Introduction

The use of organic residues in rural communities is an ancient and frequent practice. This action seeks to improve the content of organic matter in the soil to maintain its fertility and productive capacity in the crops they grow. Among the organic residues applied to the soil there are kitchen scraps, stubble, manure, coffee pulp or husks, bagasse and residues from sugar plants, among others. However, in most cases the incorporation of residues such as manure to the plants does not include any prior management. The adequate processing of residues decreases the loss of nutrients that are useful to the plants and which reduce the negative effects of air, soil and water pollution. One of the most highly recommended techniques is vermiculture (^{Colomer and Gallardo, 2007}; ^{UACH, 2006}; ^{Calderón et al., 2003}; ^{Martínez, 1995}), which allows the stabilization of organic residues with low technological requirements and financial investment. Vermiculture is a biotechnological procedure that uses the earthworm (Eisenia foetida) to accelerate the decomposition of organic matter; the results are high quality fertilizers, such as humus and earthworm leachate (^{UACH, 2006}; ^{Calderón et al., 2003}), and the possibility of using the earthworms to breed small species such as fowl. In addition, it allows for organic residues not to become a source of contamination (^{Colomer and Gallardo, 2007}). Because of its technological ease and low level of financial investment, vermiculture has been recommended in areas of subsistence agriculture (^{Gupta et al., 2014}; ^{Del Toro and Limón, 2011}).

Adoption has been defined in several ways. ^{Mosher (1979)} defines it as a new way of doing something or a practice of change. ^{Aguilar and Ortiz (2004)} define adoption of technology as the appropriation and application of technology by the final user. For ^{Leeuwis (2000)} adoption is a process of appropriation that considers a cognitive change as prerequisite. ^{De Schutter (1986)} argues that adoption of agricultural technology undergoes the farmer’s change of knowledge, where the producers understand the technology; the second change is of attitude, when the farmer puts the new technology to the test; and, finally, when the producer uses the new technology habitually. According to ^{CIMMYT (1993)}, the adoption of a new technology can be defined in several ways, so the criteria to define what constitutes the adoption in a concrete study should be specified, because sometimes it is enough to have a record of the number of producers who use the technology and other times the actual surface where it is applied must be determined.

Various authors (^{Galindo et al., 2002}; ^{Feder et al., 1998}; ^{Feder and Umali, 1993}; ^{CIMMYT, 1993}) point out that numerous variables influence the adoption of a technology, such as: cognitive change, level of cosmopolitism, contact with agricultural and livestock institutions, participation in external projects, contact with input suppliers, age, schooling, gender, ethnic group, attitude towards innovation, exposure to communication media, income from outside the farm, standard of living, level of training, workforce available, equipment and machinery available, crops in the system, land tenure, occupational profile, available financial resources, relationship with agents of change, hectares cultivated on slopes, agroclimate environment, years of living in the zone of residence, relevance of technology, initial level of technological knowledge, and index of initial technological adoption; however, the technology is adopted for its relevance (^{Kurwijila, 1981}), understanding by this the ability to be useful or indispensable, for which technology should be suitable to the circumstances of the producer.

The studies on adoption of technologies have been used to explain the motives for the farmers’ decisions; they are useful to refine the generation of technology, to optimize a strategy of technology transfer; to improve the flow of information between researchers, extension agents and policy makers; and to evaluate the effects of technology generation or extension (^{CIMMYT, 1993}).

The National Commission for the Development of Indigenous Peoples (Comisión Nacional para el Desarrollo de los Pueblos Indígenas, CDI), with the Program of Productive Organization for Indigenous Women (Programa Organización Productiva para Mujeres Indígenas, POPMI), through the Unit of Coordination and Contact in Nochixtlán, Oaxaca, implemented training courses in vermiculture and financial supports to groups of women from diverse indigenous communities in the region of Nochixtlán. During these, technical training and advising was provided for the adequate management and exploitation of the organic solid residues that are produced daily in the rural communities. However, there are no evaluations of the level of adoption of technology by indigenous women, of the benefits of this technology and of recommendations to improve the teaching of this type of courses. Therefore, this study has the objective of determining the effect of the technological, economic, social and cultural factors on the adoption of vermiculture.

Research methods and techniques

The research was carried out in the communities of Cumbre Ixtaltepec, San Isidro Apazco, Almacén Apazco, Yutanduchi de Guerrero, Ranchería Buenavista and San Antonio Nduayaco from the District of Nochixtlán, Oaxaca, in the region called Oaxaca’s High Mixtec. The geographical coordinates are located between parallels 16° 55’ and 170° 54’ of latitude north and 97° 00’ and 97° 13’ of longitude west, an altitude above the sea level of 2080 m (^{Plan Nochixtlán, 1998}). In physiography, it is characterized by a mountainous landscape with strong slopes (^{Frausto and Ortiz, 1986}; ^{Rodríguez et al., 1987}). The main types of soil are Regosol, Lithosol, Phaeozem, Chernozem and Vertisol, with Regosols and Lithosols dominating (^{Frausto and Ortiz, 1986}). The mean annual temperature is 15°C on the northern and central part of the District, increasing toward the south until reaching a mean of 22°C. The minimum temperatures vary from 4 to 7°C. The mean annual precipitation is 700 mm on the northern part of the area, and 400 mm in the central part (^{Plan Nochixtlán, 1998}). According to ^{García (1988)}, the predominant climates are temperate dry C(s) and temperate sub-humid C(w). There are three water currents throughout the year, with reduced volume flow rate, which are the Chachoapan River, Grande River, and Tilantongo River (^{Plan Nochixtlán, 1998}). The most important types of vegetation are pine-oak forest, juniper forest, scrubland and grassland (^{Miranda and Hernández, 1963}). The main agricultural and livestock activities are rainfed maize growing; maize associated with bean, broad bean and squash, cajete maize, rainfed wheat, irrigation wheat, shrub bean, alfalfa, as well as breeding sheep (^{INEGI, 2010}).

The CDI under the POPMI has as one of its objectives training women through sustainable projects, for which they provide the initial financial support for the establishment of vermiculture modules. In 2007, the POPMI program supported groups of women from the communities of Cumbre Ixtaltepec, San Isidro Apazco, el Almacén Apazco, Yutanduchi de Guerrero, Ranchería Buenavista and San Antonio Nduayaco from the District of Nochixtlán, Oaxaca. The support consisted in financial resources for the training, construction of vermiculture modules, and earthworm purchase. The process of training and construction of vermiculture modules was divided into two stages. The first was from July 12^th to 29^th, 2007, when 34 indigenous women from the communities of Cumbre Ixtaltepec, San Isidro Apazco and Almacén Apazco were trained. The second was from August 2^nd to 19^th, 2008, and 40 women from the communities of Yutanduchi de Guerrero, Ranchería Buenavista and San Antonio Nduayaco were trained.

The course consisted in a theoretical stage and a practical stage. The first was imparted in two days and consisted in a presentation, introduction, exposition of general information and basic elements of the technology, types of exploitation and technology for production. The second was three days and the way to set up a vermiculture module was shown in a practical manner, as well as how to build an earthworm bed, prepare the substrates (food), carry out the box or biological test, spread the earthworm, irrigate outside and inside the beds, separate or divide the earthworms, and harvest humus and earthworm leachate. In total, 74 vermiculture modules were built in the six communities, that is, one module per producer, and two kilograms of earthworms of the species Eisenia foetida were delivered to each module.

The adoption of vermiculture by indigenous women of Oaxaca’s High Mixtec region was studied taking into account technological, economic, social and cultural factors. The technological ones included the variables Knowledge of technology, Relevance of technology, and Contact with the information. The economic ones included the variables Costs, and Auto-consumption. The social ones included the variables Age, Schooling, Size of family, and Total surface; and the cultural ones the variables Origin, Language, and Social organization. Next, each one of the variables studied, their operational definition, and the criteria for measurement are described.

The adoption of vermiculture or of the technology was calculated with an index of adoption of the vermiculture technique, and it was measured in an ordinal scale from 0 to 1.0. In order to calculate the Adoption Index of vermiculture, the variable of knowledge of technology was considered with the components: bed construction, substrate preparation, box test, earthworm spreading, earthworm feeding, irrigation of earthworms in beds, separation of earthworms, and harvest of humus and earthworm leachate. Each component was assigned a value according to its relative importance in the success of the production of vermi-composting. The relative importance of each component was pondered based on the opinion of three experts in vermiculture. The sum of the value of all the components resulted in an ordinal index number that represented the degree of technology adoption for each observation. The variable Degree of technology adoption was built with the sum of the indicators and defined with the following formula: Technology adoption=Bed construction (0.07)+Substrate preparation (0.17)+Box test (0.15)+Earthworm spreading (0.06)+Earthworm feeding (0.26)+Irrigation of earthworms in the beds (0.23)+Earthworm separation (0.04)+Harvest of humus and earthworm leachate (0.02).

The compilation of information was divided into two stages: the first was recording information from secondary sources and the second was obtaining direct information from primary sources, that is, a questionnaire was applied to each one of the indigenous women trained from the six communities under study.

The questionnaire contemplated 108 questions and was made up of closed questions and of questions with open responses. The questions were divided in two: 21 questions regarding personal and family data, and 87 about general data of the technology.

A census was carried out and the number of interviewees was 74 indigenous women, of which 12 belonged to the community of Cumbre Ixtaltepec, eight to San Isidro Apazco, 14 to Almacén Apazco, 14 to Yutanduchi de Guerrero, 13 to Ranchería Buenavista and 13 to San Antonio Nduayaco, to whom the questionnaire was applied directly through a personal interview with each of them. Due to the distance from each of the communities, the application of the questionnaire was divided into two periods. The first period was from December 15-23, 2010, and contemplated the communities of Cumbre Ixtaltepec, San Isidro Apazco and Almacén Apazco, and the second period, from January 14-22, 2011, included the communities of Yutanduchi de Guerrero, Ranchería Buenavista and San Antonio Nduayaco. The time of application of the questionnaires in each of the communities was three days. The average duration of the application of each questionnaire was 45 minutes.

The data obtained were codified and captured in a spreadsheet to generate a database and proceed to the statistical analysis through the use of the Excel and SPSS 15.0 software (Statistical Package for the Social Sciences).

The level of adoption of technology, distribution of frequencies and percentages, was calculated with the Excel software. The Pearson Correlation coefficient (r) was calculated with the SPSS software to determine the degree of association between the adoption and the parametric variables. Spearman’s correlation coefficient was used to determine the degree of association between the adoption and the variables with ordinal scales (Relevance of technologies, Contact with information and Organization) and the Multiple linear regression, to model the mathematical form, the behavior of the dependent variable Adoption with the independent variables, which resulted in a higher level of correlation with the variable Adoption of technology.

With the aim of studying how the adoption of technology behaves in function of the dependent variables with higher correlation: Age, Schooling, Language, Relevance of technology and Knowledge of technology, the data were adjusted to a polynomial of first order. Through the multiple linear regression, the estimators of the regression coefficients were produced and later an estimated function. The model used was the following:

y _i =β ₀+β ₁ Age+β ₂ Schooling+β ₃ Language+β ₄ Relevance+β ₅ Knowledge+e _i

where y _i : represents the value of the variable adoption of technology for the i-th observation β ₀: represents the value of the origin ordinate; β ₁: is the regression coefficient for the independent variable age; β ₂: is the regression coefficient for the independent variable schooling; β ₃: is the regression coefficient for the independent variable language; β ₄: is the regression coefficient for the independent variable relevance of technology; β ₅: is the regression coefficient for the independent variable knowledge of technology; e_i: random error due to the error of the i-th observation.

With the model once adjusted, a variance analysis was performed to test whether it was adequate to explain correctly the response in function of the independent variables studied. Later, with the Student’s t-test the significance of each of the coefficients that defined the response equation was tested, in order to determine if they contributed and to what degree in explaining the function of adoption of technology.

Results and Discussion

The results indicated that the degree of adoption of vermiculture technology by indigenous women in Oaxaca’s High Mixtec region was 89.2%, corresponding to a high level; 10.8 % adopted it at the medium level and none at the low level. The mean in adoption of vermiculture technology was 0.86 and is located in the high range of adoption (Table 1).

Table 2 presents the correlations between the variables studied; it shows that the dependent variable Adoption of technology shows a direct relationship with the independent variables Knowledge of technology (r=0.588), Schooling (r=0.452), Language (r=0.413), Relevance of technology (r=0.243) and an inverse relationship with Age (r= -0.529); in all of the previous cases the correlation coefficients were at least significant (p≤0.05). With the variables Size of family, Surface, Contact with information, Social organization, no relationship was found.

Other combinations of variables that showed regression coefficients which were at least significant (p≤.05) were the following (Table 2): the variable Schooling revealed a direct correlation with Language (r=0.502), Contact with information (r=0.370) and an inverse correlation with Age (r=-0.434). Age showed a negative correlation with Language (r=-0.481), Knowledge (r=-0.288), Relevance of technology (r=-0.277) and Size of family (r= -0.266). Size of family had a direct correlation with Social organization (r=0.374) and Language (r=0.242). Language showed direct correlation with Relevance of technology (r=0.278), Contact with information (r=0.282) and Knowledge of technology (r=0.258). That of Surface used for the technology presented a direct correlation with Social organization (r=0.364) and Relevance of technology with Organization (r=-0.229).

The multiple linear regression analysis used as explicative variables the following independent ones: Schooling, Age, Language, Relevance of technology, and Knowledge of technology, which were significantly associated to the dependent variable Adoption of technology in the correlation analysis. The results were adjusted to the multiple linear regression model proposed, and it was found in the variance analysis of the regression equation obtained that the model used proved to be highly significant (F=15.968; p=0.001), which indicated that at least one of the coefficients β _j , con j=1...5, is different from zero, and that the model used explained well the response in function of the variables studied. The R squared showed that 0.540 of the variation in Adoption of technology is explained by the variables Age, Schooling, Language, Relevance of technology, and Knowledge of technology, where Knowledge of technology is the most important one followed by Age and Schooling.

The determination of the significance of each one of the parameters that define the response equation (Table 3) showed that the estimators that were significant (p≤0.05) were Age, Schooling and Knowledge of technology; therefore, they contribute to explaining the function. The variables Language and Relevance of technology were not significant.

With these results each one of the relationships suggested was proven. The data evidenced that the indigenous women interviewed present a high level of adoption of the technology (Table 1). This means that the theoretical-practical course imparted had positive results in the six communities of Oaxaca’s High Mixtec region. Next, the results are discussed based on the hypotheses set out.

Conclusions

In general, 100% of the women evaluated adopted vermiculture, 89.2 % were at a high level of adoption, the remaining 10.8% were at a medium range, and none were at the low level.

Of the technological factors studied, Knowledge of technology (r=0.59) and Relevance of technology (r=0.24) presented direct correlations with Adoption of vermiculture technique, although Relevance of technology was not a defining factor. With regards to Contact with information, there was no relationship with Adoption of vermiculture technique.

Of the economic factors explored, Costs and Auto-consumption did not show a relation with the levels of Adoption of vermiculture technique because the installation costs of the modules and the training were financed by the POPMI and they were the same for all the participants.

The social factors that were related to the Adoption of vermiculture technique were Age (r=-0.53) and Schooling (r=0.45). Age showed an inverse relation with Adoption of technology so there is a trend in which as the Age of the indigenous women is higher, the Adoption of vermiculture technique decreases. Schooling presented a direct relation with Adoption of technology, indicating that among indigenous women, as more years of school are studied there is a higher level of Adoption of technology. Both variables were determinant factors that influence the Adoption of technology. The Size of the family and the Total surface did not relate to the levels of Adoption of technology.

Of the cultural factors, Origin, Language and Social organization, only Language (r=0.41) manifested a direct relation of dependence with the level of Adoption of vermiculture technique, so that among indigenous women the more they speak Spanish the higher Adoption of technology there is. Origin and Total surface did not show dependence with the levels of Adoption of technology.

It is quite likely that with time the factors that today turned out not to be defining in adoption of technology will be later, when women and other members of the family diversify their infrastructure and the destination of the products from vermiculture; therefore, further studies with the same population are recommended.

The study performed and the results found are considered important, because they refer to poor indigenous women, in marginalized localities, and are about the adoption of a low-cost technique for investment and operation, which, in addition, is a sustainable practice that contributes to caring for the environment and to the efficient use of the scarce resources available.

Acknowledgement

To the National Commission for the Development of Indigenous Peoples (Comisión Nacional para el Desarrollo de los Pueblos Indígenas, CDI) for the support provided for graduate studies. To the women from the six communities: Cumbre Ixtaltepec, San Isidro Apazco, Almacén Apazco, Yutanduchi de Guerrero, Ranchería Buenavista and San Antonio Nduayaco from the District or Nochixtlán, Oaxaca, for the conveniences granted.