Technology appropriation helps reducing workload, improves the life quality of farmers, and increases farm performance (Tse et al., 2018). Although the use of technology has increased in recent years, its adoption rate by the dairy sector has been slow compared with other industries (Russell and Bewley, 2013).
Investing in dairy farming technologies implies overcoming several challenges. It is necessary to consider the reasons argued for investing or not (Steeneveld and Hogeveen, 2015), the technical efficiency (Steeneveld et al., 2012), and its economic consequences (Bijl et al., 2007), among others. Lack of analysis of the specific conditions and needs of farmers is a relevant factor that discourages investment (Luvisi, 2016).
Information about the adoption and use of technology by dairy farmers in developing countries is scarce (Janssen and Swinnen, 2017). This includes Colombia, where such lack of information could be related to its low rate of technology adoption (Barrios et al., 2019). Therefore, it is necessary to investigate the technology adoption processes by dairy farmers in this country. The results could improve decision making and productive performance in a sector that is still immature regarding organizational issues and business management (Vásquez-Jaramillo et al., 2018). The objective of this study was to establish the factors associated with the adoption of technologies by dairy agribusiness. The findings could help increase the effectiveness of research and policy-making agencies for supporting extension programs.
MATERIALS AND METHODS
A survey that included 45 questions distributed in two sections was carried out to determine the factors related to technology adoption. The first section provided the information required for a general understanding of demographics and productive characteristics of dairy agribusiness. The second section focused on those factors that farmers considered important to make decisions on whether or not to adopt a technology. Data were collected between May and December 2018.
A total of 280 farmers across eight municipalities of Antioquia province, Colombia, responded to the survey. Stratified sampling by size (Sorge et al., 2016) and municipality (Milán et al., 2003) was used. The municipalities were: Bello (14), Belmira (25), Donmatías (25), Entrerríos (29), San José de la Montaña (7), San Pedro de los Milagros (61), Santa Rosa de Osos (76), and Yarumal (43).
A Likert response scale with levels from 1 ("Not important") to 5 ("Very important") was used to evaluate the relationship between the process of technology adoption and the farm operational and management variables. The statistical procedure included an exploratory factor analysis using the psych library (Revelle, 2017) of the R-project software (R Core Team, 2018) and a model of structural equations. According to the Schmid-Leiman procedure (Revelle, 2017), only variables with Cronbach’s alpha values higher than 0.70 and a factorial load higher than 0.25 were included in the model. Model fit was validated by a Root Mean Square Error of Approximation (RMSEA) less than 0.1 and both Comparative Fit Index (CFI) and Goodness of Fit Index (GFI) greater than 0.9 (Cangur and Ercan, 2015) using the lavaan library (Rosseel, 2012) of the R-project software (R Core Team, 2018).
RESULTS AND DISCUSSION
The average age of farmers was 47±12 years, and they had 25±13 years of experience in dairy farming (Table 1). Both traits are known to favor productivity, considering that experiential knowledge facilitates decision-making (Cuartas-Martínez et al., 2018). However, this knowledge was not accompanied by academic training; on average, they attended through eighth grade, which means they did not complete high school. This finding is known to discourage "Management by Competencies" and limits individual and organizational learning (Pardo and Díaz, 2014). This could explain why the Colombian dairy sector has been focused for years on its survival rather than growth and business development (Barrios et al., 2016).
Regarding organizational characteristics, 70% of producers work in their own farms (Table 2). This factor, added to the fact that in 62.85% of cases there was a successor to the business, could promote the adoption of new technologies by this type of organization since there is certainty about the fate of the property in the long term. It could be related to the fact that 74.64% of farmers have used medium or long-term financing, a figure higher than that found by Rodríguez et al. (2015), who reported, for the same region, a 38% credit-access rate. In that study, they also reported that the technical assistance rate was 50%, meaning that coverage of technical assistance and technology transfer programs have improved in recent years, reaching 89.64% for the surveyed organizations. These results are positive because such programs help to guide farmers towards appropriate decision-making processes (Cerón-Muñoz et al., 2015).
Improvement of milk quality, Pastures, and Herd genetics were the most important aspects that influence a farmer’s intention to adopt technologies, averaging 4.58±0.59, 4.57±0.61, and 4.52±0.63, respectively (Table 3).
A tendency to favor the adoption of technologies related to purely technical aspects is frequent in the dairy sector, where it is common to find higher adoption rates of "hard" technologies in comparison to those associated with knowledge management and improvement of procedures and management methods (Barrios et al., 2016).
Human management was the least relevant variable when deciding on the adoption of technologies (3.09±1.41). This result disagrees with the report by Steeneveld and Hogeveen (2015), who found that investment in dairy technology significantly reduces labor, decreases production costs, and improves the life quality of farmers.
Variables with no statistical significance were eliminated after the exploratory analysis. Thus, only ten variables grouped into two factors were included in the structural equation model. This helped to identify the structure of the relationships between variables and conformed factors (Table 4). According to the common characteristics of variables grouped in each factor, it was possible to name Factor 1 as a production-related factor, while Factor 2 included variables related to business management.
The structural equation model resulted in a Cronbach’s alpha value higher than 0.7 for the proposed factors (Table 4), with 0.072 RMSEA, and fit indexes of 0.921 and 0.948 for the CFI and the GFI, respectively. This indicates the internal consistency of the scale and a good fit of the model (Cupani, 2012).
Factor "Production" included the following variables: Pasture, Herd genetics, Production costs, Equipment, and Milk quality (Figure 1). Pastures and Herd genetics were the most representative variables, with 0.7- and 0.65-factor loads, respectively. This could be due to the fact that forage quality and genetic improvement are related to dairy herd planning (Múnera-Bedoya et al., 2018; Cerón-Muñoz et al., 2017) which is considered a strategic tool linked to technology adoption. It is important to mention that variable Production costs (with 0.59 factorial load) was one of the aspects that determined the adoption of technology, which could promote the analysis of costs in this sector, considering that this industry has presented historically low rates of economy diagnostic at the organizational and sectorial level (Barrios and Olivera, 2013).
The second factor, "Management," was impacted by variables: Administrative management, Technical procedures, Human resource management, Supplies, and Farm recognition in the market. The variables with the greatest weight in the factor were Administrative management (0.69 factorial load) and Technical procedures (0.66 factorial load). These variables are related to document management, which denotes the importance of information traceability for adequate analysis and subsequent decision making (Londoño et al., 2016).
Although variable Farm recognition in the market had the lowest factor load (0.43), it is important to highlight how the decision to adopt technology is positively influenced by the image that the farm could project in the sector. This result is somewhat unusual, considering that this market has a regional oligopsonic structure where, in normal conditions, the total production of milk is sold (Von Keyserlingk et al., 2013). This could be associated with having a productive system with technology levels in line with or higher than the industry standard, which in the long term can be considered as a sustainability strategy for the organization (Adefulu, 2015).
CONCLUSIONS
The intention to adopt technologies by dairy farmers is influenced by factors inherent to production and business management traits. Improving technical aspects such as pastures, herd genetics, production costs, equipment, and milk quality can affect the overall production factor. Additionally, the Management factor is influenced by the intention to improve administrative and technical processes, human resource management, the supply process, and the recognition of the farm in the market.