texto en 
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
Traditional foods are produced from local natural resources using specific processing techniques. This specificity gives unique chemical and sensory characteristics to this type of product, in addition, they have a relationship closely linked to their culture of origin. They are key elements in the elaboration of different local gastronomic dishes that show representations of socio-cultural meanings and provide specific nutrients to the people who include them in their diet1. The production and consumption of traditional foods have been empirically passed down through generations. They have an impact on the economic development of the region of origin where they are produced, creating jobs and supporting local producers2. The promotion of these foods can therefore contribute to rural development and improve the economies of small-scale producers.
Cheeses stand out internationally as a traditional food group of great importance, they are recognized by consumers for their attributes such as protection of origin, type of process, quality indicators, safety labelling and price3; over time, attempts have been made to identify the main valuation attributes of cheeses, for this purpose different valuation techniques have been used and applied, such as discrete choice experiments, from which it is possible to estimate the willingness to pay (WTP) for different traditional cheeses and their attributes. Previous research has identified consumer behavior towards traditional cheeses, such as Colonna et al4, who identified preference and purchase of specialty cheeses from the US; Menozzi et al5, who identified preference for labelled Parmigiano Reggiano and Comté cheeses; Mazzocchi et al6 assessed consumer attitudes towards sustainably produced mountain cheeses; and Menozzi et al7 investigated the factors that influence the choice of PDO-labelled cheeses from France and Italy.
Braghieri et al3, indicated that there is a relationship between consumer attitudes, valuation and willingness to pay for traditional cheeses; this is influenced by different factors such as: origin, beliefs and attitudes of consumers, as well as level of education and income. On the other hand, Rodrigues et al8 defined that food choices are related to different points of perception and the main factor in a purchase choice is influenced by physiological or nutritional needs, however, Köster9 indicated that in addition to biological determinants (hunger, appetite and sensory aspects), there are environmental determinants (cultural, social and economic environment) and individual determinants (age, gender, personality and food experience); Therefore, it is important in this research to know what are the factors that influence the food choice of a product such as Seco Encerado Cheese.
It is important to note that in Mexico there is an important culture of cheese consumption and production, Villegas de Gante et al10 describe the production of around 40 Mexican artisanal cheeses that are considered traditional due to the set of characteristics acquired from their respective production environments; one of these products is Seco Encerado cheese, produced in the region of Istmo de Tehuantepec, Oaxaca, Mexico11.
This cheese is characterized by the fact that it is artisanal produced with raw milk from zebu cows (Bos taurus indicus), as the first phase of the production process reception operations are performed (filtering and manual skimming in some dairies), then proceed to the curdling, which is done with rennet obtained from cattle and carried out at room temperature (30 - 40 ° C) for 30 min, once the gel is formed, the curd is cut and drained through blanket cloth bags. Once the curds are salted, they are placed in cubic molds (35 x 35 cm) made of Huanacaxtle wood (Enterolobium cyclocarpum), where the cheeses are pressed with river stones (weighing approximately 10-20 kg). The cheeses are subjected to two ripening processes under the typical environmental conditions of the Isthmus (high relative humidity and temperature): the first process lasts 7 d and is carried out in Huanacaxtle moulds in order to ensure complete draining and avoid internal fractures in the cheese; The second process lasts about one month, during which the cheeses are removed from the molds and placed on wooden shelves, where every third day the cheese is turned over and the fat released is spread on the different sides of the cheese, this process is repeated until the 30-d period is over. Once the cheese has matured successfully, it is covered with a layer of wax or paraffin to maintain its shelf life and prevent possible physical or biological contamination. The final piece of Seco Encerado cheese weighs approximately 10 kg and is sold by the cheese producer to the local retail distributor, who oversees cutting the cheese into small triangular prisms with a weight of 50 g and a price of around 20 MXN, a common presentation of the product and offered to the final consumer. Seco Encerado cheese due to its sensory characteristics of aroma and intense fermented flavor, salty flavor and crumbliness, stands out as an ingredient in various typical dishes of the region such as: breads ((bollito de queso o quesadilla de arroz), garnachas istmeñas, tostadas, empanadas rellenas de queso, pure de papa horneado estilo istmeño, garnachas, enchiladas, among others, however, Seco Encerado cheese is also a perfect complement for some local foods such as: totopos istmeños, pites or corn tamales, beans or among other dishes12.
Despite the importance of traditional Mexican cheeses, there are currently few studies that evaluate consumers' valuation preferences and willingness to pay for them, however, the following consumer-related research on Mexican cheeses was identified: Sensory analysis of “Bola de Ocosingo” Cheese13; Perceptions of cheese authenticity14,15; Segmentation of cheese consumers according to their lifestyle16. The objective of this research was to determine consumers' willingness to pay and valuation of Seco Encerado cheese through discrete choice experiments using the following valuation attributes: origin, type of process, aging time, label, and price, with the purpose of exploring consumers' perceptions of Seco Encerado cheese.
Material and methods
Sample of consumers
A sample (n ≥ 355) of Mexican consumers of Seco Encerado cheese was selected according to Equation (1) described by17; for an infinite population (N= 99,999), a coefficient of reliability of 95% (z= 1.96), a coefficient of variation of 50% (Vy= 0.5) and a margin of error of 5 % (
Discrete choice experiment design (DCE)
The discrete choice experiment involves the construction of an experimental design that aims to evaluate experimental variables that contain two or more levels (attributes) and the effect of the levels of these attributes on the stated preference18.
Attribute and level selection
The attributes were selected according to the criteria that characterize traditional foods according to Guerrero et al19: geographical origin, type of process, product information and price (Table 1).
Table 1 Selectionof attributes for the survey using discrete choice experimental methodology
| Research criterion |
Attributes | Levels | Description |
|---|---|---|---|
| Geographic origin |
Place of origin | Tapanatepec, Oaxaca | Places associated with the production of Seco Encerado cheese. |
| Zanatepec, Oaxaca | |||
| Process type | Process | Natural (the rumen of the cow is used as a coagulating agent) | Rennet extracted from cow rumen, processed by salting and solar drying. |
| Not natural (industrial microbial rennet is used as a coagulant) | A blend of coagulation enzymes marketed under the CUAMEX® brand. | ||
| Process type | Ripening time | Fresh | Fresh cheese fit for human consumption from 0 to 48 h. |
| Semi-dry | Semi-ripened cheese (7 to 15 d). | ||
| Dry | Cheese ripened for at least 30 days (T ≥ 35 °C and RH ≥ 60 %). | ||
| Product information |
Label | With information | The label of the cheese must include nutritional and origin information. |
| No information | The cheese label must not contain any nutritional or origin information. | ||
| Price | Price per kg of cheese ($ MXN) | 400 | Amount to be paid in Mexican pesos per kilogram of Seco Encerado cheese. |
| 300 | |||
| 200 |
$ MXN= mexican peso; T= temperature; RH= relative humidity.
Selection of combinations
A full factorial design was applied based on the five attributes to be evaluated for Seco Encerado cheese and their respective levels (Table 1), from this experimental design 72 different combinations were obtained; implementing a design of this type with too many combinations would imply fatigue and a cognitive challenge for the respondents20, therefore it was proceeded to apply a fractional factorial design with an IV1/5 arrangement to obtain 16 combinations of representative attributes, these possible combinations were randomized to form eight selection sets.
Survey design
The survey consisted of three sections: 1) Introduction: respondents were presented with a general description of Seco Encerado cheese and its importance, as well as the purpose and implications of the survey; 2) Sociodemographic variables: consumers were asked to provide their sociodemographic data, gender, age, education level, occupation, monthly income and place of residence; 3) Choice experiment: consumers were presented with the 8 choice sets, which consisted of two purchase alternatives and one no-purchase alternative (Table 2).
Data analysis
Discrete choice experiment and willingness to pay
The data obtained were analyzed using econometric estimation with the mixed logit model (with and without interaction effects), as it has been considered the most appropriate model when dealing with aspects related to attributes21. This model considers the diversity of preferences by allowing the parameters to be randomly distributed in the sample population, so that everyone has their own preferences22. In this study, and to analyze consumer preferences in the consumption of Seco Encerado cheese, a mixed logit model was applied, in which the data from the discrete choice experiment were analyzed within a randomized utility framework, in which an individual (i) chooses the alternative that provides the highest utility among (j) alternatives in each choice occasion (t). The utility function takes the following form (Eq. 2):
Where:
For the logit model without interaction, the following variables were used: a) dependent variable: consumers' choice among the three alternatives (option A, option B and the no purchase alternative), which took values of 1 if an alternative was selected and 0 if the two alternatives were not selected; b) independent variables: attributes coded with their respective levels according to the experimental design. On the other hand, for the mixed logit model with interaction, the following variables are used: a) dependent variable: consumer's choice of the three alternatives; b) independent variables: attributes coded with their respective levels according to the experimental design.
Once the model parameters have been estimated, it is possible to calculate the marginal WTP for each attribute considered (the marginal substitution ratio between the attribute and the price). WTP is calculated using the equation already described22-24 by taking the quotient of the coefficient of the attribute (
Latent class logit model
A latent class logit model (LCLM) model was used to classify consumers into groups according to attribute preferences. This model is mainly applied to discrete variables, from the LCLM it is assumed that a constant vector β within each, identical homogeneous coefficients and heterogeneity between different groups of individuals, β that varies between groups25.
The LCLM model is mainly applied to studies with discrete variables25, from the generation of coefficient estimates it is possible to determine the membership of an individual
Where:
The LCLM was applied to the consumer variables of the three alternatives (option A, option B and no purchase alternative), obtaining two main classes characterized by the sociodemographic variables of the individuals belonging to each class. To compare the differences between the individuals of the classes, a K-proportions test was applied with X2 as the test statistic, following the procedure of Marascuilo. Analyses were performed using the XLSTAT statistical analysis software, version 2023.1.1 (Addinsoft®, USA).
Results and discussion
Sociodemographic characteristics of the sample
In Table 3, the descriptive statistics of the participants' sociodemographic information (n = 355) are presented. In the gender category, 36.62 % of the respondents are male, 61.41 % are female and 1.97 % preferred not to say; in the age category, the largest number of participants fell within the 18 to 35 yr range (40 %); the average income of the participants (68.45 %) is less than $10,000.00 MNX; in the occupation category, the majority of the participants are self-employed (30.14 %); while 72.96 % of the participants are from the South of Mexico. The analyzed sample exhibited a similar demographic distribution to Mexico’s general population. In Mexico, 52 % of the population is female and 48 % male, the average age is 29 yr, most individuals have a basic level of education, and the average income is $5,750.00 MXN27. The participation of consumers from southern Mexico is related to the production area of Seco Encerado cheese, located in the Istmo of Tehuantepec, Oaxaca, Mexico, a region that mainly includes Oaxaca, Chiapas, Tabasco and Veracruz28.
Table 3 Descriptivestatistics of the main sociodemographic variables reported by in the study (n= 355)
| Sociodemographic variables | Categories | Sample (n) | Sample (%) |
|---|---|---|---|
| Gender | Male | 130 | 36.62 |
| Female | 218 | 61.41 | |
| I prefer not to say | 7 | 1.97 | |
| Age | 18 - 35 | 142 | 40.00 |
| 36 - 45 | 130 | 36.62 | |
| > 46 | 83 | 23.38 | |
| School education | Low level of education | 223 | 62.82 |
| University studies | 132 | 37.18 | |
| Occupation | Self-employed | 107 | 30.14 |
| Unemployed | 13 | 3.66 | |
| Government employee | 81 | 22.82 | |
| Pensioner | 6 | 1.69 | |
| Private employee | 73 | 20.56 | |
| Student | 75 | 21.13 | |
| Income ($ MXN) | ≤ 10,000.00 | 243 | 68.45 |
| 10,001.00 A 20,000.00 | 92 | 25.92 | |
| > 20,000.00 | 20 | 5.63 | |
| Residence | North | 8 | 2.25 |
| West | 15 | 4.23 | |
| Centre | 73 | 20.56 | |
| South | 259 | 72.96 |
Mixed logit model on the valuation attributes
Table 4 shows the analysis of variance of the mixed logit model applied to the attributes and their different levels, this model was selected according to the criteria described29,30, who indicate that: i) the coefficients of the variables should have the expected signs, ii) the coefficients of the independent variables should be significant with a certain acceptable level of reliability, iii) the Log likelihood of the model should be large, and iv) the model should have a better fit in terms of the MacFaden R2. In this model, the main goodness-of-fit statistics were: the Log (likelihood) with a value of 8835.14 and the pseudo-R2 (McFadden) with a value of 0.210; according to Menozzi et al31, a Pseudo R2 (McFadden) value of 0.2 to 0.4 indicates an excellent model fit, while Valdivia et al30 indicated that these values (0.2 to 0.4) are equivalent to an R2 of 0.70 to 0.90 in an ordinary least squares model, indicating a good model fit.
Table 4 Analysisof variance of the logistic regression model used
| Source | DF | Chi- square (Wald) | P > Wald | P > LR |
|---|---|---|---|---|
| Price | 1 | 3.687 | 0.055 | < 0.0001 |
| Origin | 2 | 136.137 | < 0.0001 | < 0.0001 |
| Process | 1 | 5.913 | 0.015 | < 0.0001 |
| Price * Origin | 1 | 4.083 | 0.043 | < 0.0001 |
| Price * Process | 1 | 1.734 | 0.188 | < 0.0001 |
| Price * Label | 1 | 39.426 | < 0.0001 | < 0.0001 |
| Log (Likelihood) | 8850.14 | |||
| R2 (McFadden) | 0.210 |
Significant attributes (P≤0.05) identified were: price, origin, process and the of price with origin, process and label, these attributes are considered important in the valuation of the product by the consumer. Menozzi et al5, indicated that food quality systems, production methods, organic and origin labels are attributes that provide strategic elements to differentiate traditional cheeses from their competitors. Therefore, the valuation that consumers give to the significant attributes of Seco Encerado cheese constitutes an essential differentiating element for this product.
Table 5 shows consumers' preferences for specific attributes and their willingness to pay for each attribute, as well as their significance. In the case of the origin of the cheese, it was significant for both municipalities (βTapanatepec= 1.43; βZanatepec= 1.98). Consumers are willing to pay a price premium of $11.00 MXN per kg of cheese if it is produced in Tapanatepec, Oaxaca; however, if the cheese is produced in the Zanatepec, Oaxaca, there is a greater willingness to pay an additional $ 15.60 MXN per kg of cheese. The process attribute (β = 0.70) is important in the consumer's purchase choice and there is a willingness to pay an extra $ 4.80 MXN per kilo of cheese if the cheese is produced using an artisanal production process. The interaction price*label with information (β= 0.191) was found to be positive, consumers are interested and consider it positive that the product has a packaging label that provides information on the main characteristics associated with the product and nutritional information; the consumer is willing to pay an extra $ 1.60 MXN for a label that provides this information.
Table 5 Consumer preferences and WTP for Seco Encerado cheese
| Attribute | Coefficient β | Standard error |
P > Chi2 | WTP ($ MXN) |
|---|---|---|---|---|
| Price | 0.12 | 0.06 | 0.055 | |
| Origin (Tapanatepec) | 1.43 | 0.15 | < 0.0001 | 11.0 |
| Origin (Zanatepec) | 1.98 | 0.28 | < 0.0001 | 15.6 |
| Process (Artisanal) | 0.70 | 0.29 | 0.015 | 4.8 |
| Price* Tapanatepec | 0.38 | 0.19 | 0.043 | -3.2 |
| Price* Artisanal | 0.24 | 0.18 | 0.088 | 2.1 |
| Price* Label with information |
0.19 | 0.03 | < 0.0001 | 1.6 |
WTP= willingness to pay.
In this study, positive WTP in terms of origin, process and labeling, these aspects are closely related to consumers' eating habits, while for the interaction price*Tapanatepec there is a negative WTP, so consumers are not willing to pay a premium for these attributes because they are indifferent to them32. In addition to various psychological and sociodemographic factors, these habits are also influenced33. Nowadays, the food market has diversified to satisfy consumer needs34. For this reason, it is necessary to implement strategies that allow traditional foods to satisfy consumers, thus improving territorial development and generating economic benefits for producers35. In the case of Seco Encerado cheese, it is essential to promote its origin and introduce informative labels.
Valuation attributes and willingness to pay
Table 6 shows the results of the consumer profiles obtained through the logit analysis of the latent classes; two classes of consumers were identified. Cluster 1 ' Artisanal cheese enthusiasts', who value the attributes: Process (artisanal), ripened time (dry), label (with information) and Price* ripened time (dry), and have a positive WTP of $ 2.79 MXN, $ 7.32 MXN, $ 5.26 MXN and $ 22.02 MNX, respectively, for each kilogram of cheese. For this group of consumers, the value of the artisanal process and the 'dry' ripening process stand out; they prefer the cheese for its indigenous quality characteristics and are willing to pay a premium for these characteristics.
Table 6 Consumer ranking profiles and WTP ($ MXN) for different attributes of Seco Encerado cheese
| Source | Cluster 1 'Artisanal cheese enthusiasts' | Cluster 2 'Cheese consumers' | ||||
|---|---|---|---|---|---|---|
| Coefficient β | P> Chi2 | WTP | Coefficient β | P > Chi2 | WTP | |
| Price | -1.45 | < 0.0001 | ||||
| Origin (Tapanatepec) | 8.42 | < 0.0001 | -6.8 | |||
| Origin (Zanatepec) | 2.72 | < 0.0001 | -2.87 | |||
| Process (Artisanal) | 4.37 | 0.0153 | 2.79 | 6.71 | < 0.0001 | -5.62 |
| Ripening time (semi-dry) | -8.30 | < 0.0001 | 4.71 | |||
| Ripening time (dry) | 4.83 | 0.0003 | 7.32 | -3.16 | < 0.0001 | 1.17 |
| Label (with information) | -1.25 | 0.0018 | 5.26 | |||
| Price*Origin (Zanatepec) | 2.70 | < 0.0001 | -2.9 | |||
| Price*Process (Artisanal) | -3.98 | < 0.0001 | 1.74 | |||
| Price* Ripening time (semi-dry) | 4.16 | < 0.0001 | -3.86 | |||
| Price* Ripening time (dry) | -4.21 | < 0.0001 | 22.02 | |||
WTP= willingness to pay.
The higher valuation and WTP for attributes related to the process (artisanal) and ripening time are related to consumers' knowledge of traditional products, resulting in a higher valuation of local products and conceptualizations that associate artisanal food with higher quality36. Regarding consumers' preference for information labels, Stiletto et al37 defined that consumers consider the label as an important part of their buying decision, as it depends on the label whether the food is considered: "healthy", "nutritionally valid" or "frequently consumed". The use of different food labels can help consumers of Seco Encerado cheese to influence their purchase choices, satisfy their motivational needs (ethical, environmental and directly link the product to its origin.
In class 2 'Cheese consumers', who value positively the attributes: ripening time (semi-dry), ripening time (dry) and price*process (artisanal), there is a positive WTP of $4.71 MNX, $1.17 MNX and $1.1.17 and $1.74, respectively, for each kilo of cheese, this group of consumers is willing to pay a price premium for the cheese but is not willing to consume the cheese with the artisanal ripening process that it undergoes.
In the results of the sociodemographic characterization of the classes (Table 7), the main differences found were that the class of 'Artisanal cheese enthusiasts' is composed of a greater number of men. Most of the members of this class have a low level of school education, their source of employment is self-employed (they are usually housewives, farmers, transport workers, etc.) and they are consumers originally from the south of Mexico and can therefore be called native consumers of this cheese.
Table 7 : Sociodemographiccharacterization of consumer groups
| Categories | 'Artisanal cheese enthusiasts' (n= 185) | 'Cheese consumers' (n= 170) | |
|---|---|---|---|
| Gender | Male | 73 b | 57 a |
| Female | 108 | 110 | |
| I prefer not to say | 4 | 3 | |
| Age | 18 - 35 | 70 | 72 |
| 36 - 45 | 72 | 58 | |
| > 46 | 43 | 40 | |
| School education |
Low level of education | 122 b | 101 a |
| University studies | 63 | 69 | |
| Occupation | Self-employed | 61 b | 46 a |
| Unemployed | 5 | 8 | |
| Government employee | 41 | 40 | |
| Pensioner | 4 | 2 | |
| Private employee | 39 | 34 | |
| Student | 35 | 40 | |
| Income ($ MXN) | ≤ 10,00.00 | 133 b | 110 a |
| 10,001.00 a 20, 000.00 | 40 | 52 | |
| > 20,000.00 | 12 | 8 | |
| Residence | North | 5 | 3 |
| West | 7 | 8 | |
| Centre | 34 | 39 | |
| South | 139 b | ||
ab Different letters indicate difference (P≤0.05) between the categories.
'Artisanal cheese enthusiasts' are mainly men who express a preference for mature cheeses, and this preference can be derived from the typical sensory characteristics of these cheeses, which are related to intense flavors and aromas38. In addition, men pay more attention to the sensory characteristics of the product, as their preferences are mainly related to hedonic attributes39, while women, who prioritize utilitarian attributes, related to price and nutrition40.
The relationship between the origin of the consumer and the origin of the product is strongly linked, a phenomenon known as consumer ethnocentrism, which is mainly observed in regions with a strong local identity and a high demand for local products41. This group of local consumers exhibits similar behavior to that described42, where dairy consumers show a higher preference for products of a specific origin, perceiving them as authentic and higher quality. The psychological relationships that consumers associate with cheese and the knowledge of cheese are the main factors that allow this group to show a higher willingness to pay for Seco Encerado cheese.
In this research, different consumer associations with the territory of origin of the product have been identified, Zhe et al43 define these associations as attachment to place, in this concept influence different characteristics of consumer satisfaction (symbolic differentiation, sense of belonging, continuity, security and sense of belonging that traditional foods have), so it is important to study the associations that consumers have with the territory, paying special attention to its promotion and economic legal protection44.
Conclusions and implications
The attributes that directly influenced the valuation of the cheeses were: price, origin (Tapanatepec and Zanatepec), process (artisanal) and the interaction between price and process (price*artisanal) and process and label (price*label with information), for which consumers are willing to pay a price premium. Based on the above, it can conclude that consumers are looking for a traditional product produced in an artisanal way, they also showed interest in knowing the nutritional information of the product because they prefer a cheese with a label that includes the characteristics of the product to make a correct purchase decision according to their lifestyle. The latent class analysis identified two classes: "'Artisanal cheese enthusiasts'", who value the attributes Process (Artisanal), Ripening (Dry), Label (With Information) and Price*Ripening (Dry) and prefer to consume Dry Waxed Cheese with a traditional process, including the ripening to which it is subjected; while the second class "Cheese Consumers" consider the attributes Ripening (Oreado) and Price*Process (Artisanal), this group of consumers is willing to consume any type of cheese, except ripened cheeses. This study presents two main limitations: the first is related to the non-probabilistic sampling method, as it is not usually representative of the entire sample of cheese consumers. The second stems from the DCE, since the stated preference methods used in the experimental design may lead to overestimations or underestimations of the declared WTP values. Despite these limitations, this research contributes to a better understanding of consumer behavior and their WTP for Seco Encerado cheese.
