Explicit and implicit attitudes towards brands and brand
extensions :
An experiment about the effect of new information and brand
extensions on explicit and implicit attitudes
Charlotte Sophie Stienstra 10579877
charlottestienstra@student.uva.nl
26-06-2014
Master thesis
Master Persuasive Communication Graduate School of Communication
Universiteit van Amsterdam
2
Abstract |
The current research tested different attitudes of consumers, by measuring them both explicitly and implicitly, and indicated whether negative or positive information about brand extensions influenced these attitudes most. The research also identified explicit and implicit attitudes toward a parent brand and relative brand extensions. The implicit attitude was measured by the Affective Misattribution Paradigm which measures feelings and emotions towards Chinese signs, but is rarely used in communication and marketing research. The study also looked at the direct influence of information valence on the explicit and implicit attitudes as well as at its indirect influence on the effect of brand extensions’ and the parent brand’s similarity on attitudes. The similarity of the parent brand and the brand extension did not seem to have an effect on the explicit attitude towards the parent brand and implicit attitudes. Neither the information valence played a moderating role in this effect. An
individual’s experience with the brand showed that this act as more of a positive influence on people’s explicit attitudes toward the brand. However, the brand extension similarity did have a positive effect on the explicit attitude towards the brand extension. This explicit attitude was more positive when a close brand extension was showed. Also, the information valence did have a direct effect on the explicit attitude towards the parent brand and the brand extension, however not on the implicit attitude. The more positive the information was about the brand extension, the more positive the explicit attitude was, compared to the negative information. This extended to the influence of information valence on the purchase intention: the more positive the information was the higher the purchase intention was to buy the brand and the brand extension. Thus, information about a brand extension did not only give the consumer information about the brand extension but also about the parent brand which lead to explicit attitude change towards both of them.
3
Introduction |
Brand extensions are not a new phenomenon for businesses. Brand extensions were first introduced by luxury fashion brands, and it has now become custom for many
companies, especially for fast moving consumer goods (Kapferer, 2004). Virgin is a perfect example for brand extension introduction. It started as a music label, and diversified to coffee shops, telecommunication providers, and even an airline. Preferable to introducing a new brand name, the name of the existing or core brand like Virgin is used to enter new markets. A brand extension is therefore regarded as the use of an existing brand name to enter a new product market with new products (Aaker & Keller, 1990; Franzen, 2009; Park, McCarthy & Milberg, 1993). Above all, launching a brand extension is less expensive and less risky than launching a totally new brand, and gives a higher likeability to success (Aaker & Keller, 1990; Franzen, 2009; Tauber, 1981). By introducing a brand extension, the awareness of the parent brand is used to transfer the reputation onto the new product (Kahn & Rahman, 2009). In this way it is utilized to increase the sales of the new product as well as promoting the parent brand.
Significant research on brand extensions is undertaken by investigating what the effect of a brand extension is on the attitude of consumers towards it and the parent brand (Aaker & Keller, 1990; Ahluwalia & Gürhan-Canli, 2000). These studies on brand extensions have been classified by Yorkston, Nunes & Matta (2010)into (1) research on various
conceptualizations of perceived fit or similarity (Bridges, Keller & Sood, 2000), (2) studies of the effects of context variables (Ahluwalia & Gürhan-Canli, 2000), (3) studies of
communication strategies for brand extension success (Puligadda, Cronley & Kardes, 2013), and (4) investigations of consumer-level differences (Yorkston et al., 2010). The current study will research the effect of context variables and the effect of similarity between the brand extension and the parent brand. Previous research with explicit measurements will
4 therefore be replicated, but with adding implicit measurements, a new view with regards to brand extensions will be given with positive and negative information as context variable.
The similarity of the brand extension with the parent brand has always played a central role in brand extension research. A far extension tends to include little similarities between the parent brand and the extension. On the other hand, a close extension holds several similarities with the parent brand, in other words, they fit together (Franzen, 2009).
Revisiting the example of Virgin, a music label and an airline might be seen as a far fit, yet a music label and a radio station are seen as a close fit of brand extension similarity. The fit between the parent brand and the extension tends to be a good predictor of the success of the extension, as found by Aaker and Keller (1990). To go deeper into this finding, the current study sets out whether the fit between the brand extension and the parent brand has an influence on attitude, as well on explicit as on implicit attitudes.
From social psychology science it is known that attitudes and behaviors can be automatically activated outside conscious awareness (Bargh, 2006) and that our evaluations and attitudes are context sensitive (Schwarz, 2008). Giving people information about a brand is giving them a context wherein they form an attitude. Such primes can influence
consumers’ attitudes (Yorkston et al., 2010). Whether the information valence is positive or negative may make a difference in how the attitude is formed. Therefore, the current study tests what the moderating role of positive and negative information is on the effect of brand extension similarity on the explicit and implicit attitude towards the parent brand and brand extension, and whether the information valence has a direct effect on explicit and implicit attitudes.
In communication and marketing science, little implicit research is done to understand attitudes and behaviors (Ratliff, Swinkels, Klerx & Nosek, 2012). Most previous studies are limited to measure the effect of brand extensions with explicit measurements, such as recall
5 and recognition in self-reports. Explicit measurements collect attitudes and associations by directly asking participants about their thoughts of a specific object, and as a result, the conscious memory is tested, which may prejudice how consumers see the brand extension and its parent brand (Crano & Prislin, 2008; Duke & Carlson, 1993; Petty, Fazio & Briñol, 2008). Therefore, implicit measurements are important and interesting to measure as they test the underlying causes of people’s attitudes and associations which are crucial indicators of people’s choices and behavior (Gibson, 2008; Greenwald & Banaji, 1995). The implicit tests measure automatic attitudes and give data to analyze information processing about the unconscious memory of consumers (Duke & Carlson, 1993; Greenwald & Banaji, 1995). These automatic attitudes are indicators of spontaneous actions (Devos, 2008), and can guide consumers behaviors (Olson & Kendrick, 2008) which are important factors of purchasing behaviors among consumers.
In summary, this study will indicate whether explicit and implicit attitudes differ for far and close brand extensions, and their parent brand, when consumers are exposed to negative or positive information. While building upon previous studies by adding implicit
measurements, the validity of research about brand extensions will improve and give new insights on how consumers evaluate brand extension similarity and information valence. This will provide marketing and communication science and practice more in depth insights into consumers and their underlying psychological processes when confronted with brands (Maison, Greenwald & Bruin, 2004). To investigate the above, the following research question is proposed:
RQ: ‘To what extent does brand extension similarity have an influence on the explicit and
implicit parent brand and brand extension attitudes, and is this effect moderated by positivity of information about the extension?’
6
Theoretical Framework |
The use of implicit measures in this study is to obtain more insights in consumers’ attitudes about brand extensions and the parent brand. The dual attitude model states that people hold contrasting implicit and explicit attitudes toward the same product. Namely, the dual attitudes (implicit and explicit) are seen as having separate mental representations of products that are kept in separate parts of the brain, because they derive from different mental processes (Petty, Fazio, & Briñol, 2008). Whereas the expliciy attitudes stem from relatively thoughtful propositional processes, the implicit attitudes stem from more automatic processes (Greenwald & Nosek, 2008). Because attitudes are not inevitably accessible to acquire via self-observation, explicit measurements (e.g. self-reports) do not always give the true estimation of people’s evaluations and feelings. With explicit measurements the participants have to use judgments that come to mind at that specific time to think of an answer (Schwarz, 2008). Also with measuring attitudes by self-reports, social desirability, impression
management, or demand characteristics can influence the outcomes of the tests. Therefore, implicit measurements are used in this study. The attitudes obtained via implicit
measurements are qualitatively different from attitudes obtained via self-reports (Devos, 2008). This is because the implicit measurement does not involve introspective access and minimizes the role of conscious commands. It can capture attitudes of people that they are unaware of and that they are unwilling to report (Gattol, Sääksjärvi, & Carbon, 2011).
As indicated by the dual attitude model, the implicit and explicit attitudes of consumers could also vary for a brand extension. For this, it might make a difference how similar the brand extension is compared to the parent brand. Previous research showed that whenever the brand extension similarity is close, the more positive the explicit attitude is toward the extension as well as the more positive the explicit attitude is toward the parent brand (Aaker & Keller, 1990; Ahluwalia, 2008; Park, Milberg & Lawson, 1991). On the other hand, an
7 insufficient fit results in negative beliefs and attitudes towards the extension and the parent brand (Völckner, & Sattler, 2006). To replicate these studies, the following hypothesis will be tested:
H1: Explicit attitude is more positive for a close brand extension than for a far brand extension.
Typically, previous studies mostly used explicit measurements. What kind of influence close and far extensions have on the implicit attitude is not yet known. Therefore the current study will replicate previous studies but with adding implicit measurements. It is perceived that the implicit attitudes towards the parent brand are not likely to change by a close or far brand extension, because implicit attitudes are more established attitudes and once an attitude is formed toward the brand this attitude will remain stable (Ranganath & Nosek, 2008; Ratliff, Swinkels, Klerx, & Nosek, 2012). In other words, implicit attitudes are created through traces of past behavior which influence feelings, thoughts and actions towards an object (Greenwald & Banaji, 1995). Therefore, it could be that implicit attitudes do not change when an insufficient fit between the extension and the parent brand is made and do not differ for far and close extensions. Because it is expected that there will not be a difference between close and far brand extension, the following sub question is suggested:
SQ1: What is the effect of close and far brand extensions on the implicit attitude?
The information valence, whether the information is positive or negative, can have different effects on attitude. It is expected that positive information will lead to a more positive attitude, and that negative information will lead to a more negative attitude towards
8 the parent brand and brand extension (Schwarz & Bless, 1992). Schwarz and Bless (1992) found that the inclusion of positive information will result in a more positive attitude towards the object; the opposite will be observed for negative information. This is because the
temporary information that people read results in assimilation effects, whereby people take the related information to mind and include this information in their opinions. When people read negative extension information, this will lead to a negative attitude towards the parent brand, because they take this extension information into mind which directly influences their explicit attitude. On the other hand, positive extension information will then lead to a more positive explicit attitude. It is expected that the explicit attitude towards the parent brand and the brand extension differs for positive and negative information. The positive information will lead to a more positive explicit attitude and the negative information will lead to a less positive explicit attitude:
H2a: Explicit attitude is more positive for positive information than for negative information.
The effect of a brand extension on the explicit attitude can be influenced by the way new information about the extension is portrayed (Ahluwalia & Gürhan-Canli, 2000). In the case of close and far brand extensions, the extension information will weigh more heavily for the close extension than the far extension because the similarity between the parent brand and the extension will impact the way in which the extension information is perceived in
compliance with the parent brand (Romeo, 1991; Zhang & Taylor, 2009). Then, negative information about extensions that are closely related to the parent brand may be more
detrimental to the parent brand image than negative information about extensions that are not closely related to the brand name (Romeo, 1991). This effect might be due to the fact that
9 negative information about close extensions is incongruent with the parent brand’s image. For far brand extensions the brand products image does not play a big role in this, because the brand extension is less related to the parent brand.
This was also found by Zhang & Taylor (2009), who noted that that negative
information about the brand extension has a dilution effect on the parent brand. The attitudes toward the parent brand before and after the respondents were exposed to the negative information, changed significantly. Zhang and Taylor (2009) state that when negative
information about the brand extension is given as a prime, consumers are more likely to have a negative attitude towards the parent brand. When the brand extension is close to the parent brand, the negative information will create an even greater dilution effect towards the parent brand, than when there is a far brand extension. When there is a close brand extension, the information about the extension will be considered as more relevant to be related to the parent brand, than when there is a far brand extension (Zhang & Taylor, 2009). This was also studied by Ahluwalia & Gürhan-Canli (2000), who found that negative information about the extension was only of influence on the parent brand evaluation when it referred to close extensions. That means that when there is a close extension, negative information will lead to a more negative attitude towards the parent brand than when there is a far extension. This is because the information is seen as less relevant in relation to the extension and the parent brand. Ahluwalia and Gürhan-Canli (2000) suggest that this pattern of findings was the consequence of that positive information about a close brand would not be very indicative of the parent brand quality, but negative information about a close brand extension will clearly show a low quality brand.
Taking the above findings together, the similarity between the brand extension and the parent brand will lead to different explicit attitudes when people are exposed to information about the brand extension. Namely, this information will have more influence on the explicit
10 attitude whenever there is a close fit. These expectations lead to the following hypotheses:
H2b: The effect of information valence on explicit attitude, is stronger for close brand extensions than for far brand extensions.
When negative or positive information is given, this can initiate a difference wherein the implicit and explicit attitudes have divergent outcomes (Greenwald & Banaji, 1995). In other words, the negative and positive information about the extension can have a different influence on the implicit versus explicit attitude. Nosek, et al. (2007) found that, whenever people are exposed to negative information about an object, consciously they might agree with this negative information and have an explicit negative attitude towards the object. But unconsciously, and thus implicitly, positive attitudes may be stored and imply a positive attitude towards the object, while the person was exposed to negative information. This is coherent with the perspective that people may have associations about objects that they would evaluate to be false when they are asked to consciously evaluate the object (Gawronski & Bodenhausen, 2006) and with research showing that implicit attitudes change more slowly than explicit attitudes (Gregg, Seibt & Banaji, 2006). Then, the implicit attitudes are not directly influenced by new information, because the implicit attitudes are formed out of previous associations (Ranganath & Nosek, 2008). For the following example, people have a positive attitude towards Virgin mobile, but then receive negative media information about the company. The explicit attitude might change negatively, because these attitudes are influenced by the new information. But implicitly, the attitude remains the same for Virgin music, and will be as positive as before the new information was given, because the new information does not have a great impact on the formation of the implicit attitude. In summary, whenever new negative information is given, the explicit attitude might change to a
11 negative attitude, while the implicit attitude will remain positive.
Another study by Ratliff et al. (2012) found different outcomes between explicit and implicit attitudes when people were confronted with new information. They researched brand extensions and measured both the implicit and explicit attitudes and showed insight in the influence of positive and negative information. They found that when people take new information into account it influenced their explicit attitude. On the other hand, the implicit attitude remained positive after seeing new negative information about the brand extension. This indicates that explicit attitudes change more quickly than implicit attitudes. Nonetheless, Ratliff et al. (2012) did not include the focus of close versus far brand extensions, and they only looked at the attitude for the brand extension itself.
For brand extensions, the amount of similarity between the extension and the parent brand may be of influence on implicit and explicit attitude. When there are close extensions and new information is given about the extension, the implicit attitude will not change because it was formed out of previous associations (Ranganath & Nosek, 2008). But when there are far extensions, the implicit attitude will be influenced by the new information. At this point, people will not interrelate the brand extension and the parent brand, because the brand extension and the parent brand are not alike (Ratliff et al., 2012). Therefore, the new information will be used to form new associations and new implicit attitudes. In other words, it is not unlikely that when there is a far extension the implicit attitude is more influenced by the new information, because people might not make a connection between the parent brand and the brand extension, but see the brand extension as a product on its own as if it is a new product.
To measure what the influence of brand extensions and information valence is on the implicit attitude, the following hypotheses are proposed:
12
H3a: Implicit attitude is more positive for positive information than for negative information.
H3b: The effect of information valence on implicit attitudes, is stronger for far brand extensions than for close brand extensions.
Figure 1. Conceptual Framework
Method | Design and participants
This study consisted of a 2 (close/far brand extension) x 2 (positive/negative
information) between-subjects design, with one additional control condition (close-neutral). Hence the study consisted of five conditions. Beforehand, an online pretest was executed to indicate which stimuli and which brand would be best to use in the main research (see pretest, p. 17). The main research comprised an online experiment and was executed by young adults. In total, 342 people participated in the study. However, 65 participants did not finish the survey. Several participants were deleted for further analysis: 11 were too old (>30), 1 was too young (<18) and 11 had an eccentric long participation time, over 24 hours. Also the 4
13 respondents who needed more than 22 minutes to finish the survey were deleted (M = 6.12,
SD = 6.42, criteria: M –/+ (2.5 x SD)). The final sample existed of 249 participants (18 – 30
years old, M = 23.65, SD = 2.71), of which 71.1% was female. Highest education level achieved was distributed as followed: 6,0% HAVO, 4,8% VWO, 4,0% MBO, 33,3% HBO and, 51,8% WO.
Procedure
The participants were acquired via convenience sampling. In the main research the respondents were randomly assigned to one of the five groups, wherein a short text was presented about a consumer magazine which did an opinion test about a product of Vans (Ahluwalia, Burnkrant & Unnava, 2000; Ahluwalia & Gurhan-Canli, 2000; Ratliff et al., 2012) (See Table 1). Namely, 55 (22,1%) participants were exposed to a positive message about a close brand extension, 54 (21,7%) participants were exposed to a negative message about a close brand extension, 54 (21,7%) participants were exposed to a positive message about a far brand extension, 50 (20,1%) participants were exposed to a negative message about a far brand extension and 36 (14,5%) participants were exposed to the control message.
All participants were asked to read the text carefully, and afterwards they were asked to think about their view of Vans before proceeding. Directly after the stimulus the implicit attitude was measured and, subsequently, the explicit attitude. After the explicit attitude questions a manipulation check and debriefing question were included. At the end the participants were asked about their demographics.
14 Table 1
Stimuli main research
Close-Positive
“Beeld je in dat Vans hardloopschoenen introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van deze hardloopschoenen vinden. De mensen waren er erg enthousiast over. Zij zeiden dat de hardloopschoenen erg stijlvol zijn en er heel duurzaam uit zien. Velen zeiden dat ze heel graag deze hardloopschoenen zouden willen hebben.”
Close-Negative
“Beeld je in dat Vans hardloopschoenen introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van deze hardloopschoenen vinden. De mensen waren er helemaal niet enthousiast over. Zij zeiden dat de hardloopschoenen niet stijlvol zijn en er niet duurzaam uit zien. Velen zeiden dat ze nooit deze hardloopschoenen zouden willen hebben.”
Far-Positive
“Beeld je in dat Vans een aktetas introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van deze aktetas vinden. De mensen waren er erg enthousiast over. Zij zeiden dat de aktetas erg stijlvol is en er heel duurzaam uit ziet. Velen zeiden dat ze heel graag deze aktetas zouden willen hebben.”
Far-Negative
“Beeld je in dat Vans een aktetas introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van deze aktetas vinden. De mensen waren er helemaal niet enthousiast over. Zij zeiden dat de aktetas niet stijlvol is en er niet duurzaam uit ziet. Velen zeiden dat ze nooit deze aktetas zouden willen hebben.”
Controle
“Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van Vans sneakers vinden.”
Pretest
One pretest was conducted, which held three parts. The pretest consisted of an online questionnaire and was executed by young adults between 18 and 25 years old (N = 30, M =
15 22.5, SD = 1.33) of which 80,6% female. The first part identified whether new products were either high or low in their similarity to the products of a sneaker brand. The perceived
product similarity, to indicate whether an extension is a close or far extension, was measured by four 7-point scales, namely “inconsistent/consistent”, “unrepresentative/representative”, “atypical/typical”, and “different/similar” (Loken & John, 1993). Ten different products were measured compared to a sneaker brand called Brand X: soccer shoes, a watch, jeans, running shoes, sports apparel, a briefcase, a wallet, jewellery, skis and a sports bag. Based on this first part of the pretest, running shoes were chosen as the close extension of Brand X sneakers (M = 5.56, SD = 1.18) and briefcases were chosen as the far extension of brand X shoes (M = 1.37, SD = .51) The running shoes were considered more similar to sneaker brand X than the briefcases, t(29) = 15,41, p = .000.
The second part of the pretest assessed the negativity and positivity of the information messages about the extension. Positive and negative information messages were developed by using the examples of Ahluwalia, Burnkrant & Unnava (2000), Ahluwalia & Gurhan-Canli (2000) and Ratliff et al. (2012). The two messages each dealt with information about the extension and what consumers thought about the extension (See Table 2). Subjects were exposed to either a positive extension message or a negative extension message. Participants were asked how favorable or unfavorable the message was towards the brand and the
product, as well as what they thought of the believability, length and persuasiveness of the messages (Ahluwalia, Burnkrant & Unnava, 2000). The answers were ranged on a 7-point scale from 1 (= extremely disagree) to 7 (= extremely agree). The positive message was rated as more favorable than the negative message for the brand (Mpositive = 5.73, SD = 1.10;
Mnegative = 1.87, SD = 1.12), t(29) = 9.251, p = .000, as well as for the product (Mpositive = 5.87,
SD = 1.12; Mnegative = 1.33, SD = 0.61), t(29) = 7.977, p = .000.The messages were not rated as equivalent in their believability (Mpositive = 4.67, SD = 1.34; Mnegative = 4.60, SD = 1.45),
16
t(29)= 18.433, p = .000, neither on their length (Mpositive = 2.80, SD = 1.08; Mnegative = 2.30,
SD = 1.17), t(29)= 12.385, p = .000, and persuasiveness (Mpositive = 4.33, SD = 1.54; Mnegative = 4.53, SD = 1.68), t(29)= 15.267, p = .000). Still, the messages were maintained the same for the main research, while the positivity of the messages was significantly different.
The third part of the pretest was about the participants’ familiarity, attitude towards sneaker brands, and their purchase intentions. These three constructs were measured to indicate which brand would be the most useful in the current study, namely a brand that is very familiar but also possesses an average positive attitude such that the manipulations in the main study could create variations in attitude. A moderately evaluated brand name had to be chosen to decrease a ceiling effect. The brand should not have a too positive evaluation, to overcome only positive parent brand attitudes. Familiarity was measured by two items by the scale of Sengupta and Johar (2002) (α= .79), brand attitude was measured by three items of the scale of Martin and Stewart (2001) (α= .95), and purchase intention was measured by two items of the scale of Baker & Churchill (1977) (α= .81). The used scales ranged from 1 (=
extremely disagree) to 7 (= extremely agree). Based on this third part of the pretest, Vans was
chosen as the most familiar product (M = 5.21, SD = 1.56), t(29) = 5.995, p = .000 with the neutral point of the scale test value 4, with a more positive attitude than neutral attitude (M = 4.30, SD = 1.56), t(29) = 2.874, p = .000 with test value 4, for a sneaker brand. The purchase intention for this brand was below average (M = 2.81; SD = 1.51), t(29) = -2.467, p < 0.01 with test value 4. All the items used in the pretest were translated in Dutch, to conduct the test in the participants’ mother language.
Independent variables.
Brand extension similarity. Participants in the close extension condition read that Vans
had introduced new running shoes. In the far extension, the participants read that Vans had introduced briefcases.
17
Information valence. Like Keller and Aaker (1992), the positive and negative
information was given via a scenario. For the negative information, the text was about a test of a consumer magazine which indicated that consumers did not like the new product. On the other hand, for the positive information the scenario was about a test of a consumer magazine which indicated that consumers like the new product and that they would like to have this product. At the end of the stimulus text, the participants were asked to think about their impression of Vans and its new product before proceeding.
For an overview of the independent variables see Table 1.
Table 2
Stimuli pre-test
Positive message
“Beeld je in dat merk X een nieuw product introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van dit product vinden. De mensen waren erg enthousiast over het nieuwe product. Zij zeiden dat het erg stijlvol was en er heel duurzaam uit zag. Velen zeiden dat ze heel graag dit product zouden willen hebben.”
Neem een momentje om over je indruk na te denken over het nieuwe product van merk X, voordat je verder gaat.
Negative message
“Beeld je in dat merk X een nieuw product introduceert. Een consumenten magazine heeft een test gedaan om te kijken wat consumenten van dit product vinden. De mensen waren helemaal niet enthousiast over het nieuwe product. Zij zeiden dat het niet stijlvol was en er niet duurzaam uit zag. Velen zeiden dat ze nooit dit product zouden willen hebben.”
Neem een momentje om over je indruk na te denken over het nieuwe product van merk X, voordat je verder gaat.
18
Dependent measures
Implicit attitude. For the implicit attitude measurement the Affective Misattribution
Paradigm by Payne, Cheng, Govorun, and Stewart (2005) was used. Instantly after the stimulus, participants were confronted with four Chinese signs, which were randomized. The participants were asked to rate these signs on beauty, on a scale from (1) very ugly to (10)
very beautiful. The average of these items was used to measure the implicit attitude towards
the stimuli (α= .82, M = 5.30, SD = 1.86).
Explicit brand attitude. Five items used in Bosmans & Baumgartner (2005) were used
to test explicit attitude. One item of Bruner, Hensel and James (2001) was added to test high and low quality as an addition for attitude. On a 7-point semantic differential scale, the items “dislike/like”, “negative/positive”, “bad/good”, “low quality/high quality”,
“undesirable/desirable”, and, “boring/interesting” were shown. These measures were averaged to form an explicit brand attitude index (α = .92, M = 5.00, SD = 1.07). The item “low quality/high quality” let the reliability for the explicit brand attitude decrease a bit, but the scale in its fullness already had a high reliability and all loaded on one factor (varimax), EV = 4.32, R2 = 72.10%. Therefore the item was maintained.
Explicit product attitude. The same items as for the explicit brand attitude were used.
On a 7-point semantic scale the respondents were asked what they thought of the product in the text (stimuli). The items together, which loaded on one factor (varimax) were averaged to form an explicit product attitude index (α = .94, M = 4.00, SD = 1.22), EV = 4.67, R2 = 77.84%.
Purchase intention brand. The intention of purchasing Vans was indicated by two items
on a 7-point semantic differential scale ranging from “I would like to try/I would not like to try” and “I would not consider to buy/I would consider to buy”. These items were developed
19 by Baker and Churchill (1977), and indicated the participants intention to buy Vans (α = .93,
M = 5.00, SD = 1.61).
Purchase intention product. The intention of purchasing the brand extension which was
mentioned in the stimuli was measured by the same items as the purchase intention brand variable (α = .95, M = 3.50, SD = 1.64).
Manipulation check brand/product/believability/persuasiveness. The participants were
asked to rate the extent to which the information was “positive or negative” towards the brand and the product, whether the information was “believable/unbelievable” and whether it was “persuasive/not persuasive” (Ahluwalia, Burnkrant & Unnava, 2000). These items were rated on a 7-point semantic differential scale and were separately checked.
Control variables
Purchase experience. The purchase experience was measured by one item, which was
also used in Doney and Cannon (1997). This item measured whether the participants ever purchased products of Vans in the past, with the options, (1) yes, many times, (2) yes, a few times or (3) no, never. The purchase experience variable was mean centered.
Debriefing. At the end of the survey participants were asked whether they knew the
goal of the study. 13 respondents were coded as knowing the goal (coded as 1) because they knew the survey was about negative or positive information and its influence on attitudes (e.g. ‘Wat voor invloed een positieve recensie oid heeft op het beeld van een merk/product.’
and ‘Om te kijken hoe je tegen bepaalde beelden / tekens aankijkt nadat je een negatief
bericht hebt gehoord. Dat je houding ten opzichte van het eerste wat je weer tegen komt beïnvloedt is.’).
Age. At the end of the survey participants were asked to type in the two numbers of
20
Gender. Participants were asked to tick the right option male or female after the
question: ‘What is your sex?’.
Education. Participants were asked what their highest education level is. They could
choose from (1) Basisschool, (2) MAVO, (3) HAVO, (4) VWO, (5) MBO, (6) HBO, and (7) WO.
Results | Co-variates check
The three dependent variables were tested for the five control variables (purchase experience, debriefing, age, sex and education) (See Table 3). The spearman tests showed that purchase experience was significant with the explicit brand attitude, rs = -.52, p = .000, and the explicit product attitude, rs = -.15, p = .013. Whenever the purchase experience increased, the explicit brand and product attitude increased as well. This control variable may have had an influence on brand attitude, because people already experienced the brand and created a positive attitude towards it in the past. The ANOVA test showed that the groups did not differ significantly on purchase experience, Brand Extension: F(2, 248) = 1.17, p = .311; Information valence: F(2, 248) = 2.26, p = .106. However, because this control variable correlated with the explicit brand attitude and explicit product attitude, the variable was taken into account in all analyses.
Manipulation check
The manipulation check was executed by a one-way analysis of variance (ANOVA). The positive, negative and control messages differed significantly on positivity from each other for the manipulation check brand, F(2, 248) = 59.72, p < 0.001, as well as for the manipulation check product, F(2, 248) = 146.453, p < 0.001. The post-hoc tests for the brand and the product showed that the negative message was rated as being less positive than the control message (Brand: M = 1.25, SD = .26, p < 0.000; Product: M = 1.62, SD = .25, p <
21 0.000 ) and the positive message (Brand: M = 2.06, SD = .18, p < 0.000; Product: M = 3.09,
SD = .18, p < 0.000 ) . Also the control message was rated as less positive than the positive
message for brand as well as the product (Brand: M = .81, SD = .26, p = .007, Product: M = 1.46, SD = .25, p = .000). As expected, the positive, negative and control message differed from each other, indicating that the manipulation was efficient.
Further on, the messages were examined on their believability and persuasiveness. The positive, negative and control messages did not differ significantly on the manipulation check of believability and persuasiveness as expected, Believability: F(2, 248) = .814, p = 0.444; Persuasiveness: F(2, 248) = .615, p = .541.
Table 3
Correlations of covariates with dependent variables
Explicit Brand Attitude Explicit Product Attitude Implicit Attitude Purchase experience -.522** -.157* .080 Debriefing .018 -.079 -.044 Age -.088 .004 .045 Sex .102 .088 .-.209 Education -.029 -.200 -.051
*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed).
Hypotheses
For all analyses to test the hypotheses, multivariate of analysis of covariance tests (MANCOVA) were used. The explicit brand attitude, explicit product attitude and implicit attitude were the dependent variables with brand extension similarity (close vs far) and information valence (positive vs negative) being the between subjects independent variables and purchase experience was the control variable. Also the control stimuli (close-neutral) was
22 taken into account, to see whether the effect of information valence was driven by positive or negative information. Purchase intention was also measured, to indicate if people had the intention to buy the brand or product after they read information about Vans. Purchase intention for the brand and purchase intention for the product were included as extra dependent variables, and mentioned in the following section when significant results were found.
To begin, no main effects were found for the brand extension similarity on explicit brand attitude, F(1, 208) = .41, p = .520, η2 = .00, nor on implicit attitude, F(1, 208) = .06, p = .809, η2
= .00. Therefore for SQ1 it is to say that close and far brand extensions do not have an effect on the implicit attitude. Nevertheless, a strong significant main effect was found for brand extension similarity on explicit product attitude, F(1, 208) = 5.87, p = .015, η2 = 0.02. The Levene’s F-test showed no significance, F(1, 211) = 2.00, p = .158, hence there were equal variances. Participants who were shown a close brand extension had a more positive explicit attitude towards the product (M = 4.26, SD = 1.24) than the participants who were shown a far brand extension (M = 3.91, SD = 1.13). Therefore hypothesis H1 is partly confirmed, namely for the effect of brand extension similarity on explicit product attitude. This means that whenever there was a far brand extension, the less positive the explicit product attitude was and whenever there was a close brand extension, the more positive the explicit product attitude was.
The brand extension similarity also showed a main effect on purchase intention for the product, F(1, 208)= 14.87, p = .000, η2 = .06. The Levene’s F-test showed no significance,
F(1, 211) = 2.81, p = .095, thus there were equal variances. Whenever participants were
shown a close brand extension they had a higher purchase intention to buy the product (M = 3.85, SD = 1.66) than the participants who were shown a far brand extension (M = 3.04, SD =
23 1.51). This showed that the purchase intention is less high for far brand extensions than for close brand extensions.
Further on, a main effect was found of information valence on the explicit brand attitude, F(1, 208) = 4.32, p = .039, η2 = .02, and on the explicit product attitude, F(1, 208) = 37.10, p = .000, η2 = .15. For both the effects, the Levene’s F-test was not significant, Brand:
F(1, 211) = .99, p = .319, Product: F(1, 211) = .00, p = .988. The explicit brand attitude was
more positive whenever participants were shown positive information (M = 5.10, SD = .96), and was less positive whenever participants were shown negative information (M = 4.81, SD = 1.02). For the explicit product attitude counts the same. Namely, whenever positive information was shown, the participants explicit product attitude was more positive (M = 4.55, SD = 1.13). Whenever they saw negative information, the explicit product attitude was less positive (M = 3.62, SD = 1.09). This showed that the explicit brand and product attitude is more positive in the positive condition than in the negative condition. Therefore, the positive information leads to a more positive explicit brand and product attitude than the negative information.
A one-way analysis was executed to investigate whether the positive or negative message drove the effect of information valence. Here, the positive, negative and neutral messages about the close extensions were taken into account. Then, the information valence also had an effect on the explicit brand attitude, (Mpositive = 5.14 , Mnegative = 4.67, Mneutral = 4.57), F(1, 143) = 4.75, p = .031, η2 = .03, and the explicit product attitude, (Mpositive = 4.80 , Mnegative = 3.72, Mneutral = 4.22), F(1, 143) = 23.29, p = .000, η2 = .14. But, for the explicit brand attitude post-hoc Bonferroni tests showed no significant differences between the negative and positive condition (Mdiff = .47, p = .093), and between the negative and neutral condition (Mdiff = .09, p = 1.000), neither between the positive and neutral condition (Mdiff =
24 .56, p = .063). These tests suggested that the effect of information valence on explicit brand extension was more driven by the close-positive message than the close-negative message.
For the explicit product attitude post-hoc Bonferroni tests showed no significant differences between the negative and neutral condition (Mdiff = .50, p = .164), and between the positive and neutral condition (Mdiff = .58, p = .068). But the test did show a significant difference between the negative and positive condition (Mdiff = 1.08, p = .000). By these tests, it was suggested that also the effect of information valence on explicit product extension was more driven by the close-positive message than the close-negative message. The positive and negative message differed significant from each other, which indicated that the positive message was had more influence on the effect than the negative message.
In summary, hypothesis 2a can be confirmed, because this effect accounts for the explicit attitude towards the brand as well as for the product. Withal, no evidence was found for hypothesis 3a. Because no main effects were found for the information valence on implicit brand attitude, F(1, 208) = .15, p = .696, η2 = .00.
The information valence also showed a main effect on purchase intention for the brand as well as for the product, Brand: F(1, 208) = 5.76, p = .017, η2 = .02; Product: F(1, 208) = 12.11, p = .001, η2 = .05. However, the Levene’s F-test showed no significance for the purchase intention for the product, F(1, 211) = 2.68, p = .103, it did for the purchase intention of the brand, F(1, 211) = 5.32, p = .022, hence there were no equal variances. Nevertheless, the purchase intention to buy the brand Vans, or the product, was less positive when
respondents were shown a negative message (Brand: M = 4.70, SD = 1.63, Product: M = 3.08,
SD = 1.52) than when they were shown a positive message (Brand: M = 5.23, SD = 1.39,
Product: M = 3.82, SD = 1.67). This indicated that the purchase intention for buying the brand Vans or a brand extension product was more positive in the positive information condition than in the negative condition.
25 The control variable purchase experience showed a main effect for the explicit brand attitude, F(1, 208) = 59.34, p = .000, b* = -.47, η2 = .22, and the purchase intention for the brand, F(1, 208) = 61.45, p = .000, b* = -.48, η2 = .22, but not for the explicit product attitude, F(1, 208) = 1.70, p = .194, b* = -.10, η2 = .00, nor the implicit attitude, F(1, 208) = 2.81, p = .095, b* = -.07, η2 = .01. This demonstrated that experience with the brand had an influence on the explicit brand attitude and the purchase intention for the brand, so that when the more experience people had with the brand, the more positive their attitude towards the brand was, and the more likely they were to buy the brand.
No interaction effects of brand extension similarity and information valence were found for the explicit brand attitude, F(1, 208) = 1.05, p = .305, η2 = .00, explicit product attitude,
F(1, 208) = .916, p = .340, η2 = .00, and the implicit attitude, F(1, 208) = .02, p = .887, η2 = .00. Therefore, for hypotheses 2b and 3b there is no evidence to confirm the expectations.
In summary, in relation to the explicit brand attitude, it was found that that the attitude about the brand was not influenced by whether the brand extension was far or close, but was influenced by the information valence and whether people had previous experience with the brand. The explicit brand attitude was more positive when positive information was given to the participants, and was also more positive when they had previous experience with Vans.
On the other hand, the explicit product attitude was influenced by the information valence and the brand extension similarity. In other words, whenever the information was positive or whenever the brand extension was close made the attitude towards the product more positive than whenever the information was negative or the brand extension was far. The information valence also had an influence on the purchase intention for the brand and the product. The purchase intention for the brand and product increases when a positive message was shown, and decreases when a negative message was shown. Also the purchase experience had an influence on the purchase intention for the brand, and the brand extension
26 similarity had an effect on the purchase intention for the product. For implicit attitudes, the brand extension similarity did not have an effect upon it. Nonetheless, it is not possible to say whether the information valence had an influence on the implicit attitudes. In the end, no interaction effect was found of brand extension similarity and information valence on the dependent variables. The independent variables had only separately an effect on several dependent variables.
Conclusion |
This study was conducted to examine the effect of brand extensions on explicit and implicit parent brand and brand extension attitude, and tested if this effect was moderated by information valence (e.g. the positivity of information). Whilst a lot of companies introduce brand extensions, the impact of the extensions is not researched prior on implicit attitudes. The difference in similarity between the core product and the brand extension, which makes it a far or close brand extension, can have a different effect on the attitudes. Also new information about the brand extension can be of influence on attitudes, and can even have a greater effect when the brand extension is close to the parent brand (Ahluwalia & Gürhan-Canli, 2000; Zhang & Taylor, 2009). An online experiment was utilized to give more insights into these statements, such as what effect far and close brand extensions have on the explicit and implicit attitudes (H1 & SQ1), and what the effect of information valence is on explicit and implicit attitudes (H2a & H3a), as well as whether the effect of information valence on explicit and implicit attitudes is stronger for far or close brand extensions (H2b & H3b). The sneaker brand Vans was used in this quantitative study, with running shoes as a close brand extension and a briefcase as a far brand extension. A positive, negative or neutral message about Vans and its extension was given as a means of manipulation.
The outcomes showed an effect of brand extension similarity on the explicit product attitude, but not on the explicit brand attitude. When the message was about a close brand
27 extension, the explicit product attitude was more positive than when people read a message about a far brand extension. This is partly in line with previous research, where it was stated that close extensions will lead to more positive explicit attitudes towards the extension as well as towards the parent brand (Aaker & Keller, 1990; Park, Milberg & Lawson, 1991). That the current study did not found this effect for the explicit brand attitude, could be due to the purchase experience that people have with the brand. The current study found that
whenever people had experience with the parent brand, the explicit brand attitude was more positive than when they did not experienced the brand Vans before. Thus, the previous experience said more about the explicit brand attitude then whether the brand extension was close or far. This can be due to the fact that people have more information about the brand on which they base their explicit attitude towards it, than on only the information they had about the far or close brand extension (Phelps & Thorson, 1991). Also because for a familiar brand, people already have many associations because they experienced the brand before (Campbell & Keller, 2003).
The information about the brand extension that was given to the participants in the study was either positively, negatively or neutrally formulated. This information valence showed to have an effect on both the explicit brand and the explicit product attitude. Information valence noted that whenever the information is positive, it led to more positive explicit attitudes than the negative information. This finding is in line with previous research of Schwarz and Bless (1992). Thus, people bore the new information to mind which
influenced their attitude towards the brand Vans and its brand extension. Next to this, when taking the neutral message into account, it is suggested that the effect of information valence was more driven by the positive information than the negative information, and that the positive information had a greater influence. However, previous studies stated that the negative information would be of more impact because negative words are better
28 remembered and are more persuasive (Albarracín, Wang, Li & Noguchi, 2008), as well as that people give more weight to negative information than to positive information (Romeo, 1991). That the current outcomes suggested that the positive information was of more influence, could be because of the fact that people who were shown a neutral message, saw this message as more positive than negative which leaded to more overall positive attitudes. Future research could improve this by creating a stimuli that is even more clear in what is positive, negative and neutral.
In the current study it was not found that there was an interacting effect of information valence and brand extension similarity on explicit attitudes. Hence the suggestions of
Ahluwalia and Gürhan-Canli (2000) could not be applied by the current study. They suggested that positive information about a close brand would not be of much influence on the consumers, but negative information about a close brand would because it shows a low quality brand. This could be due to the fact that Vans is already a known brand, thence the new information about the brand extension is seen as less relevant to the parent brand, which then will not create an adjusted attitude towards the parent brand. Nevertheless, in the current study it was found that the information valence did not influence the effect of the similarity between the parent brand and the brand extension, but the brand extension
similarity and the information valence separately influenced the explicit attitude for the brand and product. Stated differently, whenever consumers were confronted with new positive information about a far brand extension, this positively influenced their explicit attitude as well as when they were confronted with positive information about a close brand extension.
For implicit attitude it was found that the brand extension similarity and the information valence did not have an effect on it, as well as that the interaction of the two variables did not have an effect on the implicit attitude. It was expected that previous experience with the brand would be of influence on the implicit attitude because it was considered that implicit
29 attitudes are formed through traces of past behavior, such as past buying behavior, which influence feelings, thoughts, and attitudes towards a product and brand (Greenwald & Banaji, 1995). However, what was found is that the brand extension similarity and information valence did not have an influence on the implicit attitude, but also the experience with the brand was not of influence on the implicit attitude. However, implicit measurements could indicate little effects and differences about affect and attitudes (Bargh & Chartrand, 1999), they did not come to light in the current study. Therefore, for the effect of the implicit attitude it is not defined as to how this attitude is created for brands and products.
The study also researched what influence brand extension and information valence had on the purchase intention of the brand and the product. This because it is very interesting for practice, because the most important thing of the work of marketers and advertisers is whether consumers buy their products (Kapferer, 2004). It is found that both the brand extension similarity and the information valence had an influence on the purchase intention of the product. The outcomes showed that when the brand extension was close, the purchase intention to buy this brand extension was higher than when the brand extension was far. Also, when the information was positive the purchase intention to buy the brand extension was higher than when the information was negative. These statements showed that close extensions, and positive information about a brand extensions will probably sell the product better than negative information.
On the other hand, the purchase intention for the brand is influenced by the information valence and the previous purchase experience. Whether the brand extension was close or far did not have an influence on the purchase intention for the brand. When the information was positive, this had a more positive effect on the purchase intention for the brand than when the information was negative. Moreover, the previous purchase experience was also of influence on this purchase intention. Hence, partly the information valence had an influence on the
30 purchase intention to buy Vans, as well as the previous experience with the brand. This indicated that positive information given about the brand extension also positively influenced the purchase intention. In practice, this is an important insight, because the new information given about the brand extension does not only influence the intention to buy the product but also the brand.
In conclusion, whether information is positive or negative and whether the brand extension is close or far did not have an influence on the implicit attitude. However, these constructs did have an effect on the explicit attitudes. The fact that the explicit attitude towards the Vans brand or a brand extension of Vans changed when receiving a positive or negative message about the brand extension is a positive outcome for marketers and
advertisers. The messages in the current study were about a magazine which did a test on consumers opinions about Vans’ new products. Their opinions where either positive or negative. When other consumers find the product positive, marketers and advertisers will have a better option to obtain positive explicit attitudes and higher purchase intentions. However, marketers and advertisers should be aware of negative information in consumer reports that is around about their brand extensions to keep the less positive attitudes as low as possible. It also appears that information in relation to a close extension creates more positive explicit product attitudes toward the product than a far extension, which indicates that it is better to communicate about the close extension along with the parent brand. For far extensions it might be better to keep the brand extension and the parent brand apart in informational messages.
Discussion |
This research is not without limitations. The current research studied Dutch participants between 18 and 30 years old, hence the study is definitely generalizable for this population in The Netherlands. However, because the research only tested for the brand Vans, this
31 decreases the generalizability of the study for other brands and products. Future research should include different kinds of brands from diverse product categories to improve the current study and create a more overall view on explicit and implicit attitudes on brand extensions. While Ahluwalia and Gürhan-Canli (2000) did not found differences between an athletic shoe company and an electronic products company for explicit attitudes, it is
suggested that for future research with implicit attitudes this has to be investigated once again. Because implicit attitudes are not created the same as explicit attitudes, as what the dual attitude model states that implicit and explicit attitudes can be contrasting towards the same product (Petty, Fazio, & Briñol, 2008), the attitudes might also differ for fast moving consumer goods and slow moving consumer goods.
The main reason why no effects were found for the implicit attitude, might be due to the use of the Affective Misattribution Paradigm by Payne, Cheng, Govorun, and Stewart (2005). So far, this implicit test is mostly used in psychology to test misattribution and affect towards pictures and social problematical cases such as questions about politics and race differences. This indirect measure is very useful when people do not have readily available attitudes which they can report immediately (Gattol et al., 2011). For attitudes about brands like Vans the attitudes might be readily available, and therefore the implicit test may not have discovered effects of brand extension similarity and information valence on the implicit attitude. Nevertheless, to measure the implicit attitude towards brands and products, different ways of implicit measuring could be used to improve the current results. For example, the Implicit Association Test (Greenwald, McGhee & Schwartz, 1998) might bring about differing and more enhanced results, because this test relates more to the product and its associations than the Chinese signs which are used in the Affective Misattribution Paradigm. Future research could work with this method to increase the insights in implicit attitudes and
32 their changeability by brand extension similarity and information valence, where consumers are exposed to.
For future research the stimuli of the current study could also be improved. A longer text might draw the participants more into the text, which may lead to stronger results. This could be when participants read more in-depth materials about the brand extension and the product. The information valence then might be clearer and have more influence on the changeability of explicit and implicit attitudes. The form of the information in the current study was short and related to a consumer magazine. Other forms of information, either positive and negative but poured into another funnel like an advertorial, might improve the results. Therefore, different forms of information and what its effects are on attitudes toward parent brands and their brand extensions are still interesting to study.
Nonetheless, despite of the limitations, this study shows that future research is necessary on brand extensions and information valence. New information about products is available to consumers, and marketers have to be aware of the influence of it. More research should indicate if the information valence has an influence on the implicit attitude, and whether the attitude on forehand about the parent brand changes by the information valence and the information about far and close extensions. When the previous attitude toward the brand is also tested, it is possible to see whether the attitudes change when people are exposed to new information. The current study indicated that people's implicit attitudes are not influenced by positive or negative information about far or close extensions, but the explicit attitudes are. Therefore, giving information about a brand extension does not only influence the attitude towards that specific product but also towards the parent brand. Thus, marketers and advertisers should be conscious about what information is given about brand extensions, to not let attitudes towards their parent brand change in a negative direction. To
33 overcome this, it is better to create close brand extensions than far brand extensions, because close brand extensions seize more positive attitudes and higher purchase intentions.
Reference list |
Aaker, D. A. & Keller, K.L. (1990). Consumer Evaluations of Brand Extensions. Journal of
Marketing, 54(1), 27-41.
Ahluwalia, R., Burnkrant, R.E., & Unnava, H.R. (2000). Consumer response to negative publicity: The moderating role of commitment. Journal of Marketing Research, 37, 203–214.
Ahluwalia, R., & Gürhan-Canli, Z. (2000). The effects of extensions on the family brand name: an accessibility-diagnosticity perspective. Journal of Consumer Research,
27(3), 371-381.
Ahluwalia, R. (2008). How far can a brand stretch? Understanding the role of self-construal.
Journal of Marketing Research, 45(3), 337-350.
Albarracin, D., Wang, W., Li, H., & Noguchi, K. (2008). Structure of attitudes: Judgments, memory, and implications for change. Attitudes and attitude change, 19-40.
Baker, M. J., & Churchill Jr, G. A. (1977). The impact of physically attractive models on advertising evaluations. Journal of Marketing research, 14(4), 538-555.
Bargh, J. A. (2006). What have we been priming all these years? On the development, mechanisms, and ecology of nonconscious social behavior. European journal of
social psychology, 36(2), 147-168.
Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being. American
psychologist, 54(7), 462.
Bridges, S., Keller, K. L., & Sood, S. (2000). Communication strategies for brand extensions: enhancing perceived fit by establishing explanatory links. Journal of Advertising,
34
29(4), 1-11.
Bosmans, A., & Baumgartner, H. (2005). Goal-relevant emotional information: When
extraneous affect leads to persuasion and when it does not. Journal of Communication
Research. 32(3), 424-434.
Bruner, G. C., Hensel, P. J., & James, K. E. (2001). Marketing scales handbook. Chicago: American Marketing Association.
Campbell, M. C., & Keller, K. L. (2003). Brand familiarity and advertising repetition effects.
Journal of Consumer Research, 30(2), 292-304.
Crano, W. D., & Prislin, R. (Eds.). (2008). Attitudes and attitude change. New York, London: Psychology Press.
Devos, T. (2008). Implicit attitudes 101: Theoretical and empirical insights. Attitudes and attitude change, 61-84.
Doney, P. M., & Cannon, J. P. (1997). An examination of the nature of trust in buyer-seller relationships. Journal of Marketing, 61(2), 35-51.
Duke, C. R., & Carlson, L. (1993). A conceptual approach to alternative memory measures for advertising effectiveness. Journal of Current Issues & Research in Advertising,
15(2), 1-14.
Franzen, G. (2009). Brand portfolio and brand architecture strategies, SWOCC book series 49. Amsterdam: SWOCC.
Gattol, V., Sääksjärvi, M., & Carbon, C. C. (2011). Extending the Implicit Association Test (IAT): Assessing consumer attitudes based on multi-dimensional implicit
associations. PloS one, 6(1), 1-11.
Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change.
35 Gibson, B. (2008). Can evaluative conditioning change evaluations of mature brands? New
evidence from the Implicit Association Test. Journal of Consumer Research, 35, 178– 188.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: attitudes, self-esteem, and stereotypes. Psychological review, 102(1), 4.
Greenwald, A. G., McGhee, D., & Schwartz, J. L. K. (1998). Measuring individual difference in implicit cognition: The Implicit Association Test. Journal of Personality and Social
Psychology, 6, 1464-1480.
Greenwald, A. G., & Nosek, B. A. (2008). Attitudinal dissociation: What does it mean.
Attitudes: Insights from the new implicit measures, 65-82.
Gregg, A. P., Seibt, B., & Banaji, M. R. (2006). Easier done than undone: Asymmetry in the malleability of automatic preferences. Journal of Personality and Social Psychology,
90, 1-20.
Kapferer, J. N. (2004). The new strategic brand management: Creating and sustaining brand
equity long term. London: Kogan Page.
Keller, K. L., & Aaker, D. A. (1992). The effects of sequential introduction of brand extensions. Journal of marketing research, 35-50.
Olson, M. A., & Kendrick, R. V. (2008). Origins of attitudes. Attitudes and attitude change, 111-130.
Maison, D., Greenwald, A. G., & Bruin, R. H. (2004). Predictive validity of the Implicit Association Test in studies of brands, consumer attitudes, and behavior. Journal of
Consumer Psychology, 14(4), 405-415.
Nosek, B. A., Smyth, F. L., Hansen, J. J., Devos, T., Lindner, N. M., Ranganath,
K.A., …Banaji, M. R. (2007). Pervasiveness and correlates of implicit attitudes and stereotypes. European Review of Social Psychology, 18(1), 36-88.
36 Park, C. W., Milberg, S., & Lawson, R. (1991). Evaluation of brand extensions: the role of
product feature similarity and brand concept consistency. Journal of consumer
research, 18(2), 185-193.
Park, C. W., McCarthy, M. S., & Milberg, S. J. (1993). The effects of direct and associative brand extension strategies on consumer response to brand extensions. Advances in
Consumer Research, 20(1), 28-33.
Payne, B.K., Cheng, C.M., Govorun, O., & Stewart, B.D. (2005). An inkblot for attitudes: Affective misattribution as implicit measurement. Journal of Personality and Social
Psychology, 89(3), 277-293.
Petty, R. E., Fazio, R. H., & Briñol, P. (Eds.). (2009). Attitudes: Insights from the new implicit
measures. New York, Hove: Psychology Press.
Phelps, J., & Thorson, E. (1991). Brand familiarity and product involvement effects on the attitude toward an ad-brand attitude relationship. Advances in Consumer Research,
18(1), 202-209.
Puligadda, S., Cronley, M., & Kardes, F. (2013). Effects of advertising cues on brand extension evaluation: A global versus focused processing style account. Journal of
Brand Management, 20(6), 473-487.
Ranganath, K. A., & Nosek, B. A. (2008). Implicit attitude generalization occurs
immediately; explicit attitude generalization takes time. Psychological Science, 19(3), 249-254.
Ratliff, K. A., Swinkels, B. A., Klerx, K., & Nosek, B. A. (2012). Does one bad apple (juice) spoil the bunch? Implicit attitudes toward one product transfer to other products by the same brand. Psychology & Marketing, 29(8), 531-540.
Romeo, J. B. (1991). The effects of negative information on the evaluations of brand extensions and the family brand. Advances in consumer research, 18(1).
37 Schwarz, N. (2008). Attitude measurement. Attitudes and attitude change, 41-60.
Schwarz, N. & Bless, H. (1992). Scandals and the public’s trust in politicians: Assimilation and contrast effects. Personality and Social Psychology Bulletin, 18, 574-579. Tauber, E.M. (1981). Brand franchise extension: new product benefits from existing brand
names. Business Horizons, 24(2), 36-41.
Völckner, F., & Sattler, H. (2006). Drivers of brand extension success. Journal of Marketing,
70(2), 18-34.
Yorkston, E. A., Nunes, J. C., & Matta, S. (2010). The malleable brand: The role of implicit theories in evaluating brand extensions. Journal of Marketing, 74(1), 80-93.
Zhang, L., & Taylor, R. D. (2009). Exploring the reciprocal effect of negative information of brand extensions on parent brand. Marketing Management Journal, 19(1), 1-15.
