Creating Brand Associations
Creating Brand Associations
Creating Brand Associations
3; 2016
Received: April 6, 2016 Accepted: April 29, 2016 Online Published: May 10, 2016
doi:10.5430/ijba.v7n3p140 URL: http://dx.doi.org/10.5430/ijba.v7n3p140
Abstract
Being able to determine the equity of a service or goods brand is of paramount importance to marketing practitioners
and scholars alike. One way to determine the equity of a brand is through the measurement of brand associations.
Few studies have constructed brand association measures in the context of either goods or service brands and most
have been constructed to measure traditional goods brands. As Vargo and Lusch (2004) have indicated, the
characteristics of a good differ greatly from those of a service. Therefore, the purpose of this study was to construct a
separate brand association measure for the services and goods contexts. This study employed a mixed methods
approach to generating and empirically testing brand association measures in both contexts. Thought-listing
procedures with a sample of consumers (n=72) were utilized to generate survey items for the service-based brand
associations measure. To test the psychometric properties of the brand association measure, a sample of service
consumers (n=459) was utilized. Factor analysis procedures via MPlus 3.1 were utilized to examine the
dimensionality of the service-based as well as the goods-based brand association measures. The result of this study
was the construction of brand association measures that can be utilized in the goods or services contexts. It represents
one of the first attempts to construct and compare brand association measures in both the goods and services contexts.
Goods and services marketers can utilize these measures to determine how their consumers view their brand and they
can track the success of their positioning efforts.
Keywords: brand associations, brand equity, service brand, goods brand, free thought association
1. Introduction
The importance of branding has been evident for traditional goods and service firms since the late 1980’s (Srinivasan,
Park, & Chang, 2005). Research has consistently shown that there are numerous benefits of strong brands for both
consumers and firms alike. For consumers, strong brands reduce the perceived risk and search costs, enhance the
likelihood of future consumption, and can strengthen social identity by linking brand identity to their self-concept
(Cobb-Walgren, Ruble, & Donthu, 1995; Kayaman & Arasli, 2007; Keller, 1993; Underwood, Baer, & Bond, 2001).
With a strong brand organizations can charge a premium price for their products, have the ability to gain market
share, can maintain customer loyalty, offer successful brand extension opportunities and can influence consumers to
spread positive word of mouth (Brady et al., 2008; Cobb-Walgren et al., 1995; Sangster, Wolton, & McKenney,
2001).
Due to the importance of effective brand management for an organization, scholars have hypothesized and examined
different sources of brand equity in order to determine how brand equity is established. Among these sources, brand
associations have been a significant research area for numerous marketing scholars. Although brand association
measures have been constructed utilizing multiple product categories as the basis for their construction, (see Aaker
1997; Bauer, Sauer, & Exler, 2008; Checchinato, Disegna, & Vescovi, 2013; Faircloth, Capella, & Alford, 2001;
Low & Lamb, 2000; Romaniuk & Gaillard, 2007; Uggla, 2004), scale items for measuring brand associations have
not been validated separately in the goods and services contexts in the marketing literature. The current study
addresses that limitation in the marketing literature.
A second limitation of previous research involves the contemporary conceptualization of brand associations which
consists of brand attitudes and the overall brand image that are consumers’ evaluative responses to brand
associations themselves and should not be included in a measure of brand associations. Brand associations are
considered any thoughts linked to the brand in the mind of the consumer (Aaker, 1991). Therefore, the purpose of
this study was to construct separate brand association measures for good and service product categories. The specific
direction of this research was guided by the strengths and weaknesses of previous work. The following section
examines conceptual background of brand associations and the strengths and weaknesses of previous measures.
2. Literature Review
2.1 Defining Brand Association
Keller (1993) labeled brand image as “a set of associations linked to the brand that consumers hold in memory” (p.3).
A brand image is composed of a set of associations regarding the brand in the mind of the consumer. Brand
associations are the informational nodes that contain the meaning of the brand in the mind of the consumer (Keller,
1998). Brand associations are considered to be any thought linked to a brand in the mind of the consumer (Aaker,
1991; 1996). Marketers use brand associations for product positioning purposes while consumers use brand
associations to help guide their decision-making (Low & Lamb, 2000). Scholars have held differing opinions on
what constitutes a brand association as well as how these associations should be categorized.
Brand associations were classified into three categories by Keller (1993): attributes, benefits, and attitudes. Attributes,
descriptive features that characterize a product or service, are divided into product and non-product related categories.
Product-related attributes are the ingredients necessary for the performance of the good or service sought by the
consumer. Non-product attributes are the external aspects of a product such as price, packaging, and usage imagery.
Benefits are the personal value consumers attach to the product or service attributes and are categorized into three
types of benefits: functional, experiential, and symbolic. Functional benefits are intrinsic advantages of the product
related to product attributes. These benefits are often associated with basic motivations, such as safety and
physiological needs (Maslow, 1970). Experiential benefits are the feelings associated with the use of the product.
These benefits relate to pleasure seeking and are closely associated with hedonic values. Symbolic benefits refer to
extrinsic benefits such as social status or prestige tied to the good or service. These benefits correlate with
non-product benefits and include social status, prestige, and an enhancement of self-esteem and self-concept.
Keller defines attitudes as the consumers’ overall evaluations of the brand and these often form the basis for
consumption behavior. Attitudes are often characterized by their favorability, strength, and uniqueness. Favorability
refers to how the product satisfies the consumers’ needs or wants. The level of strength is determined by the extent to
which brand image enters and is maintained in the mind of the consumer. Uniqueness simply is how much that
information recall relates to the particular brand in question.
The theoretical foundation of brand associations as set forth by Keller (1993) aided marketing researchers in
determining the importance of brand associations from a consumer behavior standpoint. The following section
details research that has examined the importance of brand associations from a marketing perspective.
2.2 Why Study Brand Associations?
The importance of brand associations is evident when examining the literature on brand equity. Brand equity is the
value added to a product by a particular mark, name, or symbol (Aaker, 1991). It has been posited that strong brand
equity is related to outcomes such as an increase in customer loyalty, willingness to accept brand extensions, and
willingness to pay a premium price for the target brand (Keller, 1993). Keller (1993) proposed that brand knowledge
(brand equity) was comprised of two dimensions: brand awareness and brand associations. Due to the fact that
building brand equity can have such positive results for an organization, it stands to reason that creating favorable
associations on behalf of consumers is a primary objective of brand managers.
The impact of brand associations on building brand equity has been strongly suggested by researchers. Keller (1993)
believed that brand associations represent the most important component of brand equity. Berry (2000)
conceptualized service brands in his model of brand equity and concluded that the meaning of the brand in the mind
of the consumer (overall brand image) had a direct impact on the level of brand equity for a service brand. He
posited that the brand meaning, a compilation of brand associations that lead to the overall meaning of the brand for
the consumer, disproportionately affects brand equity in comparison to brand awareness. The perception of the brand
is more critical to brand equity then the mere presence of the brand in the mind of the consumer. Strong brand
associations combine to form the overall image or meaning of the brand for the consumer (Yoo, Donthu, & Lee,
2000). Further, the personality of a brand can have a significant impact on the consumer’s final decision to purchase
the brand (Kim, Lee, & Lee, 2008).
Del Rio, Vazquez, and Iglesias (2001) found that positive brand image was correlated with consumer’s willingness to
pay a price premium, recommend the brand to friends, and the willingness to accept brand extensions. These
consequences have been supported in the literature as outcomes of brand equity and have served as proxy measures of
brand equity. By utilizing the benefits aspect of brand associations, the authors constructed four types of brand
functions from the perception of the consumer: guarantee, personal identification, social identification, and status. It
was believed that these brand functions were representative of the utility and benefits that brands offered to consumers.
In addition to the aforementioned benefits of crafting a positive brand image, a positive image positively influences
brand loyalty (Moon, Park, & Choi, 2010).
As marketers and researchers alike of come to understand the potential benefits of brand associations, scholars have
engaged in scale construction to develop brand association measures for multiple product contexts. The following
section details these attempts at measurement as well as the limitations of each.
2.3 Brand Association Measurement
Although Low and Lamb (2000) found support for their multi-dimensional conceptualizations of brand associations,
their results are in direct conflict with the conceptualization of brand associations put forth by Keller. Brand
associations are merely thoughts linked to the brand in the mind of the consumer (Keller, 1993) and as a result, the
associations do not carry an evaluative or emotional component to them. Therefore, the inclusion of brand attitude
and brand image as conceptualized in the proposed studies stands in stark contrast to the manner in which brand
associations have been defined. Although Keller conceptualizes brand attitudes as a type of brand association, it will
be argued here that brand attitudes constitute a consumer’s cognitive response to the associations they hold for a
given brand.
A consumer’s emotional and rational responses to a brand as well as the overall judgment are not specific aspects of
brand associations. Rather, brand attitude and overall brand image are responses to specific brand associations in the
mind of the consumer. Therefore, brand attitude and overall brand image should be positioned as responses that
occur following the formation of brand associations. A more valid conceptualization of brand associations was put
forth by Faircloth, et al. (2001). They examined the impact of brand associations on consumers’ attitudes and overall
image. They found that specific brand associations can have a positive impact on the overall brand image that, in turn,
influence brand attitudes. This finding contradicts Keller’s (1993) assertion that brand attitudes represent a specific
type of brand association. Instead, this study shows that brand associations are separate from brand attitudes and that
these associations aid in the formation of attitudes toward the brand by consumers.
The preceding literature illustrates the conceptual confusion regarding the dimension of brand attitude. Due to the
evaluative nature of attitudes, this dimension should not be included as a specific type of association. More likely,
brand attitudes are closely related to an attitudinal outcome of brand associations or a moderator between brand
associations and consumer outcomes such as attitudinal loyalty or behavioral intentions. In fact, Bauer et al. (2008)
found that brand attitudes shared the strongest relationship with brand loyalty and brand attitudes were closely
related to the attitudinal aspect of loyalty. Therefore, brand attitudes should be viewed as a response that is formed
due to the brand associations that exist in the mind of the consumer. Therefore, in the context of this study, brand
associations will be defined as the mere thoughts that consumers hold for a given brand and can be characterized by
two dimensions: attributes and benefits.
With this study, the researchers attempt to construct brand association measures that can be utilized in multiple
goods and services contexts. This will be achieved by directly eliciting brand association from consumers in multiple
goods and services contexts. Few previous studies have utilized a free thought-listing procedure to generate the items
for a brand association measure (see Nelson, McEvoy, & Dennis, 2000; Supphellen, 2000; Till, Baack, & Waterman,
2011). Keller (1993) and Supphellen (2000) have argued that brand associations reside in the mind of the consumer
thus a direct elicitation of associations from the consumer would provide the most accurate measure. Furthermore,
Supphellen posited that the only valid manner to establish brand associations is through a qualitative inquiry directly
from consumers. Therefore, this study will directly elicit brand associations from consumers as a foundation to build
separate measures for the goods and services context. The following section details the procedures utilized to
construct the brand association measure for the goods and services.
This study mirrored the scale development procedures put forth by Gwinner, Gremler, and Bitner (1998) in their
examination of relational exchange in the service context. Two studies were completed in order to construct separate
brand association measures for goods and services that are believed to be necessary for brand managers that manage
both tangible and intangible products. First, a qualitative study (Study 1) was completed from which the items for the
brand association measure were developed. Study 1 was a direct elicitation of brand associations from consumers in
order to generate specific items for the measures. Study 2 tested the brand association items via a survey
questionnaire and examined the relationship among the items and their psychometric properties by means of factor
analysis.
3. Study 1
3.1 Method
Brand associations were generated from two goods contexts (athletic footwear & automobiles) and two service
contexts (amusement parks & restaurants). Besides the goods and services distinction, there were additional criteria
that influenced our product choices. The goods context is comprised of a low involvement (athletic footwear) versus
high involvement (automobile) product as well as two products that may be consumed for both hedonic and
utililitarian reasons. In regards to the services context, the amusement park context was chosen as a highly
experiential service context where consumers may be driven by hedonic motives to consumer the product. In contrast,
the restaurant context may contain elements of hedonic consumption motives but more than likely, the decision to
consume is a utililitarian one.
A separate brand association measure for each product setting was established. This was necessary due to the diverse
nature of brand associations. Brand associations were elicited directly from consumers by means of a thought-listing
procedure. The thought-listing form was constructed to assess the associations respondents held for one of four
different products. For the goods context, associations were identified for automobiles and athletic footwear. The
services context was comprised of amusement parks and restaurants. Respondents were asked to think of their
favorite brand in the respective product category and list the first thoughts that came to mind.
3.2 Sample
Seventy-two thought-listing forms for the goods context were administered to undergraduate students in sport
management courses at a large southeastern university. Sixty-nine forms were deemed usable for analysis for a
response rate of 95.8%. Of these sixty-nine forms, thirty-four respondents completed a thought-listing form in the
context of their favorite athletic footwear brand while thirty-five respondents completed a thought-listing form in the
context of their favorite automobile brand. Seventy-five thought listing forms for the services context were
administered to a different group of undergraduate students at the same university. Seventy-one forms were utilized
for analysis which constitutes a response rate of 94.6%. Of these seventy-one forms, thirty-four respondents
completed a thought-listing form in context of their favorite restaurant brand while thirty-seven respondents
completed a thought-listing form in the context of their favorite amusement park.
Thought-listing forms were discarded when respondents failed to complete the form or when they did not follow the
directions specified. For example, two respondents did not provide thoughts related to the brand on the form and as a
result, their forms were discarded from analysis. When determining the adequate sample size, 20-30 observations are
needed to adequately capture approximately 95 % of responses by consumers (Griffin & Hauser, 1992). Accordingly,
the number of completed thought listing forms for each product category was deemed adequate.
3.2.1 Goods
A total of 34 athletic footwear brand thought-listing forms were completed. Respondents were comprised of 25
(73.5%) males and nine (26.4%) females. The average age was 20.9 years; the sample included 19 (55.8%) juniors
and 15 (44.2%) seniors. The racial makeup was 28 (82.3%) Caucasian, three (8.8%) African-American, two (5.8%)
Latino, and one (2.9%) Asian.
A total of 35 automobile brand thought-listing forms were completed. Respondents were comprised of 26 (74.3%)
males and nine (25.7%) females. The average age 20.6 years; the sample included three (8.6%) sophomores, 17
(48.6%) juniors, and 15 (42.9%) seniors. The racial makeup was 24 (68.6%) Caucasian, nine (25.7%) Latino and two
(5.7%) Asian.
3.2.2 Services
A total of 34 restaurant brand thought-listing forms were completed. Respondents were comprised of 22 (64.7%)
males and 12 (35.3%) females. The average age was 21.2 years; the sample included 24 (70.6%) juniors and 10
(29.4%) seniors. The racial makeup was 28 (82.4%) Caucasian, four (11.8%) African-American, one (2.9%) Latino
and one (2.9%) Asian.
A total of 37 amusement park thought-listing forms were completed. Respondents were comprised of 27 (73%)
males and 10 (27%) females. The average age was 20.7 years; the sample included one (2.7%) freshman, six (16.2%)
sophomores, 13 (35.1%) juniors, and 17 (46%) seniors. The racial makeup was 33 (89.2%) Caucasian, two (5.4%)
African-American, and two (5.4%) Latino.
3.3 Data Analysis
The purpose of the thought-listing procedure was to determine the most salient thoughts that consumers held for their
favorite brand in two goods (automobile & athletic footwear) and two service (amusement park & restaurant)
categories. A content analysis procedure was employed involving two coders and one of the researchers. The two
coders were doctoral students in a graduate program at a large southeastern university. The three coders analyzed the
data independently. The coders were instructed to categorize the brand associations for each good and service and
provide broad labels and operational definitions for every category identified.
3.4 Results of Study 1
3.5 Goods
In order to determine the total number of thoughts for each product category, all thoughts were counted in the
descriptive analysis. The 69 respondents listed a total of 107 thoughts regarding their favorite brand of athletic shoe.
For the automobile brand context, 110 thoughts were listed; duplicate thoughts were included in the count.
For the athletic footwear category, the three coders agreed upon 7 out of 8 categories for an agreement level of
87.5%. For the automobile category, two out of the three coders identified eight categories while the third coder only
identified seven categories. Those seven categories were in agreement with the results of the other two coders; the
eighth category for the two coders was different. The three coders agreed upon 7 out of 9 categories for an agreement
level of 78%. When there was disagreement among coders, those differences were reconciled through discussion
between the three coders.
Since the overall objective was to build a brand association measure based on the elicitation of associations from two
different goods, the results of the independent codings and subsequent thoughts were merged to develop a final list of
items for the goods context. There were 16 items that could be generalized across the automobile and athletic
footwear contexts (see Table 1). Those items were utilized in the survey questionnaire in Study 2. It is important to
note that a significant number of thoughts were discarded due to the lack of agreement across both the athletic
footwear and automobile contexts. A majority of the brand associations simply were not commonly held across both
product categories. Also, a majority of the thoughts held for athletic footwear or automobiles were consistently
repeated by multiple respondents which tended to inflate the overall number of thoughts held for each brand.
Therefore, the 16 items that were utilized for further analysis represent the common thoughts that consumers held for
automobiles and athletic shoes.
3.6 Services
The 71 respondents listed 156 thoughts regarding their favorite brand of amusement park. In regards to their favorite
restaurant, the respondents listed a total of 146 thoughts; duplicate thoughts were included in the analysis. For the
content analysis procedure, duplicate thoughts were included because it is vital to determine which thoughts were
most frequently occurring in the context of each brand category. The coders were instructed to identify the most
commonly listed thoughts for each product category level (i.e. amusement park and restaurant). The coders next
organized the thoughts into broad categories for the content analysis procedure. This was done in an effort to
compute the reliability coefficient as well as compare the brand associations in their respective service contexts. The
common categories and thoughts could be merged to form the basis for a brand association measure in the service
context.
For the amusement park context, two of the three coders identified and agreed upon eight categories while the third
coder identified nine categories, eight of which were agreed upon by all three coders. The intercoder reliability for
the amusement park category was 89%. For the restaurant category, two of the coders identified seven categories and
one coder identified six categories, which all three coders agreed upon. Seven total categories identified, six of which
were agreed upon by all three coders for an 85.7% level of agreement.
Since the objective was to establish a general measure of service brand association based on the thoughts elicited
from the two service contexts, the subsequent items were compared across the two services and similar items were
utilized in the survey questionnaire for Study 2. A total of 23 items were identified and deemed usable for Study 2
(see Table 1). It is important to note that a significant number of thoughts were discarded due to the lack of
agreement across both the amusement park and restaurant contexts. A majority of the brand associations were not
commonly held across both service brands. Also, similar to the goods context, a majority of the thoughts held for
amusement parks or restaurants were consistently repeated by multiple respondents which increased the overall
number of thoughts held for each brand. Therefore, the 23 items that will be utilized for further analysis represent the
common thoughts that consumers held across both services contexts.
4. Study 2
4.1 Method
The next step was to test the brand association items in order to uncover their latent factor structure and to assess
their psychometric properties. In Study 1, 23 items for service brand associations and 16 items for goods brand
associations were identified. Since the objective was to create a separate brand association measure for goods and
services, a separate questionnaire was used for each product category and two separate data collections were
conducted. For this study, a sport product was selected as a focal brand category for the goods and services context.
The respondents were asked to select their favorite professional sports team brand for the services context or their
favorite brand of athletic footwear in the goods context.
Data were subjected to exploratory factor analysis by SPSS 15.0. The exploratory factor analysis utilized principal
axis factoring with oblique rotation. Exploratory factor analysis was deemed appropriate due to the fact that there
was a lack of previous empirical evidence to suggest how many factors would emerge based on the categories
generated from the thought-listing task in Study 1. A determination of factor retention centers on keeping the factors
that account for the most variance. Therefore, a number of criteria were utilized to decide how many factors to retain.
First, the Kaiser criterion was employed whereas factors that have an eigenvalue greater than one are considered
common factors (Nunnally, 1978). Catell’s (1966) scree test was a second method utilized to determine factor
retention. Finally, a number of identifiability criteria were employed. These criteria included the number of
significant loadings (>.50) on a given factor, common loading variables sharing similar conceptual meaning,
different loading variables having distinct conceptual meaning, and the degree to which the rotated factor structure
illustrates a simple structure (Suhr, 2006).
4.2 Sample
A convenience sample of 500 undergraduate students in physical education courses at a large southeastern university
completed either a goods or services questionnaire. The reasoning for selecting a student sample was two-fold.
Students are an easily identifiable and accessible segment of the population. Also, college students are significant
consumers of spectator sports and sporting goods. Of the 250 surveys administered in the services context, 231 were
deemed usable for analysis for a response rate of 92.4%. In the goods context, 228 survey questionnaires were
deemed usable for analysis for a response rate of 91.2%. In both contexts, uncompleted surveys and surveys with a
lack of variability in responses were discarded from analysis.
The sample of respondents (n=231) for the services survey questionnaire included 161 (69.6%) males and 70 (30.3%)
females. The average age of the respondents was 21.1 years. In terms of race, there were 184 (79.6%) Caucasians, 33
(14.2%) African-Americans, seven (3%) Latino, and seven (3%) Asian. For the goods context (N=228), there were
148 (64.9%) males and 80 (35%) females. The average age of the respondents was 20.8 years age. There were 193
(84.6%) Caucasians, 25 (10.9%), 7 (3%) Latino, and 3 (1.3%) Asian.
4.3 Results of Study 2
4.4 Goods
An examination of the eigenvalues for the initial exploratory factor analysis (see Table 2) indicated that four factors
had eigenvalues greater than one (Hair et al., 2006). Cumulatively, the four factors accounted for 65.85% of the total
variance. The scree test revealed there were two points on the scree plot that can be characterized as “elbows”. The
first elbow point appears after the extraction of the second factor while the second elbow point exists after the point
on the fourth factor. Catell’s (1966) guidelines call for the retention of factors before the elbow joint and to discard
the factors beyond the “elbow” point. In this case, the scree test was inconclusive due to the existence of two “elbow”
points.
The Kaiser criterion indicated that four factors should be retained after accounting for the factor rotation while the
scree test indicated that two or four factors should be retained. Since the results were mixed, it was necessary to
consider further criteria for factor retention. First, an examination of the number of significant factor loadings for
each extracted factor was conducted. For a factor loading to be considered significant, it needs to have a value
greater than .50 (Hair et al., 2006). An examination of the pattern matrix, as seen in Table 3, revealed there were five
items with factor loadings greater than .50 for factor 1. Factor 2 had four items with factor loadings above the .50
cutoff. Factor 3 had two items with factor loadings greater than .50, while factor 4 only had one item with a factor
loading which exceeded the .50 cutoff. Factor 3 contained two items with factor loadings greater than .50 as well as a
common conceptual meaning. In regards to factor 4, although only one factor loading was larger than .50, this factor
was also retained due to the conceptual commonality among the items because the common conceptual meaning
regarding the items was an important determinant in retaining the factor.
An examination of the conceptual meaning of each latent factor was employed as another criterion in factor retention.
This means that each retained factor should retain items that share a similar conceptual meaning. The five items that
represent factor 1 were performance from product use, comfort, durability, enjoyment from product use, and quality.
Conceptually, these items all represent benefits from the use of particular good. Benefits are defined as the personal
value consumers attach to the product or service attributes (Keller, 1993). Durability and quality can be viewed as
functional benefits of the product. Performance from product use, comfort, and enjoyment can be characterized as
experiential benefits. Factor 2 included four items: brand logo, name of specific product, brand name, and color.
These items all relate to specific attributes of a particular good. Attributes are defined as descriptive features that
characterize a product or service (Keller, 1993). Brand logo, brand color, brand name, and name of specific product
are non-product related attributes related to packaging and imagery. This indicates that the goods brand association
measure is represented by two factors: benefits and attributes. As alluded to above, factor 3 contained two items with
common conceptual meaning regarding price. Price has long been considered a brand association that can be
characterized as a non-product related attribute (Keller, 1993). Specifically, these items represent competitive pricing
or a discount for the focal goods brand. In regards to factor 4, Design and fun are important elements for capturing
the intangible image and feeling related to the brand while expensive relates to the intangible positive image
attributable to the brand.
A final criterion utilized was an examination of the rotated factor matrix for signs of a simple structure. A simple
structure exists when items have high loadings on one particular factor and low loadings on all other factors. This
can be achieved by examining the factor loadings in the pattern matrix. Table 3 illustrates the factor loadings. An
analysis of the pattern matrix revealed that all the items for factor 1 had high loadings while having low loadings on
all other factors. The items for factor 2 had high loadings while having low loadings on all other factors. Factors 3
and 4 also do not display evidence of item cross-loading. This provided evidence that the retained four factors had a
simple structure. Collectively, four factors emerged and were interpreted as benefits, attributes, discount, and
imagery.
4.5 Services
The results of the initial exploratory factor analysis indicated that five factors emerged with eigenvalues greater than
1 and should be retained as illustrated in Table 4. The total amount of variance explained by the five factors was
63.82%. The scree test did reveal a clear “elbow” point on the graphic. The test indicates there are five meaningful
factors extracted from the data.
Table 4. Variance explained for each factor by exploratory factor analysis in a services context
Extracted Sums of Rotated Sums of
Squared Loadings Squared loadings
Factor Eigenvalue Cumulative % Total
1 8.167 35.508 6.846
2 2.364 45.786 2.249
3 1.709 53.217 3.791
4 1.352 59.097 3.212
5 1.087 63.823 3.148
The Kaiser criterion and scree test indicated that five factors should be retained after accounting for the factor
rotation. Although the initial tests were in agreement, further criteria were evaluated for factor retention. First, an
examination of the number of significant factor loadings for each extracted factor was conducted. An examination of
the pattern matrix as seen in Table 5 revealed there were seven items with factor loadings greater than .50 for factor
1. Factor 2 had three items with factor loadings above the .50 cutoff and factor 3 had four items with loadings greater
than .50. Factor 4 and factor 5 each had three items with factor loadings of >.50. An examination of the conceptual
meaning of each factor was necessary to ensure that each item representing the latent factor shared significant
conceptual meaning. The six items that represent factor 1 were fun, fun atmosphere, excitement, spending time with
friends, spending time with family, and enjoyment. Entertainment was dropped from factor 1 due to its cross loading
with factor 3. Conceptually, these items all represent benefits from the use of particular service.
An examination of factor 2 revealed that two items related to the price of a service brand and one item dealt with
quality. For many service firms and retailers, the utilization of price as a heuristic cue is a keystone element of their
brand image and a method to establish a competitive advantage over their competitors. Therefore, factor 2 was
retained and relates to the concept of “discount” as a brand association. Factor 3 included four items: friendly
employees, helpful employees and knowledgeable employees, and entertainment. Aside from entertainment, these
items represent the characteristics of service employees, a key aspect of the service experience. This factor was
retained with the exclusion of the item related to entertainment.
Factor 4 contained five items which were brand name, brand logo, brand color, name of product, and specific
product offering. Since the two items pertaining to a specific product did not have sufficient factor loadings, they
were dropped from this dimension. Although the brand color item did not reach the cutoff point, its exclusion would
fundamentally limit the conceptual meaning of the factor thus it was kept as the third item in factor 4. Collectively,
these three items represent the specific attributes of a given brand mark. Specifically, these items can be
characterized as non-product attributes. Although factor 5 had three items with large factor loadings, most of the
items cross-loaded with factor 1. This stands to reason since the items in both factors deal with the specific benefits
of service brand consumption. Recognizing the limitations with factor 5, we decided to retain this factor for further
analysis. Since two of the items dealt with social interaction, the factor was entitled social benefits.
Based on the results of the exploratory factor analysis for the services context, five factors were retained. The items
in factor 1 clearly capture the benefits of consuming a service product. For the most part, these benefits can be
described as experiential benefits of service product consumption. The items in factor 3 describe specific
characteristics of service employees, an integral part of the service experience. Factor 4 contains items that deal with
the imagery of a service brand such as the logo, name, and colors. These items can be considered non-product
attributes. Collectively, the services context contained three retainable factors: benefits, service, and attributes. The
factor names and respective items for the goods and services context can be viewed in Table 6.
Discount Service
Affordable Helpful employees
Cheap Friendly employees
Knowledgeable employees
Imagery
Expensive Attributes
Design Brand name
Fun Brand color
Brand logo
Social Benefits
Relaxation
Socializing
Friendly atmosphere
and services categories. It is imperative for managers to understand what their consumers think about their brand and
this measurement scale for both product categories is an initial step in that direction. Specifically, it is important to
contrast the brand associations consumers hold between goods and services. For both contexts, the results of the
exploratory factor analysis illustrated most consumers hold brand associations related to the attributes and benefits of
a product. The benefits in the goods context were very tangible and functional attributes such as quality, comfort, and
performance. By contrast, most consumers thought of visceral or experiential benefits in the services context such as
excitement, enjoyment, and fun. This stands to reason when we consider one of the key differences between goods
and services which is the intangible, experiential experience of consuming a service (Berry, 2000). Furthermore,
most respondents indicated that characteristics of the service employee were also significant associations they held
toward service brands. This result illustrates the importance of the interaction between service employees and
consumers.
This study provides a solid foundation for managers of goods and services brands. Managers can utilize these results
when crafting marketing and communication messages regarding their product. For example, a new line of athletic
footwear should focus on communicating the functional attributes of the product such as quality, comfort, and
performance. It is also important to construct a meaningful brand name, colors, and logo due to the fact that these
associations were front and center in the mind of the consumers. In the services context, brand managers need to
focus their marketing communications on the experiential benefits of their product such as fun, excitement, and
enjoyment. In the service environment, it is important to create an experience that appeals to the senses and leaves a
lasting impression on consumers. Similar to the goods context, attention needs to be paid to the construction and
communication of the brand imagery. Finally, service managers need to be cognizant of the training and
implementation of front-line service employees. Respondents in this study consistently thought of the friendliness
and knowledge level of among this group of employees. Due to the subjective, inconsistent nature of service
exchanges, it is paramount that service employees are personable, adequately trained, and provide an exemplary
level of customer service. Achieving these objectives will influence the strength and favorability of brand
associations that consumers hold toward the service brand.
From a measurement standpoint, this research endeavor illustrated the homogeneity of brands from the consumer
perspective. One of the glaring limitations of this study was the lack of associations that could be generalized across
service or goods brands. Since this study elicited associations from two different goods and services brands and
merged the results of this elicitation, it was clear that only a few associations were evident in both settings. Since
consumers’ thoughts are so diverse depending on the brand, it is clear that a general service or goods brand
association measure is not attainable. Instead, the focus of future research should be to construct a measure of brand
associations that can be generalized across a specific product category. For example, in the athletic footwear
category, a brand association measure can be constructed by eliciting consumers’ thoughts for numerous brands (e.g.,
Nike, Adidas, Puma, and Asics) within this product category. Therefore, future research should continue to assess
brand associations in both the goods and services contexts.
A related point on this subject is the argument between constructing a parsimonious measurement model and a model
that captures the wide array of brand associations that consumers hold for a specific brand. One of the criticisms
levied against the other brand association scales is that they contain too many items for practical application. The
brand association measure in this study is very parsimonious for the goods and services context yet it clearly does not
capture the full range of thoughts that consumers hold for a brand. Since brands are so diverse, thus creating a
multitude of associations in the mind of the consumer, brand association research suffers from a “black hole”
phenomenon. Therefore, future research needs to balance this apparent contradiction between creating a measure
with construct validity and parsimony. Researchers should focus on constructing measures that can only be
generalized to the product category level. Likewise, in the context of this study, there were a number of unique brand
associations that did not generalize across product categories (i.e. athletic footwear and automobiles). Therefore, the
results in this study suggest that a general goods or service brand association measure is not an attainable goal due to
the lack of brand parity across product categories. Instead, the thoughts generated from Study 1 could form the basis
for a measure in each of the four contexts studied (automobile, athletic footwear, restaurants, amusement parks).
Further consumer elicitation can compliment the results of this study to form a measure with greater construct
validity.
The current study was driven by important research questions including what brand associations exist in each of
goods and service contexts and how factor structures differ across the two settings. The results indicated that brand
associations in a goods context consisted of four dimensions, whereas five dimensions emerged in a service setting.
The current study represents an initial effort to provide brand managers with more holistic information pertaining to
the specific components of brand associations based both on tangible goods and intangible services. The results merit
further marketing research with respect to formulating an explanation of what factors contribute most to driving
brand equity through tangible and intangible products.
References
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New York: Free Press.
Aaker, D. A. (1996). Building strong brands. New York: McMilian.
Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34(3), 347-356.
http://dx.doi.org/10.2307/3151897
Bauer, H. H., Sauer, N. E., & Exler, S. (2008). Brand image and fan loyalty in professional team sport. Journal of
Sport Management, 22(2), 1-25.
Berry, L. (2000). Cultivating service brand equity. Journal of the Academy of Marketing Science, 28, 128-137.
http://dx.doi.org/10.1177/0092070300281012
Brady, M. K., Cronin, J. J., Fox, G. L., & Roehm, M. L. (2008). Strategies to offset performance: The role of brand
equity. Journal of Retailing, 2, 151-164. http://dx.doi.org/10.1016/j.jretai.2008.04.002
Catell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research.
http://dx.doi.org/10.1207/s15327906mbr0102_10
Checchinato, F., Disegna, M., & Vescovi, T. (2013). Does country of origin affect brand associations? The case of
Italian brands in China. Journal of Global Scholars of Marketing Science, 23(4), 409-421.
http://dx.doi.org/10.1080/21639159.2013.818281
Cobb-Walgren, C. J., Ruble, C. A., & Donthu, N. (1995). Brand equity, brand preference, and purchase intent.
Journal of Advertising, 24(3), 25-37. http://dx.doi.org/10.1080/00913367.1995.10673481
Del Rio, A., Vazquez, R., & Iglesias, V. (2001). The effects of brand associations on consumer response. Journal of
Consumer Marketing, 18, 410-425. http://dx.doi.org/10.1108/07363760110398808
Faircloth, J. B, Capella, L. M., & Alford, B. L. (2001). The effect of brand attitude and brand image on brand equity.
Journal of Marketing Theory and Practice, 9(3), 61-75. http://dx.doi.org/10.1080/10696679.2001.11501897
Griffin, A., & Hauser, J. (1992). The voice of the customer. Marketing Science Institute, 12(4), 92-106.
Gwinner, K. P., Gremler, D. D., & Bitner, M. J. (1998). Relational benefits in services industries: The customer’s
perspective. Journal of the Academy of Marketing Science, 26(2), 101-114.
http://dx.doi.org/10.1177/0092070398262002
Hair, J. F., Black, W.C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis (6th ed.).
Upper Saddle River, NJ: Prentice-Hall.
Kayaman, R., & Aralsli, H. (2007). Customer-based brand equity: Evidence from the hotel industry. Managing Service
Quality, 17(1), 92-109. http://dx.doi.org/10.1108/09604520710720692
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer based brand equity. Journal of Marketing,
57(1), 1-22. http://dx.doi.org/10.2307/1252054
Keller, K. L. (1998). Strategic brand management. Upper Saddle River, NJ: Prentice-Hall.
Kim, Y., Lee, J., & Lee, Y. (2008). Relationship between brand personality and the personality of consumers, and its
application to corporate branding strategy. Journal of Global Academy of Marketing Science, 18(3), 27-57.
http://dx.doi.org/10.1080/12297119.2008.9707516
Low, G. S., & Lamb, C. W. (2000). The measurement and dimensionality of brand associations. Journal of Product
and Brand Management, 9(6), 350-368. http://dx.doi.org/10.1108/10610420010356966
Maslow, A. H. (1970). Motivation and personality. New York: Harper & Row.
Moon, B., Park, W., & Choi, S. (2010). Relationships among brand equity components: An exploratory study of the
moderating role of product type. Journal of Global Academy of Marketing Science, 20(1), 98-108.
http://dx.doi.org/10.1080/12297119.2010.9707348
Nelson, D. L., McEvoy, C. L., & Dennis, S. (2000). What is free association and what does it measure? Memory and
Cognition, 28(6), 887-899. http://dx.doi.org/10.3758/BF03209337