THE CONNECTION BETWEEN SELF-IMAGE CONGRUENCE AND BRAND PREFERENCE FOR STORE BRANDS : A STUDY IN PORTUGAL

When examining consumer behavior, self-image, brand image and their congruity are some of the drivers. Past research assumed and then successfully proved these concepts to impact on product/brand choice, purchase intentions, brand preference and so forth. The purpose of the research is to study the self-concept and its relation to brand preference. A comparison will be made between store brands PingoDoce and Continente. Data were collected through survey method. The perceptions of 34 respondents about their own traits, the traits of a typical consumer of the store brandand brand preference for them were obtained. Examinations were made regarding the relationships between variables mentioned above. The findings suggest there are no significant differences between how store brands Continente and PingoDoce are perceived by consumers according the attribute ratings. Brand preference was very similar for both brands, which did not allow the test of whether the individual will prefer a brand which is perceived by him to be more rather than less similar to his own self-concept.


Introduction
The modern individual is more and more preoccupied with his self-image, image that he perfects as he can.The current paper is an attempt to identify the relationship between the self-image congruence and brand preference for store brands, by defining the concepts and empirically testing the relationship.The study begins with the literature review, including the notion of "brand preference", "store brands" and the "self" concept.After examining the theoretical background, the paper proposes several hypotheses, which are later tested.The reasons of using this specific methodology to test the hypothesis are discussed.After, an analysis follows, examining the hypothesis made according to literature review.The paper ends with concluding implications, limitations of the study and future possible research on the subject.
This subject is of interest in the present environment because an understanding of selfconcept and self-image can give clearer directions in developing better marketing programs.
These notions are also crucial in determining the relation between the person's image and the final purchase behavior.Studies done in the past have argued that self-congruence is an important influencing factor in consumer preference formation.The self-congruity theory (SIRGY, 1982) suggests that consumers compare their image of themselves with the image of the product.Thus, they are expected to prefer a product that has a similar image to theirs.
People prefer a certain brand because they see themselves as similar to the type of individuals that they generally thought to use this product.Self-image congruence is supported by several studies (GRUBB;HUPP, 1968;DOLICH, 1969;LANDON, 1974;HONG;ZINKHAN, 1995).
The main goal of this research is to test one assumption that seems obvious when we think of consumer behavior (even though there are contradictions in previous studies): people buy certain products if they are similar, coincide, or increase the image they have about themselves.

Conceptual background: brand preference
If you have a good story and enough money to make it heard, you could build awareness on any matter rather quickly.It is a slower process, however, to build brand preference.Aristotle said we are what we repeatedly do.Excellence is then not an act, but a habit (apud Durant, 1926).Transposed to brands, it is important not only to attain, but to sustain preference, loyalty and attitude.This will increase the company's market share, make more revenues and get the company one step ahead its competitors.

24
For the current paper, the definition of brand preference is accepted as offered by the Online Business Dictionary: measure of brand loyalty in which a consumer will choose a particular brand in presence of competing brands, but will accept substitutes if that brand is not available.It is a biasness toward a certain company's brand, in which the consumer favors it over another.Brand preference is manifested differently depending on the salient beliefs that are present at a certain time; the consumer biasness towards it; the extent to which a consumer favors one brand over another (EBRAHIM, 2011).
Through this concept we understand the consumer's predisposition to evaluate in a specific way a product compared to other products.It can be said the consumer formulated an attitude towards each of the available brands, after which he chose the one that fulfilled his conditions.In this process, the consumer expressed a multitude of preferences.These factors can be of cognitive nature (all the beliefs based on knowledge), affective nature (positive feelings and emotions, indifference or negative feelings which a product generates) and conative nature (making a conclusion by evaluating the goods and manifesting the desire to act).Individuals are ready to pay for a specific brand, even though its physical features are not much different from other brands.Research shows that most of consumers typically buy a single brand of beer, cola, or margarine (DEKIMPE et al. apud BRONNENBERG;DUBE;GENTZKOW, 2012), even though the price may change significantly in the course of time, and consumers often cannot tell their preferred brand apart from others in blind "taste tests" (e.g.Pepsi Challenge, blind tasting between Pepsi and Coca-Cola).It happens because consumers want to ease their purchasing process.They develop brand preferences so they wouldn't have to analyze all available goods every time.
The associations consumers have with brands make them distinctive.It goes beyond perceived quality that the brand promises.It relates to the intangible properties.In the minds of consumers, Coca-Cola is "All American" and Mercedes is "'prestigious" (AAKER, 1997).
Brand is a distinguishing feature of a product and is often important to customers purchasing the product.
Usually, brand preference is build when the company has been for a long time on the market.Research shows that individuals could prefer a good sold by a company, just because that company has a name on the market during its long presence (DINLERSOZ; PEREIRA, 2007).Consumers can exhibit little brand preference toward a new and unfamiliar brand when exposed to it.

Self-concept
The root of the self-image congruence comes from social psychology, where interpersonal relations are analyzed.From that area, a lot of studies show that people perceive others they like to be more similar to themselves than the people they don't like.The opposite relation was also proven: people like more the people that are similar to them and like less those who are different.If we transpose these conclusions in the context of consumer behavior, we could predict that people would prefer brands which they think people similar to them use (NEWCOMB apud ROSS, 1971).
According to Rosenberg (apud SIRGY, 1982), self-concept is the totality of the individual's thoughts and feelings having reference to himself as an object.The self-concept is the totality of all the beliefs an individual holds about himself.It is the way he defines himself.The self-concept might serve as an anchor for comparison and evaluation.
Individuals are searching to maintain and enhance their self-concept (GRAEFF, 1996), and want to be in a position where their actual self-image is so good, that it matches their relevant ideal self (HIGGINS, 1987).In the present environment, brands act as symbols, means through which consumers can express their identity and interact easier with others of similar interests.(SIRGY, 1982).Research in the past shows that consumers of a specific brand have a similar self-concept with others that use the same brand, and a significantly different self-concept than those that use a different brand (GRUBB;HUPP, 1968).This implies that consumers seek brands that would match their self-image (SIRGY 1982).
The self-image of the individual is formed based on the reactions he gets from the surrounding people: friends, family, colleagues etc.It is natural for the person to want positive reactions from these groups of people.But the interaction between them does not happen in an isolated environment.Rather, they are affected by the environmental setting and personal opinions of each of the persons involved.A person tries to transmit information about himself through the use of products that can be used as symbols (GRUBB; HUPP 1968).
In the literature, that are more perspectives over the notion of "self" concept.As described by Sirgy (1982), it has been seen through different lenses, such as psychoanalytic theory (a system with a conflict between what we are and what we want to be), behavioral theory (the totality of the responses to stimulus), cognitive theory (a conceptual system that processes information about the self), and as symbolic interactionism (a system formed from the totality of interpersonal interactions).The current paper treats the "self" concept as a construct with different types of selves (SIRGY, 1982).Some presume that "self" is conditioned by consistency (because the individual tends to behave consistently with his view of himself) and by esteem (because the individual tends to improve his image).

Self-image congruence
For consumers, brands have symbolic attributes, which construct the brand-user image, meaning how would a typical user of this brand look like and behave.To decide if they like the brand, individuals try to match their own perceived image with this that of the typical user.This is called "self-image congruence".If these two have a lot of things in common, the individual will have a positive attitude towards the brand.Everybody has a particular identity, preference and habit that they are aware of.And once their "self" is set, individuals tend to protect them (KRESSMANN, et al., 2006).
In literature, Levy (1959)  What he found is that we buy not only for the utility of the product, but also for what it means, what it symbolizes for us.He presumed that consumers preferred goods with a perceived image that matched theirs.As this topic became more and more popular in the academic environment, more research shed light upon the matter.Birdwell (1968) was the first one to actually try and prove these arguments (up until this point academics only hypothesized).Having chosen to analyze car brands, he selected a sample of 100 car owners and divided them into four groups, corresponding to four car brands.Respondents were given a questionnaire containing 22 bipolar scales (e.g.sophisticated-unsophisticated, exciting-dull etc.).He found there are significant differences in each ownership group's perception of cars and that there is a high degree of congruity of how respondents perceive their cars and themselves.
Other researchers (GRUBB; HUPP, 1968) tried to replicate the study.They chose Volkswagen car owners and Pontiac car owners (they assumed the projected images for these two brands are significantly different) and asked them to rate themselves and the two different brands according to 16 bipolar scales.The findings were similar to that of Birdwell (1968): consumers of the two different brands of autos perceived themselves significantly different one from another and they had specific stereotype perceptions of the owners of each brand.Also, they perceived themselves to be like others who owned the same make car and quite different from owners of competing brands.
However, there is one limitation in the studies mentioned above, pointed out by Evans (apud SIRGY, 1982).Both studies analyzed consumers that already owned cars included in the study.It is a rational assumption that product ownership may have influenced on the product image or the self-image, resulting in a higher relation.Dolich (1969) tested the relationship on a sample of 200 students and found that preferred brands of products were perceived to be more similar to self-concepts than least preferred product brands.Graeff's research (1996) has very similar findings: the greater the degree of congruence between brand image and self-image, the more favorable were subjects' brand evaluations.
There is also research that diminishes or denies the importance of self-image congruence altogether.Hughes and Guerrero (1971) argue that you cannot put all types of consumers into one basket.They make the point that brand preference of some individuals (especially innovators) can be explained better by incongruity rather than congruity, as they want to stand out of the crowd.They suggest examination not only of self-congruity, but social congruity, social incongruity, and self-incongruity as well.In his research, Green (apud SIRGY, 1982) fails to confirm the relationship between self-congruity and consumer choice at all.However, self-image congruence is generally supported in the academic environment Further, the paper examines recent studies on self-image congruence and its impact other different pre-consumption and post-consumption variables similar to brand preference: brand personality, purchase intention, intention to recommend.Kressmann et al. (2006) researched brand loyalty by asking 600 respondents to complete a questionnaire with respect to their car, by choosing from the 15 brand personality facets (honest, down to earth etc.) that were appropriate for them.He found a direct positive effect of self-congruence on brand loyalty.
Li, Wang and Yang (2011) studied the effects of self-image congruence on purchase intention.Their sample was made of 477 respondents from four major cities in China.They found that the more congruent an individual's self-image is with corporate-brand image, the higher the purchase intention tends to be.Usakli and Baloglu (2011) analyzed tourist destinations instead of regular consumer products and used another research instrument.While most of the research in the 80's employed semantic differential, this one uses a 5-point Likert scale with 29 items.They found that self-congruity has a positive impact on tourists' behavioral intentions, intention to return and to recommend.Hosany and Martin (2012) also analyzed a service, not a consumption good, to see if the assumptions still hold.Their sample was mode of the passengers of a cruise.The research employed a 7-point scale to rate cruisers' self-concept, experiences, satisfaction, and intention to recommend.Their findings confirm the general accepted ideas: higher congruity between customers' self-concept and perceived images of other tourists taking the same cruise contribute to overall favorable experiences.
Choi and Rifon (2012) focused their attention on goods that were endorsed by celebrities.They analyzed how the consumers' self-image related to that celebrity can influence the individual's attitude.Respondents had to watch an ad where a celebrity endorsed a product.Then they completed a questionnaire about the ad.Findings suggest that congruence between consumer's self-image and celebrity image plays an important role in endorsement process.More than that, the positive attitude towards the ad was higher when the celebrity's image was more congruent to the product image.
Peng, Wong and Wan (2012) made an attempt to compare genuine and counterfeit products.They measured brand attitude, self-image and product's image to test for possible differences.Results show that the greater the congruence level between brand image and selfimage, the more favorable were subjects' product evaluations.Findings refer to both genuine and counterfeit products.Wu (2011) studied the effect of self-image congruence as a moderating variable between brand evaluation and brand crisis (e.g.Nike's brand crisis).Wu found that self-image congruence is a moderating variable in brand crisis.Consumers with high self-image congruence tend to be less influenced by the negative information available, and still hold preference to that brand.
As can be seen from literature review, there is a significant amount of evidence for self-image congruence's impact on brand preference, brand attitude, brand evaluation and brand personality not only on consumer goods, but on services as well.

Store brands
A store brand is defined as a product owned, controlled, and sold exclusively by a retail chain store.That brand can be the retailer's own name or a name created exclusively by 29 that retailer.In some cases, a retailer may belong to a wholesale group that owns the brands that are available to only the members of the group.
Store brands existed so far as 100 years ago, with initial products such as tea.Over the last decades, there is an increase in the market share of store brands, drawing more attention from researchers and practitioners (KOTANI; SUMITA, 2013).Now, the influence of store brands is increasing in many cathegories."There is a tremendous opening for store brands to exceed name brands.The consumer understands how much money goes into this marketing and they want alternatives.They don't believe in the ethic of it" (OLDENBURG, 2005).
In the last decade, the rapid evolution of store brands has led to converging competencies, and now store and national brands are at war for the same consumer dollars (BIRWADKER, 2011).For the consumer, a store brand is the choice and opportunity to purchase quality products at savings compared to manufacturer brands on a permanent basis.
Store brand products have the same or even better ingredients than manufacturer brands.The consumer is confident in the quality standards and specifications because the retailer's name or symbol is on the package.
According to the Private Label Manufacturers Association, the latest market share data shows that the popularity of private label continues to spread across Europe.PLMA recently published its 2014 International Private Label Yearbook.Number show retailer brands now account for at least 30% of all products sold in 15 countries, the greatest number ever.Private label continues to account for more than half of all products sold in Switzerland (53%) and Spain (51%).Both countries showed market share gains in volume and value.In five countries -United Kingdom (45%), Portugal (45%), Germany (44%), Belgium (41%) and Austria (40%)private label now accounts for at least four of every ten products sold.
Market share for retailer brands also posted gains in Denmark, Norway, Hungary, Turkey and Italy (PRIVATE LABEL MANUFACTURERS ASSOCIATION, 2014).More details are shown in Figure 1.With a market position of highly competitive prices, Continente combines variety, customer care and services with a promotional approach which enables it to give excellent value for money, an approach which is much appreciated and well known among Portuguese consumers.It has a network of stores across Portugal offering an extensive and varied range of products and services with average sales areas of 9,000 m 2 .The solid and steady path followed over the last 25 years, with frequent innovative and high impact promotional and social initiatives, has earned the confidence and empathy of the Portuguese people (Sonae Website, Continente brand web page).
PingoDoce is also one of the largest supermarket operators in Portugal with 376 stores.

Methodology
According to literature review of the self-image congruence, brand preference and store brands, the following hypothesis are formulated: H1: The individual will prefer a brand, which is perceived by him to be more rather than less similar to his own self-concept.
H2: There are significant differences between how store brands Continente and PingoDoce are perceived.
There are two common methods of measuring the self-concept: use of semantic differential scales (and then compute the difference between attributes chosen for respondent and for brand) (MALHOTRA, 1981), or measure it directly (SIRGY et al., 1997).The current paper will use semantic differential scale of Malhotra (1981) instrument, as other researchers did as well (PENG; WONG; WAN, 2012;BEERLI;MENESES;GIL, 2007;KASTENHOLZ, 2004).
The attributes selected to be included in the questionnaire were: simple-complicated, follower-leader, dominant-submissive, popular-unpopular, extravagant-economic, unsuccessful-successful, weak-strong, conformist-nonconformist.As can be seen in appendix 1, the first part assesses a typical buyer of Continente brand, the second onea typical buyer of PingoDoce brand, and the third oneself-assessment according to the same attributes.By comparing the profiles, it could be identified the match/gap between the person's self-concept and his/her perception of the personality of the brand.
The study was exploratory in nature and the survey method was used for data collection.The sample of the current study consists of respondents, whose activity is related to Universidade Fernando Pessoa (most of them are students, and some are teachers).
Questionnaire is distributed personally on a paper format.This choice insured the respondent understood the questions and there were no missing values.A sample of 34 respondents was collected.
Most of the respondents are students.Because of time, distance and money considerations, students are often used as quick and convenient sources of information.
Although use of student samples and "young consumers" is limited in representing the broader population, it has been heavily debated that student samples are appropriate for theory  1972;SHUPTRINE, 1975;KHERA;BENSON, 1970).

Discussion and results
The software SPSS was used to analyze the data.
Most of the respondents were students aged 18-25, as can be seen in Figure 2. As the sampling method was random, out of the 34 people 41% were female and 59% male.

Female 41%
In order to test hypothesis 2, a comparison was made between the ratings of typical consumers of Continente and PingoDoce over the range of the instrument's 8 pairs of attributes.The results can be seen in the table below.The average discrepancy between the respondents' ratings (calculated as the average absolute value of the difference between ratings of one store brand and the second one) is 0.35.The maximum difference between the 2 store brands was 0.56 over the pair of attribute simple-complicated, and the minimum difference was 0.03 over the pair of attributes weakstrong.
Table 2 shows brand preference for the analyzed store brands.There are not significant differences between them either.The results suggest there are no significant differences between how store brands PingoDoce and Continente are perceived.Thus, hypothesis 2 is rejected.
Self-image congruence was calculated over each pair of attributes as the absolute difference between how respondents perceived themselves and how they perceived a typical buyers of the store brand.So, if the value of the difference is high, congruency is low.The results are shown in figure 4. Highest congruence between store brands and personal image are over the pair of attributes 4 (popular-unpopular) and 6 (successful-unsuccessful). Lowest congruence occurred over the pair of attributes 1 (simple-complicated) and 5 (extravaganteconomic).
Figure 4 -Congruence of attribute pairs between store brands evaluation and self-evaluation Source: made by authors.
Table 3 shows the average self-image congruence for the two store brands analyzed.It is calculated as the average discrepancy (difference) between personal ratings and store brand ratings (an average of Table 3).As can be seen, the results are not significantly different.Moreover, the correlation between self-image congruency of respondents with Continente store brand and self-image congruency of respondents with PingoDoce store brand is 0.604 (significant at 0.01 level, 2-tailed).It means that not only the brand preference of these two store brands is similar, but the self-image congruence for them is also not significantly different.Thus, there is no evidence for hypothesis 1.

Conclusions
The current paper examined whether store brands PingoDoce and Continente have different images in the eyes of the consumers.There were found no significant differences between them over all of the eight pairs of attributes.Also, the question whether respondents will prefer brands to which they are more congruent was examined.Based on the current sample, brand preference of both brands was overall similar, and the self-image congruence correlation between these two store brands was 0,604.It means the respondents equally preferred Continente and PingoDoce store brands and they perceived their images to also be very similar.So, no support was found for this hypothesis.
As any other study, the present research paper has some limitations that readers have to take into account when they consider the findings:  Due to budget and time constraints, the sample size consisted of 34 respondents.
Further research could be conducted on a larger sample, which will increase the study's robustness.
 The use of students may have increased the sample's homogeneity of perceptions.
This might not happen if a representative sample would be used.
 Brand knowledge could be a potential moderator in the relationship between brand preference and self-image congruence.Future research could take that into consideration as well.
 This study measured self-congruity through a semantic differential scale (Malhotra, 1981).However, there has been a considerable debate about whether to use the direct score formula (Sirgy's instrument) or gap score formula (Malhotra's instrument) in measuring self-congruity.Future studies could measure selfcongruity using both the direct score and gap score formulas in order to compare the results.
 Future studies could employ more pairs of attributes, to increase the study's accuracy.The intention in using 8 pairs of attributes was to avoid respondent's boredom.
 Future research should attempt to make a comparison between store brands and national brands.
 Finally, we recognize that there are other unexplored factors influencing brand preference, such as personality characteristics, social contexts, and peer pressure.
Further research could offer a more complex understanding by including the examination of these variables as well.

3
SELF-IMAGE CONGRUENCE AND BRAND PREFERENCE FOR STORE BRANDS R. Bras.Planej.Desenv., Curitiba, v. 4, n. 1, p. 22-39, jan./jun.2015.25 Relationship between self-image and product image was among the first to mention the relation between selfimage and product image.He focused his research upon what image different products had.

Figure 1 -
Figure 1 -Private label share by country (volume) It belongs to the Portuguese group Jeronimo Martins and the Dutch group Ahold (PingoDoce website).Its value proposition is based on the following (Jeronimo Martins, PingoDoce brand web page): (1) Assortment of high quality perishables; (2) Competitive prices, with the opportunity for immediate savings; (3) Pleasant store environment, recreating the atmosphere of traditional markets; (4) Great service; (5) A long-lasting relationship of trust with its customers.

Figure 2 -
Figure 2 -Age of respondents

Figure
Figure 3 -Gender of respondents

R B P D THE CONNECTION BETWEEN SELF-IMAGE CONGRUENCE AND BRAND PREFERENCE FOR STORE BRANDS
Any sample is relevant if it permits operationalization within the domain of the theory.For an MA candidate, the university supplies entire classrooms of potential respondents that are available at little or no cost.They generally follow instructions rapidly and accurately.Academics agree that usefulness of student subjects depends, in part, upon the context of the research-its problem, objectives, and hypotheses (ENIS; COX; STAFFORD, R. Bras.Planej.Desenv., Curitiba, v. 4, n. 1, p. 22-39, jan./jun.2015.32 testing.

Table 1 -
Average ratings of typical buyers of Continente and PingoDoce Brands

Table 2 -
Average brand preference of Continente and PingoDoce brands Source: made by authors.

Table 3 -
Average self-congruence of store brands Source: made by authors.