The symbolic value of wine, moderating and mediating factors and their relationship to consumer purchase intention

Abstract

Purpose

The symbolic value of wine is a relevant research topic and raises the interest in studies in both the enological and market areas. In this context, this study aims to understand the role of the symbolic value of wine and its relationship to the product purchase intention.

Design/methodology/approach

The study is based on a survey of 269 wine consumers from Brazil. The basic theoretical framework includes three latent constructs (symbolic value, consumer attitude and product-norm experience) and three moderators (consumer involvement, willingness to pay and consumer preference). Relations between these are analyzed using the confirmatory factor analysis (CFA) and the moderated mediation analyses using Haye’s process.

Findings

This research identified that the symbolic value is totally mediated by consumers’ attitudes and product-norm experiences. However, such a relationship occurs directly for consumers with higher involvement with the product, higher willingness to pay, and who assume that wine is their preferred alcoholic beverage.

Originality/value

One of the contributions is to emphasize the symbolic value of wine and highlight how the relationship with different factors can interfere and explain consumer purchase intention and can influence the strategies, actions and investments of companies in the sector.

Citation

De Toni, D., Pompermayer, R., Lazzari, F. and Milan, G.S. (2022), "The symbolic value of wine, moderating and mediating factors and their relationship to consumer purchase intention", International Journal of Wine Business Research, Vol. 34 No. 2, pp. 190-211. https://doi.org/10.1108/IJWBR-01-2021-0006

1. Introduction

Wine, a millennial beverage, carries various utilitarian meanings and symbols, translating it into the “drink of the Gods”. Among the utilitarian meanings are the shape of the bottle, its flavor, its aroma and the grape type, among other aspects. On the other hand, symbolic meanings carry a variety of attributes linked to pleasure, well-being and customer satisfaction (Beckert et al., 2014; De Toni, 2005; Kim and Cho, 2020). Therefore, wine is an intense product in signs or symbols (Atkin and Thach, 2012; Beckert et al., 2014; Bruwer and Buller, 2013; Lockshin and Corsi, 2012) , translated it into attitudes, experiences and behaviors (Calvo-Porral et al., 2019; Danner et al., 2020).

The symbolic value of wine can lead to a more favorable attitude and more intense sensory, emotional, cognitive and social experiences (Bernritter et al., 2017). According to the theory of planned behavior (Ajzen, 1985), individual intention and consumer behavior are determined by their attitudes (positive or negative evaluation), subjective norms (for example, the perception of social pressure), perceived behavioral control (for example, ease or perceived difficulty to perform some behavior) and intention which is seen as the main influencer of a certain behavior (Ajzen, 2015). Thus, such a theory has been considered a critical decision-making model to predict a wide range of intentions and behaviors in consumer behavior research (Xu, 2020).

As wine is a product associated with cultural and symbolic aspects (Beckert et al., 2014), it is relevant to investigate, in different contexts, considering its particularities, the traits and possible behaviors of the wine consumer (Meler et al., 2016), factors that may explain the purchase intention (Barber et al., 2010a; Barber et al., 2010b; Barber and Taylor, 2013), the decision-making process and consumer behavior (Perovic, 2014; Danner et al., 2020).

More specifically, through this research, it is sought to generate empirical evidence in relation to aspects intrinsic to the symbolic value, attitude and product-norm experience regarding their intention to purchase wines (Monteiro et al., 2019; Bonn et al., 2020). In addition the consumer’s involvement with the product, their willingness to pay and their preference for drinking as moderating constructs were investigated to broaden the understanding of the formation of the product purchase intention (wines). In this sense, this study aimed to understand the role of the symbolic value of wine and its relationship with product purchase behavior.

Among the research contributions, identifying the importance the symbolic value assumes in the wine consumer behavior bulges. Besides, it was observed that the symbolic value, when interacting with the consumer attitude, experiences, involvement, willingness to pay and the position the product occupies among the preferred alcoholic beverages, interfered in a distinct but complementary way the (re)purchase intention. This study also brings managerial implications, introducing the symbolic elements of wine and allowing a better understanding of this product purchase journey, generating insights to understand better consumer behavior, which may underpin more assertive strategies, actions and investments for companies and their managers operating in the wine market.

2. Literature review

2.1 Purchase intention

Consumer purchase intention is defined as consumers’ conscious plans to buy a product, resulting from a personal process and an evaluative and a normative judgment (Visentin et al., 2019). The consumer choice process involving wine is known as something complex. Ajzen (1985, 2015) provides some aspects on how attitude, perceived behavioral control, subjective norm and intention have an impact on certain behavior (Caliskan et al., 2021).

It involves several extrinsic factors, namely, varied attributes that do not necessarily change the product, such as prices, brand, packaging and labeling, and intrinsic factors, such as aspects related to taste, acidity and wine coloring, directly associated with the product itself and considered for the customer choice (Batt and Dean, 2000). Therefore, consumers purchase goods basically for two basic reasons: first, consummatory affective (hedonic) gratification from sensory attributes; and second for instrumental, utilitarian reason (Batra and Ahtola, 1990). In this way, the purchase intention of wine can be for a hedonic reason, such as pleasure, status, satisfaction, accomplishment, etc. or also for a utilitarian reason, such as health, thirst, wine pairing with some specific food, etc. Thus, it is possible to say that the purchase intention (in this paper the wine purchase intention) is a complex activity that involves understanding the customers’ processes throughout pre-purchase, purchase and post-purchase processes (Lemon and Verhoef, 2016).

According to Alawan (2018), the purchase intention is positively influenced when the product information content is relevant for the consumer. In this sense, Zubsek et al. (2017) presented arguments supporting the assumption that consumer behavior patterns tend to represent their product preferences, which can and should be considered by marketing professionals to improve companies’ offers.

2.2 Consumer attitude

Arnould et al. (2004) affirmed that the attitude consists of three interrelated components, the affective, the cognitive and the behavioral dimensions. The affective component considers feelings, states of mind, emotions and evaluations about the object. The cognitive component includes knowledge and evaluative beliefs about the object. On the other hand, the behavioral component involves actions and intentions of future behavior concerning the object. In this sense, attitudes are personal nature dispositions that bring an individual closer to or away from an idea or concept, involving affection and action, which directly influences behavior (Pimenta et al., 2009). According to the theory of planned behavior, attitude is a “degree in which a person has a positive or negative judgment of acting a behavior” (Caliskan et al., 2021, p.3). Homer (2006) identifies that the consumer attitudes can be nature bidimensional in which the consumer behaviors are driven by hedonic (or affective) gratification (e.g. agreeable/disagreeable, pleasant/unpleasant) and utilitarian (or instrumental) motives (e.g. worthless/valuable, harmful/beneficial). The hedonic dimension is the result of sensations derived from the experience of using products; the utilitarian dimension derives from the function performed by products. Hedonic and utilitarian are two distinct dimensions of brand attitude and they work independently without a predefined order (Voss et al., 2003).

2.3 Relationship between consumer attitude and purchase intention

Studies have concluded that attitudes resulting from product stereotypes can affect a person’s willingness to purchase them (Johnstone and Hooper, 2016). Thus, consumers can form attitudes about a particular product even if they have not directly experienced the product. In turn, the intentions demonstrate how much the individual is willing to try a particular result or to engage in a particular attitude (Dreger et al., 2017).

According to Han et al. (2018), the component elements of attitude and purchase intention refer mainly to the individual objectives related to a particular behavior. Furthermore, a lot of research indicated that attitudes toward the brands/products could influence future behaviors (Glasman and Albarracin, 2006; Voss et al., 2003). Howcroft et al. (2002), the study of attitudes is important in consumer behavior because it plays a vital role in the purchasing decision-making process, besides helping marketing professionals to understand consumers and predict and influence their purchasing behavior. Concerning consumers' attitudes towards wines, favorable attitudes or beliefs can positively impact this product purchase behavior. The Theory of Planned Behavior (Ajzen, 1985) identified a direct and causal relationship between consumer attitudes and behavioral intention. The more favorable the attitudes, the higher the purchase intention (Xu, 2020). That said, the first research hypothesis presented is as follows:

H1.

The consumer’s favorable attitude towards wine positively impacts the product purchase intention.

2.4 Product-norm experience

Lemon and Verhoef (2016) conceptualized the consumer experience as a multidimensional construct that focuses on the customer's cognitive, emotional, behavioral and social responses to the firm offerings throughout the purchase journey. Gentile et al. (2007) emphasized that the consumer experience is strictly personal and implies several perception systems levels: the rational or cognitive, emotional, sensory, physical and spiritual levels.

In this paper the term product-norm experience was used representing a perception more in the rational level of the experience. Sharma and Patterson (2000, p. 6) identify that the product-norm experience “represents prior product knowledge and information about how a product would perform”. Therefore, these norms are limited by the width (variety of brands/products experienced) and depth (extent or number of times each brand/products was used) of a consumer´s experience with a variety of products and brands. In this way, consumers may develop certain norms about the performance of the wine or winery based on the experience they have accumulated, and they develop confidence in assessing the performance of the wine that they will consume.

2.5 Relationship between product-norm experience and purchase intention

Consumers are no longer only interested in buying products or services for their functional benefits, but they are also focused on the fundamental aspects of the experience (Özer and Köse, 2013; Zarantonello and Schmitt, 2010). The consumer experience is one of the most important research challenges for the coming years due to the increasing number and complexity of points of sale and the belief that creating intense positive experiences at the time of purchase will improve company performance through higher effective purchase rates, customer retention or even loyalty and positive word-of-mouth (Homburg et al., 2015). Companies such as Amazon and Google have managers who specifically take care of creating and managing their customers’ experiences. Managing consumer experiences opens an opportunity to create a strong and lasting relationship with customers (Lemon and Verhoef, 2016).

Therefore, companies’ orientation to provide a positive experience to the consumer allows a differentiation against competitors, establishing emotional connections, stimulating the retention or even loyalty of consumers and the positive word-of-mouth of their products and brands (Candi et al., 2013). For Puccinelli et al. (2009), various consumer behavior elements (such as objectives, memory, involvement, attitudes, affection and atmosphere) influence the consumer experience perception and play an important role in all phases of the consumer purchase decision process.

Sharma and Patterson (2000), confirm that the product-norm experience is an important factor in the development of a long-term relationship, therefore customers with high experience will intensively evaluate the performance of product/service determining their trust and satisfaction, and consequently form their commitment, loyalty or purchase intentions. But also, their intention to buy is limited by the variety of brands known by the customers and the extent or number of times that each brand has been used by them. Thus, the second research hypothesis emerges:

H2.

The product-norm experience of the consumer with the product (wine) positively impacts the purchase intention.

2.6 Symbolic value

People buy products not only for their functional utility but also for the meaning that this product or brand has. The symbolic value can represent for the individual stands out, as it can awaken positive and/or negative emotions, in addition to the aspects related to the connections with the individual’s life purpose (Allen, 2006). The same way, the symbolic value refers to the brands ability to signal social status and enhance the self-concept of its users (Park et al., 1986; O’Cass and Frost, 2002), where status is typically defined as relative respect or esteem. The main element of the symbolic value of a product or service is related to the feeling of power or lack thereof. Consumers who deem the symbolic value as important are more concerned with this social sign and are more sensitive to other attributions (Ferreira and Coelho, 2015). In a scale validation or perceived value, Sweeney and Soutar (2001) say the value has four dimensions: emotional, functional (value for money), functional (performance/quality) and social. However, it is possible to infer that the symbolic value has an emotional and social dimensions. The emotional dimension means the utility derived from the feelings or affective states that a product generates (e.g. feel good, pleasure, enjoy, status […]) and the social dimension mean the utility derived from the product’s ability to enhance social self-concept (feel acceptable, social approval, power) (Sweeney and Soutar, 2001).

Both power and status are considered relative variables of symbolism. In this way, feeling powerless increases consumers´ desire to acquire products associated with status and their willingness to pay more for this product. In this way, status is often one signal of power, that being so, obtaining or demonstrating status is one way to obtain or restore power, and products are a means to signal one’s status (De Toni and Mazzon, 2014; De Toni, 2019; Rucker and Galinsky, 2008).

Cezanne and Saglietto (2014) considered that symbolic value creates a relationship between consumers and their community and described this connection as signs of emotional and moral belonging to a cultural and social group defined as a reference. For Levy (1981), products are used symbolically, and talking about their uses is a way of symbolizing life. In this way, many famous brands are purchased not only because of any inherent functional value but because they signal a form of social status or power.

Liu et al. (2012) reinforced that symbolic value could be attributed to a product based on its relevant cultural and social meanings. Accordingly, Smith and Colgate (2017) stated that symbolic value is the measure in which consumers attribute or associate psychological meaning to a product and/or service. High-priced, high-quality brands are associated with high prestige that enhances their self-concept or symbolic value (Lichtenstein et al., 1993; Vigneron and Johnson, 1999), these brands can become part of the extended self (Belk, 1988) and thereby move their perceived actual self-closer to their ideal self (Sirgy, 1982).

The imaginary about wine, complex and plural, make it a rich set of living collective representations that articulates tradition and modernity because wine is the target of different investments, whether symbolic or monetary (Siqueira and Coelli, 2019). Famous wine brands associated with higher price and higher quality induce the consumer to a higher perception of value through its benefit component (cognitive factor) and higher status (affective factor).The perception of wine quality, for example, is only partially derived from its material and sensory characteristics. Its value derives mostly from the symbolic qualities attributed to products based on the subjective variables of interpretation used by the consumer (Beckert et al., 2014).

2.7 Relationship between symbolic value, consumer attitude and product-norm experience

The product complexity has been emphasized by authors who demonstrated there are more intangible factors to wine than the tangible qualities (Mcintyre et al., 2016). For Benaim (2018), symbolic value is a key factor for competition, especially in some sectors/activities capable of generating, absorbing and spreading meaning. For Rodriguez (2017), goods perception and social acceptance make them more desirable to be acquired by individuals, adding symbolic value to products, such as wine. Both attitude and experience impact consumer purchase intentions or purchases.

According to Bernritter et al. (2017), perceptions of a brand or product symbolic value relate to consumer attitude and vice versa. Furthermore, Leigh and Gabel (1992) stated that the symbolic value could increase or decrease purchase probability, depending on whether the consumer identifies or maintains positive or negative attitudes towards the reference group. Studies have shown that psychological states of power or status have important consequences for how people behave (Rucker and Galinsky, 2008). Therefore, the symbolism wine represents to the individual has a strong power to manifest favorable attitudes, consequently impacting their purchase intention. Thus, the third hypothesis of the research is presented:

H3.

The symbolic value with the product (wine) positively impacts on the consumer attitude.

On the other hand, experience as the act of experiencing and observing an event, as a way of learning, knowledge, or acquired ability, also interferes with how the individual interprets and behaves about the object (Hoch, 2002). To Puccinelli et al. (2009) a lot of elements of consumer behavior (such as memory, involvement, attitude, etc.) influence the consumer´s experience and is an important factor in the process of consumer´s decision making. The experience acquired can produce many different results and feelings, depending on the consumer objective, which may vary according to the purchase motivation or occasion (Puccinelli et al., 2009).

The product-norm experience can result from the values and symbolism that a product or service represents to the individual. Nagle and Holden (2003) highlighted that consumers acquire products with a symbolic appeal because they want to communicate to others that they can afford them, and this results in a greater power of social insertion and involvement. This way, it is possible to present the fourth hypothesis of research:

H4.

The symbolic value with the product (wine) positively impacts on the product-norm experience of the consumer.

2.8 Moderating factors

Zaichkowsky (1994) defined consumer involvement with a particular product and/or service as the degree of consumer’s perceived relevance to an object based on inherent needs, values and interests, demographic and socioeconomic variables. In this sense, Bruwer and Buller (2013) analyzed the wine consumer behavior based on their involvement with the product, enhancing brand loyalty and countries of origin preference.

For Borgogno et al. (2015), symbolic values are associated with products that the consumer has a high level of familiarity with, which is linked to the level of consumer involvement with such products. Consequently, involvement has an indirect effect on consumer loyalty that is dependent on the price perception that mediates this relationship, indicating that the increase in involvement leads to an increase in loyalty, directly and indirectly, via the positive effect of price perception (Ferreira and Coelho, 2015). In other words, higher involvement with wine may lead consumers to invest more time examining the product alternatives and attributes and make more positive associations about price perception, also impacted by the symbolic value.

Highly involved consumers tend to use more product suggestions in their purchasing decisions and are interested in learning more about wine, while low-involved consumers tend to simplify their choice decisions by relying on strongly price-based suggestions (Hollebeek et al., 2007). In this sense, Qi and Tang (2011) identified that the association of symbolic value with highly involved products could inspire a higher desire to buy or consume the product. Therefore, the following research hypothesis may be presented:

H5.

Consumer involvement with the product (wine) moderates the relationship between the symbolic value and the purchase intention, and this relationship is stronger for consumers with higher product involvement.

The willingness to pay is the maximum amount a consumer is willing to pay for a product or service (Vock et al., 2013). Studies by Ramirez and Goldsmith (2009) and Goldsmith et al. (2010) indicated that higher levels of involvement, innovation and brand loyalty lead to a lower price sensitivity level. Consumers are willing to pay more when they are more involved, innovative and loyal to the brand or a specific product category.

However, without the ability to judge a product from its intrinsic attributes, consumers with high product involvement or concerned with quality tend to differentiate the product quality from extrinsic tips, in which price is a source of information, beyond the company’s brand and reputation (O’Neill and Lambert, 2001).

Rucker and Galinsky (2008) in a study about power, status and willingness to pay, identify that consumer in a state of low power might desire to acquire products that naturally signal status to others (e.g. executive pens, some specific wine) and this is evidenced by an increased willingness to pay for such goods. In this sense, it is possible to identify that the willingness to pay and consequently the buying intention for a specific product might increase when this product has a symbolic value (status and power) for the consumers.

On the other hand, for Reutterer et al. (2006), the product intrinsic factors, often disregarded by the offerer, can generate important impacts on the consumer willingness to pay. Many studies identified a positive relationship between the willingness to pay and the symbolic value. Consumers generally tend to be willing to pay more for products with higher symbolic value (De Toni et al., 2017; De Toni et al., 2020). The relationship between symbolic value and purchase intention tends to be more strongly related when there is a higher willingness to pay for the product (Grankvist et al., 2019). Consequently, the sixth research hypothesis follows:

H6.

The willingness to pay for the product (wine) moderates the relationship between the symbolic value and purchase intention, and this relationship is stronger for consumers with a higher willingness to pay for the product.

Another significant moderator that influences the relationship between symbolic value and purchase intention is the position the product occupies in the consumer choice spectrum and decision. For Rodríguez Díaz (2014), similar products have been acquiring different forms of consumption and various symbolic meanings throughout history, closely linked to the moment spirit. The symbolic value provides a connection between the consumer and the community. Sociologists describe these signs as emotional and moral signs belonging to a cultural and social group defined as a reference (Cezanne and Saglietto, 2014). Therefore, the preference for certain beverages, in the case of this research, wine, over others, is strongly linked to the social and cultural context in which the individual belongs, impacting on their purchasing and consumption preferences (Cezanne and Saglietto, 2014). This relation between symbolic value and purchase intention can be related to the consumer´s preference. For example, when the product (in this case wine) represents status or power this product can be the preferred alcoholic beverage and consequently be stronger the relation between symbolic value and purchase intention (Rucker and Galinsky, 2008). Accordingly, the seventh research hypothesis is presented:

H7.

The consumer´s preference of the product (wine) over the other types of alcoholic beverages moderates the relationship between the symbolic value and the purchase intention, and such relationship is stronger for consumers who have wine as their favorite alcoholic beverage.

3. Research method

The research method is quantitative-descriptive, implemented through a single cross-section survey (Hair et al., 2018. Malhotra et al., 2012). The population considered for this research includes wine consumers in general, without specifying the wine vintage, from different age groups and regions of Brazil.

3.1 Data collection instrument

Validated scales were converted to the Portuguese language through the reverse translation technique to build the data collection instrument (Malhotra et al., 2012). For the purchase intention construct, it was used a four-item scale (PI_02 to 05) adapted from Mittal et al. (1998) and Wu et al. (2014). For the consumer attitude, an eight-item scale (CA_06 to 13) was adapted from Homer (2006) and Voss et al. (2003). However, Homer (2006) used a bidimensional scale (hedonic and utilitarian attitude), also the study of Voss et al. (2003) identified that beer and/or alcoholic beverages predominantly have high hedonic attitude and low utilitarian attitude in a semantic differential scale. This research used only the hedonic dimension because of the nature of this study that seeks to identify the symbolic elements of the wine also was used seven-point Likert scale. Using the seven-point scale instead of the semantic differential scale didn´t affect the finding once the result presents good reliability (>0,7) and average variance extracted (AVE) higher than 0,5 (Table 1). The Likert-type scale has been widely used in research using SEM (Byrne, 2016), as well as the psychometric scale most commonly used in measurements that require self-reporting, that is, the method of self-completion (Malhotra et al., 2012). The same way, other studies also use the Likert scale to measure the consumer attitude and present a good reliability (Caliskan et al., 2021). The product-norm experience of wine consumer is based on product knowledge and was measured using a four-item scale (PE_14 to 17), adapted from Sharma and Patterson (2000).

The symbolic value was measured with a four-item scale (SV_18 to 21), adapted from De Toni and Mazzon (2014). This scale was chosen first because it reflects well the symbolic value of the construct associated with the social value of power and status (Rucker and Galinsky, 2008; Sweeney and Soutar, 2001) and second because it has already been tested and validated in the Brazilian context by Scopel (2014), De Toni (2019), Rizzon (2017), Graciola et al. (2018) and adapted to this study.

It is worth noting that all scales were composed of closed-ended questions, evaluated according to a seven-point Likert type scale (Bearden et al., 2011), with the extremes “1. I totally disagree” and “7. I totally agree”. From the exploratory factor analysis with the Bartlett test of sphericity and the convergent validity of the constructs, we observed that variable 7 “For me consuming wine is refreshing” of the Consumer Attitude construct the factor loading were not sufficiently high, (Table 1) less that 0.50 (Hair et al., 2020). After knowledgeable expert analysis, such variables would not compromise the construct contents validity, and we opted for their exclusion.

The discriminant validity was measured applying the Fornell and Larcker (1981) method. The results obtained confirm the discriminant validity between the constructs since the extracted variance reached values higher than the shared variance, as shown in Table 2. It was also made the test of discriminant validity with the Heterotrait–Monotrait method. The result shows that all correlations are less than 0,60. This confirms, according to what the literature shows (less than 0.8), that the construct has a good discriminant validity (Kline, 2011).

A 17-item scale adapted from Zaichkowsky (1986) was used to measure the moderation of consumer involvement with the product. After checking the construct unidimensionality, applying exploratory factor analysis with one factor, all items remained. The variable average was calculated, and the database was divided based on the higher or lesser degree of involvement. The first and second quintiles (40% of the sample) were considered low involvement, and the fourth and fifth quintiles (40% of the sample) and high involvement, disregarding the 20% of the respondents belonging to the third quintile. Accordingly, the database was divided into two groups: 107 cases for “low involvement” (mean < 5.4) and 104 cases for “high involvement” (mean > 6.13).

The sample was divided into people (consumers) willing to pay up to R$30 for a bottle of wine (58 cases) and people willing to pay more than R$30 for a bottle of wine (211 cases) to measure the moderation of willingness to pay. The option for the cut of R$30 was mainly due to the strategic positioning of wine prices, researched in retail and up to R$30 are considered everyday wines while wines above R$30 are considered noble wines with superior quality. According to Silveira et al. (2020), it is noticeable that most of the participants who do not have a higher income analyze the price and quality at the moment of acquiring the wine, and depending on the moment, the quality prevails and not the price. Regarding the position that wine occupies as a preferred alcoholic beverage, 130 respondents have wine as their first alcoholic beverage option, and for another 139 cases, the wine was not considered the first alcoholic beverage option.

3.2 Data collection, processing and analysis

The data collection was performed digitally by sending the survey link to Brazilian wine consumers in February and March 2020. In total, there were 20 questions related to the construct proposed (Table 1), 17 questions related to the involvement (Zaichkowsky, 1986) and 8 questions related to the demographic profile. The time to answer the questionnaire was between 10 and 15 min.

Data processing was done using IBM SPSS Statistics and AMOS software. Confirmatory factor analysis (CFA) was used for construct evaluation, and SEM was used to test the proposed structural model. To identify moderated effects between the constructs, mediation tests and the linear regression technique were applied, based on the procedures indicated by Preacher and Hayes (2004) and Zhao et al. (2010). For the study, the bootstrapping test was used, with a 95% confidence interval.

The premises of multiple regression analysis were verified to guarantee greater robustness of the sample. Regarding data normality with the variables measured, the assumption of univariate normality was tested (from data asymmetry and kurtosis) (Malhotra et al., 2012. Hair et al., 2018). The asymmetry values were between −2.18 and 0.11, and kurtosis between −0.11 and 4.95. The results show that the sample presents adequate normality.

The homoscedasticity condition was analyzed based on the Box M test and the Levene test, and the linearity condition was evaluated based on the standardized residue graph (Hair et al., 2018). By checking the scatterplots, the variables of the studied model presented linear relationships. Finally, multicollinearity was analyzed using the tolerance test, and all of them presented acceptable levels, with tolerance between 0.104 and 0.48 and, for the VIF, the values were between 1.66 and 8.22. Recent research has indicated that the VIF value should be lower than 3 (Hair et al., 2020). An alternative approach for the same author is to examine the level of multicollinearity calculating bivariate correlations between the formative indicators. In this way, the result shows that the bivariate correlation was between 0.294 and 0.462 (lower that 0.50), indicating that the levels of multicollinearity in this sample were low (Hair et al., 2020). With this multicollinearity it is not a problem

4. Results and discussion

4.1 Sample characterization

After the preliminary analyses of the data, the final sample resulted in 269 respondents or valid cases (n = 269), 51.30% (138) female and 46.84% (126) male. Regarding age, 36.8% (99) of respondents are between 20 and 30 years old, 26.02% (70) are between 31 and 35 years old, and 37.2% (100) are over 36 years old. Concerning the frequency of wine consumption, 50.19% (135) consume from 1 to 4 times a month, 17.47% (47) from 5 to 8 times a month, and 30.84% (83) consume more than 9 times a month.

Among the respondents, 54.53% (144) drink wine alone, and the others (45.47 or 125) drink in the company of friends or family. As for the position wine occupies among the favorite alcoholic beverages, 48.33% (130) indicated that wine is their favorite alcoholic beverage, while the remaining 51.70% (139) prefer other beverages, and wine is their second or third option. About willingness to pay, 12.6% (58) indicated they are willing to pay up to R$30 for a bottle of wine, while 78.4% are willing to pay more than R$30 for a bottle.

4.2 Validation of the structural model and hypotheses test

Before testing the hypothesized structural model, the validity of the measurement model was assessed. Table 3 shows results for different fit criteria. The indexes lead to the conclusion that the quality of the model is satisfactory, considering the reference values proposed in the literature (Hair et al., 2018). Therefore, it can be stated that the theoretical model is appropriate considering the indicators obtained.

The hypotheses testing helps to support the proposed structural theory (Hair et al., 2018). Hypotheses were examined based on the magnitude and significance of the estimated path coefficients. Table 3 presents the consolidated results of the tests with the direct and indirect effect of the relations between the constructs.

From the hypotheses test, it can be observed that the attitude has a positive and significant impact on the intention to purchase wines (β = 0.553, p = 0.001), supporting H1. In this case, consumer attitude plays a role in people’s decisions to purchase products or services because it generates the influence on the purchase desires and the consumer purchase decision processes. The attitudes are based on the beliefs about a particular object or an action that can be translated into an intention of concretizing such act, being a general evaluation that the consumer makes about something (Schwartz, 1992).

Regarding the impacts of experience on the purchase intention, its effect is positive and significant (β = 0.24, p < 0.001), supporting H2. The consumer experience is an important factor that opens an opportunity to expand and strengthen the consumer relationship and involvement with the product, brand, or organization (Lemon and Verhoef, 2016).

As observed, the symbolic value has a direct and significant impact on the attitude (β = 0.152p = < 0,001) and the consumer experience (β = 0.353p < 0.001), supporting H3 and H4. In that manner, it is possible to confirm that the consumer attitude and experience have an indirect effect in the relationship between the symbolic value and the purchase intention, in this case, of the product wine. (Indirect effect β = 0.165, CI = 0.104 to 0.247, p = 0.004).

Aiming to test the moderate hypotheses H5, H6 and H7 shown in Figure 1, we applied the procedures proposed by Hayes (2013) through PROCESS for SPSS (model 5). The advantage of this process is the use of the bootstrapping technique. The technique is based on the assessment of the paths, and it provides the significant calculus of the effects through the theory test with normal distribution (significant coefficient “p”) and non-normal distribution (CI upper and under), for values of −1 D.P., average and +1 D.P of the moderator M (Prado et al., 2014). For the results, as presented in Figure 2, we used a confidence interval of 95% calculated employing 5.000 resamples in the bootstrapping procedure. As a result, the model variables explain 60% of repurchase intention (Figure 2).

Either indirectly or directly, the symbolic value that a product or service represents to the consumer may interfere with their purchase intentions (Bernritter et al., 2017). However, the presence of some moderators may interfere with this relationship. Therefore, the test of hypotheses H5, H6 and H7 verified, respectively, how much the consumer involvement, the willingness to pay, and the position that wine occupies in their preference for alcoholic beverages interfere with the relationship between the symbolic value of wine and the consumer purchase intention.

Regarding the consumer involvement, the results show (Figure 2) that for consumers with higher involvement (high involvement) with the product (wine), there is a significant effect on the relationship between symbolic value and purchase intention (βhigh = 0.16, p = 0.01). In contrast, this effect is not significant for consumers with low involvement (βlow = 0.009, p = 0.91).

In a more detailed analysis, presented in Figures 3, 4 and 5, we divided the sample from consumers with low purchase intention (135 respondents, first and second quartiles and median of 5 on a scale from 1 to 7) and high purchase intention (134 respondents, third and fourth quartiles and median above 5). In the Figure 3, it is possible to observe that consumers with high involvement with the product have a higher perception of symbolic value (Mlow purchase intention = 3.27 and Mhigh purchase intention = 4.35) than consumers with low involvement (Mlow purchase intention = 2.59 and Mhigh purchase intention = 2.23), having a significant difference between low and high involvement at the level of p < 0.05. Therefore, H5 is confirmed, as consumer involvement with the product (wine) moderates the relationship between the symbolic value and the purchase intention, and this relationship is stronger for consumers with higher involvement with the product. Thus, a higher symbolic value and higher purchase intention are more strongly associated with consumers with higher involvement with the wine.

The H6 test (Figure 2) evidenced that for consumers with a higher and lower willingness to pay for wine, the relationship between the symbolic value and the purchase intention in both situation is not significant (βlow = 0.08, p > 0.05, βhigh = 0.01, p >0.05) H6 is not confirmed since the willingness to pay for the product (wine) do not moderates the relationship between the symbolic value and the purchase intention.

In Figure 4, it is also possible to identify that the willingness to pay does not moderate the relation between symbolic value and purchase intention. Consumers with higher purchase intention, in low willingness to pay as well as in high willingness to pay do not significantly vary in the perception of symbolic value (Mlow willingness to pay = 4.03 and Mhigh high willingness to pay = 3.67, p = 0.424). Even not supporting the H6, it is interesting to observe in both situations related to the willingness to pay (low and high), evidencing that the higher the perception of the symbolic value of the product (wine), the higher the consumer purchase intention.

In the H7 test (Figure 2), consumers who ranked wine first among alcoholic beverage options presented a significant effect of symbolic value on the purchase intention (βfirst = 0.15, p = 0.01). In contrast, this effect is not significant for consumers who do not consider wine their first option among alcoholic beverages (βanother = 0.06, p = 0.29). Therefore, H7 is confirmed since the preference of wine over other alcoholic beverages moderates the relationship between the symbolic value and the product purchase intention, and this relationship is stronger in consumers who consider wine as their preferred alcoholic beverage.

Figure 5 shows that when wine occupies the first position in the preference of alcoholic beverages, the consumers who have a higher intention of purchasing attribute to it a higher symbolic value (Mfirst = 3.96 versus Manother = 3.35 p < 0.05).

4.3 Academic and practical implications

The proposed and tested theoretical model and their respective hypothesized relationships have an important academic implication that contributes with directions to wine producers and marketers to improve their marketing and management strategies associated with the product (wine). Understand the moderators and mediators’ factors gives a holistic view of “how” and “why” the winés consumers make their choice to buy this product. The use of the three different moderators shows that the relation between symbolic value and purchase intention can be moderated for the involvement that the consumer has with the wine and the consumer preference. The result of this study showed that the willingness to pay does not moderate the relation between symbolic value and purchase intention, but the study evidencing that the consumers with higher willingness to pay are more involved, have a favorable attitude and have a higher purchase intention (p < 0.05). In this context, the use the moderate factors can increase the understanding about the wine consumer behavior and help the marketing management of the wine companies.

As managerial implications, the theoretical model shows that symbolic value is an important construct that has an indirect impact on the purchase intention through consumer attitude and product-norm experience and his moderators (involvement and consumer preference). In this way, it is important for the company searches to increase the symbolic value that the wine occupies in the consumer´s mind. Highlights in the social symbolic value as status and power of wine can give to consumer enhance social self-concept (Sweeney and Soutar, 2001). As specific activities or strategies to improve the symbolic value of wine, the company can use important personalities of the artistic world to promote the winés consumer. Besides that, the company can use some luxury events to promote the relation of wine with refinement and sophistication. Another strategy is making the association with luxury products as cars, jewelry or some objects that are related with wine as cork opener are some strategies that the companies can develop to increase his symbolic value. Strategies that use only the price and bottle label to improve the symbolic value of the winés brand is also important but insufficient to increase and strengthen the association of wine as a beverage that gives more status and power to the consumer.

According to Ledgerwood et al. (2007), the process of creating symbolic value represents the meaning and role of representations that companies promote in relation to consumers and how consumers or their reference groups relate to and identify with them. Thus, wine-producing companies can take actions not only in traditional communication channels to stimulate the creation of symbolic value in their consumers, but also of a consistent communication work in social media, with the production of relevant content and directed to the target audience, reinforcing its brand image and market positioning.

Promoting consumer-norm experience with the wine or winery is an important strategy for increasing your symbolic value and purchase intention. For example, promote a visit to the winery showing the manufacturing process, teaching about the wine composition are different ways to evaluate the wine quality. Otherwise, the wineries can promote courses of testing and/or culinary events harmonize with wine. This is some practical activities that can be a good way to increase his mental representation and give them a consumer experience that focuses on the customer´s cognitive, emotional and behavioral in the purchase journey (Lemon and Verhoef, 2016)

Furthermore, it is important the participation of wineries in some events where it is possible to promote tastings, workshops, lectures, merchandising actions that can encourage consumers to become more involved with the wine world. Experienced consumers can evaluate more accurately the technical aspects of wine, influencing their involvement with the product and their attitude. Wine producers and traders can enhance the consumer experience to increase consumers’ knowledge and involvement with this market and create a positive relationship with the brand, consequently, increase the (re)purchase intention, loyalty (Kaufmann et al., 2016), the positive word of mouth (WoM) (Wallace et al., 2014) or even the electronic word of mouth (e-WoM) and satisfaction.

5. Conclusions

Wine, as a millenary beverage, assumes an important economic and social role. In addition to its utilitarian value, it has a relevant symbolic value for its consumers. In this context, this study aimed to understand the role of the symbolic value of wine and its relation to product purchase intention. Among the research implications, it is highlighted how the symbolic value interacts or affects the consumer purchase intention. The symbolic value, in general, did not directly impact the purchase intention (β = 0.009, p > 0.05). However, when mediated by consumer attitude and product-norm experiences, its effects were significant. This occurred because the symbolic value had a significant impact on the attitude and product-norm experience (Table 3) and these two constructs directly impacted the purchase intention. The research results show that the consumer attitude, just like the studies of Voss et al. (2003) and Homer (2006) and the product-norm experience, just like the study of Sharma and Patterson (2000) assume a preponderant role in the wine purchase intention with a coefficient of determination R² = 0.443 demonstrating there is a high effect (Cohen, 1988) and significant influence on the product purchase intention.

Another significant theoretical implication was identifying moderating factors that influence the relationship between the symbolic value and the purchase intention. The study showed that a higher involvement with wine, a higher willingness to pay and the fact that wine ranks first among alcoholic beverage options significantly moderated the relationship between these two constructs (symbolic value and purchase intention). It is noteworthy that people with higher involvement with the product (wine) have a higher symbolic value (Mhigh involvement = 4.35 versus Mlow involvement = 2.23, p < 0.05) and higher purchase intention. As observed by Andrews et al. (1990), the higher is the consumer involvement, the more complex is the decision-making process, while consumers less involved trust in simple heuristic tips, such as price or brand (Ferreira and Coelho, 2015).

Also, a relevant finding refers to the relationship between the symbolic value and willingness to pay. As shown in Figure 4, the higher is the symbolic value, the higher is the probability of the consumer buying the product, but not necessarily the higher is the willingness to pay. Regarding the preference that wine occupies among alcoholic beverages when wine presents itself as the first option, the relationship between the symbolic value and purchase intention is significant. Consequently, the symbolic value is more relevant for people who have wine as their first option (First option = 3.55 versus other options = 3.10, p = 0.038). Therefore, symbolic value, in general, does not impact consumer purchase intention. However, when analyzed in a segmented way, there are differences in their relations with other constructs, showing themselves to be more significant for consumers with higher involvement, higher willingness to pay and for those who have wine as their preferred option among alcoholic beverages.

5.1 Limitations and future research

Some limitations and future research should be noted. Firstly, the convenience sample, with wine consumers in the south of Brazil. The Brazilian market is very diverse, with different cultures, temperatures, distances and because of this, future research in the Brazilian market can consider a sample in the different states of Brazil and check how the proposed theoretical model works. Moreover, future research can check the proposed theoretical framework with different segments and moderators of the wine consumers, for example with expert vs not expert consumers, with low vs high consumers.

Another limitation of this study is in relation to the period of data collection. The data was collected in February and March 2020, summer in the South of Brazil. Although the wine is available throughout the year, it is known that the consumption of wine presents some seasonality, with a higher consumption in the winter. Therefore, future research can investigate the relations proposed in the framework in different seasons of the year.

This study investigated the wine in general, not categorizing the kind of wine (for example, varietal, with European grapes, or with American grapes). New research can investigate if the proposed framework is valid for different kinds of wine, if the symbolic value changes with the kind of wine, or if the kind of wine moderates the relation between symbolic value and buying intention.

The other limitation is the test hypothesis for moderate analyses. In this paper the procedures proposed by Hayes (2013) were used through PROCESS for SPSS (model 5). New research aiming to test the moderation of the proposed model could also consider using PLS-SEM as a research method.

Other studies can include extrinsic variable as the country origin. It is known that the country image affects the quality, symbolic value and price perception and consequently the consumer’s attitude (Bassani et al., 2018; Brijs et al., 2011; Zeugner-Roth and Diamantopoulos, 2010). In such a way, the effects of country of origin and/or designation of origin can give an important contribution to a better understanding the wine consumer behavior.

Thus, this study constitutes a significant contribution to understand the effects of symbolic value on the wine consumer purchase intention. It is also evident that wine symbolism concerning consumer attitudes and product-norm experiences, willingness to pay, consumer involvement with the product and position in their preferences of alcoholic beverages have different and complementary impacts on their future purchase intentions. Furthermore, the relationships tested can contribute to the better understanding of wine consumers’ purchase journey and consumption.

Figures

Figure 1.

Proposed Theoretical Model

Figure 2.

Direct and indirect effects and their moderators

Figure 3.

Symbolic value, involvement, and purchase intention

Figure 4.

Symbolic value, willingness to pay, and purchase intention

Figure 5.

Symbolic value, purchase intention, and preference of wine as an alcoholic beverage

Table 1.

Constructs measurement variables

Constructs and Measurement Variables	Factor loadings
Purchase Intention (PI) (α = 0.73; AVE = 0.50 ) - Mittal et al. (1998), Wu et al. (2014)
2. I will buy larger quantities of wines in the coming months	0.632
3. I will consider wine as my first purchase option concerning other categories of beverages	0.610
4. I intend to increase the volume of wine consumption	0.755
5. I will encourage friends, neighbors and/or relatives to buy and consume wine	0.704
Symbolic Value (SV) (α = 0.96; AVE = 0.79 ) – De Toni and Mazzon (2014), De Toni (2019) Graciola et al.. (2018)
18. Wine makes me feel powerful	0.787
19. Wine gives me a prominent position in society	0.935
20. Wine contributes to increasing my status	0.919
21.Wine will favorably improve other people’s perception of my own self	0.846
Consumer Attitude (CA) (α = 0.89; AVE = 0.60) – Voss et al. (2003), Homer (2006)
6. For me consuming wine is fun	0.599
7. For me consuming wine is refreshing	0.317
8. For me consuming wine brings satisfaction	0.681
9. For me, wine tastes good	0.764
10. For me, wine is pleasant	0.847
11. For me, wine is relaxing	0.779
12. For me, wine is pleasurable	0.827
13. For me, wine is exciting	0.602
Product-Norm Experience (PE) (α = 0.96; AVE = 0.87 ) - Sharma and Patterson (2000)
14. I can understand almost all the aspects of purchasing wine	0.885
15. I possess good knowledge of wine	0.890
16. I am quite experienced in purchasing wine	0.865
17. I can understand very well the wine technical aspects	0.881

Table 2.

Discriminant validity

Constructs	Purchase intention	Consumer attitude	Symbolic value	Product-norm experience
Purchase Intention	0.500
Consumer Attitude	0.376	0.600
Symbolic Value	0.145	0.084	0.790
Product-Norm Experience	0.465	0.237	0.086	0.872

Table 3.

Hypothesis test

Hi		Effect	β - Path Coefficient	95% Confidence interval of the direct effect	p-value	Results
H1	CA → PI	Direct	0.533	[0.309|0.806]	0.000	Supported
H2	PE → PI	Direct	0.240	[0.167|0.344]	0.000	Supported
H3	SV → AT	Direct	0.152	[0.092|0.216]	0.000	Supported
	SV → CA → PI	Indirect	0.165	[0.104|0.247]	0.004
H4	SV → PE	Direct	0.353	[0.184|0.502]	0.000	Supported
	SV → PE → PI	Indirect	0.165	[0.104|0.247]	0.004

Notes:

model fit: CMIN/DF = 2,240; GFI = 0.900, NFI = 0.932, RFI = 0.919, TLI = 0.954, IFI = 0,961 CFI = 0.961; RMSEA = 0.068

Obs. CA = Consumer Attitude, PE = Product-Norm Experience, SV = Symbolic Value, PI = Purchase Intention