for being too narrow as it does not address power imbalances and unequal access to livelihoods assets. Nevertheless, it offers a systematic way of understanding livelihoods and the underlying causes of poverty. 2.4.3 The role of extension services in rural development

The SLA also considers the influence of external structures and processes (policies,

institutions, and organizations) on assets and livelihood opportunities. Extension services play an essential role in rural development through the stimulation of technical innovations in agriculture. Christoples (2010) defines it as ‘’all the different activities that provide the information and advisory services that are needed and demanded by farmers and other actors in agri-food systems and rural development (p.2)’’. Extension services are commonly referred to as non-formal education systems aimed primarily at adult farmers. Extension is

(16)

multidisciplinary; it combines technology transfer methodologies, advisory, and information services in promoting agricultural and rural development. Its function is to build farmers' knowledge and skills and to advise and help them to overcome the challenges they face while farming (Swanson, 2008). The recommendations generally proposed by extension agents involve the adoption of technologies and new farming techniques. Cristoples (2010) argues that these help to increase farmers' agricultural productivity and efficiency. The next section operationalizes theory on extension services in more detail by assessing their effects on livelihoods assets.

2.5 Operationalizing the livelihoods of smallholder farmers and the influence of extension on their livelihood assets

The primary agricultural development objectives aimed for in extension program are technology transfer (1), expansion of human capital (2), building social capital (3), and

educating farmers to use natural resources in a sustainable way (4). Extension agents believe that these help to achieve livelihood outcomes preferred by smallholders, such as improved productivity, increased income, and food security (Swanson, 2008). The following sections operationalize the different livelihood assets of smallholders and assess how extension services can expand these. The operationalized livelihood assets help to examine the impact of extension services on smallholders' farmers' livelihood strategies and outcomes. The livelihood approach is chosen because it uses the household as a level of analysis. The theory has stressed that extension programs specifically aim to expand human and technological capital. At the core of extension ideology is that improved knowledge and increased adoption of farming technologies also impacts the other livelihood assets.

2.5.1 Human capital

This asset is concerned with skills, knowledge, and ability to engage in farming activities. Extension services increase the technical and managerial skills of farm households. Aside from its intrinsic value, the expansion of human capital is also needed to leverage all other

livelihood assets. Improved skills and knowledge of farming technologies and techniques can result in increased soil fertility, productivity, and income (UNDP, 2017).

2.5.2 Natural capital

Natural capital includes natural resource stocks and environmental services, which are used by smallholders to sustain their livelihoods. Expansion of this asset has been at the core of rural development efforts. Improved management practices and technologies promoted by extension program can increase productivity and soil fertility. In this way, sustainable land use among smallholder farmers can be assured (UNDP, 2017).

2.5.3 Financial capital

Financial capital includes the financial resources that people use to achieve their livelihood objectives. It includes money inflows as well as stocks. Smallholder farmer faces several challenges, such as crop failures and low market prices that can influence their regular inflow

(17)

of capital. Improved knowledge on crop protection management and the marketization of crops can increase productivity, and appropriately farming income (UNDP, 2017).

2.5.4 Physical capital

This asset comprises the producer goods needed to engage in farming activities. Access to agricultural inputs and productive assets is crucial for smallholder farmers as these can

increase productivity. Extension agents can inform farmers about the benefits and application of new technologies (UNDP, 2017).

2.5.5 Social capital

Social capital refers to the networks and connections on which farmers rely to achieve their farming objectives. Social capital is closely related to human capital; it may include knowledge sharing networks and access to markets. Increased participation in social networks may improve farmers' capabilities and opportunities. Extension services can facilitate in this by developing knowledge sharing mechanisms that enhance innovation. Additionally, extension programs can improve farmers' connectedness to markets by providing training services that focus on the marketization of crops. Increased understanding of the functioning of the market can give the smallholders a better bargaining position and new selling opportunities (UNDP, 2017).

2.6 Extension programs and poverty reduction in the context of inclusive agricultural development

Inclusive development addresses the importance to alleviate the livelihoods of vulnerable or marginalized groups in agriculture. The term has been coined in academia and development practitioners to emphasize the need to ensure that the benefits of growth are equitably shared across all groups of society, particularly to the most vulnerable. Traditionally, the development paradigm has adopted a merely economic perspective in which growth has been measured through increases in income. However, more recent strands of literature stress that development encompasses more than just growth. Poverty reduction policies must be “growth with as much inclusiveness as possible, and with as much inclusiveness of the poorest as possible (Rauniyar and Kanbur, 2010, p. 8)''. It is challenging to address the dimensions of inclusive development; as a result, scholars have not been able to agree upon a clear-cut definition.

Hickey et al. (2015) define inclusive development as following: “It is a process that occurs when social and material benefits are equitably distributed across divides within societies, across income groups, genders, ethnicities, regions, religious groups, and others (p. 5)”. Household economics has been a central focus of inquiry in poverty analysis and unfolds in different ways in different contexts. Concerning the agricultural sector, productivity growth has been at the core of pro-poor rural development programs. Appropriately, poor farmers have been forced into roles that embrace livelihood strategies that are rather economically orientated. The mechanisms proposed that help to achieve these development objectives are primarily based on the modernization of farming production processes (Knickel et al., 2009). In the early years of this century, agricultural extension services had a substantial impact on

(18)

the modernization of farming production processes. Agrarian extension programs aimed to tackle poverty through the stimulation of technical innovations within farming communities. They adopted a growth narrative and emphasized the need to adopt modern technologies to improve farming productivity and reduce poverty within farming communities (Swanson, 2008). However, during the past decade, the work of extension services has often become more diversified. Its primary focus remains on agricultural productivity; however, strong efforts have been made to reach out to small, resource-poor farmers. However, increased inclusion of vulnerable groups in competitive agrarian systems can also have adverse effects. 2.7 Adverse incorporation and inclusive development in the context of subsistence farmers

Within rural contexts, subsistence farmers are still vulnerable in terms of opportunities and access to livelihood capitals. Agricultural production by these households has been

characterized by low yields. They are socially and economically disadvantaged as their access to markets, and economic resource endowments are limited. Despite efforts put by extension providers in trying to increase farm productivity, the marginal benefit of surpluses production remained limited in subsistence farming communities (DFID, 2005). This indicates that

eradication of poverty goes beyond efforts to increase productivity. Subsistence farmers often operate outside of formal market structures; as a result, they are not able to convert their surpluses into capital, which could ultimately be used to invest in more advanced

farming technologies. Furthermore, high input prices and low-bargaining position discourages subsistence farmers from entering commercial farming sectors. In this way, the inclusion of vulnerable farmer groups in agricultural extension programs does not automatically translate into sustainable livelihood outcomes (Include, 2017).

To understand why some farmers have not been able to benefit from these development interventions, the underlying explanatory paradigms of poverty have to be revealed. Theory on adverse incorporation and social exclusion (AISE) introduced by Hickey and du Toit (2013) describe how marginalization and deprivation occurs and why individuals, groups and

communities become socially and economically disadvantaged. The concepts are

heterogeneous; however, they adhere to different approaches. Social exclusion discourse has adopted a residual approach. ''It views poverty as a consequence of being left out of the development processes and contains the assumption that development brings growth and that what is required is to integrate people into markets (Hickey and du Toit, 2013, p.5).'' The rational approach, adopted in adverse corporation discourse, on the other hand, ''emphasizes the extent to which ‘development’, growth and the workings of markets can also produce poverty (Hickey and du Toit, 2013, p.5)''.

AISE further broadens the conceptualization of causes of disadvantage and challenges the assumption that social inclusion is relational. Inequality is not only rooted in social, political, and economic inclusion and exclusion, but also global drivers of exclusiveness have an impact on the local level (Hickey and du Toit, 2013). Neo-Marxists critique development programs for being too capitalistically orientated. For example, in the extension programs, farmers are required to use expensive hybrid seeds that cannot be propagated from the crop from one season the next. Munyi and de Jonge (2015) blame that extension programs for feeding the international seed value chain at the cost of vulnerable groups. Moreover, Gupta et al. (2015)

(19)

stress that ''sometimes, nested circles of exclusion may make farmers worse off through adverse incorporation in value chains (p.547).'' Therefore, a relational approach to inclusiveness stresses that extension projects should carefully analyze the needs of each targeted group to ensure efficient and equal distribution of the benefits of their interventions (ABDG, 2010). The following section operationalizes the variables related to inclusivity and assesses how extension program can contribute to inclusive development at the household level.

2.8 Operationalizing inclusive development

From a rural development perspective, livelihood approaches have been put forward to asses development within farming communities. The SLA helps to assess the impact of institutional interventions on the livelihoods of farming households. Extension programs have a rather output oriented design; they understand poverty as a deprivation of farming capabilities and thus of the substantive assets' farmers need to pursue their livelihood strategies. The

diversity of farmers in rural areas makes it challenging to introduce well-designed

interventions that are tailored to the local needs for each category of a farmer. Inclusion of various marginalized groups in the organizational design and implementation is the first step to inclusivity. However, more importantly, inclusive development is about whether

development interventions lead to a qualitative change in people's livelihoods. This section uses a relational approach to assess the nature, forms, patterns, and outcomes of inclusion within extension programs. It adopts a neo-Marxists approach to critically assess how

development projects with specific market orientated designs address rural inequalities. This thesis focusses specifically on the role of subsistence farmers, as they are typically part of pro-poor programs that promote inclusive agriculture.

2.7.1 Inclusiveness in the project plan

The organizational design, implementation, and the scope and rate of adoption among vulnerable farmers groups provide insight into SEVIAs contribution to inclusivity in the Moshi and Hai districts. This section specially addresses the inclusion of subsistence farmers and assesses how the extension programs attempt to empower them through training on agricultural practices and technologies

2.7.2 Outcomes of inclusiveness

This section examines how the extension services and promoted technologies meet the farmers' needs and if these help to achieve their livelihood objectives. It seeks to understand how subsistence farmers perform in comparison with other farmers in terms of improved productivity, changes in income, and market access.

2.7.3 Relational approach to inclusive development; links between asset (dis)accumulation and processes of adverse incorporation

In this section I seek to understand how the project responds to subsistence farmers marginal position in formal market systems and I critically assess the general belief that expansion of livelihoods assets results in improved livelihood outcomes. Extension programs specifically

(20)

aim to expand human and technological capitals and stress that the expansion of these is vital for achieving development at the household level. However, to what extent do the links between assets contribute to patterns of inclusive development? Adverse incorporation looks at the relational contribution of assets to farmers livelihoods and emphasizes how the

workings of the market can produce poverty. It assumes that the absence of solid financial and social assets proceeds in further marginalization of vulnerable groups. The workings of capitalistic markets can induce processes of adverse incorporation as subsistence farmers are generally economically and socially disadvantaged.

2.9 Conceptual scheme

To decide which dimensions could be useful to assess the impacts of extension services and new farming technologies on smallholders' farmers' livelihoods and rural development patterns, I adopted an interdisciplinary approach. Studying the factors of adoption and livelihood capitals allows for a deeper understanding of the choices made by farmers to engage with new farming practices or not. By combining the more relationally based theory of inclusive development that analyses how development programs are designed and implemented with the livelihood's framework, I seek to analyze how development

interventions in agriculture affect the livelihoods of vulnerable groups. The SLA allows for strong local-level fieldwork, whereas inclusive development addresses issues from a conventional perspective. It requires research into how development programs aim to end poverty recognizing the contributions of different social groups in agriculture. The conceptual scheme below is best read from left to right. It starts with the combination of livelihood assets, which are potentially influenced by extension programs through training on new farming practices and technologies. Although smallholder farmers pursue similar livelihood strategies, notable differences in asset holding deter some farmers from technology

adoption. By studying the outcomes among different social groups, one can assess whether these interventions have led to inclusive development or adverse incorporation.

(21)

Chapter 3: Methodology

This chapter elaborates on the data collection methods and other methodological

considerations that were used during a fieldwork period of nine weeks in Northern Tanzania. 3.1 Selection of research location and case study

I collected empirical data during my fieldwork research period in Northern Tanzania. In this research, I have adopted a case study approach with the SEVIA project as selected case, and the Moshi and Hai districts as research location. This thesis uses the project to assess the impact of extension services on smallholder farmers' livelihoods. In collaboration with SEVIA, which's training and research center is based close to Moshi, the districts above have been selected. They have been chosen considering their proximity to the SEVIA station, and the possibility to work closely together with the extension officers responsible for both districts. The SEVIA project is a well-established player in the development of Tanzania's agricultural sector. Since 2012 it has been operating as a research center and extension provider in 11 regions throughout Tanzania. Development academics and practitioners have studied the effects of extension services on the agricultural output of farmers, but often overlook the social changes that occur after technology adoption. Therefore, this study provides a relevant research case for this thesis on smallholder farmers livelihoods and inclusive development. One has to be aware that this case study is not representative of other development and extension programs. Nevertheless, it navigates through the field of technologies and extension services and is critical of its performances. A more detailed description of the research location and the SEVIA project will be given in chapter 4.

3.2 Research design

This research uses a critical interpretive epistemology to study the smallholder farmers that are engaged with the extension programs of SEVIA. According to Bryman (2016), this epistemology grasps the meaning of social action and is critical of the application of

development programs such as the SEVIA project. Interpretive approaches assume that the rate and scope of inclusivity within development programs can be very distinct, and therefore makes it an interesting field to study. Ontologically, this research takes a critical realism approach. It assumes that smallholders operate in highly subjective realities that are constructed on the basis of their socioeconomic, institutional, and natural environment. Therefore, I assume that this study requires understanding of farmers perceptions of local contexts and realities.

A case study approach, with the SEVIA project as selected case, has been used to study the livelihood of farmers in a particular geographical context. This approach has been chosen as it allows for an in-depth analysis of the highly contextual nature of agricultural development. This thesis makes use of a mixed method approach. It has a strong emphasis on qualitative methods with quantitative data consisting of 97 surveys complementing the qualitative data. The qualitative data has been suitable to study the livelihoods contexts of farmers through the lens of perceptions and experiences. In contrast, the quantitative data helped to test these perceptions and has been used to support and or reject hypotheses concerning

(22)

technology adoption. For triangulation purposes, it makes use of several data collection methods such as semi-structured interviews, focus groups, document analysis, and surveys. 3.3 Sampling criteria

The unit of analysis has been centered specifically around the extension services offered by SEVIA, which formed a network from which participants have been sampled via snowballing. Two extension officers covering the Hai and Moshi districts have been very helpful in getting access to research participants. The respondents of the survey were approached during field days. The participants of the survey were randomly selected as every participant attending the field day was asked to fill out the questionnaire. Attendants were coming from the surrounding villages and were very diverse in terms of sex, age, and social group. The

participants of the semi-structured interviews and focus groups discussions were selected by the extension officers. As a researcher, I do not know how random their selection was, nevertheless, I encouraged them to introduce me to farmers from different sexes, ages, and social groups. A purposive strategy ensured that the voices of marginalized individuals were heard, allowing me to see if and how the extension program contributes to inclusive

developments patterns in the region.

3.4 Methods of data collection and analysis

The primary data was collected from 26 in-depth interviews, two focus group discussions, and 97 surveys. The variety of data collection methods captures different dimensions of the same phenomenon by looking at it from different perspectives. According to Olsen (2004),

triangulation of data helps to deepen the researcher's understanding and improves the validation of research findings. The research participants are divided into two groups, SEVIA participating farmers and non-SEVIA participating farmers. The first group of smallholders served as respondents for the qualitative data collection methods, and the second group as respondents to the questionnaire. The next sections give an extended description and explanation of the research participants and the data collection methods.

3.4.1 Semi-structured interviews

The interviews with 26 farmers who participated in the SEVIA project were centered primarily on their attitudes towards technology adoption and how the implementation of farming innovations and the extension program has influenced their livelihoods. The respondents have been identified with the help of SEVIAs extension agents working in the Hai and Moshi districts. In the first stages of the fieldwork period, I have been shadowing these extension officers to make a varied selection of respondents. The interviews were conducted on the farms of the smallholders themselves. The interviews were conducted in Kiswahili, the local language spoken in this region, with the help of a translator and typically lasted for half an hour to an hour. All research participants were asked for their consent to be recorded with my phone before the start of the interview. Of the respondents, 22 out of 26, agreed to be recorded, and I took detailed notes for the four participants who did not agree to be recorded. Furthermore, I assured the research participants that their information would be held confidential. Additionally, I also employed a research assistant to transcribe the interviews.

(23)

In addition to the interviews with farmers, in-depth interviews were conducted with four SEVIA staff members: two extension officers, the general manager, and the extension manager. The SEVIA respondents provided insightful information on the project's program, planning, design, and implementation. Furthermore, two experts in the field, namely the general manager of East West Seed, a multinational seed company, and the program leader of the World Vegetable Center, an international non-profit research and development

institute, have provided contextual information on the development of Tanzania's agricultural sector. The key informant interviews were conducted at the earlier stages of the fieldwork period to gain a deeper understanding of farming practices and livelihoods in my study location that later has been very useful when interviewing farmers. These interviews were conducted in English and lasted for an hour. Annex 2 gives an overview of all the respondents who participated in the semi-structured interviews.

3.4.2 Focus groups

One of the main issues of focus in this research, inclusive development, stresses the participation of marginalized groups in extension programs. Therefore, focus groups discussions with the vulnerable groups are very relevant and worthwhile to the research. They provide a better understanding of social mobilization in a group or community and encourage people to speak openly. Bryman (2016) argues that focus group discussions help to analyze the perceptions and experiences of different groups about sentiment in a specific community. This method allows people to interact with each other and explore different perceptions and experiences of individuals. I organized two focus group discussions with smallholder farmers in the Hai and Moshi districts. One focus group consisted of four farmers under the age of twenty-six, and the other one consisted of a group of six subsistence farmers ranging in age and sex. The focus groups were arranged by my research assistant and took place on the field of a farmer participating in the group discussion. To glean participants' perspectives, my role as a facilitator was very moderate; I explained my role as a researcher and introduced the topic to the participants. After this introduction, the focus group

discussions were moderated by a Kiswahili speaking research assistant who was trained and informed beforehand about my general topics of interest. The assistant led the discussions, and after each topic covered, he summarized the main points of discussion in English to me. In this way, I was able to follow the group discussions closely without and interrupting the flow of the conversations. The focus group both had a duration of over an hour.

3.4.3 Surveys

I used this method to examine the factors of adoption of new farming technologies among smallholders' farmers who did not participate in the SEVIA project. In these surveys, I collected data on the farmer characteristics and reasons to adopt technologies or not. The benefits of surveys identified by Bryman (2016) are that it provides high representativeness developed in less time compared to other data collection methods. With the help of a SEVIA employee, the questionnaire consisting of mostly closed-end questions was translated into Kiswahili and has been distributed on paper during field days organized by SEVIA. Ninety-seven farmers participated in the self-administered surveys and were randomly selected. I am aware that this method limits me to understand the respondents' answers as the results do

(24)

not gain insight into the cultural and socioeconomic environment the farmers live in. Nevertheless, these surveys helped to gain a more general view of the socioeconomic characteristics of the respondents and their reasons to adopt new farming technologies or not. The quantitative data derived from the surveys have been used in addition to qualitative data and has been used to make a comparison between SEVIA participating farmers and non-SEVIA participating farmers. This methodological approach is equally important in answering the first research questions concerning the different factors of adoption between the two farmer groups.

3.4.4 Document analysis

Two reports and two documents produced by SEVIA and one report published by the Dutch Ministry of Foreign Affairs have been used to complement other data collection methods. These documents provided valuable additions to data on the organizational design of the SEVIA project and the implementation of its extension services. The reports helped to get a clear picture of how the organization is running and how it fares over time. To analyze the levels of inclusiveness, the project plan, and annual and quarterly reports on the program's planning and implementation are used as a verification of findings from other data sources. Moreover, they add data on production and profits records that have been forgotten by the participants of the semi-structured interviews. Some of these have been published publicly, and some belong to the internal archive. Regarding the documents that have been internally stored, the managing director of SEVIA has given me access to these. The confidentiality of these documents has to be considered as he could have been able to withhold information that would shine a bad light on the project. Nevertheless, the documents are mainly used to convey the performance of SEVIA, emphasizing the selection of farmers in the project. 3.4.5 Fieldnotes

I have taken fieldnotes during visits and interviews with farmers to record descriptive information, behaviors, and actions that contribute to the data being obtained. I have used observations to triangulate farmers own perceptions on the number of technologies adopted, with watching their farm structures during on-farm visits. These observations and

post-interview reflections have enriched this study with contextual information in addition to the other methods of data collection.

3.4.6 Data analysis

Qualitative data derived from semi-structured interviews and focus group discussion has first been transcribed by a research assistant from Kiswahili translated to English. After the

transcribing process, the key thematic codes concerning technology adoption, livelihood changes, and inclusive development have been identified through the Atlas.ti qualitative data management program and fitted into the coding scheme below.

(25)

The last phase of the data analysis consisted of the classification of thematic patterns; for example, the frequency of themes, unanticipated themes, differences between interview participants and focus groups. The data derived from guided surveys have been analyzed by using Microsoft Excel. Demographic characteristics and reasons to adopt new technologies or not have been identified quantitatively.

3.5 Ethical considerations and limitations

First, I would like to address my positionality within this research. My position as a researcher has been critical as I am a young Western female in an African country. Presuppositions that have to do with the image of a 'helper' can cause awkward situations. Sumner and Tribe (2008) discuss the role of a researcher in complex situations and argue that development studies have an ethical and instrumental point of departure. As a researcher in this field I seek to do good; however, my research in itself will not necessarily improve the participants' lives. Therefore, informed consent has been up-held throughout the field work period to avoid the risk of misinterpretation and expectations. For example, respondents might ask for help or give answers that are expected from them and which are desired by researchers. Before each interview, I informed my participants about the study's aim, topic, and future use and

stressed that participation in interviews and focus groups was voluntary. In this way, the participants could understand my role as a researcher and not confuse me with other practitioners in the field. I have explained the consent orally as I felt that a written consent form could deter the farmers from participating.

Confidentiality is another important ethical concern in this research. I guaranteed the

respondents that their information would not be used for anything other than study purposes and will not be shared with others. I also told them that if they wished not to answer a

question, they could do so and that they could withdraw answers. I expressed this again at the end of the interview to remind the participant of his/her rights to delete information. The only issue that I encountered was that the translator could pass on information to others without the interviewee or me knowing it. To overcome this, I trained him very well and made him promise to be confidential and trustworthy.

Safety was ensured by asking the farmers where they would be comfortable to be

interviewed, in all cases on their own farms. In this way, trust was created easily as they were in a familiar place. No other people were present, except for the interviewee, translator and me. The translator is also employed as an extension officer in the SEVIA project and kept a good relationship with the respondents. This was very beneficial to the interviewing

processes; thanks to him the building of rapport happened very easily, and the farmers very speaking very openly about their lives and also the issues they encountered while farming.

(26)

3.6 Methodological reflection and limitations

A challenge concerning the mixed-method design is the need to combine qualitative and quantitative data in a meaningful way. The solution applied in this thesis is that the research design is constructed in a way so that the quantitative and qualitative data address the same variables. The research has multiple techniques to triangulate different data sources against each other. As a result, the validity and quality of the study has been assured. Further assessment on the quality of this research is done on the basis of alternative criteria for evaluating qualitative research developed by Lincoln and Guba and presented in Bryman (2016), respectively trustworthiness and authenticity.

3.6.1 Trustworthiness

This section first examines how close the research comes to reality, which is commonly referred to as credibility. As the research focuses on smallholder farmers perceptions, the study presents its realities. The interviews were intended to measure the change in livelihoods; as a result, I had to deal with issues of self-reporting bias. To overcome this, before starting with the actual data collection processes, I first shadowed SEVIAs extension officer. In this way, I was able to get to know the respondents and to build up rapport before starting to interview them. Also, the questions were constructed in an open-ended way to allow the respondents to include more information, including experiences and perceptions of the subject and their lived experiences. This has allowed me to better access the participants' true perceptions of technology adoption and livelihood outcomes. To ensure that the

research was credible in terms of good practice I asked all respondents for their consent before participating in the interview or survey, and I clearly explained my role as a researcher to avoid the risk of misinterpretation and expectations, as explained in a previous section. Secondly, this sub-criterion assesses the research's transferability. I believe that this research can hold to a certain degree in other times; however, not in another context. The research is context specific in terms of its connectedness to the SEVIA project and the cultural and socioeconomic situation of the respondents. The experiences, perceptions, and beliefs are deeply rooted in the social environments the respondents live in; however, these are not static; everything is subject to change.

Finally, this section evaluates dependability. To ensure this, I have been transparent about my research steps, and I included a list of all the respondent's interviewed. As discussed, during the process, I had a research assistant who challenged my assumptions and my choices. Confirmability has been reached through discussion with my research assistant and SEVIAs managerial staff.

3.6.2 Authenticity

The criteria assessing authenticity relate to the broader social and political impact of the research (Bryman, 2016). The political impact of the research lies in the examination of the successfulness of SEVIAs extension program in terms of livelihood outcomes. It assesses the effects of technology adoption and extension services on smallholder farmers livelihoods. It

(27)

looks beyond the changes in agricultural output, as it aims to understand the social changes that occur after adoption and participation in the extension program.

The use of SEVIAs extension officer as a translator has helped me to get access to a varied research population; however, it can also limit the respondents to speak openly about their perceptions and experiences regarding the extension program. To minimize the impact of the translator, I have carefully explained my research intentions and the importance of honesty in their answers. This relates to the ontological authenticity of this study; I have given the respondents the freedom to interrupt the interviewing process to ask for clarification and verification. The educative authenticity of the study was not applicable. I adopted a neutral stance and avoided to get involved in discussions on the usefulness and functioning of the SEVIA project.

(28)

Chapter 4: Empirical context

This chapter briefly discusses the empirical context of this study. This chapter demonstrates SEVIAs suitability for this study and provides practical information surrounding the role of its extension program, that may contribute to the findings of this research. It firstly introduces the study area and describes the livelihoods of Tanzania's rural population. It then discusses Tanzania's agricultural extension environment and its contributions to rural development and technology adoption. Thereafter, it gives an overview of the SEVIA project in which the project's planning, implementation, and successes and are analyzed. The information for this chapter derives from secondary data, and internal documents belonging to the SEVIA project. 4.1 Introduction to the study area

Tanzania is a country in East Africa bordering on the Indian Ocean with a population of 56 million of which 72 percent lives in rural areas. In Tanzania, the agricultural sector employs about 75 percent of its people, and overall production accounts for nearly 30 percent of national GDP (Leyaro and Morissey, 2013). This research took place in the Hai and Moshi districts located within the Kilimanjaro Region of Northern Tanzania. The Northern highlands provide favorable climatic conditions, and a high density of farming activities characterizes the region. Several ethnic groups coexist in this region, of which the Chagga community is dominating the areas around the slopes of the Kilimanjaro. The Chagga have amassed considerable economic wealth coming from the favorable climate of the area and excellent agricultural methods. Moreover, their long contact with European models of education and Christianity may have given them an economic advantage over other ethnic groups (Moore and Purrit, 2017). Today, their livelihoods are still merely based on agriculture, with bananas as their staple crop. The other main food crops produced in this area are maize, rice, wheat, and vegetables. Their production is mainly used for domestic consumption and cash.

Throughout the country, the agricultural sector has not been performing to its potential. Farmers have not been able to increase their productivity; yields are low, risks for farmers are high, and profitability is generally poor. For example, the yields are about 3,2 tons per acre, while the estimated potential is almost 5 ton per acre (TNBS, 2017).

Development programs emphasize the potential of technology to increase agricultural production. Much of the land in Northern Tanzania is relatively fertile but is not used

optimally. Technologies such as drip irrigation, improved seeds, and fertilizers are seen as key drivers to ensure higher yields. Improvements in agricultural productivity contribute to the subsequent increase in income levels and improved livelihoods within rural communities. However, to make use of these advanced technologies, several conditions apply. Firstly, these technologies are relatively expensive. As a result, smallholder farmers with few financial resources are not able to use these advanced technologies on their lands. Secondly, knowledge of the potential of these technologies is required to successfully integrate new farming innovations (Leyaro and Morrissey, 2013). The importance of the latter has been emphasized in extension programs and is being discussed in the following sections.

(29)

2.2 Agricultural extension in Tanzania

''Extension services are crucial to putting farmers’ needs at the center of rural development, ensuring sustainable food security and poverty reduction, and dealing with risks and

uncertainty (GFRAS, 2012, p. 2)''. The Global Forum for Rural Advisory Services (2012) stresses that knowledge sharing is critical for sustainable development. More and better agricultural extension and advisory services allow the knowledge-based infrastructure to adapt to changes in agriculture. Extension includes the dissemination of technical knowledge and facilitates the mechanisms of knowledge sharing.

The field of extension services is characterized by its pluralistic nature in which several actors operate. All of these actors have different interests, and therefore prioritize different

outcomes of their promoted extension strategy. In Tanzania, several extension actors are present; government extension officers represent the public sector. According to the African Soil Health Consortium (2015), government officials are considered to be the leading

providers of extension services in the country. In 2012, Tanzania had 10,891 public

agricultural extension officers. The average ratio of agricultural extension officers to farming households was 1:630, and only ten per cent of the farming households were reached by agricultural extension officers. Besides public extension provision, NGOs and bilateral donors, private sector firms and farmer-based organizations and cooperatives comprise the extension field in Tanzania. In this thesis, the SEVIA extension program has been used as a case study and will be further analyzed in the following section.

2.3 SEVIA

The SEVIA project can be considered as a non-profit organization funded by two world leaders in vegetable seeds: East-West Seed and Rijk Zwaan, and by the Dutch Ministry of Foreign Affairs. Wageningen University - Applied Plant Research is the Dutch knowledge partner. The project has started in 2012 and will end in December 2019 and aims to contribute to the development of the vegetable industry in Africa and food security (SEVIA, n.d.a). SEVIAs operational center (see image 2) is located at Lambo Mferejini, which is very close to Moshi, and its main activities are concerned with research and demonstrations. The station serves as a research center where new techniques and varieties are tested. The research findings and observations help to develop, implement, and disseminate new farming innovations.

Moreover, it screens genetic vegetable resources which are used to make recommendations to smallholder farmers in Tanzania. The SEVIA station also serves as a managerial center for the organization of its extension program (SEVIA, 2017). This program is operational

throughout Tanzania; in 2017, fourteen extension officers were covering 19 districts in 10 regions of Tanzania (see image 3). This study focusses on the Hai and Moshi district, which are located in the Kilimanjaro region (see image 4). To have a clear understanding of SEVIA as an organization and its extension work, the project's planning and implementation are now briefly assessed.

(30)

Technology adoption and extension services for inclusive rural development : assessing the livelihoods of smallholder farmers in the Hai and Moshi districts (Tanzania)