S
S
t
t
o
o
r
r
y
y
t
t
e
e
l
l
l
l
i
i
n
n
g
g
a
a
n
n
d
d
B
B
r
r
a
a
n
n
d
d
A
A
t
t
t
t
i
i
t
t
u
u
d
d
e
e
T
Th
he
e
Ro
R
ol
le
e
of
o
f
C
Co
on
ns
su
u
me
m
er
rs
s’
’
L
Le
ev
ve
el
l
o
of
f
I
In
nv
vo
ol
lv
ve
em
me
en
nt
t
a
an
nd
d
A
Av
va
ai
il
la
ab
b
le
l
e
T
Ti
im
me
e
Eline van Ballegoijen (10049681) Msc. Business Studies (Marketing track)
Master thesis, 27 January 2014
Supervisor: Dr. F.B. Situmeang Second assessor: Dr. R.M. Singh
A
A
B
B
S
S
T
T
R
R
A
A
C
C
T
T
Storytelling is a powerful tool in which brands communicate their story to build a relationship with a consumer. Some research has been done regarding the effects of a brand story on consumers’ attitude towards the brand. However, the assumption being made here, is that consumers are involved with the story and have time to elaborate on it. According to the Elaboration Likelihood Model (Petty & Cacioppo, 1986), these are two important factors to become persuaded. In practice, consumers are exposed to thousands of advertising messages and will not extensively process all the messages. Therefore, consumers’ level of involvement and available time need to be taken into account when explaining the effect of a brand story on brand attitude. No research has examined these factors in the context of storytelling yet.
A total of 159 Dutch consumers participated in the research. Participants were randomly divided among 5 conditions: high involvement/no time restriction, high involvement/time restriction, low involvement/no time restriction, low involvement/ time restriction and a control group. They received different instruction letters, and were asked to read a story about the lifestyle sock brand “Alfredo Gonzales”. Afterwards, questions followed that assessed consumers’ attitude towards the brand.
The results showed that differences in brand attitude between conditions exist. High involved consumers who have time available are more positive towards the brand after reading the story than consumers that lack motivation and/or time. For lowly involved consumers, factual information about the brand is equally effective. Furthermore, the level of involvement moderates the effect of time on brand attitude. The more involved the consumer is, the more important it is that they have time to elaborate on the brand story.
T
T
A
A
B
B
L
L
E
E
O
O
F
F
C
C
O
O
N
N
T
T
E
E
N
N
T
T
S
S
INTRODUCTION ... 5
LITERATURE REVIEW ... 8
2.1. The essence of stories ... 8
2.2. Storytelling and branding ... 8
2.3. Impact of brand stories ... 10
2.4. Attitude towards the story and brand ... 11
2.5. Consumers' level of involvement and available time ... 13
2.6. Conceptual framework ... 17 RESEARCH METHOD ... 18 3.1. Design ... 18 3.2. Sample ... 18 3.3. Stimulus material... 19 3.4. Pre-test ... 21 3.5. Procedure ... 21 3.6. Measures ... 23 3.6.1. Dependent variables ... 23 3.6.2. Independent variables ... 24 3.6.3. Control variables ... 27 RESULTS... 28 4.1. Respondent collection ... 28 4.2. Manipulation checks ... 30 3
4.3. Testing of hypotheses ... 33
4.3.1. Brand story versus factual information ... 33
4.3.2. Differences across conditions ... 32
4.3.3. Level of involvement, available time and interaction ... 36
CONCLUSION AND DISCUSSION ... 43
5.1. Summary of findings ... 43
5.2. Theoretical contribution and managerial implications ... 45
5.3. Limitations and directions for future research ... 46
REFERENCES ... 49
APPENDICES ... 54
7.A. Stimulus material... 54
7.B. Instruction letters ... 57
7.C. Questionnaires ... 59
7.D. Output ... 65
I
I
N
N
T
T
R
R
O
O
D
D
U
U
C
C
T
T
I
I
O
O
N
N
Every brand has a story and nowadays, companies acknowledge the value of communicating their story (Marzec, 2007; Merchant, Ford, & Sargeant, 2010; Mossberg, 2008; Papadatos, 2006). Storytelling is used to bring a brand to life, giving them a perspective and personality (Keller, Parameswaran, & Jacob, 2011). It is a brand’s DNA and explains who you are and what you stand for. Also, in the academic literature there is a strong belief in the benefits of storytelling in branding (Marzec, 2007; Love, 2008; Lundqvist, Liljander, Gummerus, & van Riel, 2012). Why? Storytelling is a way to differentiate your brand from competitors and making sure that you are not “just another commodity”. It is a way to connect with consumers.
Storytelling is a popular topic in marketing and branding, and has been there for ages. It is one of the oldest, most powerful modes of communication (Kaufman, 2003). From childhood, we are used to read, listen to and tell stories. In this way we make sense of the things that happen around us. A well-structured story can take a reader on a journey and they can be left changed (Shankar, Elliott, & Goulding, 2001). That is why stories have always fascinated people and are often more easily remembered than facts (Lundqvist et al., 2012).
Storytelling has reached a significant status in companies’ branding efforts (Love, 2008). As the modern society becomes more communicative and markets saturated, companies acknowledge that it is necessary to put extra effort in differentiating their brands (Fog, 2010). They focus on building long-term relationships with consumers instead of catchy slogans and temporal flirts. The idea is that loyal consumers can be created through storytelling, because this makes it possible for the consumer to identify with who you are and what you do (Lundqvist et al., 2012). This indicates that practitioners have embraced storytelling. Advertising agencies specialized in storytelling are hired to unravel memorable events that can be
communicated on websites, packaging, in commercials and so on. Every company that wants to do something with storytelling can find a great variety of “best practices” on the internet.
Consequently, there has been an increase in storytelling research (Lundqvist et al., 2012; Mossberg, 2008; Woodside, Sood, & Miller, 2008). In a recent study, Lundqvist et al. (2012) investigated how a brand story influences consumers’ brand experience, and learned that consumers who were exposed to a story described the brand in much more positive terms. Mossberg (2008) found that certain settings are more suitable for a consumer to become transported into a story and to have an extraordinary experience.
Most brand story research investigates consumer experiences, but the academic research into the actual effects on the consumer’s attitude is limited. These two brand concepts are closely related, but not identical (Schmitt, Zarantonello, & Brakus; 2009). Moreover, in past research, the assumption is made that consumers are involved with the story and have time to elaborate on it (Lundqvist et al., 2012). These are, according to the Elaboration Likelihood Model (ELM), two important factors to become persuaded (Kim, Lim, & Bhargava, 1998). This well-known theory explains that an individual will only process a message at a deeper level and change their attitude, when they are involved and have the ability to do so. In addition, Mossberg (2008) proposes that involvement is an important facilitator for a consumer to be immersed in the story. In practice, consumers do not always have the opportunity to elaborate on a message. In daily life, they are exposed to thousands of advertising messages and they cannot and/or will not extensively process all of them. Time is restricted and not every message is personally relevant (Carr, 2011). When explaining the actual effect of a brand story on brand attitude, these factors need be taken into account. Yet, there is no research that examines consumers’ level of involvement and available time in the context of storytelling. Therefore, this research will contribute to the current
knowledge of storytelling by investigating whether the effect of a brand story is dependent on the situation of the consumer. This leads to the following research question:
RQ: “How does consumers’ level of involvement and available time affect the relationship between a brand story and the attitude towards the brand?”
By answering this research question, the study contributes to brand management research and practice by demonstrating the power of storytelling on consumer response. More specifically, it investigates the applicability of two factors of the Elaboration Likelihood Model in the context of storytelling. It examines if storytelling is equally effective across different situations.
This research is also important for practitioners because it demonstrates how brand stories can be used to affect consumers’ attitude towards the brand. With the explosive growth of social media and content marketing, the opportunities to tell a story increase. This means that consumers are exposed to an overload information, and have to make choices regarding on which messages to elaborate (Carr, 2011). The role of consumers’ level of involvement and available time in processing a brand story is therefore also an important issue for marketers and brand managers.
The thesis is structured as follows. The first part is a literature review on storytelling, brand attitude and consumers’ involvement and time. This sets the stage for the conceptual framework, in which the hypotheses are presented. Chapter three outlines the research methodology. Chapter four contains the research findings, where the hypotheses are tested and interpreted. In the final part of the thesis, a conclusion of the findings is given and the limitations and directions for future research are discussed.
L
L
I
I
T
T
E
E
R
R
A
A
T
T
U
U
R
R
E
E
R
R
E
E
V
V
I
I
E
E
W
W
2.1. The essence of stories
Bennet and Royle (2004, p. 55) defined stories as ‘a series of events in a specific order, with a beginning, a middle and an end’. Stories and storytelling help us to make sense of our lives (Shankar et al., 2001). We need stories to create meaning, to understand what goes on and to manage our relationships (Keller et al., 2011). From the moment we are born, our parents begin to tell us stories and we become familiar with understanding the narrative form. A story answers questions like who, what, why, when and how (Shankar et al., 2001). It is comprised of a theme and often conveys a deeper meaning. A life without storytelling is hard to imagine, whether listening to, or watching, reading and telling them (Gergen & Gergen, 1988).
Stories are more easily remembered than facts (Lundqvist et al., 2012). This is because the human memory is story-based: information is indexed, stored and retrieved in the form of stories (Schank, 1999). When we experience something new, we connect the information to existing information in our mind. Our brain prefers to believe that there are connections between the things that it sees, because that makes it easier to interpret the bigger picture. In general, people create stories to organize their experience, create order, explain unusual events and make evaluations (Escalas, 2004).
2.2. Storytelling and branding
Storytelling is a powerful tool that is widely used in marketing and branding. Brand managers know the brand is no longer just the words and pictures; the brand is about the total delivery of a promise to the consumers (Papadatos, 2006). In today’s
competitive world, it is not only about the product or service itself. Competitors can copy all of this. A story, on the contrary, often forms an ideal basis for a powerful and memorable mission and is harder to imitate (Kaufman, 2003). It is a way to differentiate your brand and to connect with consumers, because consumers seek experiences that appeal to their emotions and dreams (Love, 2008). Stories help to create such experiences, and therefore it is important to expose the story behind the brand (Fog et al., 2005).
A brand story is a narrative that portrays the heart and soul of the brand, its DNA (Fog et al., 2005). It tells something about the heritage, the founder, the highlights and crises, the mission and values, the functional and emotional benefits (Fog et al., 2005). And by sharing such a story, a brand can position itself and defines “who we are” and “what we stand for”. With storytelling, a brand can build upon its identity. The goal is to create a positive image of the brand in the mind of all stakeholders, particularly consumers (Van Riel & Balmer, 1997).
Storytelling has the power to strengthen a brand both internally and externally (Fog et al., 2005). It is used because of many reasons: to engage employees, to enhance fundraising or recruitment or to connect to consumers (Kaufman, 2003). Different stories can co-exist, but most enduring brands have one core story (Papadatos, 2006). For example, Nike represents the “will to win”. Every marketing communication is structured to that one story and it creates consistency in all communications (Fog, 2010). Whether you are shopping in a Nike store, visiting the website or watching a commercial, you will recognize this message everywhere.
2.3. Impact of brand stories
The persuasiveness of stories is never been doubt. Researchers have found that storytelling is far more convincing to an audience than rational arguments, statistics or facts (Kaufman, 2003). A story makes a topic more real and relevant for individuals 9
than rational persuasion, because it appears in an easy to understand format to which anyone can relate (Kaufman, 2003). McKee (2003) advocates that the best way to persuade someone, whether a friend, colleague or consumer, is by telling a compelling story. In his opinion, individuals can be persuaded by using conventional rhetoric, but individuals will only remember it when the idea is united with an emotion in a story (McKee, 2003). Stories thus help convey ideas more effectively than the use of mere facts (Goodman, 2006).
A recent study (Schmitt et al., 2009) showed that well-told stories regarding the origin of a brand appear to have the potential to influence how consumers think about the brand. The information in the story may add strong, favorable and unique associations to a brand, which can increase brand equity (Keller, 1993). In other words: the value of a well-known brand name. According to Keller et al. (2011), in the end, the power of the brand lies in what consumers have learned, felt, seen and heard about the brand. And one way to evoke associations about the brand, which might reside in the mind of the consumer, is by telling a compelling story. It is important that unique associations are communicated, because those help to differentiate your brand from other brands in the product category (Fog, 2010). Besides, it is important that consumers can appeal to a story. Consumers that can identify with the values in the story can function as brand ambassadors and are more likely to spread positive word-of-mouth about the brand (Lundqvist et al., 2012).
In these subchapters, (brand)stories and their essence, the purpose of stories for individuals, and the role of storytelling in branding have been discussed. In the following section, the impact of storytelling on brand attitude, and consumers’ situation when processing a message are considered.
2.4. Attitude towards the story and brand
A well-structured story has the ability to affect attitude formation (Green & Brock, 2000). In some situations, consumers lose themselves in a story and then change their attitudes and intentions to reflect that story (Van Laer, Ruyter, Visconti, & Wetzels, 2012). This is called “narrative transportation” (Green & Brock, 2005). This theory is emerged to explain the persuasiveness of narratives and yet has key divergences from well-known dual processes communication theories (Green & Brock, 2000), like the Elaboration Likelihood Model (Petty & Cacioppo, 1986).
Many researchers have shown interest in this narrative theory due to the impact a story can have on consumers’ beliefs and attitudes (Green & Brock, 2000; Mazzocco et al., 2010). Attitude towards the brand represents an important research topic for practitioners, because this psychological construct is tightly correlated with purchase intention (MacKenzie & Lutz, 1989; Spears & Singh, 2004). It means that consumers can think about a brand in more positive or more negative terms. What brands want to achieve with communicating their stories, is trying to change consumers’ attitude regarding the brand in a positive way (Fog, 2010).
Although a lot has been written about effect of a brand story on consumer experience, little empirical evidence exists of their effect on the attitude of the consumer. Brand experience and brand attitude are closely related, but not identical (Schmitt et al., 2009). Fishbein and Azjen (1975) describe attitude as ‘general evaluative judgments about the brand (e.g. “I like this brand”) based on beliefs or affective reactions. In contrast, brand experiences include specific feelings, such as enthusiasm, satisfaction, or curiosity, not an overall “liking” (Schmitt at al., 2009). In a recent study, Lundqvist et al. (2012) investigated how a brand story influences consumers’ brand experience, by comparing the brand experiences of two groups of consumers. The comparison revealed remarkable differences between the 11
two groups. “Consumers who were exposed to the story described the brand in much more positive terms and were willing to pay more for the product than those consumers that were exposed to factual information” (Lundqvist et al., 2012, p. 292). These results show that exposing consumers to a story can have positive effects. Consumers give the products a higher rating and are willing to pay more than consumers that were exposed to factual information about the brand.
Lundqvist et al. (2012) used brand experience as dependent variable. Because this brand concept is closely related to brand attitude, the same findings are expected for the latter. A brand story is expected to be more persuasive than factual information about the brand. The following hypothesis is formulated:
H1a: Participants exposed to a brand story have a more positive attitude towards the brand than participants who were exposed to factual information about the brand.
An indicator for brand attitude might be the attitude towards the brand story itself. Gardner (1985) indicates that brand attitudes may be affected by consumers’ attitude towards the advertisement itself. Attitude toward the ad is defined as a “pre-disposition to respond in a favorable or unfavorable manner to a particular advertising stimulus during a particular exposure occasion” (MacKenzie & Lutz, 1989, p. 48). Other researchers have also proposed that attitude towards the ad is a mediating influence on brand attitude (Petty & Cacioppo, 1986; MacKenzie et al., 1986; Park & Young 1986).
Although consumers do not recognize a brand story as a typical advertisement, it is indeed a way to advertise (Simmons, 2006). It is used as part of the company’s corporate branding strategy and as sales promotional tool (Fog, 2010). The following hypothesis is formulated to see whether a more positive attitude towards the story itself is associated with a more positive attitude towards the brand advertised.
H1b: Consumers’ attitude towards the story is associated with their attitude towards the brand advertised in the story.
The relationship between storytelling and brand attitude is examined by testing these two hypotheses. However, the main goal of this study is to see whether consumers’ level of involvement and available time might affect this relationship. These circumstances are further discussed in the next section.
2.5. Consumers’ level of involvement and available time
One prerequisite for being persuaded is that the reader can be transported (i.e. absorbed or immersed) in the story being told (Green & Brock, 2005). In practice, consumers do not always have the opportunity to give full attention to every communication, and are trained to filter communications options (Carr, 2011). In the end, it is the choice of the consumer if they want to give attention to a brand story. The effectiveness of storytelling, especially in situations where individuals lack motivation or ability, can therefore be questioned.
Past research has demonstrated that individuals who are more transported into a story exhibit greater attitude and belief change in response to stories (Escalas, 2004; Green, 2004; Green & Brock, 2000, Van Laer, 2012). Transportation is sometimes more or less likely to occur. In general, a well written story will elicit more transportation than a poorly written story (Green, Brock, & Kaufman, 2004). Some people are in general more sensible for transportation than others, because it requires mental processes including emotional responding, character identification and narrative comprehension (Green & Brock, 2002). More importantly, transportation requires a certain level of effort and attention to the story, and not everyone is able to deliver that. As Mazzocco, Green, Sasota, and Jones (2010) maintain, individual differences regarding the motivation and ability to become transported are likely to exist.
To become transported and adjust beliefs or attitudes, it is essential that consumers are involved with the message content (Van Laer, 2012). They have to find the message personally relevant and/or important. The relationship between
involvement and consumers’ reaction to persuasive communication has been studied extensively by researchers (Park & Young, 1986; Petty, Cacioppo, & Schumann, 1983). A well-known theory that is used in this context is the Elaboration Likelihood Model, or ELM (Petty & Cacioppo, 1986). It is a model that explains attitude change and the effectiveness of persuasive communications. With “elaboration” the authors mean to what extent a consumer thinks about the relevant arguments in the message. The elaboration likelihood is said to be high when consumers process the message on a deeper level (Petty & Cacioppo, 1986).
In daily life, consumers are exposed to thousands of advertising messages and it is not possible to extensively process all these messages. The Elaboration Likelihood Model distinguishes two ways in which a consumer can process a message: central route processing or peripheral route processing (Petty & Cacioppo, 1986). For the message to be centrally processed, a consumer must have the opportunity (depending on i.e. time availability) and they must be involved with the message. According to Laczniak, Muehling, and Grossbart (1989), message involvement is high when consumers pay attention and perform a brand-processing strategy (1989). This means that they read the message to evaluate the brand (or product) advertised. When they pay attention to the message for other purposes (for example, evaluating the style or grammar), a non-brand processing strategy is said to be employed (Laczniak et al., 1989).
When employing a brand-processing strategy, a consumer is motivated to elaborate on the relevant arguments in a message (Kardes, 1994). It is likely that the consumer will follow the central route and this can lead to a permanent change in their attitude (Laczniak et al., 1989). In situations where consumers are not motivated to process a message, it is more likely that they take the peripheral route (Petty & Cacioppo, 1986). Here, peripheral cues play a role in attitude formation. This is a simple cue, like attractiveness or credibility of the speaker, that affects attitudes in the
absence of argument processing (Petty et al., 1983). Permanent attitude change in this situation is less common because consumers base their beliefs on other things than the relevant arguments (Gardner, Mitchel, & Russo; 1985).
Although the level of involvement is the first step towards the central route, a person must also have the ability to think about the message. A consumers’ ability to process a message depends on different factors, for example: other distracting stimuli, the complexity of the message, and the amount of prior experience with the topic (Petty, Wells, & Brock; 1976). But the most simple and straightforward factor that determines ability is time availability (Petty et al., 1983). When consumers lack time to read a brand story, they neither have time to notice or elaborate on the relevant arguments. In practice, there are situations in which the motivation to process is present, but the ability is lacking or the other way around. In situations where both the involvement and ability are high, the elaboration likelihood is said to be high and consumers are likely to follow the central route (Petty & Cacioppo, 1986).
These two factors, level of involvement and available time, are important ingredients to become transported and to change or adjust attitude or beliefs regarding a brand. In the recent study of Lundqvist et al. (2012), however, the assumption is made that consumers have all time and are motivated to elaborate on brand stories. In practice, consumers need to deal with an overload information and cannot pay attention to everything (Carr, 2010). They are restricted in time and therefore used to scan information. Offering richer content in the form of a brand story with the goal to connect consumers sound reasonable, but in some situations, catchy slogans that are easy to process might be a more effective strategy. A story can be extremely well-crafted, but when a consumer does not have time to read it or is not interested in the brand at all, the story might be equally (or even less) influential as factual information about the brand (Stichting Wetenschappelijk Onderzoek Commerciële Communicatie, 2013).
Following this reasoning, two main hypothesis are formulated that will answer the research question: “How does consumers’ level of involvement and available time affect the relationship between a brand story and the attitude towards the brand?”. As discussed above, a consumer will only be transported into the story and follow the central route to persuasion when they are involved and have the ability to process the message (Kardes, 1994). The effect of a brand story on brand attitude is therefore expected to be highest for high involved individuals with no time restriction, and is expected to be least for low involved individuals with a time restriction.
H2a: High involved consumers who have time available are more positive towards the brand after reading the story than consumers that lack involvement and/or time.
H2b: For low involved consumers that are restricted in time, factual information about the brand is equally effective as a brand story.
The elaboration likelihood depends on the level of involvement and ability. In some situations, the involvement with the message is high whilst the ability is low, and vice versa. In situations where both the involvement and ability are high, it is likely that consumers follow the central route and change or adjust their attitude towards the brand permanently (Petty & Cacioppo, 1986). It is expected that the interaction of these two factors affects the persuasiveness of a brand story.
H2c: The interaction between consumers’ level of involvement and available time affects the relationship between a brand story and attitude towards the brand.
2.6. Conceptual framework
When all hypotheses are combined, a conceptual model for the research can be constructed (figure 1). Hypotheses 1a and 1b are included to replicate and test prior research. The main focus of this study will be hypothesis 2.
I IVV:: B Brraannddssttoorryy Time availability & Involvement
DDVV:: B Brraannddaattttiittuuddee H H11aa H H22 Attitude towards the story HH11bb 17
R
R
E
E
S
S
E
E
A
A
R
R
C
C
H
H
M
M
E
E
T
T
H
H
O
O
D
D
3.1. Design
The purpose of this research is to examine the effect of a brand story on brand attitude, and the possible moderating role of consumers’ level of involvement and available time. A 2x2 between subjects experimental design was applied to investigate the effect of a brand story on brand attitude. Two factors are manipulated (involvement and time), with two levels of each variable (high/low involvement and time restriction/no time restriction). This leads to four experimental conditions (table 1). A control group is added to measure the brand attitude without exposure to a brand story.
Table 1 - Experimental design, N (%)
No time restriction Time restriction
High involvement 30 (18,9%) 30 (18,9%)
Low involvement 29 (18,2%) 32 (20,1%)
Control condition: 38 (23,9%).
3.2. Sample
The sample in this study consists of Dutch consumers. A total of 159 participants between the age of 16 and 59 (M= 27.22, SD= 9.27) completed the questionnaire. The majority (61%) of the participants was between 20 and 30 years old, which is the target group of the brand that was used as stimulus material. The group consisted of 74 men (46,5%) and 85 women (53,3%). Of all participants, 4,4% has had a low-level education
(i.e. primary school or VMBO), 20,8% has an average educational level (i.e. HAVO, VWO or MBO) and 74,8% of the participants was highly educated (i.e. HBO or WO). The participants are more or less equally distributed across the five conditions (table 1). Due to limited resources, a non-probability sampling technique has been used to recruit the participants. A self-administered questionnaire was spread by email and through Facebook to the social network of the researcher. Participation was voluntary and participants did not receive any compensation for their answers. To increase the reach of the survey, a snowball sampling was applied by asking the respondents to share the link with their friends between the age of 15 and 60. No further sampling criteria were used.
3.3. Stimulus material
In all four experimental conditions, participants were exposed to the same stimuli: a story about the brand “Alfredo Gonzales”. This is the first lifestyle sock brand in the world. The brand is considered suitable for stimulus material because of several reasons. Firstly, it is a neutral product, and applicable to a broad audience, as everybody wears socks, whether young or old, man or woman. Secondly, the brand has built its marketing strategy around a brand story. This makes it possible to study a real brand with a real story, rather than an invented brand and invented story. Thirdly, the brand Alfredo Gonzales was founded in 2012 and is only available online, and is not so familiar yet. To check this, an item has been added at the end of the questionnaire to measure the participants’ familiarity with the brand prior to the research. Only seven participants did know the brand, and are subsequently deducted from the data.
The brand story of Alfredo Gonzales is in fact a story within another story. It starts with a short fairy tale about a local Mexican legend named Alfredo who abandoned his town to start working in a sock factory in New York, USA. He customized his own socks because he did not have any money to buy new pairs. The
founder of Alfredo Gonzales used this fairy tale accordingly to tell the brand’s story. In the story, he explains what inspired him in founding the company, how he built it up, and what the company’s mission and value proposition is. The story also includes qualitative arguments about the brand, which encompass the following:
• Design socks with unique colors and patterns;
• The socks are only available in some premium stores; • The high-quality socks are made in Italy;
• The socks are durable, green (made of bamboo) and comfortable; • The collection is unisex.
These arguments are “hidden” in de story, and participants need to be involved and be able to read the whole story to notice them. Participants that have to deal with a time restraint need to scan the information, and therefore it is unlikely that they will elaborate on all these arguments. The stimulus material exists of text and pictures, by which participants can follow both the peripheral and central route (Areni & Lutz, 1988).
The participants in the control condition are not exposed to a brand story. They are exposed to the same pictures and only factual information about the brand and socks. The qualitative arguments are presented without any background information or explanation, but are comprehensive enough to give them enough information to form an opinion about the brand.
The final version of the brand story and the information presented to the control group can be consulted in Appendix 7A.
3.4. Pre-test
In order to assess the amount of time participants get in the time restriction conditions, a pre-test is done to figure out what would be a reasonable limitation. Participants (with different educational backgrounds and ages) were asked to focus on the story and read it from beginning to end, while keeping note of their reading time. Hahn, Lawson and Lee (1992) used a similar method in their research into the effects of time pressure and information load.
Ten individuals (five women, five men) between the ages of 18 and 55 participated in the pre-test. Two were poorly educated (i.e. primary school or VMBO), four had an average educational level (i.e. HAVO, VWO or MBO) and four were highly educated (i.e. HBO or WO). The slowest reader took four minutes and two seconds, where the fastest only needed one minute and 50 seconds. Seven participants needed more than three minutes (M = 169,9 seconds, SD = 37.2 seconds).
Based on this information, the time restriction is set on 60 seconds. This means that even the fastest reader will not succeed in finishing reading the story in time. The time restriction is pre-tested by three other participants. All of them admitted that they needed to make choices in what to read and what not to read, and that they scanned through the information.
3.5. Procedure
Following the link that has been provided in the Facebook messages or emails sent, individuals were directed to the online questionnaire program Qualtrics. The questionnaire was in Dutch. Participants in both the control group and experimental conditions were exposed to the same introductory page that informed the participants about the research. The lay-out of another faculty (Social and Behavioral Sciences) is used for the questionnaire to distract the participants from the goal of the research. 21
The randomizer function in Qualtrics is used to assign participants to either the control group or to one of the four experimental groups. Each of the four experimental blocks existed of two pages: the instruction letter and the brand story. The brand story is the same across all experimental conditions, where the instruction letters differ.
In the high involvement, no time restriction condition, participants were asked to carefully evaluate the brand and to imagine that they are considering buying the product (brand processing strategy). Instructions were given to elaborate on the content, and to read the story from beginning to end. In the high involvement, time restriction condition, participants also received instructions to evaluate the brand, however pointing out that they had only 60 seconds to do this. They were instructed to scan through the information to obtain a general view of the brand.
In the low involvement, no time restriction condition, participants were given general reading instructions and they were told to focus on the overall writing style and readability (non-brand processing strategy). Nothing was said about timing. In the low involvement, time restriction condition it was pointed out that participants are automatically forwarded to the questions after 60 seconds. The instruction letters can be found in Appendix 7B.
Qualtrics provides a customer timer that allows to display an object only for a period of time. This timer was set on the time restriction conditions (2 and 4) to make sure that participants were automatically forwarded to the questions after 60 seconds. In the no time restriction conditions (1 and 3), the “move on” button was hidden for 90 seconds. Submitting was disabled to make sure that participants at least paid more attention to the story than in the time restriction conditions.
In the control group, participants received instructions to evaluate the brand based on some facts and pictures. These facts were also communicated in the brand story. The same pictures were used and no customer timer was set.
The manipulation checks followed directly after the brand story. Participants in the experimental conditions answered questions about their level of attention, brand processing and non-brand processing strategies. After these questions, statements about their attitude towards the story and the brand followed. To determine the participants’ attitude towards Alfredo Gonzales and the story itself, differential rating scales were used. Participants in the control condition did not read the brand story and, therefore, only see the questions about their attitude towards the brand.
On the next page, participants were asked to indicate their purchase intention for the socks displayed. All five groups have been exposed to these questions. Following, questions that assessed participants’ attitude towards storytelling in general were posed. Furthermore, participants’ buying habits regarding the product category were checked. The questionnaire closed with some demographic items, specifically gender, age, nationality and education. Respondents were thanked for their participation and the Qualtrics software automatically processed the answers.
3.6. Measures
This part will explain which items have been used to measure the different variables. All items are measured on a 7-point Likert scale, a scale which is commonly used to measure attitudes (Likert, 1932). The final questionnaire can be consulted in Appendix 7C.
3.6.1. Dependent variables
Brand attitude
Brand attitude is defined as consumers’ overall evaluation of a brand (Fog, 2010). The participants’ evaluation of the brand was assessed with a three 7-point scale with bipolar items, specifically good versus bad, pleasant versus unpleasant, like versus 23
dislike. This measure was first used by Gardner (1985) and by many other academic researchers subsequently. One item (“My attitude towards the brand Alfredo Gonzales is negative”) was added to the scale and recoded afterwards. For analyses, these four items were averaged (α = .87) to represent brand attitude.
Purchase intention
Attitude towards the brand is correlated with purchase intention (MacKenzie & Lutz, 1989; Spears & Singh, 2004). Regardless of it not being a main variable that is hypothesized in the conceptual framework, it is interesting to see whether a brand story affects purchase intention. Purchase intention is an individual’s conscious plan to make an effort in purchasing a brand (Fishbein & Ajzen, 1975). It can be measured with a single item measuring the likelihood attribute (Bergkvist & Rossiter, 2007). The question: “How likely is it that you will buy Alfredo Gonzales socks in the future?” is answered on a scale from 0% to 100%.
Besides likelihood, it is interesting to see whether reading a brand story would increase the perceived value of the brand. Lundqvist et al. (2012) found that consumers were willing to pay more for a facial cream after hearing a story about the origin of the brand, than when they did not hear that story. Therefore the question “What do you think is the retail price of this gift box including five pairs of Alfredo Gonzales socks?” is included in the questionnaire. Furthermore, participants were asked if they are willing to pay the price they have indicated.
3.6.2. Independent variables
Involvement
Involvement with a message has a considerable impact on how brand attitudes are formed or changed. According to Lazcniak et al. (1989), a manipulation of message involvement (and its corresponding manipulation checks) must capture at least two key dimensions: (1) the degree to which individuals pay attention to the message, i.e. message attention, and (2) individuals’ processing strategies, i.e. brand vs. non-brand evaluation.
Manipulation of the involvement variable was accomplished by giving different experimental instructions to the groups (see Appendix 7B). Participants in the high involvement conditions were given instructions to maximize attention to the brand story and the amount of brand processing (Laczniak et al., 1989), and were asked to imagine a purchase in the product category (Gardner, 1985; Higie, Feick, & Price, 1991). These instructions should lead to higher scores on involvement. In the low involvement conditions, participants were given general instructions to scan the information and evaluate the overall appearance, to invoke the use of a non-brand processing strategy (Gardner et al., 1985; Leigh, 1984; Wright, 1973, 1975). (1) Message attention. To determine if these manipulations adequately impacted participants’ involvement with the message, multiple-item manipulation checks of Laczniak et al. (1989) have been used. A total of three items were used to assess message attention:
• I concentrated hard on this message; • I carefully considered the information; • I spend effort to read the whole story.
Responses for all three items were combined into a scale (α = .82).
(2) Processing strategy. Laczniak et al. (1989) constructed two sets of items for processing strategy (i.e. brand versus non-brand processing). To check for brand processing, the following items were used and averaged for analyses (α = 0.82):
• I tried to evaluate the brand Alfredo Gonzales;
• I carefully considered the information, to decide if I want to buy Alfredo socks; • I paid attention so that I could determine what the brand has to offer.
To check for non-brand processing, the following items were used (Laczniak et al., 1989) and averaged for analyses (α = 0.77):
• I paid attention to help me determine how easy it was to read; • I tried to evaluate the writing style;
• I tried to evaluate the overall appearance of text and pictures.
These three sub scales are used to represent the variable “involvement”. In the results section, it is shown that these measures were successfully manipulated.
Attitude towards the story
A brand story is a particular form of advertising. The main attribute underlying attitude towards the ad, or in this case the brand story, is ‘affect’ or ‘likability’ (Brown & Stayman, 1992). However, attitude towards the ad is often measured with multiple items assessing specific perceptions of the ad (MacKenzie & Lutz, 1989). Dens, Pelsmacker, and Janssens (2008) measured the attitude towards the advertisement by means of three items (good, interesting, like it). One item is added to this scale (“The story appeals to me”) and one item is formulated negatively (“This story is bad”), to reduce acquiescent and extreme response bias. Responses from all four items were combined into a scale (α = .88), with higher scores indicating a more positive attitude towards the story.
3.6.3. Control variables
Attitude towards storytelling in general
Earlier research shows that the attitude towards advertising in general can influence the acceptance of one particular advertisement (Bauer, Barnes, & Reichardt, 2005). The items that are used to measure attitude towards advertising are adapted to storytelling: “Generally, I find it a good thing that brands tell their stories”, “I find corporate storytelling disturbing” and “I like stories about brands”. For analyses these three items were averaged (α = .81) to represent attitude towards storytelling in general.
Demographics
Demographics can also influence the dependent variable. Participants therefore needed to answer questions about their gender, age, nationality, and education at the end of the questionnaire.
R
R
E
E
S
S
U
U
L
L
T
T
S
S
4.1. Respondent collection
166 Dutch consumers participated in the research. Only seven were familiar with the brand Alfredo Gonzales before the research started (4,2%). These individuals were deleted from the sample because they can have predispositions about the brand that in their turn can affect the opinions of the participants. The software program Qualtrics makes it possible to force response, so every respondent – except one – answered every question. This participant was also deleted from the sample. The final sample thus holds 159 participants, which were randomly divided into four experimental conditions, and one control group. In the tables below an overview of the characteristics of the respondents.
Table 2 - Gender of respondents per condition (N) Condition 1 Condition 2 Condition 3 Condition 4 Control condition Total Male 13 19 14 16 12 74 Female 17 11 15 16 26 85 Total 30 30 29 32 38 159 28
Table 3 - Age of respondents per condition (N) Condition 1 Condition 2 Condition 3 Condition 4 Control condition Total <19 4 3 7 2 7 23 20-25 17 9 14 12 17 69 26-30 3 10 1 7 7 28 31-35 3 3 3 5 3 17 36-40 0 2 3 3 1 9 41> 3 3 1 3 3 13 Total 30 30 29 32 38 159
Table 4 - Educational level of respondents per condition (N) Condition 1 Condition 2 Condition 3 Condition 4 Control condition Total VMBO 1 1 1 1 3 7 HAVO 1 1 3 0 2 7 VWO 4 1 2 1 3 11 MBO 1 3 2 5 5 16 HBO 9 15 10 13 20 67 WO 14 9 11 12 5 51 Total 30 30 29 32 38 159 29
46,5% of the respondents are male, 53,4% are female. The respondents varied in age between 16 and 59 years old. The majority (61%) of the respondents was between 20 and 30 years old at the time of conduct, which is also the target group of the brand. Furthermore, 74,8% of the respondents was highly educated (i.e. HBO or WO).
4.2. Manipulation checks
In this paragraph several tests were conducted to see whether the sub dimensions of involvement (message attention and brand processing strategy) are successfully manipulated. An alpha level of .05 is used for all statistical tests and the measurement is based on a 7-point Likert scale.
The instruction letters were designed such that participants in the no time restriction conditions (1 and 3) should score higher on the “message attention” scale than participants in the time restriction conditions (2 and 4). Furthermore, participants in the high involvement conditions (1 and 2) must score higher on the “brand processing” scale, whilst participants in the low involvement conditions (3 and 4) must score higher on the “non-brand processing” scale. In table five the means and standard deviations per group are visible. The instruction letters can be found in Appendix 7B.
As can be seen in table five, the average scores are higher in the expected groups. A one-way analysis of variance (ANOVA) has been conducted to determine if the message attention, brand processing and non-brand processing strategies significantly differed across conditions. This is an appropriate measure to determine whether there are any differences between the means of two of more independent groups (Fielding & Gilbert, 2006).
Table 5 - Means (SD) per condition on message involvement subscales
Message attention Brand processing Non-brand processing 1: High involvement, no time restriction 5,32* (1,32) 4,64* (1,10) 3,86 (1,15) 2: High involvement, time restriction 4,07 (1,37) 4,18* (1,33) 3,39 (1,41) 3: Low involvement, no time restriction 4,98* (1,15) 3,36 (1,39) 5,67* (0,79) 4: Low involvement, time restriction 3,79 (1,45) 3,31 (1,49) 4,74* (1,07) * Means scores are higher in the expected groups.
For all three variables, data was normally distributed for each condition, as assessed by Shapiro-Wilktest (p >.05). The four outliers are modified by replacing the values with ones that are less extreme. The results show that there was no homogeneity of variances on message attention (p = .306) and brand processing (p = .235). However, for non-brand processing the assumption of homogeneity of variances was violated, as assessed by Levene’s Test of Homogeneity of Variance (p = .007).
Message attention. The level of message attention was significantly different between the four conditions, F (3,117)= 9.778, p < .001. The assumption of homogeneity of variances it not violated, and therefore Tukey post-hoc tests are used to compare possible combinations of group differences. As expected, there was a difference in average scores between condition one (M = 5.32, SD = 1.32) and condition two (M= 4.07, SD= 1.37), a difference of 1.25, 95% CI [ 0.39, 2.12], which was significant (p < .001). Also the difference between condition one and four (1.53, 95% CI (0.68 to 2.38)) was significant (p < .001). Furthermore, there was a difference in average scores between condition three (M = 4.98, SD= 1.15) and condition two (M= 31
4.07, SD = 1.37), a difference of 0.91, 95% CI [0.04, 1.78], which was significant (p = .037). The difference between condition three and four (1.19, 95% CI (0.33 to 2.04)) was also significant (p = .003). This means that the subscale message attention is successfully manipulated.
Brand processing. The level of brand-processing was significantly different between the four groups, F (3,117)= 7.179, p < .001. Participants in the high involvement conditions score significantly higher on brand processing items than participants in the low involvement conditions, t (119) = -4.440, p < .001). This means that also the brand processing variable is successfully manipulated.
Non-brand processing. The assumption of homogeneity of variances was violated for the variable non-brand processing (p = .007), and therefore the results of a robust ANOVA test are reported. The level of non-brand processing was significantly different between conditions, Welch’s F (3, 63.931)= 27.750, p < .001. The Games-Howell post-hoc test is used to determine where the differences exactly lie. The difference between the average scores of condition three and one is 5.43 (95% CI, 3.39 to 7.48), which is statistically significant (p < .001). Also the difference of 6.8 (95% CI, 4.46 to 9.20) between condition three and two is significant (p < .001). Furthermore, Games-Howell post-hoc analysis revealed that the difference between condition four and one (2.65, 95% CI (0.41 to 4.89)) was significant (p = .014), as well as the difference between condition four and two (4.05, 95% CI (1.52 to 6.59), p = .001). In summation, this means that the instruction letters are manipulated successfully. Participants in the four experimental conditions score significantly different on the three subscales of involvement. Thus, the level of involvement is successfully manipulated by giving the conditions different instruction letters. The three sub variables (message attention, brand processing and non-brand processing) themselves are not taken into account in the following analyses. They are automatically integrated because participants in the low involved conditions (3 and 4) are given the
value 1, and participants in the high involved condition (1 and 2) the value 2.
4.3. Testing of hypotheses
In this section several tests are performed to either reject or accept the hypotheses. An alpha level of .05 is used for all statistical tests and the measurement is based on a 7-point Likert scale.
4.3.1. Brand story versus factual information
Hypothesis 1a stated that participants that were exposed to a brand story (whether high or low involved, time restricted or not) would have a more positive attitude towards the brand than participants in the control condition, who were exposed to factual information about the brand. An independent samples t-test was performed in order to find differences in brand attitude between the total of experimental conditions versus the control condition. The average attitude towards the brand in the experimental conditions was 4.77 (SD = 0.93), against an average score of 4.55 (SD = 1.27) in the control condition, which is not significant (t (157) = 1.176, p = .241). Hypothesis 1a is therefore rejected. It cannot be argued that a comprehensive story about a brand has a more positive effect on brand attitude than factual information about the brand.
4.3.2. Differences across conditions
When comparing the average scores on brand attitude across the different conditions (figure 1), it is clear that participants in condition 1 are more positive towards the brand (M= 5.21, SD= 0.69) than participants in all other four conditions.
Hypothesis 2a stated that the effectiveness of a brand story is dependent on a consumer’s situation: high-involved consumers who have time available are more positive towards the brand after reading the story than consumers that lack motivation and/or time. Furthermore, it was hypothesized (2b) that low involved consumers that had to deal with a time limitation (condition 4), are equally positive towards the brand as participants exposed to factual information (control condition).
A one-way ANOVA was conducted to determine if brand attitude significantly differed across conditions. Participants in the high involvement, no time restriction condition are more positive towards the brand than participants in the other three experimental conditions and the control group, F (4, 154)= 2.461; p <.048). Consumers are the most positive towards the brand if they have time and are involved with the message. Consumers that do have time, but are lowly involved, are less positive (M= 4.77, SD= 0.93) about the brand (t (57)= .280, p = .043). Low involved consumers that have no time to read the story, are equally positive about the brand as consumers that were exposed to factual information (t (68) = .204, p = .839). 34
Hypotheses 2a and 2b are therefore accepted, and this implies that the effectiveness of a brand story is, to some extent, dependent on the situation of the consumer.
Remarkable about this result is that participants in condition 2 score lower on brand attitude (M = 4.53, SD= .91) than participants in condition 3 (M= 4.77, SD= .93), however this difference is not significant (t (57) = -1.013, p = .315. A possible explanation for the low average score in condition 2 could be that the participants have felt themselves fooled. They were given instructions to carefully evaluate the brand, but were given only 60 seconds to do so which is far from enough.
Although it is not a variable that has been hypothesized in the conceptual framework, it is interesting to see whether a brand story affects purchase intention. Purchase intention is tightly correlated with brand attitude (MacKenzie & Lutz, 1989; Spears & Singh, 2004), and therefore there should be similar results on this variable. A Pearson’s product-moment correlation was run to assess the relationship between brand attitude and purchase intention. There was a moderate positive correlation between the two variables, r (157)= .437, p < .001, with brand attitude explaining 19% of the variation in purchase intention. In the following table the purchase intention in percentages per condition can be found.
Table 6 – Purchase intention in % per condition
No time restriction Time restriction
High involvement 40,6% 24,2%
Low involvement 30,4% 27,0%
Control condition: 33,6%.
The overall purchase intention is low. It is most likely that highly involved consumers that had time to read the story buy Alfredo Gonzales socks in the future (40,6%), and least likely that low involved consumers that were restricted in time make a purchase (27,0%). This makes sense, because consumers in condition one also had the most positive attitude towards the brand. Performing a one-way ANOVA, it is shown that the differences in purchase intention are significant, F (4, 154)= 2.574; p =.040).
Remarkably is that the average score of the experimental conditions (M= 30.5, SD= 21.6) is lower than the average score of the control condition (M= 33.6, 24.3), but this difference is not significant (t (157) = -.746, p= .457). Based on this result, it cannot be argued that exposure to a brand story leads to higher purchase intention than exposure to factual information. This is in line with the finding that exposure to a brand story does not lead to a more positive attitude towards the brand.
A possible explanation could be that participants who were exposed to the brand story perceived the brand to be more expensive. Participants in the experimental conditions estimated the price of a gift box with 5 pairs of socks at €36,30 (SD= 20.06), which is close to the actual price (€45). Participants exposed to factual information about the brand thought the gift box is worth €25,80 (SD= 11.57), and this difference is significant (t (109.67) = 4.041, p< .001). The brand story thus invokes associations like expensive, premium, and luxurious, and this may account for a lower purchase intention in the experimental conditions. In general, participants indicate that they do not want to spend much money on socks, because it is a functional and utilitarian good.
4.3.3. Level of involvement, available time and interaction
It was shown that the situation of the consumer has some effect on the relationship between a brand story and brand attitude. High involved consumers that have time available have a more positive attitude towards the brand, and it is most likely that these consumers will buy the brand in the future. To further examine the role of 36
consumers’ time, involvement and the interaction between these two factors, a multiple regression analysis was conducted. This is an appropriate measure to test the conceptual model, because it allows to predict a dependent variable based on multiple independent variables (Fielding & Gilbert, 2006). With this method it is possible to determine the relative contribution of each predictor to the total variance explained.
The analysis include brand attitude as dependent measure, with involvement (high involvement = 0, low involvement = 1), time (no time restriction = 0, time restriction = 1), and the interaction between involvement and time as independent variables. Furthermore, the attitude towards the story and storytelling in general, age, and gender are included as variables. The correlations between the main variables are presented in table 7 on the following page.
There is a neutral to moderately positive correlation between time and brand attitude, r (199) = .230, p < .05). The more time a consumer has to read the story, the more positive they are about the brand advertised. The acceptance of storytelling in general is also positively correlated with brand attitude (r (199)= .274, p < .05). Consumers’ age negatively correlates with attitude towards the brand, r (199) = -.343, p< .05. The higher the age, the lower the attitude.
The strongest positive correlation is between attitude towards the story and attitude towards the brand, r (199) = .677, p < .05, with attitude towards the story explaining 46% of the variation in brand attitude. The more consumers like the story itself, the more likely it is that they also like the brand advertised in the story. Hypothesis 1b is therefore accepted: consumers’ attitude towards the story is associated with their attitude towards the brand advertised in the story. This is in line with other empirical studies that have documented the explanatory power of attitude towards the ad (Cacioppo & Petty, 1985; MacKenzie et al., 1986; Park & Young 1986).
Table 7 - Pearson correlations for main study variables
Variable name M SD 1 2 3 4 5 6 7 8 1. Brand attitude 4.77 .93 (.87) 2. Gender (1=male, 2= female) 1.49 .50 .328* (-) 3. Age 27.57 9.25 -.343* -.143 (-) 4. Attitude storytelling in general 5.40 1.01 .274* .073 -.191* (.81) 5. Attitude towards the story 4.66 1.20 .677* .281* -.229* .392* (.88) 6. Involvement 1.50 .50 .099 -.042 -.009 -.060 .048 (.82) 7. Time restriction 1.49 .50 .230* .107 -.199* .128 .178* .025 (-) 8. Interaction .01 .25 .143 .091 .012 .114 .042 .000 .001 (-) * Significantly at p <.05.
Before checking which of the variables significantly predict brand attitude, and making any conclusions regarding hypothesis 2c, the assumptions of multiple regression were analyzed (Appendix 7D). There was independence of residuals, as assesses by a Durbin-Watson statistic of 2.377. The independent variables are collectively linearly related to the dependent variable. The residuals are equally spread over the predicted values of the dependent variable, so there is homoscedasticity. None of the independent variables have correlations greater than 0.7 and all the Tolerance values are greater than 0.1 (the lowest is 0,694), so there’s no problem with collinearity.
The studentized deleted residuals were inspected to determine whether there 38
are any residuals greater than 3 standard deviations, and these outliers were subsequently filtered out of the data. Last, there were no cases that have problematic leverage values. Only seven cases had leverage values above the “safe” value of 0.2, the highest being 0,.24859, which is well under the “risky” value of 0.5. Furthermore, Cook’s Distance values were stored for each case, and none of them have values above 1, which means that there are no influential points. In order to be able to run inferential statistics, the residuals needed to be normally distributed. The points are not aligned perfectly along the diagonal line, but are close enough to indicate a normal distribution (see Appendix 7E).
To conclude, the assumptions of linearity, independence of errors, homoscedasticity, unusual points and normality of residuals were met.
Performing the multiple regression analysis, it seems that 50,6% of the variance in the dependent variable can be explained by the independent variables (adj. R2 = 0,506).
Involvement, time, interaction, attitude towards the story, attitude towards storytelling in general, gender, and age significantly predict brand attitude (F (7, 113) = 18.570, p < 0.001). The general form of the equation is:
Predicted brand attitude = 2.506 + ( .228 x gender) – ( .018 x age) – ( .016 x attitude towards storytelling) + ( .451 x attitude towards the story) + ( .132 x involvement) + ( .141 x time restriction) + ( .408 x interaction).
As illustrated in table 8, the main variables (involvement, time and interaction) did not add significantly to the prediction. Attitude towards the story is the most important predictor for brand attitude. Only this variable, and age, added statistically significantly to the prediction. The interaction of involvement and time is almost significant ( p = 0.091). This interaction is visualized in figure 2.
Table 8 - Summary of multiple regression analysis Variable B SE ß ß Intercept 2.506 .527 Gender .228 .125 .123** Age -.018 .007 -.180* Attitude towards storytelling -.016 .065 -.017 Attitude towards the story .451 .057 .586* Involvement .132 .119 .071 Time restriction .141 .112 .077 Interaction .408 .239 .111**
Note. *p< .05, ** p< .10; B = unstandardized regression coefficient; SE ß = standard error of the coefficient; ß = standardized coefficient.
Figure 2 illustrates that the level of involvement moderates the effect of time on brand attitude. The more involved the consumer is, the more important it is that they have time to elaborate on the message. This corresponds to the result that was explained earlier: consumers are most positive about a brand if they have time and are involved with the message. For highly involved consumers, a time restriction can work negatively. They are motivated and are willing to read the story, but are not able to pay the desired attention due to the time limitation. Elaboration on qualitative arguments is suppressed and the effectiveness of the brand story actually backfires. It results in a more negative attitude towards the brand.
In contrast, for consumers that are lowly involved or not involved at all, it does not matter if time is restricted or not. In both situations, they will be equally positive about the brand. This means that hypothesis 2c (“The interaction between consumers’ level of involvement and time affects the relationship between a brand story and attitude towards the brand) is supported under a significance level of p = 0.1. In table 9, a summary of the hypothesis testing can be found.
Tabel 9 - Summary testing of hypotheses
H1a: Participants exposed to a brand story have a more positive attitude towards the brand than participants who were exposed to factual information about the brand.
Not supported.
H1b: Consumers’ attitude towards the story is associated with their attitude towards the brand advertised in the story.
Supported.
H2a: High involved consumers who have time available are more positive towards the brand after reading the story than consumers that lack
involvement and/or time.
Supported.
H2b: For low involved consumers that are restricted in time, factual information about the brand is equally effective as a brand story.
Supported.
H2c: The interaction between consumers’ level of involvement and time affects the relationship between a brand story and attitude towards the brand.
Supported (p < 0.10).
C
C
O
O
N
N
C
C
L
L
U
U
S
S
I
I
O
O
N
N
A
A
N
N
D
D
D
D
I
I
S
S
C
C
U
U
S
S
S
S
I
I
O
O
N
N
In this chapter the results are further interpreted and discussed. This will lead to a thorough clarification of the research question in this study: “How does consumers’ level of involvement and available time affect the relationship between a brand story and the attitude towards the brand?” The theoretical and managerial contributions are presented in this chapter, after which the limitations, some of which lead to suggestions for further research, are displayed.
5.1. Summary of findings
The aim of this thesis was to analyze the effect of a brand story on a consumer’s attitude towards the brand. In particular, the role of consumers’ level of involvement and available time are considered. The main question was whether the effect of storytelling is dependent on the situation of the consumer. In table 9, a summary of the hypotheses testing can be found.
The results showed that there is no difference in brand attitude between participants that were exposed to a brand story, and participants that were exposed to factual information about the brand. This is not in line with earlier research of Lundqvist et al. (2012), who found that consumers exposed to a brand story describe a brand in much more positive terms, and are willing to pay more for it. However, these researchers used brand experience as dependent variable, which is closely related to brand attitude, but not identical (Schmitt et al., 2009). Another explanation could be that a brand story does not work equally well for all types of products. In general, participants admitted that they do not want to spend much money on socks. In their opinion, it is a utilitarian good, which they are using for functional rather than symbolic reasons.
When looking more closely at the brand attitude across different conditions, it turns out that highly involved consumers with time available are more positive towards the brand than consumers that lack involvement and/or time. It is likely that these individuals have followed the central route to persuasion, and based their beliefs on the qualitative arguments in the story. Low involved consumers that did not have enough time to read the story, were equally positive towards the brand as consumers exposed to factual information about the brand. The participants in the control condition could easily elaborate on the presented qualitative arguments, whilst participants in condition four might not have even noticed them due to time limitation.
Similar results can be seen on purchase intention. The overall purchase intention is low, and it cannot be argued that a brand story leads to higher purchase intention. It is most likely that high involved consumers that had time to read the story buy the brand in the future. For low involved consumers that lack time, it is least likely that they will make a purchase. This indicates that the effectiveness of a brand story is in some way dependent on a consumers’ situation.
According to the Elaboration Likelihood Model, the first step towards central route processing is the level of involvement (Petty & Cacioppo, 1986). This corresponds with the findings of this study. It was found that the level of involvement moderates the effect of time on brand attitude. The more involved the consumer is, the more important it is that they have sufficient time to elaborate on the message. In other words, only when consumers’ motivation is high, it becomes important that they are also able to pay attention to the message. On the contrary, for low involved consumers, it does not make any difference if they have time to elaborate on the story or not. In both conditions, they are equally positive about the brand.
