Acceptance of coffee by different consumer profiles using multivariate statistics

This study aimed to evaluate the acceptance of coffee by consumers using olfactory sensory analysis in two moments: without and with information about the different classifications of the products. Three different coffee classifications (rio, hard, and soft) and a commercial sample of a Traditional type of coffee were used. The research was carried out in a supermarket in Lavras – MG Brazil with 100 volunteers. Consumers were asked to evaluate the samples according to the 9-point hedonic scale without any prior information on coffee quality. Subsequently, the same test was applied after informing the consumers that there are differences in coffee quality (high, intermediate and low), which have different characteristics, including aroma. The most accepted samples were associated by consumers with coffees classified as hard and soft. The information provided about the differences in the quality of the samples influenced only females consumers with higher incomes.


Introduction
Coffee is one of the most consumed beverages worldwide and Brazil is the largest producer and exporter, also, it is the second-largest consumer of this commodity. The Brazilian Coffee Industry Association -ABIC shows that consumption remains concentrated in homes and 81% of the total consumption is of roasted and ground coffee (ABIC, 2019).
Over the years, the need to understand consumer behavior has grown, and today it is the focus of investigations, especially by industries. The sensory senses are among the mechanisms that contribute to the perception of the quality of food products by consumers, which provide multiple options for creating and modifying perceptions of quality (ASIOLI et al., 2017). In this sense, the olfactory perception is gradually being incorporated into the experience of the multisensory product and can be used as an attractive, conservation and even as a quality differentiator, already demonstrated in an application for special coffees in packaging with aromatic valves (Motoki et al., 2018).
Much of the perception of food quality is induced by the product's physical properties through the senses as well as the expectations placed concerning contextual information (Kumpulainem et al., 2018). Most packages of roasted and ground coffee in Brazil do not provide descriptions that allow careful selection by the consumer. Inferior quality coffees, adulterated with impurities or other defects are commercialized and, due to customs and lack of information on the part of the population, many consumers take into consideration only the price and the brands of preference when buying the product (Benedito, 2019).
It is known that there is a greater appreciation of products when people have prior knowledge about quality (Botelho et al., 2017). In this sense, the objective of this study was to evaluate the acceptance of coffee quality by consumers using olfactory sensory analysis in Research, Society and Development, v. 9, n. 6, e102963592, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i6.3592 4 two moments: without and with information about the different product classifications. In addition, it was possible to verify whether there is a difference in the preference of different consumer profiles and whether information about quality influences that preference.

Material and Methods
The present research can be classified as quali-quantitative, basic and experimental (Pereira et al., 2018). It was conducted in Lavras -MG, Brazil, with visits to a supermarket located in the central region of the city, whose choice was due to the intense flow of customers.
A qualitative structured questionnaire was applied, voluntarily, with 100 intervieweds, composed of closed questions, to characterize the consumer profile (Pereira, 2018). It is important to highlight that the information about being a coffee consumer was decisive to continue the sensory preference test. The volunteers were randomly chosen, at different times of the day and on different days of the week. The interaction between researcher and interviewee was minimized and only the academic nature of the research was explained.

Samples of coffees obtained from the Mogiana region, located in the state of São
Paulo, were used in three different classifications (rio, hard, and soft), carried out employing the cup tasting (Brasil, 2003), and a commercial sample of coffee of the Traditional type, as shown in Table 1. old, with about 40% earning up to 1000 reals of monthly income and only 17% over 3000 reals. From the acquired data, it was possible to carry out Principal Component Analyzes -PCA correlating the samples and the consumer profile before the information on the quality of the coffees (Figure 1).  Figure 1 shows that, before the information, samples 1 and 2 were not the most preferred of the analyzed profiles, and consumers with an income of 1000 to 3000 reals, male and age up to 55 years preferred sample 4, presented in 4th quadrant of Figure 1. Profiles with income up to 1000 reals, female and over 55 years old preferred sample 3, presented in the 2nd quadrant of Figure 1.
From the data acquired after the information on the quality of coffees was passed on to consumers, Principal Component Analyzes -PCA was also performed correlating the samples and the profile of consumers ( Figure 2).
Research, Society and Development, v. 9, n. 6, e102963592, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i6.3592 The highest acceptance of the different profiles analyzed (monthly income, age, and sex) was for the coffee samples 3 and 4, hard and soft drinks, respectively, despite the information about the quality of the coffees. Consumers with monthly incomes up to R$ 1000 and above R$ 3000, age over 55 and females presented a trend towards hard coffees ( Figure   1). Meanwhile, consumers with an income of up to 1000 reals and age over 55 years were not influenced by the information. Besides, consumers with an income above 3000 reais and females started to prefer soft coffees samples (Figure 2). Consumers with an income of 1000 to 3000 reais, aged over 35 years and males also tended to the soft coffees samples (Figure 1), maintained after the information, except for the male sex that started to prefer hard coffees samples, after information ( Figure 2).
In contrast to this study, Aschemann-Witzel et al. (2018) realized that the information was better understanding by men and low-income people than women and middle and highincome people. On the other hand, Mora et al. (2018) observed that the age and sex of