Five districts of Ayacucho, Peru were selected (Figure 1), considering their representativeness in the number of goat farmers: Ocaña, Chuschi, Accomarca, Pacaycasa, and Santillana. These districts belong to the highlands of southeast Peru, characterized by large dry areas, mountains, cliffs, and inter-Andean enclaves where animals use shrublands and grasslands, with a rainy season between December and March, and a dry season during May and October. The temperature ranges from 0.1°C in July to 28.7°C in November, with a monthly rainfall of 5.5–154.7 mm (Servicio Nacional de Meteorología e Hidrología del Perú, 2018). The cultivation of cereals (corn, peas, barley, and wheat) and animal husbandry are the main economic activities.

According to Ministerio de Desarrollo Agrario y Riego (2021), the districts of Ocaña, Chuschi, Accomarca, Pacaycasa, and Santillana present a total of 270, 208, 374, 434, and 619 smallholder goat farmers, respectively. A descriptive cross-sectional study was carried out, in which the sample was estimated using a random sampling stratified according to population size. A sampling error of 10% and a 95% confidence interval (1.96) were used.

Questionnaire-based face-to-face interviews were conducted with 91 goat farmers between May and August 2023. The interviews were distributed in agricultural units established in the districts of Ocaña (n = 13), Chuschi (n = 10), Accomarca (n = 17), Pacaycasa (n = 21), and Santillana (n = 30). The survey was primarily made up of closed-ended questions and was organized into four key sections to gather comprehensive information: 1) General information, 2) Socioeconomic data, 3) Productive data, and 4) Commercialization. The interview was conducted by a professional who perfectly understands the local languages (Quechua and Spanish), to obtain more accurate and concise declaratory information.

With the information obtained, the smallholders were classified as: infra-subsistence farmers, subsistence farmers, intermediate farmers, and remaining farmers according to small-family farming in Peru (SFF) (Organización de las Naciones Unidas para la Agricultura y Alimentación, 2017). This classification was made based on family or nonfamily nature, size or scale, productive orientation, technified breeding, and the extensions of arable land and livestock extensions owned by the farmer.

The statistical analysis of all the information was carried out using R Software (version 4.3.1). For characterization, the data were analyzed descriptively based on frequencies by contingency table and percentage. To classify the herds, 15 variables of the questionnaire were chosen, based on their relevance to the herds of small producers, and a multivariate analysis based on multiple correspondence analysis (MCA) and hierarchical classification analysis (HCA) were performed using the FactoMineR package (Lê et al., 2008). The variables chosen for the analysis of MCA and HCA were defined according to the economic, social, and productive factors; which are detailed in Table 1.

The MCA is a statistical data analysis tool used to describe, explore, and summarize, which allows the visualization of the association of a given set of categorical variables in small dimensions (Algañaraz, 2016; Sow et al., 2021). HCA is a multivariate technique that groups elements or treats variables to maximize homogeneity or difference between groups (Vilà-Baños et al., 2014). These methods are often used to analyze questionnaire data (Pagès and Josse, 2014).

The differences between clusters were evaluated using Fisher’s chi-square and exact test to assess whether the selected variables had significant differences between groups.

The proportion of interviewees was 66% women and 34% men. The majority of the farmers had completed elementary school (30%), followed by those who completed high school (20%) and had incomplete high school (18%). It was found that 14% of farmers were illiterate, and only 2% had attended university. Smallholders older than 35 years represented 83% of all the interviewees.

Women were responsible for making decisions about herd management. For smallholders, the main purpose for goat keeping was obtaining milk or meat for home consumption to ensure their food security (44%), and if there was cheese surplus, it was offered to local markets (40%).

Almost half of the farmers were involved in livestock more than 20 years ago (43%), with the priority of improving the genetic quality of their animals to increase milk and meat production. The farmers interviewed stated that they lacked technical assistance.

Most breeders use their products (milk, cheese, meat, skin, and manure) for self-consumption (80%) and only sell when there are surpluses. Producers who marketed their products did not have a sales record, which is a fundamental tool for economic assessment. More than half of breeders (63%) sold goat kids at four (13%) or 5 months (48%) of age. There was not any type of external financing to which breeders could have access.

Multiple correspondence analysis (MCA) was applied to 15 variables (Table 1). Each dimension of MCA explained the amount of variation, and only the first two dimensions (17.1% of the total variance) were retained, compared to other dimensions that had few contributions to the total variance (Figure 2).

The MCA was carried out on 15 variables and 54 modalities that contributed to the formation of 2 dimensions. Figure 3 shows the distribution of the herd modalities according to the first (8.8%) and second (8.3%) size. In the first dimension, positive coefficients were cattle ranchers from the districts of Ocaña, Pacaycasa, and Accomarca, who have lands greater than 1 hectare. Their goats graze freely in their own and communal areas; producers aim to produce milk and meat for the market, and with their main source of income besides crops and trade. The negative coefficients were the producers of the districts of Santillana and Chuschi with extensions of lands less than 1 hectare; their breeding objective was for self-consumption, with the eventual sale of milk when there were surpluses. In the second dimension, the positive coefficients are made up of subsistence and middle-class farmers, their main source of income was goat breeding, using milk and meat for sale, and their herds were dewormed 1 to 2 times a year. The negative coefficients are made up of subsistence farmers, the objective of goat breeding was for self-consumption; their main source of income was crops using technified irrigation. Some herds were dewormed when sick, while the “healthy ones” were never treated.

The representation of the variables in dimensions 1 and 2 is shown in Figure 4. The Ocaña goat breeders were better represented, followed by breeders who raised goats for self-consumption, breeders with less than 0.5 hectares of land, herds that drank water from the ditch, and those that were not dewormed. Breeders who achieved incomplete high school, with 10–20 years of breeding, and those who grazed their herds on communal and their lands, were the ones who had a low representation in dimensions.

The hierarchical classification typed three groups (Table 3; Figure 5). The chi-square test revealed a statistically significant difference between the groups for all variables (p < 0.001), except for mixed farming variables, education level, land tenure, and milk production.

Group 1 (n = 36) was typified as a system in which breeders raised goats only for self-consumption. The Santillana district breeders were classified mainly in this group (83%). According to the SFF classification, more than half of the breeders belong to the subsistence category (58%), with agriculture being their main source of income (89%). Eighty-one percent of the breeders had less than 0.5 hectares of land installed with technified irrigation (67%), The herds drank water from streams (61%), and most of them had twin births (92%). The majority of breeders in this group (71%) dewormed their goats 1 to 2 times a year.

Group 2 (n = 39) was typified as a system in which breeders raised goats for self-consumption and marketing of surplus products (cheese, milk, and meat). This group classified breeders mainly from the Accomarca (44%) and Pacaycasa (44%). Just over half of the breeders in this group were classified as subsistence producers (59%), with agriculture (51%) and livestock (28%) being their main sources of income. In this group, the breeders had larger extensions of land, between 0.5 and 2 hectares (84%), where some had access to water (46%), while others had access to technical irrigation (44%). Herds taken directly from taps with drinking troughs (62%), and more than half (67%) had twin births. Most of the breeders (74%) did not deworm their herds.

Group 3 (n = 16) was typified as a system in which the goat plays a major role in the economic livelihood of the breeders. This group consisted mainly of breeders from the Ocaña district (69%). Most breeders were in the subsistence category (88%), with goats as their main source of income (81%). The breeders kept the goats to produce cheese and goat kids. Twenty-five percent of the breeders had land extensions greater than 2 hectares. Herds drew water from natural sources such as the river (65%). The breeders of this group mostly dewormed their goats once to twice a year (94%).

Geographical characteristics influence socio-economic articulations (Torres, 2001). It is important to know the socio-economic situation of smallholders in order to improve goat farming in a sustainable way (Van Arendonk, 2011). In this study, illiteracy was below 15%; in contrast, breeders in the north of Peru exceeded 25% (Temoche, 2019), and Algeria and Senegal showed 44% and 69%, respectively (Laouadi et al., 2018; Sow et al., 2021); while in Ecuador 9% were reported (Villacres Matias et al., 2017). This might be due to the increasing number of schools in rural zones during the last years and the improvement of roads which let them get access to the cities with educational services; since education plays a determining role in the understanding and adoption of new technologies (health, feeding, reproduction and installation) that help improve the goat production system (Vásquez, 2016). Most of the goat farming was inherited through generations but the low percentage of smallholders with less than 35 years old could indicate that young people are not interested in livestock and have decided to find different job opportunities outside of their rural areas.

Unlike the present study (66%), women in northern Peru were found to be less involved in this livestock activity (Temoche, 2019). In Senegal, women (91%) are the main actors in goat breeding (Sow et al., 2021), while the opposite occurred in Algeria (99% men) (Laouadi et al., 2018). The greater prominence of women in goat activity in the sectors of study is because men developed other professions that forced them to travel to the city and develop part-time jobs to contribute to the economy of the family; while women remained mostly at home and were dedicated to care of children and raising livestock.

In this part of Peru, goats were generally raised to ensure the family’s food security; although there were breeders who sought economic sustenance with the sale of cheese and yogurt. Similar reports were found in Senegal (Sow et al., 2021) and Tanzania (Nguluma et al., 2020); while in Uganda, they only considered life insurance (Nampanzira et al., 2015). This heterogeneity in the purpose of breeding would be related to the presence of many cultural, social, and family traditions, where either the woman or the man has the main role in the goat breeding process (Missohou et al., 2004). Illiteracy also plays an important role in the current situation of small-holders’ goat production systems (Agossou et al., 2017; Laouadi et al., 2018; Temoche, 2019).

Production systems can be classified as: extensive, semi-intensive, and intensive, which are determined based on many factors that compete with the heterogeneity of production processes (Morales-Jerrett et al., 2022). In the present study, only 4% of breeders owned land greater than 2 hectares, while in the north of the country, there was higher (14%) (Temoche, 2019), as well as in Ecuador (14%) (Villacres Matias et al., 2017). The planting of crops was the main activity for goat smallholders and they preferred to allocate their lands for crops such as corn, wheat, beans, and peas, turning communal areas into the forage support of most herds.

This study reveals that breeders practice mixed parenting, similar to the methods used in northern Peru (Temoche, 2019). This involves raising different livestock (goats, sheep, and cattle) on the same farm, allowing for efficient use of natural resources, better risk management, and increased sustainability of grasslands through differentiated grazing (Sow et al., 2021). Mixed parenting is also practiced in Brazil (Rodrigues Alves et al., 2017) and other African countries (Moula et al., 2017; Laouadi et al., 2018; Nguluma et al., 2022), however, these countries have different populations, policies, and sociocultural traditions that influence the production systems.

Animal husbandry is a tool for cash flow in a family economy so in other parts of the world it is considered “saving” life security (Missohou et al., 2016; Manirakiza et al., 2020). In this study, it was recorded that few breeders have a defined market for the sale of their products, as well as in Ecuador (Villacres Matias et al., 2017). These reports differ from those of northern Peru, where breeders know about the economic value of goat breeding (Temoche, 2019), as well as in Spain and Argelia (Morales-Jerrett et al., 2022; Laouadi et al., 2018). In these areas, goat-derived products continue to be traditional, with a short production chain that is related to a weak organization of the actors involved in designing the policies of the sector (Camara et al., 2019; Sow et al., 2021) that results in an unnoticed competitiveness in the regional and national market (Laouadi et al., 2018).

Using multivariate analysis (MCA), the breeders were grouped into three groups. This typification in groups of goat breeders is differentiated by variables such as total unit area, income source, production objective, irrigation system, sources of water, and deworming period.

The breeders in Group 1 were part of family infra-subsistence farming, the cultivation of cereals in the sector was their primary activity. The harvesting of crops covered all their family’s expenses, including food, education, and clothing. Goat breeding was done solely to ensure food security. Because they have areas of less than 0.5 hectares of land, breeders are limit themselves to livestock, putting more interest in growing cereals in the sector (corn, peas, wheat, beans, wheat, and barley). The breeders mostly carried out technified irrigation due to the geographical conditions (ravines and slopes) or the type of economic activity (agriculture or fruit production). Herds that took advantage of the natural pastures of communal lands preferred to consume vegetables from shrub and arboreal strata rather than herbaceous ones, these preferences are related to the feeding behavior of the goats (browsing).

The majority of Group 2 breeders belonged to subsistence family farming. The breeders of this group consider livestock and agriculture as their main activities, which would be related to the extensions of their lands due to the geographical conditions (accessible and wide valleys). Between March and June, breeders had higher incomes due to the calving and the great availability of pastures. The use of pressurized irrigation in this group would be related to the existing agricultural and fruit activity in the sectors, however, flood irrigation still prevails. Regarding parasitic control, most breeders did not comply with the deworming schedule, reporting a higher prevalence of gastrointestinal parasites (Eimeria spp., Strongyloides, Trichuris spp.). (Mendoza, 2023). The lack of sanitary management in these herds is attributed to the absence of entities and professionals that provide technical assistance and training.

The majority of the producers in group 3 belonged to subsistence family farming. In this group, goat breeding was the primary activity for breeders, driven by strong market demand for cheese and standing kids. To support this, they practiced transhumance, moving their herds throughout the year to ensure adequate feeding, unlike the sedentary lifestyle of the other two groups. The breeders also had a long-standing tradition of effective herd health management, passed down through generations. However, there is a need to enhance the production chain in the sector.

Group 1, includes mainly producers from the districts of Santillana and Chuschi; Group 2, producers from the districts of Pacaycasa and Accomarca; and Group 3, producers from the districts of Ocaña, concluding that there are differences between the districts. Group 1 concentrates on producers who are mainly dedicated to agriculture, most of them, being subsistence category. Group 2 concentrates on producers who give importance to both livestock and agriculture, most of them, being subsistence categories. On the other hand, group 3 concentrates on producers who are dedicated only to goat activity, where almost all of them are subsistence categories. Thus, the producers in Group 1 raise goats for self-consumption, Group 2 for consumption and sale of surpluses, and Group 3 to market their products. The producers in Group 1 have smaller land extensions (<0.5 hectares) than those in Group 2 (0.5–2 hectares) and Group 3 (>2 hectares).

The Ayacucho region is divided into three ecosystems by the Andes Mountains: the southern highlands, the center of the abrupt mountains, and the northeast of the tropical forest, with altitudes ranging between 500 and 5,000 m.a.s.l. It is characterized by a rugged geography, a variety of microclimates, with a diversity of ecological sites, where there are scattered natural resources. These geographical characteristics have influenced the nature of human settlements and socio-economic articulations, attributing to agro-ecological diversity among the populations of the region (Torres, 2001). These geographical and agroecological characteristics, in addition to social, cultural, and economic factors, would be influencing the different agricultural and livestock production processes, resulting in grouping producers into 3 groups.

Group 1 concentrates producers in the subsistence category (58%) because goat farmers in the sectors do not cover their basic needs and have land <0.5 hectares, so they opt for other sources of income other than agriculture and livestock. Group 2 concentrates producers in the subsistence category (58%), where goat farmers can cover their basic needs, with land between 0.5 and 2 hectares, however, they do not have enough savings capacity. Group 3 concentrates on producers in the subsistence category (88%), where goat farmers can save their income and have the opportunity to improve their quality of life. The classification of PAFs is based on socioeconomic and ecological characteristics (land, water, and biodiversity), which have a strong impact on agricultural production systems (Fonseca-Carreño, 2019), and requires a highly disaggregated vision (Maletta, 2017).