The GGAVATT Tepetzintla

The first group that worked with organized livestock producers in a technology validation and transfer program is located in Tepetzintla (^{González, 2013}). In the Production Units of this region, in the 1970s, beef systems prevailed (cow-offspring) and double-purpose cattle (beef and milk). During those years, in Rancho Bella Esperanza, located in this municipality, the process of technological validation and adoption began, generated by INIFAP in La Posta Experimental Field, including genetic improvement of its animals through artificial insemination with semen from European bull breeds, and other activities such as dietary supplementation and double milking. In 1980 the results were communicated to livestock producers and authorities in the region (^{Rodríguez, 2010}).

Due to the interest shown by some producers to learn about the actions that were being carried out in Rancho Bella Esperanza, an invitation was made through the local Livestock Production Association in Tepetzintla to all the members for them to become organized as a group. In this way, in 1983, the Tepetzintla Livestock Program (Programa Ganadero Tepetzintla, PROGATEP) was born, with the supervision of La Posta Experimental Field and the Technical Consultancy from the Rural Development District 002 in Tuxpan, Veracruz. However, it was not until 1989 when the first GGAVATT in México was formed (^{INIFAP, 2005}).

During its first years, the GGAVATT Tepetzintla was made up of 28 producers. However, due to the frequent disputes over leadership in the group, it divided and only nine actors remain.

The GGAVATT Vía Corta

This GGAVATT is located in the municipality of Ozuluama, in the state of Veracruz. The members of this group learned about the GGAVATT model in the years 1991 and 1992, when they attended the GGAVATT Tepetzintla’s evaluations. Once they were convinced of the goodness of the model they decided to become integrated as GGAVATT Vía Corta. At the beginning there were 19 members; however, in 1997 they decided to divide into two groups, giving rise to the GGAVATT La Rivereña and the GGAVATT Vía Corta. In this way, the GGAVATT Vía Corta remained integrated by nine members.

The GGAVATT Caprinocultores Unidos Región Montañosa de Veracruz

The producers in this group belong to different municipalities of Veracruz: Coatepec, Coacoatzintla, Xico and Emiliano Zapata (^{Aguas, 2011}). In February 2002, in the Rural Development Training Unit No. 2 (UNCADER No. 2), various talks were presented concerning goat milk production and artisanal cheese elaboration. Five and a half years later, in August 2007, in the UNCADER facilities, a course-workshop was imparted - workshop on fine goat cheeses -, which marked the beginning of goat production in the state of Veracruz. One year later, in August 2008, a group of goat milk producers, officials and researchers attended the 9^th International Conference on Goats (IGA) in Querétaro, an event where diverse information about goat products was presented (^{SIPECAV, 2012}).

With this background, the first GGAVATT in Veracruz for goat milk goat producers was constituted in March 2008, made up of 11 producers that belonged to four municipalities. Currently the GGAVATT Caprinocultores Unidos Región Montañosa de Veracruz, is integrated by eight producers (^{Aguas, 2011}).

Research Instrument

To understand the interactions there are between different actors of the GGAVATTs, a questionnaire was designed which was applied from May to August 2013 to a total of 26 actors, members of these three GGAVATTs. This instrument allowed obtaining information from three temporal periods (initial T1, intermediate T2, and the one at the moment of the study T3), when the model has been applied. The instrument sought data about: the dynamics of technology adoption, the links between the producer and different actors in the network, and the socioeconomic characteristics of the members in each group.

The information gathered was organized and processed in a database in Microsoft Office Excel 2010, classifying it into the variables of: structure of the network, and social, socioeconomic interactions and of technology adoption.

To analyze the relational structure of the actors, the UCINET 6 for Windows-Version 6.374 software was used (^{Borgatti et al., 2002}). The variables analyzed were network density (number of relationships divided by the number of possible relationships), nodal degree (number of actors to which an actor is directly united), and intermediation (possibility that a node has to intermediate the communications between pairs of nodes) (^{Wasserman and Faust, 1999}; ^{Hanneman, 2000}; ^{Velázquez and Aguilar, 2005}).

On the other hand, the socioeconomic level is a total measurement that combines the economic and sociologic part of the labor preparation of a person and the individual or familiar economic and social position, with regard to other people. It includes three basic aspects: economic capital, human capital and social capital. The economic capital is measured, approximately, by family wealth or income. Human capital is measured, approximately, by education. The social capital of the family consists in the relationships between children and parents, among other actors (^{Haretche, 2011}; ^{Vera and Vera, 2013}). In this study the socioeconomic variables considered were: age, schooling, size of the production unit, and amount of livestock.

With regard to technology adoption, the questionnaire included four spheres: health, food and nutrition, reproduction and genetics, and economic-administrative aspects. In these spheres, 15 innovations were defined in total (Table 1), which have been implemented in the production units by extension workers from institutions like INIFAP, SAGARPA, SEDARPA, Universidad Veracruzana, Colegio de Postgraduados, and UNAM, among others.

Table 1 Innovations implemented in the GGAVATTs. 

Source: authors’ elaboration, based on the methodology of the GGAVATT model.

The index of innovation adoption was estimated by using the equation (^{Muñoz et al., 2004}):

where: INAM = Index of innovation adoption per module per group; ΣXi/n = Value of the indicator per group; n = Total number of variables per group.

Through the sum of adoption indexes per groups, the total adoption index was constructed, using the expression:

where: INAT = Index of total innovation adoption ; ΣIAIC_k = Sum of values of indicators per group; k = Total number of indicators per group.

For the statistical analysis of the data, an exploratory analysis of the data was carried out (^{Tukey, 1977}), using Statistica 7.1 (^{Stat Soft Inc., 2006}). On the other hand, the Spearman correlation was used, considering as independent variables the size and density of the network, the nodal degree and intermediation and the socioeconomic conditions of the actors. The dependent variable considered was the technology adoption index.

Structure of the GGAVATT network and their relationship with technology adoption

In the three GGAVATTs studied, it was observed that the size of the network increased in all of the cases in the T2; however, in the T3 it suffered a decrease in the GGAVATTs Tepetzintla and Vía Corta. In the case of goat producers, the network continues to grow, which according to what was observed in the other two GGAVATTs is normal, for in the first years there is a constant growth of the network, to later be stabilized or else decrease (Figure 1).

Figure 1 Evolution of the three GGAVATT. 

In the GGAVATT Tepetzintla the network grew substantially in the second stage (from 26 to 59), which caused a decrease in the density of the network. It was also observed that the TAIs increased. In the GGAVATT CURMV the network grew significantly in stage 2, which caused for the density to decrease. In turn, the TAIs increased significantly.

The study results are similar to the ones found by ^{Muñoz et al. (2004)}, with citrus producers from Michoacán, where they report that the propensity to adopt innovations is higher when the network is smaller and there is a high density. In this regard, the authors maintain that the relationships relevant for innovation are characterized by their weakness. According to ^{Granovetter (2003)} the reason for this effect is that individuals and organizations with close relationships between them come to have the same pattern of opinion about various issues, while the actors with weak connections tend to move in more diverse social circles, allowing them a broader access to information, and therefore a higher ability to choose. The number of weak links is increased with the development of communication systems, bureaucratization, and population density.

However, the GGAVATT Tepetzintla’s network, suffered a decrease in size during stage 3 (from 69 to 60), and in contrast the density increased; on the other hand, the TAIs had a slight decrease. In the GGAVATT CURMV, also during stage 3, the size of the network continued increasing, although the density increased slightly, suggesting that the information in the network is reaching many of the members, which is reflected in the gradual increase of their TAIs, linked to the existing information that is found circulating through the network, fostering technology adoption in most of the producers.

In the case of the GGAVATT Vía Corta, the density of the network has remained at around 10b%, for although the network has grown in size, it has done so slightly, contributing to the network remaining cohesive.

However, the case of the GGAVATTs differs from what was reported by ^{Monge and Hartwich (2008)}, in a study about agricultural innovation adoption in Bolivia, where they found a positive and significant association between the density of the network and the index of adoption. In this case the authors reported a density of 33.7 %.

In the case of the low densities found in the study networks (3.5 to 8 %), this agrees with what was reported by ^{Zarazúa et al. (2011)} in a study with strawberry producers in Michoacán, where it was 1.58 %.

For the case of technology adoption, it has increased in the three GGAVATTs in a first stage. Later they have entered a phase of stagnation and they have even suffered a slight decrease in the case of the GGAVATTs Tepetzintla and Vía Corta; in contrast, the GGAVATT CURMV continues to increase, situation that is attributed to the presence of a higher number of actors with whom there are weak links, as ^{Granovetter (2003)} argues, for whom weak links provide bridges over which information crosses the limits of social groups, while the influence on decision making is exerted primarily through the network of strong links in the core of each group.

Our study agrees with what was reported by ^{Perea (2010)} in a research study carried out in Michoacán with sheep producers, where he suggests that the social relationships favor to a higher degree the adoption of innovations through information flows between actors of the network. In social relations, in a network that is of greater size and density, the spread of knowledge within it increases. However, the author reports low densities, of 2.33 % and 3.39b%, which are similar to those found in this study.

Social interactions in the GGAVATTs and their relationship with adoption of technology

In the GGAVATT Tepetzintla, when there was a substantial increase in the nodal degree and the intermediation, the indexes of technology adoption also increased. However, when the TAIs reach a point where they have stabilized, the nodal degree and the intermediation also have remained without great changes. That is, there has not been a substantial increase in interactions with new actors, so the information that circulates through the network has been maintained with the existing intermediation (Table 2).

Table 2 Evolution of the means of the TAI, nodal degree and intermediation in the three stages of the GGAVATT studied. 

Fuente: elaboración propia.

In the GGAVATT Vía Corta, the nodal degree and the intermediation have increased slowly, but constantly. On the other hand, the TAIs increased in the first stage, but they have become stagnant in recent years because of the prevailing conditions of insecurity in the region, characterized by kidnapping and organized crime. However, the fact that the nodal degree and the intermediation continue to increase leads us to think that the TAIs can increase in the future due to the social capital present in the group.

In the GGAVATT CURMV, the TAIs have increased gradually, together with the nodal degree. However, with regard to intermediation, it increased in stage 2, though not in stage 3, which is attributed to the substantial increase of the links with different actors, which indicates that the social capital is under construction, and as it increases, the actors are in possibilities of intermediating the communication in the network.

In the three case studies, leaders who had high technology uses (70 % or more) and high centrality indexes (14 or more) were found, from the beginning, and the latter remained in the three periods, showing few changes in the TAIs. In general, it was found that in at least one of the stages evaluated it was possible to establish a positive association between the centrality indexes (nodal degree and intermediation), and the technology adoption indexes of each actor in the network. The most evident case was that of the GGAVATT Tepetzintla, where in the three stages analyzed it was observed that the higher the degree of centrality, the higher the technology adoption index; therefore it was concluded that the higher degree and intermediation, the higher the TAI is, for there is a higher amount of information circulating through the network and there are diverse possibilities of gaining access to it.

This agrees with what ^{Rogers (1995)} exposes, who states that those members of a system who are involved with the greatest number of actors in a network are opinion leaders. In this sense, the leadership of opinion is the degree to which an individual is capable of influencing the attitudes of individuals or the manifest behavior in a desired manner, with relative frequency. The opinion leaders are individuals who influence the opinions of others about the innovations.

The study results coincide with what was observed by ^{Monge and Hartwich (2008)}, who visualize the spread as a communication process, determined by factors of social cohesion such as intermediation and nodal degree, among others. In this sense, the actors with highest levels of nodal degree tend to be opinion leaders, who usually adopt the culturally acceptable innovations earlier than the rest, and they show themselves to be opponents of those culturally inacceptable. According to this, the nodal degree is a good measure of the immediate influence, that is, of the probability of “infection” as a function of the number of actors with whom the producers are linked.

This also agrees with what was reported by ^{Muñoz et al. (2004)}, who argue that the actors have a series of specific attributes and occupy a different position in the network. The actors with high degree of centrality stand out because they are the first adopters of innovations, as a result of registering a relatively high adoption index, and because they show a strong propensity to issuing information. The actors with highest degrees of intermediation stand out because of their higher relationship density, their great ability for adoption, and from a higher propensity to receive information.

The GGAVATT Tepetzintla

The mean age of the members of this GGAVATT at the moment of the study was 60 years; all the members know how to read and write, and they have a schooling average of 8 years, which is equivalent to second year of secondary. In average, the production units are 130 hectares and there are 128 heads of cattle in them.

When relating the technology TAIs with the socioeconomic indexes of stage 3 (2013), except the leading actor of the group, it turned out that actors younger than 60 years, with more than 10 years of schooling, with production units of more than 80 hectares, and who owned more than 60 heads of cattle increased the use of technology; this is explained by the need to make more efficient the resources of their production units (Figure 2).

Figure 2 Association between socioeconomic indexes and TAI. 

In this GGAVATT, the actors who increased to a greater extent their TAIs were the youngest, with higher degree of formal education, larger land surface and higher amount of cattle.

The GGAVATT Vía Corta

The average age of the producers from this GGAVATT is 54.2 years. All these actors know how to read and write, the schooling mean is 10.1 years, which is equivalent to the first year of high school. Of their income, 93.7 % comes from livestock producing activities. The production units have an average of 71.7 hectares of extension, and 66 heads of cattle (Figure 3).

Figure 3 Association between socioeconomic indexes and TAI in the GGAVATT Vía Corta. 

In this GGAVATT, when relating the TAIs with the socioeconomic indexes (except the leader), it turned out that the TAIs were higher in production units with actors younger than 60 years, with more than 6 years of schooling, with production units of less than 100 hectares, and which had less than 120 heads of cattle. In this GGAVATT, the use of their resources has been made more efficient by increasing the use of technology in their production units. Although it is possible to improve their TAIs, the regional context of insecurity from kidnapping and organized crime limits them.

The GGAVATT Caprinocultores Unidos Región Montañosa de Veracruz

The average age of producers in this GGAVATT is 54.8 years. Their mean schooling is 15.1 years, which is equivalent to third year of undergraduate studies. Of their income, 48.1 % depends on the production and transformation of goat milk. In average, there are 64 goats in each production unit. The mean of hectares in each production unit is 33.6.

As shown in Figure 4, when socioeconomic indicators are associated with the TAIs, it was found that those who adopted more technology were the actors older than 60 years, with more than 10 years of schooling, with production units of less than 30 hectares, and which had more than 80 heads of cattle. This is explained because in this GGAVATT schooling is high and its members are enthusiasts, although for some actors, for the time being their main activity is not the production of goat milk and its byproducts.

Figure 4 Association between socioeconomic indexes and TAI. 

In general terms, the actors that increased their TAIs to a greater extent were the longest-living, with higher degree of formal education, with less land surface, and less amount of cattle.