Diversity,complexity,and structure of social networks: Study of a smallholder dairy project

Abstract

Network structures have drawn a lot of research interest in recent years. However, little is known about how the diversity, complexity, and structure of the multi-actor networks associated with a project shapes information exchange and linkages in different contexts. This study, conducted in 2017, addresses this deficiency by analyzing the social networks in three villages, and it was directly set up in response to operationalizing a smallholder dairy development project. Social network analysis was used to characterize the networks and to look into network configuration and information exchange dynamics. Data were collected through household-level interviews with the project’s participant farmers and key project stakeholders, supplemented with other Participatory Rural Appraisal (PRA) techniques. Analysis focused on network visualization and estimation of specific network parameters for comparison. Results indicated that public sector actors were the key drivers in all these smallholder networks. Flow of information was favored both vertically and horizontally in the network, which was configured at a later stage, rather than in the project piloting stages, owing to unique actor alignment and multiplicity of ties. The networks were also found to vary in terms of some of the network attributes, signaling varying levels of network integration and brokerage potential. Better cohesiveness and information spread was observed in the network, which had the right mix of information acquisition and exchange networks. The study offers valuable lessons with respect to connectedness of network actors, need for multi-actor alliance, and implication of centrality measures in determining network dynamics.

Keywords

Dairy development social networks information flow linkages agricultural extension

Introduction

Livestock systems represent a potential pathway out of poverty for many smallholders in the developing world (Randolph et al., 2007). Although dairying can help the underprivileged come out of poverty, they need institutional and technological support services for sustained production (Rangnekar, 2006). Furthermore, dairy marketing and processing activities are more critical for resource-poor smallholders, who often are unable to establish linkages with markets, processors, and consumers on their own (Singh et al., 2013). For the same reason, it is extremely important to integrate and organize smallholder producers (Sinyolo and Mudhara, 2018) into agri-food chains (Reardon et al., 2009). However, they find it increasingly difficult and costly to participate in value chains (Barrett et al., 2012) and usually operate in very fragmented and disorganized chains in developing economies (Dicecca et al., 2016). Hence, they need to be organized, which could be in the form of producer cooperatives (Bijman and Hendrikse, 2003), farmer producer organizations (Vishnu and Gupta, 2017), or commodity-specific farmer interest groups (Patil et al., 2014). Nevertheless, given the predominance of smallholders who own much of the cattle wealth in developing countries (Herrero et al., 2012), it is impossible for any single body to vertically integrate all of them. Localized dairy development projects were evolved as a response to this challenge, under the patronage of development agencies, to serve small holder producers exclusively (Odeo-Waitituh, 2017; Sulaiman and Reddy, 2015; World Bank, 1991). Often these projects were driven by an alliance of actors with diverse interests, who form multi-actor networks along with the project’s beneficiary farmers (Hall et al., 2008). Smallholder interactions often emerge from such networks, conditioned by social and economic institutions that influence their form and function (Davis et al., 2008). Vamsidhar Reddy and Sulaiman (2016) argue that the entire network of actors, especially the producers and their alignment with other actors, becomes crucial for network performance. Some of the latest studies contend that, though the farmers are well connected and central in their networks, they have little influence on other actors in the whole system (Chindime et al., 2016; Weyori et al., 2017).

Another strand of literature demonstrates the importance of positioning actors in the networks, so as to support smallholders. For instance, the findings of Spielman et al. (2007) demonstrated how the public extension and service providers continue to be the prominent actors with an ability to influence smallholder access to information and services. Conversely, Asres et al. (2012) showed how the public sector actors failed to institute marketing links, despite being the prominent network actors. However, these network connections and power equations are specific for each locality (Hoang et al., 2006).

Another important dimension to discuss here is the dynamism of activities within a network. It refers to changing patterns of relations, formation of new linkages, and power positions. This is important because the performance of organizations can be related not only to their internal knowledge and their intangible assets but also to the effects of networking (Cinelli et al., 2017). An overview of a recent strand of literature suggests that studies are undertaken to explore the influence of internal as well as external organizational structures on the spread of information (Novkovic and Holm, 2012; Reed and Hickey, 2016). Also, previous studies have underscored the role of network ties in deciding the effectiveness of the value chain at the macro level (Folder et al., 2015). Then again not many studies have been reported, which explore micro-level network structures and its dynamics with respect to smallholder development. Therefore, this study is an attempt to strengthen this field of investigation. Consequently, this study analyzed the social networks of the Nature Fresh (NF) project in three villages—Kannadi, Puduppariyaram, and Akathethara in Kerala state—with respect to network structure, actor diversity, and linkages. Various important aspects, such as connectedness of network actors and centrality measures, were scrutinized in order to derive key lessons relevant for network configuration, information flow, and integration of the network.

Theoretical framework

We used the social network theory articulated by previous researchers (Guillemois, 2013; Hermans et al., 2013) to explain the formation of the network of relations in shaping the dairy value chain under the NF project. This theory is based on the tenet that social behavior is embedded within, and affected by, complex webs of social relations (Abizaid et al., 2015). The network perspective views any system as a set of interrelated actors or nodes (Borgatti and Li, 2009). It is necessary to understand that network structures shaped by local interactions are important as they are crucial to the network dynamics (Powell and Bromley, 2015). Social networks serve to promote collaboration (Cross et al., 2005), to facilitate information flow, and to lower barriers to form new links for the exchange of materials, goods, and services (Hall et al., 2003; Johny et al., 2014). Hence, knowledge and ideas within the networks are not static but rather dynamic (Eshuis and Stuvier, 2005). Further, networks are generally considered to be a kind of organizational skeleton that holds together a variety of institutional actors allowing their interaction (Ceglie and Dini, 1999; Cinelli et al., 2017). Therefore, they are widely used in organizational research in a number of dimensions. For instance, network analysis was found to be useful to explore the influence of the social organizational structure of cooperatives apropos their ability to spread agricultural innovations (Reed and Hickey, 2016).

Similarly, Chaudhury et al. (2017) used the same approach to examine community relations with outside actors, to understand better how they influence social structure. However, there aren’t too many micro-level studies exploring the internal social networks and relationships (Reed and Hickey, 2016). Within the networks, information spreads through specific linkage patterns and may be significantly controlled by key actors in leadership positions (Reed and Hickey, 2016). Similarly, collaborating actors have specific roles and functions to perform, to spread the information in these specific patterns (Hermans et al., 2013). There may even be powerful actors who can shape the composition of the network, either through providing the needed resources or by altering the structural conditions for the network to grow (Hermans et al., 2013). However, inefficiently structured networks can seriously limit network dynamics by putting constraints on the information flow mechanisms (Caria and Fafchamps, 2015). In this study, by examining network structures, heterogeneity, and relationship among different actors that form the core and peripheral structures of a network (Spielman et al., 2011), we tried to expose their role in determining the dynamics of smallholder social networks in the three villages.

The NF project, designed by the Department of Animal Husbandry (DoAH), Kerala, aimed to address poverty among rural women through the use of dairying as a tool. The project was launched as a dairy development intervention to bypass the cooperative model of milk procurement, where the milk collected at the base level is collected, transported, and sold at distant places. Under the NF model, localized procurement, branding, and sale of milk were conceptualized. The project was plugged into the social infrastructure through the women-based groups, formed by the State Poverty Eradication Mission (SPEM).¹ All the required financial and technical assistance was provided by DoAH, and SPEM took the lead in building capacity of the participants, prior to project launch. Under the project, the participants were mobilized into smaller groups (of five producers) and financial assistance was provided to purchase good quality cattle. The milk produced by these groups was labelled before its sale to ensure its traceability. Bottled fresh milk was directly sold to consumers, fetching higher payments for the producers unlike in the previously followed cooperative mode. Project monitoring committees were established to check its progress periodically, and these committees were comprised of multi stakeholders, including producers and officials from local self-government as well as DoAH. Importantly, the project addressed many of the structural issues typical of a smallholder production system, like lack of sufficient capital and market linkages for the producers (Daviron and Gibbon, 2002; De Janvry et al., 2005). Initially the project was piloted in one village (Kannadi) in 2007, and later it was replicated in two other nearby villages (Puduppariyaram and Akathethara). The series of activities in launching the project is represented in Figure 1.

Figure 1.

Series of activities in launching NF: Nature Fresh project.

Methodology

The study focused on the configuration of networks in the three study locales of the NF dairy project. The key project partners, private input dealers, and banking institutions along with the project beneficiaries, that is, smallholder dairy producers, constituted the main actors of the NF project. The network of actors connected to the project is diagrammatically represented below (Figure 2).

Figure 2.

Network of actors in the NF project. Codes of various actors—NFP: Nature Fresh Producers; DoAH: Department of Animal Husbandry; KLDB: Kerala Livestock Development Board; DDD: Dairy Development Department; PRI: Local Self Government Institution (Panchayath); SPEM: State Poverty Eradication Mission; NIA: National Assurance Agency; MILMA: Kerala State Milk Marketing Federation; SBI: State Bank of India; LMTC: Livestock Management and Training Centre; SCB: Service Cooperative Bank; CFT: Cattle Feed Traders; KF: Kerala Feeds Public Ltd; MDPT: Milk and Dairy Products Traders.

This study covered the entire network of agents at various levels—from the village to individual farm households—in each of the study villages. Primary data were collected using a farm household survey with semi-structured interviews. All the participants of the project were included in the survey by contacting them at the household level. The complete list of project participants, maintained by DoAH, was obtained for this purpose. Telephonic interviews were conducted with those respondents who were absent from their farm households during the survey. The survey basically covered questions about socioeconomic and demographic characteristics, farm household level details including dairy production and marketing, as well as questions on social network of the actors. The questions on social network were designed to capture network dynamics resulting from the interaction, information exchange relations, and linkages with other actors. These questions were (a) Who are your major information providers with respect to dairying? (b) With whom do you share your knowledge on dairying frequently? and (c) Have you shared any new information with respect to dairying in the last five years? With the intent of capturing the various features of the information system, questions were asked about the kind of knowledge they receive/share as well as the various inputs and service providers on dairying. The questionnaire was pretested with dairy producers from the non-sampled area to improve clarity and potential survey bias before the final survey. We assigned weights to the relations based on their multiplicity and type, using the equation from Chaudhury et al. (2017).

W = Σ R

where W is the weight of relation and R is the type relations—knowledge and resources (1 = relation, 0 = no relation).

Using this information as input, social networks were drawn up for the study locales using the visualizing software. Social network analysis is a methodology that provides complementary visual and statistical components for analyzing the traits and relationships of actors in a network (Chaudhary and Radhakrishna, 2018). Thus, selected centrality measures were derived to identify key actors in the network as propounded by previous researchers (Asres et al., 2012; Chindime et al., 2016) after the network visualization. These included degree, betweenness, and closeness centrality measures as well as overall network measures, such as density. (For detailed information on these parameters including mathematical expressions, please refer to Wasserman and Faust, 1994). The personal interviews with the NF participants were supplemented with three focused group discussions (FGDs) each at locations A and B and five at location C. Furthermore, key members who were directly involved in the project, and holding specific positions, were identified during the FGDs. Other important contributors included officials from the DoAH, Dairy Development Department, Panchayath Raj Institution, SPEM, NF Promotion Council (NFPC), Local Milk Cooperative Society, Input dealers, and Women Group Leaders. As the investigation was done at the micro level, most of the actors engaged in the project were identified and named during the time of FGDs, who were then contacted to obtain the required information. The entire data collection exercise was conducted between January and March 2017.

Results

Before venturing into network analysis, a comparison of some of the important socioeconomic and social network variables was made among the three villages. One-way analysis of variance (ANOVA) was used to check the difference. For a group or organization within a community to function properly, its members must have commonly shared objectives and fairly similar social/economic backgrounds (Clark, 1986). The analysis was thus performed with this rationale in mind. A significant difference was observed among the respondents with respect to their operational landholding, between the various socioeconomic parameters, and in terms of bonding as well as community association networks in the case of network variables. Details of the results are presented in Table 1.

Table 1.

An overview of important socioeconomic and social network variables in the study locales.

No.	Variable	Kannadi	Puduppariyaram	Akathethara	p Value
No.	Variable	Mean ± SD	Mean ± SD	Mean ± SD	p Value
1	Age (number)	35.3 ± 7.66	36.58 ± 9.33	34.5 ± 5.42	0.38
2	Education (number of years)	10.04 ± 2.71	9.02 ± 2.81	10.07 ± 2.28	0.07
3	Operational holding (in cents)^a	9.34 ± 4.45	7.67 ± 5.97	6.89 ± 4.72	0.04^*
4	Years of experience in dairying (number)	21.62 ± 10.87	25.84 ± 15.34	22.96 ± 11.70	0.24
5	Total livestock unit	6.7 ± 6.37	5.56 ± 4.29	5.88 ± 3	0.46
6	Bonding network (number)	1.54 ± 0.9	1.14 ± 0.92	1.92 ± 1	0.00^**
7	Extension network (number)	1.12 ± 0.88	1.14 ± 0.96	1.34 ± 1.2	0.54
8	Market network (number)	0.96 ± 1.1	0.86 ± 1.01	0.71 ± 0.89	0.45
9	Community association network (number)	2.06 ± 1.6	2.12 ± 1.22	2.75 ± 1.18	0.01^**

SD: standard deviation.

^a1 cent = 40 m².

* and ** indicates significance at 5 and 1 per cent level, respectively.

It was seen that the respondents of all three villages were almost similar with respect to various socioeconomic characteristics, except in operational landholding. The results of one-way ANOVA indicated that there was no significant difference (p value) among the respondents. As indicated by the mean values, NF participants were middle-aged and had 10 years of education on average. The operational holding varied significantly with the respondents at Kannadi having the highest mean holding (9.34 cents) and that of Akathethara, the lowest (6.89 cents). In all the villages, the respondents had more than 20 years of experience in dairying on average. Total livestock units—an indicator of the possession of cattle wealth—also were not significantly different, though the respondents of Kannadi had the highest number (6.7 units) on average. However, a significant difference was there among them with respect to some of the important social network variables. The bonding network (relatives and family members of respondents reckoned by the respondent as an important contact with respect to dairying) and community association network (acquaintances in the joint liability groups (JLGs) of respondents) were found to vary significantly (p < 0.01). But the extension network (various formal and informal extension contacts on dairying) and the market network (various information sources related to marketing) of respondents did not vary across the villages. Thus, it became interesting to draw the network maps of these three villages, and to analyze the overall network dynamics, by comparing the various network parameters.

Network structures in the study locales

The whole network of maps characterized all the actors in the respective networks in the three villages. The maps of different NF projects’ locales are presented below (Figures 3 to 5). The social networks of three project locales, Kannadi, Puduppariyaram, and Akathethara, are denoted as Net-A, Net-B, and Net-C, respectively. The legend codes to identify individual actors in the network are presented in Table 2. A visual inspection of the network maps would suggest that while Net-A is comparatively disintegrated, Net-B and Net-C are more organized with the latter having more ties in the periphery.

Figure 3.

Social network of Kannadi village. Actor codes: A1-A50 project participants; CS: official of milk cooperative society; Vet: veterinary surgeon; InD: input dealer; VNF: milk vendor; FF: friend/relative/family member/neighbor; CO: official of SRLM.

Figure 4.

Social network of Puduppariyaram village. Actor codes: A1-A50 project participants; CS: official of milk cooperative society; Vet: veterinary surgeon; InD: input dealer; VNF: milk vendor; FF: friend/relative/family member/neighbor; P: progressive farmer; AO: agricultural officer; BWS: beer waste trader; PAN: official of Panchayath.

Figure 5.

Social network of Akathethara village. Actor codes: A1-A52 project participants; CS: official of milk cooperative society; Vet: veterinary surgeon; InD: input dealer; VNF: milk vendor; FF: friend/relative/family member/neighbor; P: progressive farmer; GAC: Member of Guidance and Advisory Committee; AC: Official of Nature Fresh Promotion Council; PAN: official of Panchayath.

Table 2.

Network actors with highest centrality measures in various study locales.

Study villages	Degree centrality	Betweenness centrality	Closeness centrality	Nodes/Edges
Kannadi	Vet (20)	A30 (0.006)	A1 (1)	62/77
	InD (12)	A24 (0.005)	A44 (1)
	A24 (7)	A43 (0.003)	A20 (1)
Puduppariyaram	Vet (26)	A44 (0.023)	A5 (1)	81/147
	A33 (14)	A33 (0.018)	A16 (1)
	VNF (13)	A27 (0.014)	A33 (1)
Akathethara	Vet (49)	A46 (0.005)	A1 (1)	81/185
	AC (11)	A40 (0.005)	A3 (1)
	GAC (15)	A18 (0.005)	A8 (1)

Vet: veterinary surgeon; VNF: milk vendor; AC: Official of Nature Fresh Promotion Council; GAC: Member of Guidance and Advisory Committee.

Values in parenthesis denote the specific network measure of the particular node.

In Table 2, actors with the highest measure for a particular network parameter were classified as having an understanding of the prominence of actors in the networks. Those actors with the highest degree centrality measures have more connections with other actors—indicating their prominence in the network. Such actors were found to be at the center of a network. Moreover, those highest in betweenness are said to occupy strategic positions in the network with a higher degree of information brokerage potential and closeness to other actors in the case of higher closeness scores.

From the network diagrams and degree centrality measures, it was clear that village veterinary surgeons were the central actors in all the locales. Their position is critical as the change agent’s structural position affects their ability to introduce changes in the network. Putting it differently, actors at the core of the network have control over new information or ideas reaching others, while those who are at the periphery have little control over network dynamics. In case of other two centrality measures (betweenness and closeness), project participant farmers had the higher scores.

Going further, it became clear that the spread of information was predominantly in a vertical manner in Net-A and Net-B, whereas it was flowing horizontally as well in Net-C. This is indicated by the greater number of ties in the network periphery in Net-C as compared to the other two locales. More horizontal ties between peripheral actors could be construed as better information contacts among these actors. Hence, it was evident that interaction and knowledge flow happened more rigorously in Net-C as compared to other locales, indicating a well-connected network. However, dependency of the majority of actors on a few other actors for accessing or exchanging information can’t be overlooked, as it hints at possible constraints in the information network. However, the project participants had limited interaction with the members of village-level milk cooperatives, functioning in all the three study locales. This might be due to lack of transaction between milk producers and milk cooperatives within the operation of the NF project. Discussions with the key participants also reinforced this finding. It could be seen that each network was composed of two distinct elements, namely information acquisition network (IAN) dominated by the formal communication sources as well as the information exchange networks (IEN) led by the peer farmers, which also has bidirectional ties, as evident from the network diagrams. To gain a better understanding of this aspect, the nature of ties in various networks was derived to see the extent of utilization of the formal and informal connections for information flow in the networks. Though Net-C had the maximum number of ties (200), formal ties were present almost equally (37.5%) in Net-A and Net-C, while informal ties were more in Net-B (73.15%; see Table 3).

Table 3.

Nature of the ties in various networks.

Nature of ties	Kannadi (Net-A; %)	Puduppariyaram (Net-B; %)	Akathethara (Net-C; %)
Formal	29 (37.66)	40 (26.85)	75 (37.5)
Informal	48 (62.34)	109 (73.15)	125 (62.5)
Peer only	25 (32.47)	43 (28.86)	54 (27)
Total	77 (100)	149 (100)	200 (100)

However, the content of information exchanged within their networks was almost the same in all the networks. The content was mostly about sources and types of cattle feed, use of mineral supplements, disease management, and cattle stabling management. Further, the participation of private actors was found to be limited in all the networks, which is not a desirable feature.

Analysis of network parameters

Network density—a measure of existing ties as a percentage of all possible ties—was found to be lower in all the networks, though it was highest in Net-C (2.5%). This is also reflected by the average degree (2.22) of the actors in Net-C, which is independent of the network size. The measure is a direct indicator of the number of connections each actor has within a network. Then, we made a comparison of various centrality measures in social networks following earlier studies. The measures were calculated and compared and checked for difference with respect to their mean values using one-way ANOVA. The standardized network measures were derived so as to have a comparison among social networks. The results are presented in Table 4.

Table 4.

Network centrality measures of the study locales.

Network measures	Kannadi (A)	Puduppariyaram (B)	Akathethara (C)	p Value
Indegree centrality	1.55 ± 1.36	0.93 ± 1.78	2.54 ± 2.51	0.00^**
Outdegree centrality	1.55 ± 2.68	0.93 ± 2.11	2.54 ± 4.36	0.003^**
Betweenness centrality	0.035 ± 4.09	0.12 ± 0.37	0.075 ± 1.14	0.06
Closeness centrality	1.49 ± 0.073	1.21 ± 0.14	1.21 ± 0.09	0.00^**

Mean ± standard deviation values are given in the table.

**means values are significant at 1 per cent level.

The mean values differed significantly with respect to indegree, outdegree, closeness, and information centrality measures. Net-C had higher mean in and out degree scores, signifying the higher number of information receiving and giving ties for its members. Surprisingly, average betweenness score was found to be highest for Net-B (0.12) rather than for Net-C, following an expected correlation among various network measures. It indicated a higher degree of mediating or brokerage position owned by its members. In other words, more members of Net-B were acting as information bridges connecting fragmented actors to the network, than the other two networks. However, the measure did not vary significantly among the three networks. The mean closeness measure was found to be higher for Net-A, meaning that the distance for the network members to communicate or exchange resources is greater when compared to the other two networks. In other words, shorter distance between the nodes, both in Net-B and Net-C, indicated more productive information spread in those networks.

Discussion

The study focused on the social networks in the project locales as they have a vital role in the exchange and practice of information in agricultural systems (Isaac, 2012). The project was basically comprised of the same set of actors in all the three villages. An actor with the project implementing agency, that is the veterinary surgeon, was the predominant information source in all the networks. Their position is important as the change agent’s structural position affects their ability to introduce changes in the organization (Battilana and Casciaro, 2012). In other words, actors at the core of a network have control over new information or ideas reaching others, while those at the periphery are far removed from day-to-day operations of a network (Cross et al., 2001; Mittal et al., 2018). Ideally the project should have been driven by an alliance of actors, with the active participation of project participants in its governance (Hall et al., 2003). Many of the actors who are well-connected to the dairy producers, such as livestock input dealers, should have been brought into the alliance to realize better dynamics of the social networks. Further, Net-C was found to have an edge over the other networks because it was able to facilitate both vertical and horizontal flow of information (Pramila Krishnan, 2012; Reed and Hickey, 2016). More horizontal ties between peripheral actors could potentially increase collective actions in support of innovation (Reed et al., 2009) by channelizing better flow of information (Pretty and Smith, 2004). Moreover, these ties are crucial in determining the performance of the whole network vis-à-vis their individual capabilities and roles (Hermans et al., 2013). Moreover, the better integration of IEN and IAN network with higher mean centrality measures indicates better networking and information spread within Net-C. In short, their combination denotes the right mix of strong (IEN) and weak ties (IAN), which are complementary in nature (Granovetter, 1983). One of the respondents from Akathethara opined

Though we all were members of the SPEM groups earlier also, we rarely used to discuss topics related to dairying. Once we became the project participants, we got a new platform to discuss many of the important issues, including cattle feed management and health care. More experienced members share valuable information on cattle management while our veterinary doctor shares information on new information on scientific dairy farming. These interactions help us to remain up-to-date in dairying, and gives a platform to share our own issues and concerns in dairying with peer farmers and experienced farmers.

At the time of data collection, the project was already functional in all the villages with varying levels of actor engagement and style of functioning. The nature and pattern of network ties were almost similar, in the first two villages, but they were distinctly different at Akathethara. There, the project got institutionalized as a farmer producer organization called NFPC, primarily comprising the milk producers. The council played an instrumental role in changing network dynamics by organizing regular meetings of the project participants. Moreover, a Guidance and Advisory Committee (GAC), made up of experienced dairy farmers in the village, was constituted to support the project participants. GAC played key roles ranging from provision of dairy information support to the project members to resolving conflicts with the milk consumers, in case of quality concerns regarding the milk. In terms of information flow, these meetings catalyzed integration of formal and personal information networks. Moreover, the Council played an instrumental role in instituting new network linkages. For example, NFPC took the lead in establishing linkage with Komerala Feeds, a cattle feed firm, for the benefit of project participants. The feed was distributed to the members at a subsidized rate. Sometimes, NFPC meetings were also attended by the officials of the local cooperative bank to discuss financial matters, including availability of credit, repayment, and so on by the members. Thus, NFPC played the key role of a facilitator by catalyzing network interactions and linkages and influencing its overall dynamics.

With respect to the network parameters, higher network size and density (comparatively) of Net-C also signifies higher level of interactivity and faster communication within the network (Chaudhury et al., 2017; Helsley and Zenou, 2014), and innovativeness (Davis et al., 2008). Moreover, as indicated by the higher mean degree and closeness centrality, Net-C was found to be better organized with the presence of a greater number of ties (Beauchamp, 1965; Yin et al., 2006), whereas Net-B has more information brokers signified by higher mean betweenness value (Kwon et al., 2015). It can be deduced that mobilization of the project participants under a farmer-led organization (NFPC) has positively contributed to better information flow and functional dynamism in Net-C. When successful collective actions take place in a value chain, the social dynamics enhance the capacities of farmers connected to it (Hodge and Reader, 2007). The project, governed by NFPC, a farmer-led cooperative at Akathethara, reiterates this argument. According to one NF group leader at Akathethara,

With the formation of NFPC, the meetings became a regular affair and we could raise many of our concerns during these meetings. The presence of the village veterinarian is a boon as we could seek more information on dairying, such as feeding and health care of our animals, which was missing earlier. The presence of Guidance and Advisory Committee members is again an added advantage as they are experienced farmers who provide much vital information, including indigenous technical knowledge on dairying. Besides, they actively engage in resolving conflicts with the consumers and milk vendors. Moreover NFPC facilitates addressing our concerns like revision of milk prices and procurement of inputs.

Thus the role of NFPC was instrumental in integrating both formal and informal information sources (Demiryurek, 2010). Altogether, the results demonstrate that the network structure and project dynamics were better at Akathethara, with better actor alignment and networking that facilitated easy flow of information and other resources. However, we couldn’t look into the network structures of dairy producers before the implementation of the NF project to have an understanding on the influence of the project on network structuring, which therefore remains as the main limitation of our study.

Conclusions

This study set out to analyze the configuration of social networks formed in response to a dairy development project in three adjoining villages. The social networks displayed differences with respect to alignment of actors and dynamism, though all of them were basically comprised of a similar set of actors. From the network analysis, it became evident that the project was primarily driven by a public sector actor in all the locales. Further, the network of actors in the project tended to integrate better at Akathethara, where the project started last. The network dynamics there was stimulated by the integration of professional and personal networks to a greater extent. The networks were also found to vary in terms of some of the network attributes, indicating varying levels of network integration and brokerage potential. The network maps of project villages generated can become a key resource for project planners—to realign the existing actors or to connect new actors. This would also be helpful to deal more effectively with uncertainties such as disconnect of network actors and information bottlenecks.

Footnotes

Acknowledgement

We acknowledge the support of NF project implementing agency, SPEM, officials of NFPC, officials of dairy development department, and the project participating farmers for their cooperation during the course of this study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sreeram Vishnu

Note

References

Abizaid

Coomes

Takasaki

, et al. (2015) Social network analysis of peasant agriculture: cooperative labor as gendered relational networks. The Professional Geographer 67(3):447–463.

Agriculture Development Policy (2015) Department of Agriculture. Government of Kerala. Available at: https://www.kerala.gov.in/documents/10180/46696/AGRICULTURAL%20DEVELOPMENT%20POLICY%202015 (accessed 13 October 2017).

Asres

Sölkner

Puskur

, et al. (2012) Livestock innovation systems and networks: findings from smallholder dairy farmers in Ethiopia. Livestock Research for Rural Development 24(9):1–21.

Barrett

Bachke

Bellemare

, et al. (2012) Smallholder participation in contract farming: comparative evidence from five countries. World Development 40(4):715–730.

Battilana

Casciaro

(2012) Change agents, networks, and institutions: a contingency theory of organizational change. Academy of Management Journal 55:381–398.

Beauchamp

(1965) An improved index of centrality. Behavioral Science 10(2):161–163.

Bijman

Hendrikseen

(2003) Co-operatives in chains: institutional restructuring in the Dutch fruit and vegetable industry. Journal on Chain and Network Science 3(2):95–107.

Borgatti

(2009) On social network analysis in a supply chain context. Journal of Supply Chain Management 45(2):5–22.

Caria

Fafchamps

(2015) Can farmers create efficient information networks? Experimental evidence from rural India. CSAE Working Paper.

10.

Ceglie

Dini

(1999) SME Cluster and Network Development in Developing Countries: The Experience of UNIDO. Vienna: UNIDO.

11.

Chaudhury

Thornton

Helfgott

, et al. (2017) Ties that bind: local networks, communities and adaptive capacity in rural Ghana. Journal of Rural Studies 53:214–228.

12.

Chaudhary

Radhakrishna

(2018) Social network analysis: a methodology for exploring diversity and reach among extension programs and stakeholders. Journal of Extension 56(6):6FEA1.

13.

Chindime

Kibwika

Chagunda

(2016) Positioning smallholder farmers in the dairy innovation system in Malawi: a perspective of actors and their roles. Outlook on Agriculture 45(3):143–150.

14.

Cinelli

Ferraro

Iovanella

(2017) Structural bounds on the dyadic effect. Journal of Complex Networks 5(5):694–711. Available at: https://dx-doi-org.web.bisu.edu.cn/10.1093/comnet/cnx002 (accessed 13 March 2019).

15.

Clark

(1986) Extension methods involving community organization and local involvement. In: Jones

(ed) Investing in Rural Extension: Strategies and Goals. London: Elsevier Applied Science Publishers, pp. 193–200.

16.

Cross

Liedtka

Weiss

(2005) A practical guide to social networks. Harvard Business Review 83(3):124–132.

17.

Cross

Parker

Prusak

, et al. (2001) Knowing what we know: supporting knowledge creation and sharing in social networks. Organizational Dynamics 30(2):100–120.

18.

Daviron

Gibbon

(2002) Global commodity chains and African export agriculture. Journal of Agrarian Change 2(2):137–161.

19.

Davis

Ekboir

Spielman

(2008) strengthening agricultural education and training in Sub-Saharan Africa from an innovation systems perspective: a case study of Mozambique. The Journal of Agricultural Education and Extension 14(1):35–51.

20.

De Janvry

Sadoulet

Zhu

(2005) The role of non-farm incomes in reducing rural poverty and inequality in China. UC Berkeley Department of Agricultural and Resource Economics, UCB. CUDARE Working Paper No. 1001. Available at http://escholarship.org/uc/item/7ts2z766 (accessed 3 February 2014).

21.

Demiryurek

(2010) Analysis of information systems and communication networks for organic and conventional hazelnut producers in the Samsun province of Turkey. Agricultural Systems 103(7):444–452.

22.

Dicecca

Pascucci

Contò

(2016) Understanding reconfiguration pathways of agri-food value chains for smallholder farmers. British Food Journal 118:1857–1882.

23.

Eshuis

Stuiver

(2005) Learning in context through conflict and alignment: farmers and scientists in search of sustainable agriculture. Agriculture and Human Values 22(2):137–148.

24.

Folder

Cavaye

and others (2015) The role of social capital in Tasmanian cherry industry value chain. Rural Extension and Innovation Systems Journal 11(1):43.

25.

Granovetter

(1983) The strength of weak ties: a network theory revisited. Sociological Theory 1:201–233.

26.

Guillemois

(2013) The use of social network theory in value chain analysis. No. 309-2016-5285. 2013.

27.

Hall

Sulaiman

Clark

(2003) From measuring impact to learning institutional lessons: an innovation systems perspective on improving the management of international agricultural research. Agricultural Systems 78:213–241.

28.

Hall

Sulaiman

Bezkorowajnyj

(2008) Reframing technical change: livestock fodder scarcity revisited as innovation capacity scarcity: part 2. a framework for analysis. MERIT Working Papers 003, United Nations University - Maastricht Economic and Social Research Institute on Innovation and Technology (MERIT). Available at: https://hdl.handle.net/10568/277 (accessed 2 December 2018).

29.

Helsley

Zenou

(2014) Social networks and interactions in cities. Journal of Economic Theory 150:426–466.

30.

Hermans

Stuiver

Beers

, et al. (2013) The distribution of roles and functions for upscaling and outscaling innovations in agricultural innovation systems. Agricultural Systems 115:117–128.

31.

Hermans

van Apeldoorn

Stuiver

, et al. (2013) Niches and networks: explaining network evolution through niche formation processes. Research Policy 42(3):613–623.

32.

Herrero

Grace

Njuki

, et al. (2012) The roles of livestock in developing countries. Animal: An International Journal of Animal Bioscience 7:1–16.

33.

Hoang

Castella

Novosad

(2006) Social networks and information access: implications for agricultural extension in a rice farming community in northern Vietnam. Agriculture and Human Values 23(4):513–527.

34.

Hodge

Reader

(2007) Maximising the provision of public goods from future agri-environment schemes. Final Report for Scottish Natural Heritage, Rural Business Unit, Department of Land Economy, University of Cambridge (accessed 23 March 2016).

35.

Isaac

(2012) Agricultural information exchange and organizational ties: the effect of network topology on managing agrodiversity. Agricultural Systems 109:9–15.

36.

Johny

Wichmann

Swallow

, et al. (2014) Role of social networks in diversification of income sources in rural India. In: Selected paper at the Agricultural and Applied Economics Associations 2014 (AAEA) Annual Meeting, Minneapolis, Minnesota, 27–29 July 2014.

37.

Kwon

Y-I

Jeong

D-H

Moon

, et al. (2015) Comparing analysis study of centrality indices using paper information on secondary battery. Indian Journal of Science and Technology 8(S1):333–339.

38.

Mittal

Padmaja

Ajay

(2018) Agricultural information and knowledge network in rural India: a case of Bihar. The Journal of Agricultural Education and Extension 24(5): 393–418. DOI: 10.1080/1389224X.2018.1491871.

39.

Novkovic

Holm

(2012) Co-operative networks as a source of organizational innovation. The International Journal of Co-operative Management 6:51–60.

40.

Odero-Waitituh

(2017) Smallholder dairy production in Kenya; a review. Livestock Research for Rural Development 29(7):Article #139. Available at: http://www.lrrd.org/lrrd29/7/atiw29139.html (accessed 3 July 2018).

41.

Patil

Hiremath

Lokesh

. (2014) Economic sustainability through farmers interest groups and their linkage with institutional agencies – an evidence from Karnataka. Agricultural Economics Research Review 27: 141–146.

42.

Powell

Bromley

(2015) New institutionalism in the analysis of complex organizations. International Encyclopedia of the Social and Behavioral Sciences 2(4): 764–769.

43.

Pramila Krishnan

(2012) Neighbours and Extension Agents in Ethiopia. Who Matters More for Technology Diffusion? CEPR Discussion Paper No. DP9539. Available at SSRN: https://ssrn.com/abstract=2291340 (accessed 20 January 2015).

44.

Pretty

Smith

(2004) Social capital in biodiversity conservation and management. Conservation Biology 18(3):631–638.

45.

Randolph

Schelling

Grace

, et al. (2007) Invited review: role of livestock in human nutrition and health for poverty reduction in developing countries. Journal of Animal Science 85(11):2788–2800.

46.

Rangnekar

(2006) Livestock in the Livelihoods of the Underprivileged Communities in India: A Review. ILRI (aka ILCA and ILRAD). Available at: https://hdl.handle.net/10568/29061 (accessed 15 September 2017).

47.

Reardon

Barrett

Berdegué

, et al. (2009) Agrifood industry transformation and small farmers in developing countries. World Development 37(11):1717–1727.

48.

Reed

Hickey

(2016) Contrasting innovation networks in smallholder agricultural producer cooperatives: insights from the Niayes Region of Senegal. Journal of Co-operative Organization and Management 4(2):97–107.

49.

Reed

Graves

Dandy

, et al. (2009) Who’s in and why? A typology of stakeholder analysis methods for natural resource management. Journal of Environmental Management 90(5):1933–1949.

50.

Singh

Meena

Swanson

(2013) Extension in India by public sector institutions: an overview. Available at SSRN 2315457 .

51.

Sinyolo

Mudhara

(2018) Collective action and rural poverty reduction: empirical evidence from KwaZulu-Natal, South Africa. Agrekon 57(1):78–90.

52.

Spielman

Davis

Negash

, et al. (2007) Organic Rural Innovation Systems and Networks: Findings From a Study of Ethiopian Smallholders. pp. 225–237.

53.

Spielman

Davis

Negash

, et al. (2011) Rural innovation systems and networks: findings from a study of Ethiopian smallholders. Agriculture and Human Values 28(2):195–212.

54.

Sulaiman

Vamsidhar

Reddy TS

(2015) Policy incoherence in smallholder dairying in Bihar, India. ILRI Discussion Paper 33. Nairobi, Kenya, International Livestock Research Institute (ILRI). Available at: https://hdl.handle.net/10568/67182 (accessed 24 May 2017).

55.

Vamsidhar Reddy

Sulaiman

(2016) Smallholder Dairy Transformation and Innovation in Bihar, India. ILRI (aka ILCA and ILRA).

56.

Vishnu

Gupta

(2017) Evolving localized innovation system: the case of Imasree farmer producer company. Rural Extension and Innovation System Journal 13(2):96–105.

57.

Wasserman

Faust

(1994) Social Network Analysis: Methods and Applications. Cambridge: Cambridge University Press.

58.

Weyori

Amare

Garming

, et al. (2017) Agricultural innovation systems and farm technology adoption: findings from a study of the Ghanaian plantain sector. The Journal of Agricultural Education and Extension 24(11):1–23.

59.

World Bank (1991) Smallholder dairy development project. Project completion report (Zambia). Report no.9936. Agriculture Operations Division, Southern Africa Department.

60.

Yin

Kretschmer

Hanneman

, et al. (2006) Connection and stratification in research collaboration: an analysis of the COLLNET network. Information Processing and Management 42(6):1599–1613.