The changing role of brands in the age of transparency
Master thesis
MSc in Business Administration – Marketing
By: Steffen Douwe van der Land (10608303)
Under supervision of: J. Demmers MSc
2
Statement of originality
This document is written by Steffen Douwe van der Land, who declares to take full
responsibility for the contents of this document. I declare that the text and the work presented
in this document is original and that no sources other than those mentioned in the text and its
references have been used in creating it. The Faculty of Economics and Business is
3
Table of contents
Abstract ... 5 Introduction ... 6 Literature review ... 9 Transparency ... 9The effects of transparency ... 10
Brand equity ... 12
Transparency and its effect on the need for brands ... 13
Methodology ... 19
Design ... 19
Sample and procedure ... 19
Stimuli ... 20 Measures ... 23 Statistical procedure ... 27 Results ... 28 Discussion ... 35 General discussion ... 36
Theoretical and Practical implementations ... 37
4
References ... 41
Appendix 1 ... 47
List of figures and tables
Figure 1: Example lack of transparency ... 17Figure 2: Conceptual model ... 18
Figure 3: Path model ... 34
Figure 4: Moderation effect by product type and information source ... 35
Table 1: Quality and price levels for the laptop ... 23
Table 2: Quality and price levels for the toothbrush ... 23
Table 3: Mean utility for laptop quality ... 31
Table 4: Mean utility for laptop price ... 1031
Table 5: Mean utility for toothbrush quality ... 31
Table 6: Mean utility for toothbrush price ... 31
Table 7: Descriptive statistics utility HP and Aquafresh ... 32
Table 8: Mean, standard deviation and correlations of laptop variables ... 32
Table 9: Mean, standard deviation and correlations of toothbrush variables ... 32
Table 10: Statistics of indirect effect of transparency on HP utility . ... 264
5
Abstract
With the development of web 2.0 and the shift in the consumer awareness, more need for
transparency has arisen. This need for transparency has created a more transparent age, in
which both consumer and brands appear to benefit from transparency. This study examines
how transparency influences the role of brands in the consumer’s product choice. This is done
by comparing a consumer’s decision within different information settings, using an online
experiment. Respondents were randomly assigned to a non transparent-, a consumer
information-, or a transparent information group, and were then asked to choose between
products varying on brand, quality and price. This was done for both a durable good and a
Fast Moving Consumer Good. CBC analysis software was used to establish the utilities for a
well known branded product in comparison with a fictionally branded product in the three
different conditions. A comparison between the non transparent and the transparent condition
led to the findings that in the transparent condition the utility was significantly lower for the
known brand in the durable goods category. This decrease in utility was partly mediated by
brand trust for the unknown brand. Product type and source disclosure moderated this
relationship. No differences in utility were found for the FMCG category or when comparing
the consumer information group with the non transparent group. Overall, this research shows
that in the transparent age, brands are becoming a less important aspect in the riskier
6
Introduction
On the 7th of September 2011 McDonald’s started to display calories on their menu,
immediately followed by Burger King, KFC and Pizza Hut. The same year, Puma promised toxin free shoes, directly followed by Nike’s ‘right to know’ campaign and a toxin free announcement by Adidas. It appears that more and more firms are, often in combination with
corporate social responsibility reports, providing more information about their way of doing
business. Taking a glance at the corporate websites of multinationals, comprehensive
information can be found on employee conditions, manufactory processes and products.
Brands have become aware of the positive effect of transparency and use it as a marketing
tool to create consumer trust. Previous researched showed that transparency can increase
purchase intention, product value and brand equity (Bhaduri & Ha-Brookshire, 2011; Brady,
2003). Now that all brands are becoming transparent, the question arises whether this
marketing approach still affects the brand equity when all brands are applying this strategy?
The increasing transparency is rooted in the consumers growing need for information.
We have entered the age of transparency, where information is becoming more and more
important (Fournier & Avery, 2011; Tapscott & Ticoll, 2003). The internet has made
information easily accessible, which in turn has led to a more critical consumer with a higher
information need (Rezabakhsh, Bornemann, Hansen, & Schrader, 2006). Indeed, Web 2.0
created an environment in which information travels with unprecedented speed through social
media. Consumers are only ‘one click away’ from critical reviews, reports and blogs. This
makes it much harder for companies to hide information from consumers (Fournier & Avery,
2011). At the same time, consumers are becoming more concerned with the environment and
society. This concern has led to a higher demand of transparent and sustainable products in today’s market (Bhaduri & Ha-Brookshire, 2011). Responsibility and transparency seem to be necessary ingredients to make a brand sustainable (Brady, 2003).
7 Cohne and Wolf (2013) argue that the consumer has become more sceptical and less
trusting towards brands. Nowadays, declining trust is not solely an issue for brands that have
been involved in a scandal. An example can be found in figure 1, which displays a message
found on the authors’ social media website. The figure shows a clearly disappointed consumer
in a previous beloved brand, due to the non transparent behaviour concerning their
manufactory process. Cohne and Wolfe (2013) found that between sixty and eighty percent of
the consumers today take transparency and
honesty of a brand into consideration when
buying a product. Whilst the importance of
transparency is growing, the authors argue
that the importance of brands is decreasing.
An important role brands were fulfilling for
consumers was the role of risk reduction.
Lack of information about the quality of a
product or service before purchase made
consumers use the brand as a quality signal (Fischer, Völckner & Sattler, 2010). Brand equity
helped the consumers trust the product and service before usage. With brands using
transparency, more information is available and more trust can be created, which might lead
to a decline in the importance of brands. Thus, although transparency is increasing brand
equity, at the same time it appears to be making brands less important, giving rise to a
paradox.
However, so far little empirical proof is provided to back up this theory. This research
addresses this gap in the literature by answering the following research question: How does
transparency influence the role of brands on the consumers’ product choice?
Figure 1
8 Although transparency has gained popularity in the research field, there is clearly still
much to discover about the relationship between transparency and consumer behaviour
(Granados, Gupta, & Kauffman, 2012). This study builds on the transparency literature and
adds to the theory on transparency and consumer behaviour. More specifically, this research
brings about new insights in the effect of transparency on brands utility, taking in account
consumer trust, source disclosure and product type.
A better understanding of this effect can have a great influence on how new products
can be marketed. If it indeed turns out that brands are becoming less important because of
transparency, companies with new products can focus more on the distribution of product
information, instead of spending a great deal of resources on building brand equity.
This article will first provide a review of the relevant literature, in order to gain a
better understanding of the different concepts. Subsequently, the research design and research
methods will be discussed. Thirdly, the results of the study will be explained. Finally, a
discussion of the results is provided, consisting of conclusions, implications and suggestions
9
Literature review
This chapter provides an extensive overview of the literature on transparency and its effect on
the relationship between brand equity and consumer behaviour. First of all, the concept of
transparency is discussed. Secondly, the effects of transparency on consumer behaviour are
depicted, followed by an examination of the relationship between brand equity and consumer
behaviour. Finally, a review on the link between transparency, brand trust and brand utility is
presented, including a discussion of two extra variables, namely product choice and source
disclosure.
Transparency
The increasing importance of transparency has led to a growing body of research in the field
of transparency. Although different definitions are used, they appear to describe similar key
elements. The Cambridge Business Dictionary (2014) describes transparency as “a situation
in which business is done in an open way without secrets, so that people can trust that it is done fair and honest”. Christensen (2002) defines it as “public availability of relevant information”. These two definitions describe important elements of transparency. Indeed, Vishwanath and Kaufmann (2001) argue that for information to be transparent it needs to be
accessible, qualitative, reliable and relevant. Accessibility of information is when the
information is available and easily accessible for the stakeholders of the firm. The information
should also be qualitative and reliable, making the information understandable, complete,
honest and consistent. Finally the information should be relevant. Christensen (2002) argues
that transparency is a perception of the consumer based on the information need of the
consumer. Information provided by a firm that does not fulfil this need, consequently does not
10 There are different ways for an organization to be transparent. Hultman and Axelsson
(2007) found four different types of transparency in previous literature: Technological-,
Organizational-, Supply chain-, and Cost/price transparency. Within these different types
there are different levels of transparency. Moreover, these different types are not mutually
exclusive and therefore it could be argued that transparency is a continuum in which different
types of transparency can be used to gain a fully transparent organization.
The effects of transparency
So far various scholars have tried to provide a better understanding of the effect of
transparency on consumers. Different studies have shown that transparency can benefit the
consumer in a great way. For instance, more information can bring opportunities, comfort,
enlightenment, wealth and power (Sweeny, Melnyk, Miller, & Shepperd, 2010). It can help to
make easier and better choices (Scheibehenne, Greifeneder, & Todd, 2010) and deals with
uncertainty avoidance (Vishwanath, 2003). Furthermore, transparency leads to an overall
improvement in consumer welfare (Carter & Curry, 2010; Gu & Wenzel, 2011).
Moreover, research examined the effect of transparency on product value perception,
purchase intention and brand attitude. Carter and Curry (2010) found that price transparency
can affect purchase intention and willingness to pay directly. Their research showed that
transparency of price allocation can increase the perceived value and the buying intention of
products by triggering a social component. When consumers perceive that a fair amount of the
purchase price is allocated to a certain agent, they are willing to pay a premium price for the
product. Buell and Norton (2011) conducted a research with operational transparency. They
tested whether providing information of the online searching process had an effect by using
five different experiments within two different service industries. The results showed that
operational transparency can significantly increase consumers’ perceived value of the service.
11 can in fact favor websites with longer waiting time. The authors ascribe this effect to feelings
of reciprocity evoked by the operational transparency. Furthermore, Eisend (2006) found in
his study on two-sided advertisement that providing negative information, when provided the
right way, can also increase the purchase intention and brand attitude. This study shows that
transparency can have a positive effect even when the information is not perceived neutral or
positive. The authors describe the attribution theory as a possible explanation for this effect.
By providing negative information in the advertisement, consumers find the source more
credible. Due to the increase in trust in the firm, consumers attribute the positive arguments
made in the advertisement to be the actual product characteristics and not just arguments
created for the sole purpose of selling the product. In line with these findings, Demmers, Erbé,
Van Strijp and Wientjes (2015) found that using transparency as a marketing tool by the firm
can lead to higher purchase intention and willingness to pay. They accredit this effect not only
to the increase in brand trust, but also to the positive change in the consumers’ perception of
the disclosed information.
The positive effect of transparency on consumer behavior appears to have a positive
effect on the firm’s performance. Margolis, Elfenbein and Walsh (2007) conducted a meta-
analysis on 251 studies examining the relationship of corporate social performance and
corporate financial performance. Within this meta analysis, seventeen studies examined the
effect of information disclosure by a firm and corporate financial performance. Taken
together, the results suggested a positive relationship between transparency and firm
performance. Cohne and Wolfe (2013) argue, that for companies to be sustainable, they
should opt for full disclosure. Due to the more critical and less trusting consumer, they argue
that transparency is crucial to create trusting consumers. In their study among consumers in
UK, USA and China, the authors found that the three most important aspects for purchase
12 honesty of a company was even more important than the price of the product. The authors
conclude that while transparency is becoming more important, brands are becoming less
important in the decision making process, which leads us to the next subchapter.
Brand equity
The impact of a brand on consumer choice can be measured through brand equity. Brand
equity can be described as the subjective assessment of a brand on top of the consumers’
objective evaluation of the product (Lemon, Rust, & Zeithaml, 2001). It can be measured by
the consumers’ utility for a branded product in comparison with a similar non branded or
fictional product. This subjective assessment is determined by the consumers brand
knowledge. Brand knowledge can be divided into brand awareness and brand image. Brand
awareness is the consumer’s ability to identify a brand under different conditions. Brand
image is the brand perception that a consumer holds in memory and can be described as a set
of brand linked associations. The combined associations that form the consumers’ image of a
brand differ among consumers and brands in type, favorability, strength, uniqueness and
amount (Keller, 1993).
Up to now, brands have been an important factor in the decision making process when
buying a product (Fischer, Völckner, & Sattler, 2010; Macdonald & Sharp, 2000).
Cobb-Walgren, Ruble and Donthu (1995) established in their research that higher brand equity leads
to a significantly greater consumer preference and purchase intention. This was evident for
both the service and product category. This positive effect of brand equity can lead to better
firm performance (Kim, Kim, & An, 2003). Together with product equity, brand equity is
found to be the best predictor of a company’s future sales (Vogel, Evanschitzky, &
Ramaseshan, 2008).
A reason for this effect to occur is that brands help consumers make purchase decisions
13 shopping behaviour, Degeratu, Rangaswamy and Wu (2000) found that brands become more
important when less information about the different attributes of a product is provided. In the
offline shopping environment, more risk is experienced due to unavailability of information.
Fischer, Völckner and Sattler (2010) established that from the two major factors determining
the need for a brand, social demonstrance and risk reduction, the latter is found to be the main
factor. Most consumers have little information about the attributes of a product and use the
brand as a source of information. Brands create trust in the quality of a product (Fischer et al.,
2010). Indeed it appears that the relationship between a brand and the purchase intention is
mediated by consumer trust and perceived risk. Furthermore, the factors trust and perceived
risk are interdependent. This leads to the conclusion that trust can be created by a brand,
which can reduce the perceived risk in a purchase decision, leading to an increase in purchase
intention (Chang & Chen, 2008).
Transparency and its effect on the need for brands
Cohne and Wolfe (2013) argue in their article that in the decision making process brands are
becoming less important due to a more transparent world. In their survey they found a decline
of brand importance in the purchase decision from 43% to 27% in one year time. This decline
is paired with an increase from 53% to 66% of consumers taking transparency in account in
their purchase decision. Moreover, while the importance of transparency is increasing and the
importance of brands is declining, an increase in the importance of quality and price can be
noticed. The importance of quality among UK consumers increased from 86% to 89%. In
addition, the importance of price increased from 83% to 84%. It could therefore be argued
that due to transparency, brands play a smaller part in the decision of a consumer, and this
leads to a more objective assessment of the product with higher emphasis on the price and
quality of the product. However, these changes are small and no statistical data on
14 Although trends can be examined by this longitudinal survey, no causal relationship can be
concluded (Lewis & Saunders, 2012).
So far little empirical proof has been provided that transparency can make brands less
important in the purchase decision, despite the fact that theoretical support can be found in
support of this vision. Transparency appears to fulfil a similar role in the consumer purchase
decision as brands do, namely the creation of trust in a product. Indeed the positive effect of
transparency on purchase intention appears to be at least partly mediated by consumer trust in
a brand. First of all, scholars argue that transparency leads to an increase in trust (Bhaduri &
Ha-Brookshire, 2011; Christensen, 2002; Kanagaretnam, Mestelman, Nainar, & Shehata,
2010). Kanagaretnam et al. (2010) found that transparency significantly increased trust in one
shot interactions in a game theory setting. In this experimental investment game, the
difference in trusting behavior was measured between a group with complete information
versus incomplete information. In their article they conclude that a consumer’s lack of trust in
a company could be overcome by more information about the way the organizations conduct
their business. Bhaduri and Ha-Brookshire (2011) have researched if transparency pays off in
the clothing industry. They indeed argue that to build a trustworthy relationship with a
consumer, it is necessary to provide information about the manufactory process. Secondly,
trust can lead to an increase in willingness to pay and better brand performance
(Delgado-Ballester & Munuera-Aleman, 2001). Bhaduri and Ha-Brookshire (2011) argue that consumer
trust positively influences the outcome of purchase intention. Brady (2003) states that a
sustainable brand can be created by relationships based on trust, which can be accomplished
by being transparent.
Christensen (2002) reasons that the need for transparency is not caused by extra
interest in a brand or information, but by the role of risk reduction. A small group of brand
15 means that the information provided by transparency is not a goal, but merely a mean to reach
the goal of risk reduction (Morgan, 2009). Therefore it seems that transparency is fulfilling
the same role as brands to some extent, that is to say the role of risk reduction. Therefore, the
following is hypothesised;
H 1: The utility of a known brand name is lower under high (versus low) corporate
transparency conditions.
As discussed above it is clear that brands and transparency appear to be fulfilling a similar
role of risk reduction by building trust in a brand. Following hypothesis 1, it can be expected
that less known or unknown brands benefit more from transparency than well known brands.
Unknown brands have not build up as much trust as known brands and it is likely that a
higher need for risk reduction is present for products from an unknown brand. It is thus likely
that transparency has a bigger impact on trust concerning a product of an unknown brand than
on a product of a known brand. This leads to the following hypothesis:
H2: Consumer trust mediates the relationship between transparency and brand utility, such
that trust for the product of the unknown brand rises more than trust for the product of the
known brand.
In line with Christensen’s (2002) argument that it is the need for risk reduction, not the need
for information, which drives the consumers call for transparency, the source that includes
this information should not be forgotten. In their study on information disclosure, Demmers et
al. (2014) found that the positive effect of transparency is mediated by the source of the
16 of willingness to pay and product preference. When the information is provided by someone
other than the company, no significant relationship is present. This effect appears to be at least
partly caused by the perception of the disclosed information. When the information is
disclosed by the brand, consumers find the information to be more relevant and positive.
It can therefore be expected that when information is disclosed by the brand, rather than
another source, it leads to a greater increase in trust. Due to the higher need for transparency
of a consumer, disclosure of information by a brand is likely to be seen as a positive action.
According to the reciprocity theory, consumers reward this action with a positive response.
Trust can be used by the consumers to reward the brand. However, when the information is
not disclosed by the firm, but by a different source, feelings of reciprocity are not likely to be
evoked (Falk & Fischbacher, 2006). Thus, although it is expected that the disclosed
information alone can lead to an increase in trust in the product, when this information is
disclosed by the firm itself, an even greater increase in trust is expected. Hence:
H3: The effect of transparency on trust is moderated by disclosure source, such that it leads to
a higher increase in trust when the information is disclosed by the brand, in comparison with
information disclosed by a different source.
The importance of a brand in a consumer’s decision depends on the product type. Longer
lasting products, known as durable goods, are perceived as a more risky decision than Fast
Moving Consumer Goods (FMCG). Examples of FMCG are food and hygiene products.
Examples of durable goods are cars, computers, and mobile phones. As mentioned above,
brands fulfil a risk reduction and a social demonstrance role. The risk reduction and social
demonstrance function is found to be more important for durables goods than for FMCG
17 needed. Moreover, these durable goods are often used in a more social or public environment.
Consumers use brands as an expression of the self and to communicate this to others (Escalas
& Bettman, 2005). Although transparency is expected to have the capacity to take over some
of the risk reduction function, this is not expected to be the case for the social demonstrance
function. This leads to the possibility that, because brands play a more diverse role for the
durable goods, transparency will have less effect on brand utility for these goods.
On the other hand, as explained above, risk reduction was found to be less important
for the FMCG (Fischer, 2010). It can therefore be argued that there is less need for
transparency in a FMCG consumer decision. Moreover, it is likely that consumers are less
motivated to fully consider the transparent message for a FMCG in comparison with a
transparent message for a durable good. According to the Elaboration likelihood model by
Petty and Cacioppo (1983), consumers respond differently to messages depending on their
involvement. The involvement of the consumers depends on the ability and motivation to read
the communication. Highly involved consumers are able and motivated to think about the
communication provided. Highly involved consumers follow the central route of persuasion
in which individuals carefully consider the elements of a message in order to determine
whether it makes sense and will benefit them in some way. Low involved consumers on the
other hand, are not able and/or motivated to think about the message, and follow the so called
peripheral route. With the peripheral route, consumers use a simple decision rule to evaluate
the message. Consumers are likely to be less motivated to think about a FMCG decision
because it is a less expensive product, and is therefore accompanied with less risk. This could
lead to a smaller or no effect of transparency on brand trust, since consumers are less
motivated to evaluate the message. In this case they might use a simple decision rule, such as ‘I will pick the brand I know’.
18 H1, H2
Because the consumers evaluate the message to a lesser extent, the counterarguments
made above are expected to be less influential in affecting the consumer behaviour, as the
ELM theory. This leads to the following hypothesis:
H 4: The effect of transparency on trust is moderated by product type, such that it leads to an
increase in trust when it entails a consumer decision for a durable good, but not when it
entails a consumer decision for a FMCG.
An illustration of the hypothesized relationship between transparency and utility is presented
in the conceptual model in figure 2.
Figure 2 Conceptual model
H1: The utility of a known brand name is lower under high (versus low) corporate transparency conditions.
H2: Consumer trust mediates the relationship between transparency and brand utility, such that trust for the product of the unknown brand rises more than trust for the product of the known brand.
H3: The effect of transparency on trust is moderated by disclosure source, such that it leads to a higher increase in trust when the information is disclosed by the brand, in comparison with information disclosed by a different source.
H 4: The effect of transparency on trust is moderated by product type, such that it leads to an increase in trust when it entails a consumer decision for a durable good, but not when it entails a consumer decision for a FMCG.
Transparency Trust consumer Brand utility
Product type
Source disclosure H4
19
Methodology
Design
An online survey-based experiment with a between-within subject 3 (non transparent
information/consumer information/transparent information) x 2 (durable/FMCG) design is
used to test the hypothesized relationships. The experimental design provides a way to
examine causal relationships between the variables (Lewis & Saunders, 2012).
The remaining part of this chapter is dedicated to a further discussion of the
methodology. First an overview of the sample is provided. Afterwards, all the variables used
for examination of the hypothesized relationships, including some control variables, are
discussed. Finally, an explanation of the statistical approach is presented.
Sample and procedure
The data is collected through an online survey with the use of the Sawtooth Discover
software. Non-probability sampling techniques, including self selection sampling and
snowball sampling, have been used to reach respondents. Social media has been the main
communication tool in the search for these participants. The survey was written in Dutch and
thus only distributed among a Dutch speaking sample. A total of 207 respondents have
participated and 199 of those respondents finished the survey completely. Respondents were
randomly divided into three groups, a non transparent-, a consumer information- and a
transparent group. The average age of the non transparent group was 31 and 58% were male.
The consumer information group had an average age of 32 and 62% of them were male.
Lastly, in the transparent group the average age was 32 and 49% were male. All respondents
together were 32 years of age on average and slightly more than half was male (57%). Each
respondent had to answer 18 questions in total. After answering a set of demographic
20 a Choice based Conjoint (CBC) Analysis. Within each set of questions respondents had to
choose between products based on brand (branded/fictional branded), price and quality.
Finally, they had to answer 4 questions regarding trust. This research is conducted according
to the code of ethics of the University of Amsterdam. No names or personal data were
collected or used in the report.
Stimuli
Transparency. To measure transparency, respondents were randomly divided in three
groups, a non transparent-, a consumer information-, and a transparent group. The non
transparent group was the control group and this group received limited information about the
different brands used in the survey. The transparent group and the consumer information
group received more information about the brand and the product, before answering each set
of questions related to a product type. The respondents in the transparent information
condition and the consumer information condition received similar information. An example
of the information provided is; ‘According to the websites of both companies, no child labour
was used and all employees received a fair wage’. In the transparent information group it was
clearly stated that this information was provided by the brand, however this was not the case
in the consumer information condition. For the consumer information group it was clearly
stated the information came from a consumer website. An example of the information
provided is; ‘Research from the consumer website showed that no child labour was used and all employees received a fair wage’. The information given in each group was about both the known and the fictional brand, in line with the goal of this research, which measures the effect
of the ‘transparent age’ on brand utility.
As explained, before answering each set of questions for the different product types, the
participants where provided with information. To create a more realistic situation, the
21 A pre-test was done to assure the information for the two different product types where
perceived similar. A within subject design was used, with a sample of 20 respondents, of
which gender was equally divided. The two different messages were tested on perceived
relevance, perceived positivity and perceived transparency of the brand. Relevance was tested
because both Christensen (2002) and Vishwanath and Kaufmann (2001) describe it as an
important element of transparency. Relevance was measured using four items adapted from
the five item scale by Mishra, Umesh and Stem (1993). One question was not included,
because this question had the same meaning as one other item when translated into Dutch.
The four remaining items measured how relevant, meaningful, important and useful the
information was perceived on a likert-scale ranging from 1 (not at all) to 7 (very much).
Cronbach’s alpha of this scale for the durable and FMCG message was α = .84 and α = .92.
As for positivity and transparency, two additional questions were asked, namely; ‘how do you perceive the information’ and ‘how transparent do you perceive the brand’. Transparency was measured on the same scale used for relevance. Positivity was measured on a
likert-scale ranging from 1 (very negative) to 7 (very positive). A paired sample t-test showed no
significant difference between both messages. All the product and brand information
messages are displayed in the appendix.
Product type. This study used two different product categories, a laptop for the durable
goods category and a toothbrush for the FMCG category. Respondents were asked to choose
between products differing on brand, quality and price within each category. A laptop is used
because it is expected that most if not all participants have bought or used a laptop before.
Moreover, this category provided an easy way to group the product in three different types of
quality (low, medium and high), which was needed for the CBC analysis. The different types
22 biggest electronic store in the Netherlands, Mediamarkt. In table 1 the different quality and
price levels can be found.
Similar to the laptop, the toothbrush is picked by the author because it is assumed that
all participants use this product. Moreover, this product group can also easily be divided
between three levels of quality. Most importantly, these different levels of quality are also
expected to be easily recognized by the participants. Prices are established by an online
examination of the prices of different toothbrushes sold in the biggest supermarket of The
Netherlands, Albert Heijn. In table 2 the different quality and price levels for the toothbrush
can be found.
Three levels of quality and pricing are used as it creates a more realistic setting for the
consumer in comparison with two levels. It also provides more information about the
relationship between brand equity and product choice. For example, a consumer might choose
the laptop from the well-known brand, even if the quality is medium in comparison with a
high level quality unknown branded laptop in a situation where the price level is the same.
However, this same consumer might choose the unknown branded laptop if the quality level is
high versus a branded laptop with a low quality level. With only two quality levels, this pattern doesn’t show up in the conjoint analysis. The same phenomenon could exist for the price levels. Several scholars have used a similar way of structuring price and quality levels
while conducting a CBC (Cobb-Walgren et al., 1995; Carter & Curry, 2010).
Furthermore, for each product type there were two brands. For the product category
laptops the brand HP is picked as the existing brand and ERA as the fictitious brand. For the
product category toothbrushes Aquafresh is picked as the established brand and Dentala is the
fabricated brand. A panel of 5 students found the fictitious brand names convincing. Only two
brands where used in each product category because the main focus of this study is on
23
Table 1
Quality and price levels for the laptop
Level Low Medium High
Quality 1.5 GHZ processor speed, 250 GB memory, 4 hour battery 2.0 GHZ processor speed, 500 GB memory, 6 hour battery 2.5 GHZ processor speed, 1000 GB memory, 8 hour battery Price 599,- 699,- 799,- Table 2
Quality and price levels for the toothbrush
Level Low Medium High
Quality Normal Flex Flex and control
Price 0,90 1,60 2,30
Measures
Brand utility. A Choice Based Conjoint Analysis (CBC) is used for the measurement
of the brand utility. The CBC is a trade-off analysis that lets consumers evaluate different
product attributes (Green, Krieger & Wind, 2001). To conduct a CBC analysis, consumers are
asked to choose a product that they prefer based on different attributes. By simulating this
question multiple times using different combinations of the price and quality levels,
estimation can me made about the utility of the different attributes.
Although there are different types of conjoint analysis developed, a carefully
consideration has led to the use of a CBC analysis, over other conjoint analyses. Firstly, CBC
24 likely to get a higher response rate (Lewis & Saunders, 2012). Secondly, a choice based
conjoint analysis reflects a more realistic analysis of the consumer’s choice compared to the
standard rating-based conjoint analyses. Consumer’s product preference and choice is often
based on a comparison between products with a set of attributes, instead of a rational
evaluation of the individual attributes of a product (Huber, Wittink, Johnson, & Miller, 1992).
Finally, for the measurement of brand utility a CBC analysis is found to be a reliable tool
(Aaker, 1996; Keller, 1993). It is seen as the greatest tool for marketers to discover how
consumers make tradeoffs between different brands and attributes (Green, Krieger, & Wind,
2001). Researchers have used the conjoint analysis in their research to measure brand equity
(Cobb-Walgren et al., 1995; Rangaswamy, Burke, & Oliva, 1993) and to measure the effect of
transparency (Carter & Curry, 2010). The formula from Johnson and Orme (1996) was used
to calculate the minimal sample size needed for a CBC analysis. According to this formula
each CBC should consists of at least 63 participants1. With 65 participants in the smallest
group, this rule of thumb is satisfied.
There are three attributes for each conjoint analysis; brand, quality and price. For the
brand attribute there are two different options, an existing brand and a fictional brand. Keller
(1993) explicitly states that a non branded or fictional brand can be used to measure brand
equity. A fictitious brand is used in this research because of the more realistic scenario it
creates for a consumer. Choosing between a well known branded product and a non branded
product is not a scenario a consumer often encounters. Furthermore, participants are told that
this study researches consumer choice between different attributes such as price and quality.
1
n∙t ∙a / c ≥ 500, with n = the number of respondents, t = the number of questions (8), a = the number of options per questions (3), and c = the number of analysis cells (3).
25 If they need to choose between a non branded and branded product, there is a higher
possibility they get suspicious.
For quality there are three different options: low, medium and high quality. There are
another three possible options for price: low, medium and high price. Thus, in total there are
eighteen different options (2 x 3 x 3) per product type. Due to the design of the conjoint
analysis, respondents don’t have to compare all of these options separately. Although the
respondent is not asked to make a choice between every single option, conjoint analysis can
predict a consumer’s choice with a high accuracy (Orme, 1998). The recommended setting
suggested by the Sawtooth software is used, with 3 options per questions and six question per
CBC analysis. CBC questions for examination of the utility of the toothbrush always followed
the CBC questions for examination of the utility of the laptop. However, Johnson and Orme
(1996) found no loss of reliability for the first 20 tasks within a CBC in their research on 20
CBC studies.
With the CBC software attribute utilities are calculated on an individual level using a
Maximum Likelihood Estimation (MLE). MLE is a well known and widely used method for
estimation (Sawtooth, 2014). MLE estimates the utility of a population by using the mean and
variance of the samples as parameters and by comparing them to particular parametric values.
The parametric value offering the best fit with the data is used for the calculation of the
individual utilities (Islam, Towhidul, Jordan Louviere, and David Pihlens, 2009). The maximum
likelihood estimation for the utility of the quality and price attribute is subjected to
monotonicity constraints to increase the robustness of the analysis. These constraints are
based on expected preferences. For the price attribute for example, higher preference for the
less expensive levels is expected. When utilities violate the preference order, the utilities are
corrected. No monotonicity constraint is used for the attribute brand, because determining a
26 unordered, the Sawtooth discover software uses a preference rating question for this attribute
before starting with the product choice questions. However, this option was bypassed because
it was unlikely that respondents can determine the preference of the fictional brand, as they
have never encountered this brand and there was no neutral option available in this rating
question. In addition, Empirical Bayes is used to smooth the utilities towards the population
parameter. Using the Maximum likelihood model on a individual level, while correcting the
utilities somewhat towards the population parameter, is found to be a reliable method for
utility estimation (Sawtooth, 2014). In conclusion, respondents have to make six choices,
between three product options, for each product category. Sawtooth software is used to
calculate individual utilities for the attributes. The higher the utility, the higher the preferences
for this attribute. For the brand utility with only two possible options, a positive utility for the
known brand will automatically lead to a negative utility for the unknown brand, and vice
versa.
Trust. Brand trust was measured for each brand at the end of the survey. The items
read, ‘The feeling of trust in brand X is’. A 5-point likert-scale was used (1= very low to 5=
very high). This item was based on the control item Delgado and Munuera (2001) used to
assess their scale to measure trust. This full scale, composed of 6 questions, was not used, as
this would add 24 additional questions to the survey (4 brand x 6 questions). This would
double the length of the survey and create the possibility of a smaller response rate. A high
correlation with significance of p < .01 between the scale and the control item was found by
the authors.
Control variable. Participants were asked about their gender (1 = male, 2 = female)
and age in the beginning of the survey. The selection of these control variables was based on
27 For instance, risk reduction is more important for women than for man (Fischer, Völckner, &
Sattler, 2010).
Statistical procedure
Data was collected online using the Sawtooth discover software. For the statistical analyses,
all individual utility scores were collected and copied into SPSS. Dummy variables were
created for the transparency variable. Descriptive statistics, skewness, kurtosis and normality
tests were computed for all variables. The variables brand trust ERA and HP utility were not
normally distributed. Positive kurtosis was found for brand trust ERA in each experimental
group. This can be explained by the 5-point likert-scale which was used, with a neutral option
in the middle. Positive kurtosis was also found for HP utility and appeared to be caused by
outliers. The outliers were examined to ensure no data entry or instrument errors were made.
A normality test of the variable with exclusion of outliers showed a normal distribution.
A One-way ANOVA was used to establish the main effect of transparency on brand
equity. In order to test the moderating role of brand trust, an SPSS macro of Hayes (2012) was
used. Due to violation of the normality assumption bootstrapping was applied. The macro
computed confidence intervals for the indirect effect of transparency on HP utility. Hayes
(2012) recommendation to resample 5000 times, instead of the default of 1000 times, was
28
Results
In table 3 and 4 the mean of the utilities are displayed for the different price and quality levels
of the laptop. In table 5 and 6 the same data is provided for the toothbrush. Paired sample
t-tests were used to test if the different levels used for price and quality were perceived as
expected. The test showed significant difference between the low and medium and between
the medium and high quality level, for both the laptop and the toothbrush (p < .01). This
provides sufficient evidence that consumers indeed preferred the medium level of quality over
the low quality and preferred the high quality over the medium quality product. Moreover,
similar results were found for the different price levels. For both the laptop and the
toothbrush, respondents preferred the low price over the medium price and preferred the
medium price over the high price (p < .01).
A difference was found in the importance of quality and price between the two
different products. Paired sample t-tests were applied to compare the quality utility and the
price utility between the different products. A significant difference was found between the
quality utility of the laptop and the quality utility of the toothbrush, such that the utility for the
high quality laptop was significantly higher than utility for the high quality toothbrush (p <
.01). In addition, the negative utility for the low quality laptop was significantly lower than
the utility for the low quality toothbrush. Furthermore, a significantly higher utility was found
for the low toothbrush price compared with the low laptop price (p < .01). This leads to the
conclusion that quality was a more important aspect of the product choice in the laptop
decision and price was a more important aspect for the toothbrush decision.
Utility of the laptop brand HP and the toothbrush brand Aquafresh are presented in
table 7. Utility of the unknown brand ERA and Dentala are not included in this table, because
29 between the brand utility for the two products, that is to say, Aquafresh benefitted more brand
utility than HP (p < .01).
The mean, standard deviation and correlations of the variables used for the
examination of the hypotheses are provided in table 9 and 10. Spearman’s correlation
coefficient is applied due to the violations of normality for several variables. Spearman’s
correlation coefficient is found to be a useful correlation efficient to minimize the effect of
outliers (Field, 2013). Transparency was measured as a dummy variable with (a) transparency
(0 = non transparent group, 1 = transparent group) and (b) consumer information (0 = non
transparent group, 1 = consumer information group). Table 9 shows that transparency is
negatively related to the HP utility (rs = -.19, p < .01) and positively related to brand trust in
ERA (rs = .12, p < .10). HP utility is also positively related to brand trust in HP (rs = -.39 .17, p
< .01). However, the correlations between the discussed variables are relatively small. No
correlation between transparency and the dependent variable Aquafresh utility, or the
hypothesized mediator trust was found. Corresponding with these results, the One-way
ANOVA showed no significant relationships. No further tests were conducted for the
hypothesized relationship between transparency and Aquafresh utility, due to the lack of
significant correlations between the variables.
To test the relationship between transparency and HP utility, a One-way ANOVA was
used. Due to a violation of the assumption of homogeneity of variance (p < .04), Welch’s F
was used for robustness. There was a statistically significant effect of transparency on brand
utility, Welch’s F (2, 131) = 4.45, p < .05. Tukey post-hoc tests revealed that the brand equity
was significantly lower in the transparent group compared to the non-transparent group (p <
.01). There was no statistically significant difference between the consumer information group
and the non transparent group and the transparent group (p > .14). Therefore, hypothesis 1 is
30 such that the utility for a branded product decreases. No relationship was found when the
consumer decision entailed a FMCG (p > .10).
One-way ANOVAs were used to examine the effect of transparency on the utility of
the low and high price level. The same was done for the low and high quality level. The
outcome demonstrates that a more transparent environment leads to a greater emphasis on the
price and quality of the laptop. A significant effect of transparency on the preference for the
low laptop quality was found (p < .01). Tukey’s post-hoc test exposed a significantly lower
preference for the low quality level within the transparent group compared to the
non-transparent group (p < .01) and the consumer information group (p < .01). No significant
difference between the non transparent and the consumer information group emerged (p >
.10). A violation of the assumption of homogeneity of variance led to the use of Welch’s F to
examine the effect of transparency on price utility. A significant effect of transparency was
found on the low laptop price level, Welch’s F (2, 133) = 3.29, p < .05. Further examination
showed a marginally significant higher utility for the low price by the transparent group
compared to the non transparent and the consumer information group (p < .10). Again, no
significant effect was found between the non transparent and the consumer information group.
A marginally significant effect became evident for the high laptop price level, Welch’s F (2,
133) = 2.74, p < .10. Tukey’s post-hoc test revealed marginally significant lower utility for
the high price level in the transparent group in comparison with the non transparent group (p
< .10). No significant effect was found between the other groups. Furthermore, no effect was
31
Table 3
Mean utility for laptop quality
N Low Medium High
Not transparent 69 -101.82 6.48 95.33
Consumer info 71 -99.28 -1.86 101.14
Transparent 67 -86.49 -7.34 93.83
Total 207 -95.99 -0.85 96.84
Table 4
Mean utility for laptop price
N Low Medium High
Not transparent 69 26.66 8.59 -35.25
Consumer info 71 26.95 6.10 -33.05
Transparent 67 34.67 6.77 -41.45
Total 207 29.35 7.15 -36.50
Table 5
Mean utility for toothbrush quality
N Low Medium High
Not transparent 65 -55.77 3.35 52.42
Consumer info 68 -53.09 -0.57 53.65
Transparent 66 -49.43 -9.38 58.81
Total 199 -52.75 -2.21 54.96
Table 6
Mean utility for toothbrush price
N Low Medium High
Not transparent 65 62.18 7.95 -70.13
Consumer info 68 62.48 14.09 -76.57
Transparent 66 66.61 6.80 -73.42
32
Table 7
Descriptive statistics utility HP and Aquafresh
Utility HP Utility Aquafresh
Transparency N M SD M SD Non transparent 69 18.80 18.69 22.64 31.69 Consumer info 71 13.91 21.22 24.45 24.68 Transparent 67 6.20 29.56 18.69 28.60 Total 207 13.05 23.98 21.95 28.37 Table 8
Mean, standard deviation and correlations of laptop variables
M SD 1 2 3 4 5 6 1. Transparency 2. Consumer info 3. HP utility 13.05 23.98 -.19** .01 4. Trust HP 3.66 0.64 .10 -.15* .21** 5. Trust ERA 2.93 0.51 -.12 .03 -.39** .04 6.Gender 1.43 0.5 .10 -.08 .17* .21** .04 7. Age 31.76 11.74 .00 .02 -.04 .08 .05 -.07
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Table 9
Mean, standard deviation and correlations of toothbrush variables
M SD 1 2 3 4 5 6 1. Transparency 2. Consumer info 3. Aquafresh Utility 21.95 23.98 -.10 .09 4. Trust Aquafresh 3.66 0.64 .04 -.07 .24** 5. Trust Dentala 2.9 0.51 .09 .01 -.28** -.04 6.Gender 1.43 0.5 .10 -.08 .17* .13 .02 7. Age 31.76 11.74 .00 .02 .03 -.07 .01 -.07
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
33 To test if brand trust mediates the effect of transparency on brand utility, the SPSS
macro of Hayes (2012) was used. A dummy variable was used, leaving the consumer
information group out of the analysis. This decision was based on the previously discussed
results from the One-way ANOVA. A first examination of the dependent variables trust HP
and trust ERA as moderators, showed no significant relationship between transparency and
trust HP (p > .10). Consequently, this variable is left out in the following statistical analysis.
Results from the regression analysis, shown in figure 3, indicate that trust ERA partly
mediates the effect of transparency on HP utility. Transparency has a marginally significant
effect on trust ERA (B = 0.14, p = .06). Also, trust ERA significantly influences HP utility (B
= -15.04, p < .01). In table 10 the statistics of the indirect effect is presented. Bootstrapping is
used to correct for the violation of the normality assumption. The indirect effect is negative
and statistically different from zero (B = -2.15, BCa95 = [-4.91, -0.05]), thus providing support
for mediation by trust ERA. Moreover, figure 3 shows a negative and significant direct effect
of transparency on HP utility (B = -8.43, p = .01). Examination of the bias corrected
confidence interval in table 11 confirms this significant relationship (BC95 = [-15.05, -1.81]).
Consequently, hypothesis 2, which proposed mediation by trust on the effect of transparency
on utility, is supported. It should however be noted that there was also a significant direct
interaction between transparency and HP utility, leading to the conclusion that trust in the
fictional brand only partly mediates this interaction.
Hypothesis 3 and 4 proposed moderating effects of source disclosure and product type.
The interaction effect is visually presented in figure 4. Transparency only had a significant
effect on utility when the information was disclosed by the brand (p < .01). No such effect
was found when the information was disclosed by the consumer website. However, there was
also no significant difference between the consumer information condition and the transparent
34 Furthermore, transparency only had an effect on brand utility for the durable good.
There was no effect found for the FMCG decision. Hence, hypothesis 4 was supported.
Figure 3
Path model
Table 10
Statistics of indirect effect of transparency on HP utility
BCa 95% CI
B SE Lower Upper
Indirect effect of transparency on HP utility -2.15 1.19 -4.81 -0.05
Note: N=199. BCa: bias corrected and accelerated; 5000 bootstrap resample. Note: N =199. R² Trust ERA = .02. R² Brand utility = .15. Coefficients are presented. *p < .10; **p < .05; ***p < .01. HP utility Transparency Trust ERA -15.04 *** .14* -8,43**
35
Table 11
Statistics of direct and total effect of transparency on HP utility
BC 95 % CI
B SE P Lower Upper
Total effect of transparency on HP utility -10.58 3.51 .00 -17.52 -3.65
Direct effect of transparency on HP utility -8.43 3.36 .01 -15.05 -1.81
Note: N=199. BC: bias corrected.
Figure 4
Moderation effect by product type and information source
0 5 10 15 20 25 30
Non transparent Consumer information Transprarent
U ti li ty Laptop Toothbrush
36
Discussion
This final chapter elaborates on the findings of this study. A general discussion on the results
is provided. Next, the theoretical and practical implications of these findings are discussed.
Finally, limitations of the study and suggestions for future research are presented.
General discussion
The purpose of this research was to contribute to the transparency literature, by examining
how consumer behavior changes in a more transparent shopping environment. This study
contributes to this literature through three main findings.
First, the results show that brands are losing their importance in consumer decision on
certain products. The brand is significantly less important in the consumer decision between
different laptops. The utility for a known brand in a transparent situation is only a third of the
utility of the same brand in a non transparent situation. The findings correspond with Cohne
and Wolfe’s (2013) arguments that brands are becoming less important in the age of
transparency. Moreover, results show that consumers put more emphasis on price and quality
in a transparent environment.
Second, this decrease of importance of brands depends on product type and source
disclosure. For the consumer decision concerning FMCGs, transparency has no effect on
brand importance. A possible explanation for this phenomenon is found in the Elaboration
likelihood model by Petty and Cacioppo (1983). Consumers might be less motivated to read the
information provided in the case of a FMCG decision. When consumers are less motivated to read the message, it is highly likely the transparency has less effect on the decision. In case of a low motivated consumer, brand name can be a key aspect of the decision. Indeed, a significant difference was found between the brand utility for the two products, with a higher brand utility for Aquafresh
37 than for HP. Moreover, it was found that the brand only becomes less important in the
consumer decision when the transparent information is provided by the brand itself. No
significant difference was found between the non transparent condition and the consumer
information condition. There was also no significant difference between the consumer
information condition and the transparent condition. This leads to the conclusion that although
the information alone might not be enough to cause a significant difference, the extra
information in combination with a brand being transparent is sufficient. This corresponds with
previous findings on source disclosure (Demmers et al., 2014) and the reciprocity theory (Falk
& Fischbacher, 2006). Disclosure of relevant information by the brand can lead to an extra
increase of trust and a more positive evaluation of the information.
Finally, the results of this research show support for a partial mediation by trust of the
unknown brand on the interaction between transparency and brand utility. Consumers in a
transparent environment have more trust in the products of an unknown brand, then they do in
a non transparent environment. At the same time, transparency doesn’t lead to significantly
higher trust in the product of the known brand. Since there is higher trust in the product of an
unknown brand, but not for the known brand, established utility for the known brand is lost.
At the same time, in a transparent situation price and quality are more important. Therefore, it
can be argued that the increase of trust in the ‘unknown’ leads to a decrease in the importance
of a brand.
Theoretical and Practical implementations
The findings of this study have a number of theoretical implications. This study is one of the
first in researching the overall effect of the transparent age. So far most research has focussed
on comparing a non transparent brand with a transparent brand. However, the question remaining was, ‘what happens when all brands are transparent?’To examine this question, this research used an innovative research approach to compare a full transparent situation with
38 a non transparent situation. Multiple CBC analyses were used. CBC is a widely used tool in
marketing to measure the utility of attributes. However, this research applies and compares
the outcome of the CBC analyses in several conditions. The results, a significant change in
brand utility and an increased importance of the price and quality attribute, build on the
statement of Cohne and Wolfe (2013) that brands are becoming less important in the
transparent age.
Moreover, this relationship was found for a consumer decision entailing a durable
good, but not for a FMCG. This study is the first to examine transparency for different
product categories and one of the first studies that combines the theory on transparency with
the Elaboration Likelihood Model (Petty & Cacioppo, 1983).
In addition, this study builds on the findings of Demmers et al. (2014) on source
disclosure. Demmers et al. (2014) found that when information is provided by a company,
transparency leads to an increase of willingness to pay and product preference. When the
information is not provided by the firm however, no significant relationship is present. As
expected, this study shows a similar effect of information disclosure on brand utility. When
the transparent information is disclosed by the brand, brand utility decreases. On the contrary,
when the information is disclosed by a consumer website, no interaction is established.
Game theory setting demonstrated that transparency can increase participants trust
significantly (Kanagaretnam et al., 2010). Moreover, different scholars have named trust as a
potential mediator between transparency and consumer behavior (Brady, 2003; Christensen,
2002). However, to the best of this author’s knowledge, this study is the first to empirically
examine trust as a mediator. The transparent information increased consumer trust in the
fictional brand, but not in the known brand.
The findings have practical implications for brand managers. Up until this point,
39 transparent products. With the expectation that transparency is only increasing, the results
proof to some extent that brands are becoming less important for some consumer decisions.
Although branding still appears highly important for the FMCG, quality and price are
becoming more and more important for the durable goods category. Transparency might
provide opportunities for new brands, as long as the quality and price can compete with the
established brands.
Limitations and future research
Like most other empirical studies, this study has some limitations. Fortunately, these
limitations bring about new opportunities for future research in the young field of
transparency.
Firstly, this research uses the Sawtooth software for the conduction of the CBC
analysis. The Sawtooth software calculates the utility, which limits the ability to examine the
validity of the scale. Furthermore, utilities are calculated based on a consumer’s choice
between two different brands. Consumer decisions are likely to differ from this experimental
setting, because in daily life consumers are exposed to a greater variety of products.
Secondly, for the examination of the moderating effect of brand type, only one product
was used for each category. This affects the generalizability of the results. Products within the
two categories widely differ on factors such as price, social demonstrance and risk reduction
(Fischer et al., 2010). Future research could focus on the interaction effect between
transparency and brand utility for different products belonging to the same product category.
Moreover, examination of the reason why transparency has more effect on brand utility for
the laptop in comparison with the toothbrush was beyond the scope of this study. Possible
reasons why transparency has more effect on certain products included a lack of motivation to
elaborate on the message and the lower need for risk reduction for a FMCG. Future research
40 Finally, the amount of transparent information was limited and positive. Transparency is
the disclosure of relevant information, which is not restricted to positive or neutral
information. Questions remain on what will happen with brand equity, when negative
information is provided. Additionally, a limited amount of information was used in
consideration of the length of the survey. It could be argued that although the information
provided was enough to raise trust for an unknown brand, it was not enough to change the
already established trust for a known brand. In a more transparent age, more and more
information is provided by brands. Opportunities in the transparency research remain in the
41
References
Aaker, D. A. (1996). Measuring brand equity across products and markets. California
Management Review, 38(3), 103.
Bhaduri, G., & Ha-Brookshire, J. E. (2011). Do transparent business practices pay?
Exploration of transparency and consumer purchase intention. Clothing and Textiles
Research Journal, 29(2), 135-149.
Brady, A. (2003). How to generate sustainable brand value from responsibility. The Journal
of Brand Management, 10(4), 279-289.
Buell, R. W., & Norton, M. I. (2011). The labor illusion: How operational transparency
increases perceived value. Management Science, 57(9), 1564-1579.
Carter, R. E., & Curry, D. J. (2010). Transparent pricing: Theory, tests, and implications for
marketing practice. Journal of the Academy of Marketing Science, 38(6), 759-774.
Chang, H. H., & Chen, S. W. (2008). The impact of online store environment cues on
purchase intention: Trust and perceived risk as a mediator. Online Information Review,
32(6), 818-841.
Christensen, L. T. (2002). Corporate communication: The challenge of transparency.
Corporate Communications: An International Journal, 7(3), 162-168.
Cobb-Walgren, C. J., Ruble, C. A., & Donthu, N. (1995). Brand equity, brand preference, and
purchase intent. Journal of Advertising, 24(3), 25-40.
42 Degeratu, A. M., Rangaswamy, A., & Wu, J. (2000). Consumer choice behavior in online and
traditional supermarkets: The effects of brand name, price, and other search attributes.
International Journal of Research in Marketing, 17(1), 55-78.
Delgado-Ballester, E., & Munuera-Aleman, J. L. (2001). Brand trust in the context of
consumer loyalty. European Journal of Marketing, 35(11/12), 1238-1258.
Demmers, J., Erbé, A., van Strijp, J., & Wientjes, C. (2015). The value of transparent
marketing communication: how proactive disclosure affects consumer behavior. In P.
Williams and A. Morales (eds.), Advances in Consumer Psychology, 7. Washington, DC:
Society for Consumer Psychology.
Eisend, M. (2006). Two-sided advertising: A meta-analysis. International Journal of
Research in Marketing, 23(2), 187-198.
Escalas, J. E., & Bettman, J. R. (2005). Self‐construal, reference groups, and brand meaning.
Journal of Consumer Research, 32(3), 378-389.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.
Fischer, M., Völckner, F., & Sattler, H. (2010). How important are brands? A cross-category,
cross-country study. Journal of Marketing Research, 47(5), 823-839.
Falk, A., & Fischbacher, U. (2006). A theory of reciprocity. Games and Economic Behavior,
54(2), 293-315.