Study área

The study was developed during 2012 in five communities of three municipalities in the central region of the state of Puebla: San Lorenzo Chiautzingo and San Nicolás Zecalacoayan (municipality of San Lorenzo Chiautzingo), San Buenaventura Tetlananca and Aquiles Serdán (municipality of Tecali de Herrera) and San Felipe Cuapexco (municipality of Cohuecan). The choice of the communities was based on their contrasting soil-climate and socioeconomic characteristics (Table 1).

Table 1 Municipalities of the study region. 

Source: authors’ elaboration from data from INEGI (^{INEGI, 2010}).

As data from FAO indicate, the evaluation of the agricultural aptitudes of a specific region should consider a series of characteristics, such as rainfall, topography or relief, quality and type of soil. The municipality of San Lorenzo Chiautzingo presents the best characteristics in terms of rainfed agricultural production and some fruit trees due to its climate, soils, and sources of water supply; in turn, Tecali de Herrera is the municipality that presents the highest advantages for production under conditions of irrigation in localities of the low zone given the presence of irrigation channels and small plains, not so for the irregular slope that is part of Sierra del Tentzo. The municipality of Cohuecan presents various productive advantages, such as those that make up the Atlixco Valley, without orographic accidents, in addition to some seasonal streams and easy access to roads.

Size of the simple

The size of the sample used to obtain information from the production units was determined through qualitative sample, considering the maximum variance with a precision of 10 % and a reliability of 95 % (^{Castillo and Vásquez, 2003}). Making use of the INEGI database (^{INEGI, 2010}) regarding the characteristics of the households and homes in the state of Puebla, and taking as a sampling framework the total of households per municipality, a sample size of 95 households was established (95 questionnaires), which were selected randomly and proportionally in the five localities. It was decided to consider 120 households (120 questionnaires) to increase the reliability and the accuracy of the results.

where: N: Size of the population (Total households in the municipalities)=9833; d: accuracy=10 %; Z _α/s : reliability, Z value of a normal standard distribution 95 %=1.96; Maximum variance=p_nq_n=0.25.

Degree of Food Security

The determination of the degree of food security was through the scale of food shortage that CONEVAL uses and which consists in a battery of 12 questions about perception, where each one of the questions has two possible answers: yes and no. The affirmative answers are assigned the value of 1, and the negative ones 0. From the sum of the marks of affirmative responses and the presence of members younger than 18 years old, it is possible to estimate the degree of FS in the households at four levels, based on the cutoff points used (^{CONEVAL, 2010}). The four points are the following:

Structure of the households

Information was recorded about the number of members, age interval, and presence of members younger than 18 years old in the households.

Productive characteristics

To understand the productive characteristics and available resources of the households, variables were recorded about agricultural, livestock and backyard resources present in the households, as well as the income generated from paid work.

Statistical analysis

The data were captured in Excel (2003), from where they were exported for their analysis to the SAS statistical package (^{SAS Institute, 2003}). Descriptive statistics were used to characterize the study region and to understand the variance in the variables studied. To understand whether the differences between the variables are statistically significant, the Variance Analysis was carried out using Kruskal-Wallis ranges (p<0.05) applied to the median and a Least Significant Difference (LSD) test for the comparison between means, according to the degree of food security.

The statistical model used was:

where: Y _ij : value of the variable given by Group_i in its Household; μ: general mean; G _i : Group with values from 1 to 4; E _ij : error associated to Group_i and the Household_j.

Characterization of the households in the study región

The descriptive statistics of the study region are presented in Table 2, taking into account the three municipalities. Values of zero were found in most of the variables, which originated high variance levels. Those of backyard surface and monthly income from paid work were the ones with highest variance. The species produced in the backyard contribute directly to the availability of foods for their consumption, while the income increases the earning power of foods in the households. The study region is considered diverse in terms of productive resources.

Table 2 Descriptive statistics from the study region. 

N=120

Source: authors’ elaboration.

Food security in the households

The proportion of households in a situation of food insecurity is presented in Table 3. The greatest part of the households presented some degree of food insecurity, accumulating in all its levels 56 % of the total households studied, in contrast with 44 % of the households that are secure in terms of food. The study region was considered as insecure in terms of food. The results found in this study were lower than those reported by the ENSANUT at the national level (2012) for the strata of moderate (MFI) (22.4 %) and severe (SFI) (13.0 %) food insecurity, but they coincide with the classification of slight insecurity (SFI) (41.6 %); that is, just as at the national level, in the study region the members of these households experience preoccupation over access to food, and they can even sacrifice the quality of the family diet.

Table 3 Proportion of the degree of food security in the study region. 

S: security, A: dietary, I: insecurity, L: slight, M: moderate, S: severe.

Source: authors’ elaboration.

In Figure 1 the degree of food security is shown per community and municipality. San Lorenzo Chiautzingo was the municipality with highest degree of food security, reflecting the best conditions in terms of diet, while Tecali de Herrera was the one with greatest problems, being the locality of San Buenaventura Tlalancaleca the one that presented the highest proportion of households with severe and moderate food insecurity. More than 80 % of the households from Chiautzingo presented food security and slight food insecurity. In the three municipalities considered in the study, the latter was severe, although it had the lowest proportion. There is a difference in its degree between municipalities and between communities. The transition from slight to moderate insecurity is given by the reduction in the quality and quantity of foods consumed by members of the household and, although the difference in percentage is large for the study region, once found at this level, the possibility of belonging to the group of households with severe food insecurity is much higher. This allows recognizing the heterogeneity in the food situation of households and sustains the importance of differentiating the policies and programs in attention to food security, paying special attention to the households with slight food insecurity, with the goal of avoiding their vulnerability when falling in the category of moderate food insecurity.

Source: authors’ elaboration.

Figure 1 Degree of food security per community. 

Characterization of the households in function of the degree of Food Security

Table 4 shows the degree of significance of the variables studied in the region. The number of inhabitants in the households and the presence of children (from 0 to 6) turned out to be variables with significant statistical differences. This allows recognizing the weight that these variables have on the methodology established by CONEVAL for the determination of the degree of food security.

Table 4 Mean squares for the factor of variation groups in the variables evaluated. 

**statistical significance p≤0.01

* statistical significance p≤0.05, NS: not significant.

Family structure

Table 5 presents the characteristics according to the degree of food security of the households in the study region. The average number of members in the households was four, of which the stratum with highest presence was the one of over-age members (18 years old or more); those with severe food insecurity reported being the largest in terms of the number of members with a higher presence of underage and elderly members (60 years or more). Food security was affected negatively by the presence of under-age members, which indicates that they are the most vulnerable sector with dietary problems (^{INEGI, 2008}), and which can be attributable to the methodology used in this study, marking a greater dietary vulnerability in the households with members younger than 18 years. These results agree with what was reported by ^{Schmeer (2012)}, who found that the family structure is associated with malnutrition in children and that it can affect the quality of children’s health in México.

Table 5 Structural and productive characteristics of the households. 

^abc Different letters in columns indicate significant difference (p<0.05).

S: security, A: dietary, I: insecurity, L: slight, M: moderate, S: severe.

Note: In the variables where the measurement unit is not indicated, the data correspond to units.

Agricultural and livestock resources

Agricultural and livestock resources are presented in Table 5. Although without statistical difference, it is possible to recognize that the households with food security in the study region are the ones that have the highest amount of these resources. The average extension of land for agricultural use, which was one hectare, is larger in households with food security. This is of utmost importance when considering that the rural population represents 26.37 % in México (^{Banco Mundial, 2016}); this population depends directly or indirectly on sustenance agriculture (^{Hurst et al., 2007}), which is important because in México, 54.9 % of the territory is of agricultural use. ^{Pat-Fernández et al. (2011)} consider that the households with higher agricultural production can destine a higher amount to auto-consumption, making them less dependent on money to gain access to foods (^{FAO, 2013}).

Birds were the animals of highest presence in the households with food security (nine hens and two turkeys), and those of greatest importance in relation to other livestock species, followed by sheep and pigs; bovines were only important in households with slight food insecurity and are not considered important in terms of presence in the study region, the same as goats and rabbits.

However, in terms of food security, the lack of nutrients could be offset through the supply of foods of animal origin, since they have a high content of high-quality proteins and micronutrients, especially important for children, pregnant women and babies, improving the nutritional state of low-income households, in addition to livestock being part of the family strategy at times of risk, stabilizing the income and consumption through the regular sale of eggs, milk and livestock (^{FAO, 2011}). This indicates that food security is influenced positively by the amount of agricultural and livestock resources.

Backyard resources

The average surface destined to backyard production, as well as the type and number of species produced in this space, are also presented in Table 5, understanding the backyard as the place in the household destined to the production of medicinal and ornamental plants, vegetables, fruit trees, and where animals are bred, allowing rural households to obtain fresh and available foods to complement the family diet (^{Salcido, 2008}). The backyard was highly diverse in terms of surface and species produced in it. The households with severe food insecurity reported the largest surface (582 m2) destined to backyard production, a difference that can be attributed to the soil-climate conditions of each locality. When recognizing the contribution of the backyard to households to improve the nutritional state of the members, it was found that fruit trees were the ones with greatest diversity in terms of the number of species reported, followed by medicinal species and, lastly, vegetable species.

However, in terms of food security, the households with severe food insecurity reported a higher diversity, in contrast with what was reported by other authors (^{Melgar-Quiñonez et al., 2005}; ^{Álvarez et al., 2006}), where it has been evidenced that as food insecurity increases, the number of foods produced in the households decreases.

Financial resources

The income reported in Table 5 corresponds to the financial resource obtained through paid work. The households with food security were the ones that reported the highest amount of financial resources that enter the household ($2566 monthly), followed by those with moderate food insecurity ($2108 monthly), while those that have a greater problem in terms of food presented the lowest monthly income ($1400). Income has positive effects on the households with food security, since it allows increasing their purchasing power for a larger quantity of foods, results that coincide with what was reported in other studies (Hernández, 1995; ^{Martínez and Villezca, 2003}; ^{Hurst et al., 2007}), where it is recognized that the groups vulnerable to food insecurity are forced to change their livelihoods, and that the lack of income is one of the most important causes in keeping the population from purchasing sufficient foods. Therefore, the generation of income through employment is a key element in every intervention directed at improving food security of the households, since it can be considered as a strategy to compensate the insufficient amount of foods produced in the household, mainly of animal protein and vegetables; however, the generation of income is not an element considered in the measurement of food security through the CONEVAL methodology.

Plant resources produced in the households

Table 6 presents the main species produced in the plots and backyards of the households in the study region. The main ones were maize and bean, basic in the families’ diets; the main vegetable species were chili, cilantro and tomato, and the principal fruit and medicinal species were lime and chamomile, respectively. This indicates that the diet of people in the study region is made up mostly of basic crops and that it is complemented by other species with different nutritional contributions, which coincides with what was reported by ^{Pérez et al. (2012)}, who explain that the traditional diet consumed in the rural sector is considerably good in nutritional terms, given by the contribution of protein and carbohydrates through maize and bean, of vitamins and minerals by the consumption of fruits and vegetables; and by chili, which is consumed daily, and by the source of calcium from maize that is nixtamalizado (boiled). It also coincides with what was reported by ^{Martínez and Villezca (2003)} when considering that the Mexican population has a diet based on one or more of the following foods: a) of plant origin: maize (tortilla), vegetables (tomato), pulses (bean), roots (potato); and b) products of animal origin: eggs, milk and the consumption of relatively cheap meat, such as chicken or beef. This also agrees with what was reported by ^{Martínez et al. (2012)}, whose results indicate that the diet in northern México is based primarily on products such as bean, tortilla, egg, rice and soup. Within the framework of food security, the diversity in the diet has very important nutritional implications, especially when there is scarce variety and low intake of fruits, vegetables, and animal products (^{Ruel, 2003}).

Table 6 Principal plant species produced in the plot and backyard. 

Source: authors’ elaboration.

This allows recognizing that maize, bean, chili and tomato are the common denominator in the diet consumed by members of the households in the study region, and which are produced in their majority for auto-consumption; this is why it is suggested to center the nutritional strategies in promoting the production of complementary crops, as well as recognizing the importance of the contribution of backyard production in the diet of members of the households and the diversity of species produced in them.

The diet and nutritional state of people were influenced by the soil-climate characteristics in the municipalities, since the plant and animal resources produced in the households depend on these, which are determinant in the study region for their destination, which corresponds primarily to autoconsumption. This is shown in Table 1 and Figure 1, where it is recognized that the municipality of San Lorenzo Chiautzingo is the one with highest degree of food security and the one that has the best soilclimate characteristics.

In addition to the diversity of crops, the presence of at least one crop produced with the objective of trading to obtain with it additional income was reported for each one of the regions. In Cohuecan, amaranth and sorghum were reported, and in Chiautzingo, flowers such as rose; in the case of Tecali, vegetable species such as squash, cilantro and onion.

Productive resources and food security measured through CONEVAL

In the methodology used by CONEVAL, the four categories assigned are given primarily by the presence of the most vulnerable stratum in nutritional terms (children) and not necessarily from the lack of resources for the diet. The households with greatest lack of productive resources were the ones with moderate food insecurity and not necessarily those where it was severe, which is attributed to the presence of a higher number of under-age members in the households of this last group, which in the methodology used are a determinant factor.

Food security measured through the CONEVAL methodology allows identifying the households with lack of foods and with hunger experienced in the study region; however, it does not reflect the way in which they make use of their productive resources as a strategy to appease this situation or the contribution of animal resources, vegetables and income to the diet, which were determinant in the way in which members of the households in the study region obtain food.