The effect of engaging consumers through stories

(1)

The effect of engaging consumers through stories

Written by Lasse Goud (11147385)

Date of submission: 24 – 06 -2016

Qualification: Msc Business Administration - Marketing

Institution: Amsterdam Business School

(2)

2 Statement of originality

This document is written by Student Lasse Goud 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 responsible solely for the supervision of completion of the work, not for the contents.

Signature:

(3)

3 Acknowledgement

Herewith I would like to take the opportunity to thank the people who supported me in writing this master thesis. Firstly I would like to thank my supervisor Ed Peelen, who has been of great value in the process of creating this document. Secondly, I would like to thank Zwitersland Toerisme for granting me access to their online community in order to collect responses. Last but not least my thanks goes out to friends and family who have been a great mental support in writing this master thesis.

(4)

4

This study investigates how storytelling affects the psychological process in which customer engagement, attitude- and behavioural change take place for the tourism industry. Customer engagement consists of affective, cognitive, social and behavioural elements, which are measured through 5 dimensions (identification, enthusiasm, attention, absorption, interaction) on a 25-item scale. Little research is done on how these elements are affected by storytelling, their impact on the cognitive- and affect-based attitude, and how the attitude types consequently influence the intention to travel. Moreover, the advantage of using factual information without a story is also investigated. The difference in subjective knowledge for the two information types (story and fact) and the effect of subjective knowledge on the attitude towards the destination are measured.

An experimental survey was conducted on a sample of 212 participants. The outcomes clarify that storytelling influences the affect-based attitude via the affective components of customer engagement (enthusiasm and absorption). Moreover, it shows that the intention to travel is affected by the affect-based attitude, whereas the cognitive-based attitude has an insignificant effect on the intention to travel. Additionally this study shows that factual information causes a greater increase in subjective knowledge than storytelling. Nevertheless, subjective knowledge does not influence the attitude towards the destination.

(7)

7

1 Introduction

Stories are as old as mankind and have always fascinated people (Lundqvist, Liljander, Gummerus, & van Riel, 2013). Marketing through storytelling is gaining popularity, especially in the tourism and hospitality industry (P. Tussyadiah, Park, & R. Fesenmaier, 2010). While the power of stories has been studied widely, little empirical evidence exists on the impact of stories on consumer responses and it deserves more attention in brand management literature (Lundqvist, Liljander, Gummerus, & van Riel, 2013). Why is storytelling becoming so popular and how can its effectiveness be explained?

For this study another concept is used to come up with a possible explanation for the effectiveness of storytelling. This is the concept of customer engagement, which is recently gaining more attention by marketing scholars (So K. K., King, Sparks, & Wang, 2014).

Over the past years a growth in the range of new media channels has occurred, which enables companies more easily to connect with their customers beyond the purchase (So, King, & Sparks, 2014). Nevertheless, this growth in new media channels leads to an information overload for the consumer, which eventually makes it more challenging for marketers to connect with their customers. MSI (2010, p. 4) stated that ‘‘many firms see customer engagement as a route for creating, building, and enhancing customer firm relationships and (ultimately) improving business performance’’.

New ways of interaction with the customer have to be developed in order for companies to stand out. A possible way to accomplish this is by offering relevant and/or inspiring content, which is aimed at the customers’ needs. The increase of companies applying content marketing as part of their relationship marketing strategy shows that companies recognize the importance of it in re-connecting with their customers (Wang, Qiao, & Peng, 2015). Yet, there is a lack of understanding about the influence of content marketing on customer engagement and other constructs. This is vital since proactive consumer

(8)

8 engagement is a concept that influences the relationship between customer and company, and has shown to affect the purchase intention (Wang, Qiao, & Peng, 2015).

Previous studies have been investigating the effect of storytelling on behavioral intention towards a travel product (Akgün, Halit , Ayar, & Ebru , 2015; Hsiao, Lu, & Lan, 2013). Those studies discovered that empathy and attitude are mediators in the relationship between storytelling and the behavioural intention. This study will build on the findings of those former studies, by adopting consumer engagement and measure its effect on attitude and the behavioral intention towards a travel destination. This was done through an experimental survey. Additionally, knowledge is added as a variable to our research in order to understand the potential benefit of factual information (non-narrative) over storytelling on the attitude and intention to travel. This is addressed with the following research question:

How do storytelling and factual information affect the attitude and intention to travel towards a destination through customer engagement and knowledge?

Firstly, the literature review will provide a theoretical background for this study. Secondly, a conceptual model will visually present the constructs. Next, the methodology is described followed by the results. Lastly, the implications and limitations of the study are discussed.

(9)

9

2 Literature review

2.1 Storytelling in the marketing context

Content marketing is becoming more important nowadays for companies in attracting and retaining customers. According to Pulizzi (2012) companies need to start acting more like media companies in order to stand out in an environment where the consumer is exposed to an overload of information. Nevertheless media companies create content with the intent of making money directly from it, whereas non-media company content does not have a focus on gaining profit directly from the content, but indirectly from attracting and retaining customers (Pulizzi, 2012). Compared to traditional vehicles, which are mainly focussed on short-term achievements, content marketing is used to generate a positive behaviour from a customer or prospect of the brand (Pulizzi, 2012; P. Kotler and G. Armstrong, 2009, in Liu & Chia-yen Wu, 2011). Building relationships with current and potential consumers triggers them to become an advocate of the business (R. T. Rust, K. N. Lemon, and D. Narayandas, 2005, in Liu & Chia-yen Wu, 2011). This is more important than ever in a world where buyers have gained more control (C. Li and J. Bernoff, Groundswell, 2008, in Liu & Chia-yen Wu, 2011). Guber (2007) indicated that the challenge for firm-originated stories is entering the hearts of the listener in order to touch their emotions. Even though our minds are open, the heart is being protected, as we are aware of its power to move us. Words and ideas that engage the listeners’ emotions are the base of stories. Therefore, storytelling is a noteworthy tool within content marketing to study.

The extant literature on marketing contains a large extent of information about the traditional vehicles, yet little research is done on businesses applying storytelling (Liu & Chia-yen Wu, 2011). Stories are able to promote positive features of a product or service, without being perceived as a commercial (Lundqvist, Liljander, Gummerus, & van Riel,

(10)

10 2013). Nevertheless, little emperical research is conducted on the effects they have on consumer responses in a business setting.

2.2 The role of firm-originated storytelling

Previous researchers claim that storytelling helps companies to inform current and prospective customers what the company is about and what can be expected from it (Jensen, 1999, in Ågren & Ölund, 2007). Nevertheless, informing the consumer can not be seen as sufficient enough in creating a favorable attitude towards the brand. Good stories are able to create positive associations and can embrace the core values of a company better than traditional communication vehicles (Lundqvist, Liljander, Gummerus, & van Riel, 2013). Consumers that are exposed to a story express high levels of emotional enthusiasm, which could possibly develop into an emotional connection with the brand (Lundqvist, Liljander, Gummerus, & van Riel, 2013). Therefore the biggest challenge for companies nowadays is not the mere creation of valuable and compelling content, but the creation of engaging content (Pulizzi, 2012). This thesis suggests that storytelling can be applied as a tool to enhance the engagement and subsequently affect the attitude towards a company.

The impact of storytelling on customers’ behaviour can be explained with ‘the transportation into a narrative world’, which describes ‘‘the process of becoming fully engaged in a story’’ (Green & Brock 2000, 2002 in Green, Brock & Kaufman, 2004, p. 312). The transported individual, whom has the feeling to be departed from reality and entered the imagined world of a story, is in this particular state more prone to attitude change (Green & Clark, 2013). According to Green and Clark (2013, p.477) narratives are experienced through ‘‘high levels of cognitive and affective engagement, and may form vivid mental images’’. Immersion into a story reduces counterarguing, increases connection with the character, increases perceptions of realism and heightenes the emotional experience (Green & Clark,

(11)

11 2013). This demonstrates that narrative transportation in a firm-originated story can be seen as one of the causes for an individual to change their attitude towards a brand or product.

2.3 The elements of firm-originated storytelling

Previous research on stories has found that a story included several elements. First of all, a story should always include a beginning, a middle and an end (Aristotle, 1987, in Chiu, Hsieh, & Kuo, 2012). The beginning leads to a problem and creates a climax, which is tackled with the use of a resolution at the end of the story (Freytag. 1863, in Casebeer, 2008) Additionally, a narrative should include characters and a lesson learned (Aristotle, 1987, in Chiu, Hsieh, & Kuo, 2012). It presents product information in a more conversation-like approach than factual information (Peracchio & Meyers-Levy, 1997).

Chiu, Hsieh, and Kuo (2012) stated four elements that contribute to a good brand story. These elements are authenticity, conciseness, reversal and humor and can be used to engage readers in product evaluations and have shown to positively affect brand attitude and purchase intention. Fog, Budtz, and Yakaboylu (2005) also stated four important elements that a story should contain, namely a message, conflict, plot and characters. The authors mention that the elements are necessary to ensure that the content can be recognized as a story. It is stated that the message should cultivate positive associations with the brand, the conflict should be resolved after putting the reader out of tune for a moment, the plot should maintain audience interest, and the characters should include a hero and an adversary. According to Peracchio and Escalas (2008) a plot should contain something unusual or unexpected, which Fog, Budtz, and Yakaboylu (2005) describe as the ‘conflict element’.

Akgün, Halit , Ayar, and Ebru (2015) have recently studied the components of storytelling and the effect these elements have on travel intentions in the tourism industry. This study revealed that 3 elements, namely perceived aesthetics, narrative structure and self-reference, evoke reader empathy and change in attitude towards the product.

(12)

12 Similar research by Hsiao, Lu, & Lan (2013) on this topic exposed that perceived aesthetics (of all three elements) has the most significant total effect on intention. Making a story visible through photos, layout design and audio, and good aesthetics can have an impact on the attraction of the stories.

2.4 Storytelling as a marketing tool in the tourism industry

Research shows that vacations are evaluated more favourably when they are described through a narrative instead of a list of features (Adaval & Wyer, 1998). Therefore it is no surprise that travel companies are becoming more aware of storytelling as a marketing tool (Hsiao, Lu, & Lan, 2013). In 2009, about 30 percent of US travelers used travel-related blogs before making travel plans (US Trave Association, 2009 in Hsiao, Lu & Lan, 2013). Even though this does not inidicate whether they were influenced by stories, it does proof that creating good content is valuable.

A travel product is an experience good (McIntosh, 1972 in Hsiao, Lu, & Lan, 2013), hence it is more difficult to promote its features. This limitation could be minimized through the use of storytelling, as it enhances the perceived quality of the travel product by giving the consumer a better impression.

Previous studies show that empathy is a mediator in the relationship between storytelling and attitude in the tourism industry. Nevertheless customer engagement, which is recently gaining more attention by marketing scholars, would be a valuable variable to investigate. Narratives are experienced through high levels of cognitive and affective engagement and heighten the emotional experience (Green & Clark, 2013). Therefore, the aim of this study is to understand how customer engagement is affected by storytelling and how the impact of this effect changes the attitude and intention to travel towards a destination.

(13)

13 2.5 Customer engagement and its dimensions

According to Vivek, Beatty and Morgan (2012. p. 127) customer engagement refers to ‘‘individuals who interact with the brand, without necessarily purchasing it or planning on purchasing it, or on events and activities engaged in by the consumer that are not directly related to search, alternative evaluation, and decision making involving brand choice’’. Customer engagement is, as the MSI (2010, p. 4) states, ‘‘a behavioural manifestation toward a brand or firm beyond the purchase’’.

Non-transactional behaviors are becoming more important (Verhoef, Reinartz, & Krafft, 2010). So, King and Sparks (2014, p. 2) describe this non-transaction behavior as ‘‘a customers’ personal connection to a brand as manifested in cognitive, affective, and behavioral actions outside of the purchase situation’’. A relationship can be easily build in an online environment that engages a customer that has not bought a product yet. It can motivate consumers to do something beyond what is required in the contract. Having frequent interactions with both current and potential customers could strengthen the emotional investment the consumer has in the brand (Chaffey, 2007, in Sashi, 2012). It is very likely that by using the correct tools these interactions will positively influence the emotional bond between consumer and brand.

According to Roderick, Ilic, Juric, and Hollebeek (2013) the concept of customer engagement focuses on interactive consumer experiences. Nevertheless, many definitions have been used to describe the concept. Table 1 by Vivek S. D., Beatty, Vivek, and Morgan (2014) includes a set of definitions that describe CE and has been partialy extended to show the 4 elements through which CE can be manifested, namely cognitively, affectively, socially and behaviorally (Vivek, Beatty, & Morgan, 2012) :

(14)

14

Authors: Definition: Object/Term Used: Component:

Vivek, Beatty, and Morgan (2012, p. 133)

The intensity of an

individual’s participation in and connection with an organization’s offerings and/or organizational activities, which either the customer

or the organization initiate.

Consumer engagement Behavioral

Mollen and Wilson (2010, p. 922)

The customer’s cognitive and affective commitment to an active relationship with

the brand as personified by the Web site or other computer-mediated entities designed to communicate brand value. It is

characterized by the dimensions of

dynamic and sustained cognitive processing and the satisfying of instrument value

and experiential value.

(Online or computer-mediated

entities) Customer engagement

Cognitive and affective

Sashi (2012, p. 267)

It embodies interactive consumer experiences where ICTs such as social media act

as tools that can enable and facilitate these experiences. The level of consumer engagement is calculative and affective commitment to an active relationship with

a firm or the firm’s online community.

Consumer engagement process

Social, cognitive, affective and behavoiral

Brodie et al. (2011b, p. 260)

“[A] psychological state that occurs by virtue of interactive, cocreative customer

experiences with a focal agent/object (e.g., a brand) in focal service

relationships.”

Customer engagement Social and behavioral

Hollebeek (2011, p. 790)

The level of a customer’s motivational, brand-related and context-dependent state of mind characterized by specific levels of cognitive, emotional, and behavioral activity in brand

interactions. It includes the themes of immersion, passion, and activation. Customer–brand engagement Cognitive, affective, behavioral and social

(15)

15 Gambetti, Graffigna, and Biraghi (2012, p. 668) Customer-brand engagement appears as a multi-dimensional concept combining

such elements as attention, dialogue, interaction, emotions, sensorial pleasure,

and immediate activation aimed at creating a total brand experience with consumers.

Advertising/media engagement

but referred to as customer– brand

engagement

Social, behavioral, affective, cognitive

Van Doorn et al. (2010, p. 254)

Customer engagement behaviors go beyond transactions and are defined as a

customer’s behavioral manifestations that have a brand or firm focus, beyond purchase, resulting from motivation drivers.

Consumer engagement behaviors

Behavioral and cognitive

Higgins and Scholer (2009, p. 112)

A state of being involved, occupied, fully absorbed, or engrossed in something (i.e., sustained attention), generating the

consequences of a particular attraction or repulsion force.

Strength of engagement Affective

Table 1 Vivek s. D., Beatty, Vivek, & Morgan (2014)

The following broad definition by Roderick, Ilic, Juric, and Hollebeek (2013) was slightly adjusted in order to cover many of the definitions that were used in previous literature and defines consumer engagement and its 4 elements (Vivek, Beatty, & Morgan, 2012):

‘‘’Consumer engagement involves specific interactive experiences between consumers and the brand, and/or other members of the community. Consumer engagement is a context-dependent, psychological state characterized by fluctuating intensity levels that occur within dynamic, iterative engagement processes. Consumer engagement is a multidimensional concept comprising cognitive, emotional, social and/ or behavioral dimensions, and plays a central role in the process of relational exchange where other relational concepts are engagement antecedents and/or consequences in iterative engagement processes within the brand community’’ (Roderick, Ilic, Juric, & Hollebeek, 2013, p. 3)

(16)

16 This definition clarifies that customer engagement involves experiences between a brand and its’ customers. An example of this experience could be a story. Moreover, this definition includes the cognitive, emotional, social and/or behavioral elements of customer engagement. These elements are used as a base to understand how customer engagement should be measured. The scale that will be used within this study touches upon all these dimensions. Nevertheless, this scale and its relation with the elements will be discussed in further detail later. Storytelling as an experience involves high levels of cognitive engagement and affects emotions (Green & Clark, 2013). Therefore it can be assumed that storytelling could influence the elements of customer engagement and, through this affect, the attitude and intention of the customer. Moreover, the engaged customer develops an emotional attachement to the organisation (Gallup consulting, 2010, in Sashi, 2012; Roderick, Ilic, Juric, & Hollebeek, 2013). Hence, creating emotions through storytelling could strengthen the relationship with the company on an emotional level as well. Nevertheless, as aforementioned, an appropriate scale containing validated dimensions should be used to measure the effect of storytelling on the 4 elements (cognitive, affective, social and behavioral) of customer engagement.

Schaufeli, Salanova, Vincente, and Bakker (2002, p. 74) study employee engagement, which is defined as ‘‘a positive, fulfilling, work-related state of mind’’ and represents 3 dimensions, namely vigor, dedication, and absorption. Vigor is labelled as having high levels of energy and being resilient in the face of difficulties. Dedication is described as feelings of ‘‘significance, enthusiasm, inspiration, pride, and challenge’’ (Schaufeli, Salanova, Vincente, & Bakker, 2002, p. 74). Absorption occurs when someone is fully engrossed in an activity and time pases quickly. One loses track of the world around him/her and has difficulties ending the activity. Customer engagement is similar to employee engagement. Nonetheless, it includes a ‘‘strong behavioral focus’’ (So, King, & Sparks, 2014, p. 306). Moreover, the social element of customer engagement is not measured with this scale. Therefore other

(17)

17 studies made an effort to measure customer engagement and included dimensions that signify the behavioral and social aspect.

Vivek, Beatty and Morgan (2012) developed and validated a 10-item CE scale, which is able to measure CE on 3 dimensions, namely conscious attention, enthused participation and social connection. When comparing the dimensions of CE-scale to the EE-scale it shows that enthused participation shows similarities with vigour, because a certain kind of energy arises when being engaged. Dedication and social connection are the dimensions that focus on the social element of customer engagement. Also conscious attention and absorption concern high involvement in completing the task, without being disrupted.

The CE-scale can be used to develop effective engagement strategies, by addressing the various venues the firm considers to use (Vivek S. D., Beatty, Vivek, & Morgan, 2014). Nevertheless, So, King, and Sparks (2014) also developed and validated a measurement 25-item CE scale, which can be seen as an extension of the 10-25-item scale and consists of 5 factors: identification, enthusiasm, attention, absorption, and interaction. This scale is in line with the aforementioned definition, since it measures all 4 elements of customer engagement (cognitive, affective, social and behavioural). Enthusiasm and absorption cover the affective aspect of customer engagement (Harmon-Jones, Price & Gable, 2012; Higgins & Scholer, 2009). Identification and attention are considered to be the cognitive components of the customer engagement (So, King, & Sparks, 2014). Interaction, which is the last dimension on this scale, covers the behavioral and social aspect of customer engagment (Vivek S. , 2009). These dimensions will be discussed in further detail later.

The 25-item scale is ‘‘a useful tool for practitioners to gain insight into customer psychological behavioural connections with their brands beyond the service consumption experience’’ (So, King, & Sparks, 2014, p. 304). The effectiveness of marketing strategies can be measured by surveying customers prior to and after a certain marketing strategy with the items on the CE scale (So, King, & Sparks, 2014).

(18)

18 2.6 The effect on the cognitive and affective components of attitude

Vivek, Beatty, and Morgan (2012, p. 133) describe customer engagement as ‘‘the intensity of an indivual’s participation in and connection with an organization’s offerings or organizaiton activities, which either the customer or the organization initiates’’. As aforementioned in the definition, customer engagement can be manifested either cognitively, affectively, behaviorally or socially (Vivek, Beatty, & Morgan, 2012) . Cognitive and affective elements include experiences and feelings, and the behavioral and social elements embrace the participation of both current and potential customers.

This thesis will focus on the effect of all the dimensions of customer engagement on the affective, cognitive and consequently the behavioral elements of the attitude towards a destination. San Martin and Rodriguez Del Bosque (2008, p. 274) state that through storytelling a ‘‘mental representation of a tourist destination is formed on the basis of individuals’ belief about the place (cognitive image), as well as their feelings towards it (affective image) ’’.

A study by Millar and Tesser (1986) demonstrated that affective components lead to affectively driven behavior and cognitive components lead to cognitively driven behaviors. Hence, affect and cognition are two distinguishable components of attitude (Breckler & Wiggins, 1989). Breckler and Wiggins (1989, p. 253) define affect as ‘‘emotional responses and feelings engendered by an attitude object’’ and cognition or evaluation is described as ‘‘thoughts, beliefs, and judgments about an attitude object’’. Previous research indicates that narratives elecit stronger emotional reactions, compared to content-equivalent information (Kopfman, Smith, Ah Yun, & Hodges, 1998 ; McQueen & Kreuter, 2010). A study by McQueen and Kreuter (2010) showed that a narrative video increased the engagement, which as aformentioned could reduce counterarguing (Green & Clark, 2013) and is therefore more likely to increase the attitude. Moreover, respondents reading statistical evidence messages produce a higher number of thoughts than respondents who are exposed to a narrative

(19)

19 (Kopfman, Smith, Ah Yun, & Hodges, 1998). Hence, the narrative and the factual information will presummably differ in the impact they have on the affect-based and the cognitive-based attitude. Nevertheless, these elements of attitude will be affected differently through the 5 dimensions of customer engagement.

1. Identification

A strong consumer-company relationship can help the customer satisfy their important self-definitional needs (Bhattacharya & Sen, 2003). Hultman, Skarmeas, Oghazi, and Behesthi (2015) reveal in their study that tourists ascribe distinct personalty characteristic to destinations, which is defined as tourist-destination identification. According to the authors, the personality of a destination makes the tourist identify him/herself with the destination and leads to promotion and revisit intentions. A study by McQueen and Kreuter (2010) showed that participants were more likely to identify with cancer survivors when being exposed to a story. Moreover, exposure to stimulus materials that contain brand personality and elicit identification have a signifincant impact on the attitude towards the brand (Freling & Forbes, 2013 ; Shen & Chiou, 2009). So, King, and Sparks (2014, p. 310) state that identification is a ‘‘cognitive component that justifies consumers’ engagement behaviors’’. This is in line with the statement by Currás-Pérez, Bigné-Alcañiz, and Alvarado-Herrera (2009), which confirms that C-C identification is a purely cognitive variable. Therefore the process of identification is most likely to affect the cognitive-based attitude. This brings us to the the following hypotheses:

H1a: Storytelling elicits a higher level of identification with the destination than factual information.

H1b: Identification with the destination increases the cognitive-based attitude towards the destination more than the affect-based attitude.

(20)

20 2. Enthusiasm

Enthusiasm is the high level of energy that arises when exposed to a focal engagement object (So, King, & Sparks, 2014). Vivek S. (2009, p. 60) describes enthusiasm as ‘‘strong excitement or zeal about the focus of engagement’’. As mentioned before, consumers that are exposed to a story express high levels of emotional enthusiasm, which could possibly develop into an emotional connection with the brand (Lundqvist, Liljander, Gummerus, & van Riel, 2013). This dimension is similar to the dimensions of dedication and vigor, which are used by Schaufeli, Salanova, Vincente, and Bakker, (2002). Harmon-Jones, Price and Gable (2012) state that enthusiasm is an affective state, which narrows the cognitive scope. Therefore, it is very likely that enthusiasm will impact the affective component of attitude. Enthusiastic consumers are highly excited about getting in contact with the product or service. The idea of visiting a destination after being exposed to a story is expected to influence their affect-based attitude towards this destination. The following hypotheses are constructed on the abovementioned findings:

H2a: Storytelling elicits a higher level of enthusiasm for the destination than factual information.

H2b: Enthusiasm for the destination increases the affect-based attitude towards the destination more than the cognitive-based attitude.

3. Attention

Engaged customers have a higher interest in information that is related to the brand (So, King, & Sparks, 2014). Higgins and Scholer (2009) mention that engaged consumers are in a state of sustained attention. If an unpleasant task needs to be completed it is more likely for people to turn their attention away, which will subsequently lower their level of engagement (Higgins & Scholer, 2009). Consumers are more likely to pay attention when for

(21)

21 example being exposed to information about a tourist destination they are engaged with. The attention can possibly be created through the use of storytelling. Stories can be used to facilitate the information processing and capture the attention of the consumer, which will make the person more likely to get a better understanding and recall (Kreuter, et al., 2010). Attention can be generated through sensational value, but could possibly influence the cognitive forces to appear afterwards (Donohew, Lorch, & Palmgreen, 1998). As the story grabs the attention, more thinking could occur about the negative and the positive aspects of the destination. The following hypotheses are based on the foregoing literature:

H3a: Storytelling elicits a higher level of attention towards the destination than factual information.

H3b: Attention towards the destination increases the cognitive-based attitude towards the destination more than the affect-based attitude.

4. Absorption

Higgins and Scholer (2009, p. 102) define engagement as ‘‘a state of being involved, occupied, fully absorbed, or engrossed in something’’. This type of absorption is similar to the concept of ‘flow’, which is the experience of being engaged in an activity that brings energetic focus, creative concentration and peak enjoyment (Csikszentmihalyi, 2000).

Absorption embodies an effortless concentration in which a person loses self-consciousness, has little to no realization of time and experiences an intrinsic enjoyment. When a transported individual enters a story, he/she feels the departure from reality and is more prone to attitude change (Green & Clark, 2013). Absorption could eventually lead to a level of attraction to, or repulsion from a focal engagement object (Higgins & Scholer, 2009). The strength of absorption within engagement is aligned with how we affectively value something (Higgins & Scholer, 2009). Hence, the affect-based attitude towards the destination can be affected

(22)

22 through absorption, when the destination is used as the focal engagement object. The following hypotheses were developed:

H4a: Storytelling elicits a higher level of absorption with the destination than factual information

H4b: Absorption with the destination increases the affect-based attitude towards the destination more than the cognitive-based attitude.

5. Interaction

According to So, King and Sparks (2014, p. 309) interaction is ‘‘a customer’s online and offline participation with the brand or other customers outside of purchase’’. Interaction is also one of the dimensions on the 10-item scale by Vivek (2009). In this study interaction is described as ‘‘interchange of ideas, thoughts, and feelings with other people about the engaged consumer’s participation and the focus of engagement’’ (Vivek S. , 2009, p. 61). Vivek (2009, p. 61) ascribes a higher level of engagement to an increase in the exchange of ‘‘experiences, thoughts and feelings with others’’. People are more likely to interact by sharing information about the brand with others after being engaged. A lot of information that is stored in the memory is episodic (Fournier, 1998; Schank, 1990, in Woodside, Sood, & Miller, 2008) and it is therefore easier to recall a story about the brand, which makes it easier to share as well. Vivek (2009, p. 61) also mentions that ‘‘interaction plays a more significant role in cases where engagement generates social benefits’’. This is in line with the tourist-destination identification, which is mentioned before as part of identification. The more a person identifies with a destination, the more likely it will become that a person uses the destination in dialogue with others. These dialogues can concern either thoughts and ideas or feelings. Having frequent interactions with both current and potential customers could strengthen the emotional investment the consumer has in the brand (Chaffey, 2007, in Sashi,

(23)

23 2012). Hence, it is very likely that the consumer will share his/her feelings after being exposed to the information, which presumably increases the affect-based attitude to a higher extend. These findings are hypothesized as follows:

H5a: Storytelling elicits a higher level of interaction than factual information. H5b: Interaction increases the affect-based attitude towards the destination more than the cognitive-based attitude.

There has been a rise of interest in the concept of customer engagement since the upcoming of social media, which has led us to a search for approaches that have an effect on the engagement of the customer (Sashi, 2012). A study by Hsiao, Lu, and Lan (2013) shows that storytelling has an impact on purchase intention. Nevertheless this influence of storytelling is mediated through reader empathy and attitude and little is known about other constructs (Hsiao, Lu, & Lan, 2013; Akgün, Halit , Ayar, & Ebru , 2015). This provides the opportunity to study the effectiveness of engaging content on attitude which could be accomplished through storytelling.

2.7 The effect of factual information on cogitive-based attitude change

The aforementioned hypotheses predict that storytelling elicits higher outcomes for each of the dimensions of customer engagement than factual information does. Moreover it is expected that eventually it is more likely that the affect-based attitude will be influenced by engagement through storytelling.

Former research on storytelling is mainly biased towards the advantages of storytelling and little research is conducted on the positive effect of factual information. Therefore this study will consider a possible advantage of using factual information over storytelling. Previous studies show that affect precedes cognition in attitude formation when affective

(24)

24 means of persuasion are used; whereas cognition precedes affect when cognitive means of persuasion are used (Edwards, 1990). A study by Kopfman, Smith, Ah Yun, and Hodges (1998) demonstrates that statistical evidence messages produce more thoughts than narrative messages. Aligned with this theory, it can be assumed that factual information is a cognitive mean of persusassion and influences the cognitive-based attitude to a greater extend. Hence, other factors should be taken into account to explain an alternative affect of factual information on the cognitive-based attitude. The attitude towards an organization is affected by the knowledge one has about the organisation, which is a cognitive variable (Morgan, Stephenson, Harrison, Affi, & Long, 2008). It is very likely that people who are exposed to factual information will show an increase in knowledge, given that they become more educated about the organisation.

Previous research shows that knowledge can be measured in two different ways, either subjectively (individuals perceived knowledge) or objectively (knowledge stored in memory) (Brucks, 1985). This study will only cover the subjective knowledge given that the two videos contain different information, which will give an unreliable representation of the increase in objective knowledge. Nevertheless, subjective knowledge is likely to increase after being exposed to a video that contains factual information. People could feel more knowledgeable after receiving facts, which eventually is assumed to affect their cognitive attitude.

The following was hypothesized in order to investigate how factual information, through the construct of subjective knowledge, could be a greater influencer of the cognitive attitude than storytelling:

H6a: Factual information elicits a higher level of subjective knowledge than storytelling. H6b: High levels of subjective knowledge increases the cognitive-based attitude towards the destination more than the affect-based attitude.

(25)

25 2.8 Purchase intention

As aforementioned, Hsiao, Lu, and Lan (2013) reveal that empathy and attitude towards the brand indirectly mediate the relationship between storytelling and behavioural intention towards a travel product. Nevertheless, these former studies completely focus on the effect of empathy that arises from storytelling to explain intention, whereas other relational constructs should also be studied (Hsiao, Lu, & Lan, 2013)

With regard to engagement, a study by Wang, Qiao, and Peng (2015) has revealed that there is a positive relationship between the evaluative valence of proactive engagement and purchase intention. Additionally, it is suggested that ‘‘the evaluative aspect of the emotional process links engagement with purchase intention’’ (Wang, Qiao, & Peng, 2015, p. 197). Correspondingly, other studies found that affect and feelings have a higher impact on behavior, sometimes even leading to ‘‘twice the variance toward conative attitude’’ (Morris, Woo, Geason, & Kim, 2002, p.14 ; Farley & Stasson, 2003). Moreover, Vivek S. , Beatty, Dalela, and Morgan (2014) mentioned patronage intent as a relationship outcome of customer engagement. Hence, it can be assumed that customer engagement influences the customers’ intentions towards purchasing a travel product. This is in line with a study by Merchant, Ford, and Sargeant (2010) on charitable organizations’ storytelling, which proofs the effect of story content on both emotions and purchase intentions. Furthermore, a study by Chiu, Hsieh, and Kuo (2012) also shows that a brand story has a positive impact on brand attitude and purchase intention. Nevertheless, research needs to be conducted on how customer engagement affects the attitude and thereafter the behavioural intention, when stories are used. When looking at the dimensions it can be assumed that stories have a greater impact on attitude. This could give an additional explanation for the effectiveness of stories on behavioural intention to travel to a destination. Therefore the following hypothesis is investigated:

(26)

26 H7: Affect-based attitude towards the destination has a higher effect on behavioural intention to travel than a cognitive-based attitude

(27)

27

3 Conceptual Model

The hypotheses that derived from our literature review are visualized trough the connecting lines in the following conceptual model:

(28)

28

4 Method

4.1 Procedure and experimental conditions

A quantitative online experiment was conducted in which the participants were randomly assigned to one of two groups (storytelling vs. factual). The sample included both community members of a Swiss destination management organization, but was also distributed through snowball sampling. This created a sample that consisted of people that already had a certain degree of engagement towards the country and people who did not. The first group (fact group) was exposed to a short video providing factual information about Switzerland. The second group (storytelling group) was exposed to a video that included the elements of a story.

The factual video described the same destination, namely Switzerland. Nevertheless it did not include the aforementioned elements of a story. The factual information presented revelatory facts about Switzerland. A pre-test was conducted in order to make sure that the information did not include the elements of a story.

Both groups also received a number of questions after the experiment to ensure the internal validity. In this way confounding variables were recognized that could cause a variation in the dependent variable, which gave the opportunity to filter the results at a later stage. Examples of such factors could be an excessive pre-existing preference for Switzerland.

4.2 Measures

Customer engagement was measured according to the 5 dimensions of the 25-item customer engagement scale: identification, enthusiasm, attention, absorption and interaction. This gave a clear view on how customer engagement was affected by the story compared to factual information. Every dimension contains measurable items that were measured with a 7-point Likert scale. Nevertheless, the items had to be modified in order to measure the engagement with a destination instead of a brand.

(29)

29 Moreover items were created that tested the construct of knowledge. The items of knowledge were, just as the components of customer engagement, measured on a 7-point Likert scale in order to assure that all outcomes could be easily compared.

Subsequently, attitude was measured on a cognitive and affective level. Cognitive and affective responses were measured on a bipolar evaluative continuum anchored at both ends (Kim & Stepchenkova, 2015). Within this experiment deviation from the neutral or zero point on the 7-point semantic differential scale was used to measure whether a cognitive or an affective attitude effect had appeared for the items on the scale.

The intention of travelling towards Switzerland is a dependent variable. The effect on this variable was expected to be different for the affect-based attitude towards Switzerland than for the cognitive-based attitude. The intention to travel was measured with the purchase intent scale developed by Hsiao, Lin, Wang, Lu, and Yu (2010). Nevertheless, the items were partly modified in order to measure the purchase intention towards a destination.

4.3 Manipulation check

A pre-test was conducted in order to test whether the treatments were measuring the correct information type; either storytelling or factual information. A sample of 13 participants participated in the pre-test, in which the two videos were shown and questions were asked that measured the elements of storytelling on a 7-point Likert scale. Seven participants were exposed to the factual video and six participants where exposed to the story video. A mean variable was computed for all the scales.

An independent samples t-test was conducted in order to compare the means of the two treatments. There was a difference between the story (M = 4.98, SD = 1.19) and the factual video (M = 3.87, SD = 1.17). This means that the story video scored higher on the story elements than the factual video. Therefore, the videos were appropriate for use in the main study. Nevertheless, the difference between these two videos was not significant. This

(30)

30 is due to the small sample size (N = 13). Hence, it was decided to retain the manipulation check in the main study.

An increase in mean difference was observed for the results of the manipulation check in the main study. The mean of the story video was higher (M = 5.04, SD = 0.85) than the mean of the factual video (M = 3.82, SD = 1.03). The difference between these two videos appeared to be significant t(210) = 9.399, p<0.01. Therefore, it can be concluded that the story video did clearly include the elements of a story, whereas the factual video did this to a significantly lesser extent.

(31)

31

5 Results

In this chapter all the results that were collected from the experiment are analyzed. Firstly, preliminary steps were taken through a descriptive analysis, frequency check, and a reliability analysis. Secondly, a correlation analysis was done and the hypotheses were tested. Lastly, a check for mediation was conducted as a supplemental analysis of the main study.

5.1 Indicator level: Preliminary steps 5.1.1 Descriptive analysis

All data was collected through an online experiment that started on the 30th of April 2016 and was closed on the 16th of May 2016. 216 participants completed the survey. Moreover, after cleaning the data, 4 outliers were detected by calculating the Z-score for the total means of all items (Field, 2003). The variables cognitive-based attitude had 2 outliers and some outliers were detected in the manipulation questions. The participants with outliers were deleted, which resulted in a sample of 212 participants. 103 out of the 212 respondents (48.6%) were exposed to the first treatment, which was a video containing merely factual information. 109 out of the 212 respondents (51.4%) were exposed to the second treatment, which was a video that included a story. In the factual treatment, 95% had never seen the video before, whereas in the story treatment 73% had never seen the video before.

The gender division within the experiment was 38.7% male and 62.3% female. The majority of the respondents were Dutch (85.4%). Other nationalities that were represented are Belgium (7.5%), Swiss (2.4%), German (1.4%), Norwegian (0.9%), Surinamese (0.9%), Bosnian (0.5%), Iraqi (0.5%) and Serbian (0.5%). The Netherlands also showed to be the current country of residence for most respondents (88.7%), followed by Belgium (7.5%), Switzerland (2.8%) and England (0.9%).

(32)

32 Exactly 50% of the respondents were aged between 20 and 39 years old. Another age group that was well represented in the experiment is the group of people between 40 – 59 years old (35.8%), followed by 60 – 69 (9.4%), 70+ (3.8%) and younger than 20 (0.9%).

30.7% of the respondents travel abroad 2 times per year, followed by 20.8% who travel abroad 1 time per year, 17.9% 3 times, 11.8% 4 times, 6.6% 5 times and 2.4% 6 times. About 8.8% travels abroad 7 times or more per year. Of the 212 respondents, 29.7% has travelled between 1-5 times to Switzerland in the past 10 years and 27.8% has travelled to Switzerland more than 10 times. 26.9% has not travelled to Switzerland in the past 10 years and 15.6% has travelled between 5 - 10 times. This shows that the majority did travel to Switzerland in the past 10 years at least 1 time.

More than half of the respondents (57.5%) felt a personal connection with Switzerland and 14.2 % felt somewhat personally connected with Switzerland. 28.3 % did not feel any personal connection with the country. From these results it can be concluded that the majority of respondents somehow felt that they had a personal connection with the country.

No counter-indicative items were used in this experimental survey. We continue with a reliability analysis.

5.1.2 Reliability analysis

The next step is to test the consistency of the findings by performing a reliability analysis. Similarly to the previous preliminary steps, the two treatment groups were analyzed separately, in order to test whether the two groups had the same structure. The Cronbach’s Alpha values for the factual treatment group ranged from 0.66 to 0.94. Only two variables, namely cognitive- and affect-based attitude, showed a Cronbach’s Alpha that was lower than 0.7 (Field, 2003) and the total Cronbach’s Alpha for these variables could not be raised to a value higher than 0.7 by deleting any items. However, the Cronbach’s Alpha for all other variables was very high ranging between 0.87 and 0.94.

(33)

33 The Cronbach’s Alpha for the storytelling treatment group ranged from 0.60 to 0.95. Here we observe again that the Cronbach’s Alpha for cognitive- and affect-based attitude did not surpass the threshold of 0.7. Nevertheless, it was possible to raise the affect-based attitude to 0.70 by deleting one item. Hence, this item was deleted.

According to Field (2003) the bear minimum for an Alpha is 0.5. Therefore we decided to continue with the few variables that did not have a Cronbach’s Alpha that exceeded 0.7. The resemblance of the Cronbach’s Alpha of the two-treatment groups indicates that the two groups have a similar structure.

The Shapiro-Wilk test checked the normality of the data (Field, 2003). Apart from identification and interaction for the factual treatment group and cognitive-based attitude for both groups, all other variables appeared to differ significantly from normality. This might be an indication for the absence of normality. Nevertheless, through further analysis of the QQ-plots and the histograms we were able to see that the normality did not deviate drastically from normal. Moreover, according to Central Limit Theorem the sample is normally distributed when the sample is large (N > 30) (Field, 2003). Therefore we assume that the data meets the assumptions for parametric tests, since the sample includes 212 participants.

5.1.3 Correlation matrix

Table 1 represents the correlation matrix for the variables, including the newly computed variable for affect-based attitude (after item deletion). We can see that many significant correlations were found. With regard to our hypotheses this Table shows that there is no significant correlation between information type and three dimensions of customer engagement, namely identification (r = -0.01, p > 0.05), attention (r = 0.11 p > 0.05) and interaction (r = 0.05, p > 0.05). However, the two remaining dimensions of customer engagement, namely enthusiasm (r = -0.15, p < 0.05) and absorption (r = -0.21, p < 0.01) show a significant correlation. In search for support of the relationship between information type and knowledge we can also see that there is a very significant relationship between these

(34)

34 variables (r = 0.44, p < 0.01). Moreover, the relationship between knowledge and the cognitive- (r = 0.17, p < 0.01) and affect-based attitude (r = 0.12, p < 0.05) also shows to be significant. Lastly, we also observe that the cognitive- (r = 0.26, r < 0.01) and the affect-based attitude (r = 0.32, p < 0.01) correlate significantly with the intention to travel. Hypothesis testing will provide more information about the relationships between these variables.

Variable M SD 1 2 3 4 5 6 7 8 9 10 1. Information Type 0.49 0.5 1 2. Identification 4.29 1.38 -0.01 1 3. Enthusiasm 4.69 1.43 -.15* .78** 1 4. Attention 4.43 1.45 0.11 .80** .72** 1 5. Absorption 4.16 1.45 -.21** .63** .73** .62** 1 6. Interaction 3.83 1.48 0.05 .78** .67** .82** .63** 1 7. Knowledge 4.23 1.68 .44** .47** .44** .53** .34** .53** 1 8. Cognitive-based attitude 2.18 0.36 -0.06 .25** .27** .19** .25** .25** .17** 1 9. Affect-based attitude 2.36 0.49 -.17** .38** .38** .30** .39** .32** .12* .53** 1 10. Intention to travel 3.33 0.99 0.09 .65** .61** .72** .52** .71** .54** .26** .32** 1 * Correlation is significant at the 0.05 level (1-tailed).

** Correlation is significant at the 0.01 level (1-tailed).

Table 2 Means, Standard Deviations & Correlations

5.2 Construct level: Testing Hypotheses Construct 1

Firstly, we compared the means of the different treatments for the 5 dimensions of customer engagement. A test of homogeneity of variance showed that the samples were homogenous for all these variables. Table 2 shows the results for the effect of information type on both the dimensions of customer engagement and knowledge.

Treatment M SD t df Sig. (1-tailed)

Identification Story 4.31 0.14 0.186 210 0.43

Fact 4.27 0.31

Enthusiasm Story 4.90 0.13 2.139 210 0.02

(35)

35 Attention Story 4.28 0.14 −1.552 210 0.07 Fact 4.59 0.14 Absorption Story 4.46 0.13 3.118 210 0.00 Fact 3.85 0.15 Interaction Story 3.90 0.14 -656 210 0.26 Fact 3.76 0.14 Knowledge Story 3.52 0.15 -7.032 210 0.00 Fact 4.98 0.15 Table 3 Indepedent t-test CE Dimensions & Knowledge

The first dimension that was tested is identification. An independent t-test was applied in order to find the difference between the means of the two treatment groups. The independent samples t-test showed that there was no significant difference in means when comparing the two treatments t(210) = 0.186, p > 0.05. Two additional tests were conducted in which for the first test only female cases were selected and for the second test only participants older than 40 years old were selected. This was done, since the factual video had a youthful male character and the story video had an older female character. It could be the case that the participants identify themselves more easily with a character that is similar to their own. Nevertheless, this was not the case, as the difference between means for both tests was not significant. Therefore, hypothesis 1a is not supported.

The second dimension of customer engagement is enthusiasm. The difference between means within the two treatments for this variable was significant t(210) = 2.139, p < 0.05. The independent t-test showed that there was a higher level of enthusiasm for the storytelling treatment (M = 4.9, SE = 0.13) than for the factual treatment (M = 4.5, SE = 0.15). Hence, hypothesis 2a is supported.

The third dimension of customer engagement is attention. After running an independent t-test the results showed that the difference between these two treatments was not significant t(210) = -1.552, p > 0.05. Consequently, we will reject hypothesis 3a.

(36)

36 The next dimension of customer engagement is absorption. There was a strong significant difference between the treatments for this variable t(210) = 3.118, p > 0.01. The results of the independent t-test showed that there was a higher level of absorption for the storytelling treatment (M = 4.46, SE = 0.13) compared to the factual treatment (M = 3.85, SE = 0.15). Thus, hypothesis 4a is supported.

The last dimension of customer engagement is interaction. The independent t-test was used again. The results of the test presented no significant difference between the factual treatment and the storytelling treatment t(210) = -0.656, p > 0.05. Hence, hypothesis 5a is not supported.

The next step in the testing of hypothesis is the knowledge variable. A test of homogeneity of variance showed that the samples were homogenous for this variable as well. Moreover this variable is continuous and was therefore analyzed in a similar way as the dimensions of customer engagement, using an independent t-test. The test showed that the differences between the storytelling treatment (M = 3.52, SE = 0.15) and the factual treatment (M = 4.98, SE = 0.15) was very significant t(210) = -7.032, p < 0.01. This means that hypothesis 6a is supported.

Construct 2

The dimensions of customer engagement can either increase the cognitive- or the affect-based attitude. In this case the effect of one variable on two other variables is analyzed. The dimensions were divided for each dimension into people who scored high on this specific dimension and people who scored low. In this way, we were able to understand whether the specific dimension was the reason for the difference in attitude. Dividing the variable into two groups turned the continuous variable into a categorical variable. A Manova was applied instead of a regression analysis, since our hypotheses concerned the effect of 1 independent variable on 2 dependent variables. The partial eta squared, as a result of the Manova, explains the proportion of variance of a variable that is not explained by the other variable in the

(37)

37 Manova (Field, 2003). In order to run a Manova the predictor variable has to be categorical. Therefore dummy variables were created for each of the dimensions of customer engagement. This was also done for the knowledge variable. A value of 1 was given to cases that scored between 1 and 4 on the 7-point likert scale and a value of 2 was given to cases that scored between 4 and 7. In this way, we were able to divide the sample into meaningful groups for each of the dimensions. The Manova provides us with a partial eta square, which explains the proportion of the variance of a variable that is not explained by other variables (Field, 2003). The effect sizes are interpreted following Cohen’s (1988) standards. A partial eta square between 0.01 and 0.06 is a small effect, between 0.06 – 0.14 is a medium effect and higher than 0.14 is a large effect. Table 3 shows the outcomes of the Manova for the dimensions of customer engagement and the knowledge variable (Cohen, 1988).

Variable Dependent

Variable

df Mean

Square

F Sig. Partial Eta

Squared Identification Cognitive 1 0.52 4.11 0.04 0.02 Affect 1 3.89 17.28 0 0.08 Enthusiasm Cognitive 1 0.54 4.3 0.04 0.02 Affect 1 1.73 7.35 0.01 0.03 Attention Cognitive 1 0.62 4.98 0.03 0.02 Affect 1 1.37 5.77 0.02 0.03 Absorption Cognitive 1 0.98 7.95 0.01 0.04 Affect 1 5.94 27.60 0 0.12 Interaction Cognitive 1 1.22 9.98 0.00 0.05 Affect 1 4.07 18.14 0 0.08 Knowledge Cognitive 1 0.41 3.27 0.08 0.01 Affect 1 0.46 1.89 0.17 0.01

(38)

38 The first dimension that was investigated is identification. 38.7% of the sample did not feel any identification towards Switzerland after seeing the video, whereas 31.1% did identify him or herself with the destination after seeing video. According to the Manova, identification has a significant effect on the cognitive-based attitude and the affect-based attitude. Nevertheless, the affect-based attitude seems to explain the variance better than the cognitive-based attitude. No support was found for the hypothesis; since identification seems to affect the affect-based attitude to a greater extend than the cognitive-based attitude. Therefore we reject hypothesis 1b.

The second dimension of customer engagement is enthusiasm. 31.3% showed no enthusiasm after being exposed to the video and 68.9% did feel enthusiastic after exposure. Both types of attitude showed to be significant. Support for hypothesis 2b could be found in the fact that enthusiasm shows to have a greater impact on the affect-based attitude than the cognitive-based attitude. Nevertheless, the effect sizes were small and did not deviate much from each other. Thus, we partially support hypothesis 2b.

The third dimension that we analyzed was attention. 37.3% of the participants did not have an increase in attention towards the destination and 62.7% of the participants did show an increase in attention towards the destination. The effect sizes of attention on the cognitive-based and affect-cognitive-based attitude are both significant. However, similarly to enthusiasm, the effect sizes were small and did not deviate much. Nevertheless, we reject hypothesis 3b, given that the effect of attention on the cognitive-based attitude was smaller than on affect-based attitude.

The next dimension is absorption. Of all respondents 42% did not feel absorbed when seeing the video, whereas 58% did experience absorption. Both attitude types were significantly affected by absorption. However, we can see that absorption has a much higher effect on the affect-based attitude compared to the cognitive-based attitude. Hence, hypothesis 4b is strongly supported.

(39)

39 The last dimension of customer engagement is interaction. Interaction did not increase for the majority of the respondents (54.7%) and it did increase for 45.3% of them. Interaction is significantly affecting both cognitive- and affect-based attitude. Nevertheless, the effect of interaction on the attitude types showed a medium sized effect on the affect-based attitude, whereas the effect on the cognitive-based attitude is small in size. Therefore, hypothesis 5b is supported.

Next, the effect of knowledge on the attitude towards the destination is analyzed. 58% of the participants felt more knowledgeable after exposure to one of the videos, whereas 42% did not experience an increase in knowledge at all. Additionally, the effect of knowledge on the attitude types showed not to be significant. No support for the hypothesis was found, since none of the attitude types were affected, which makes it impossible to compare them. Consequently, hypothesis 6b is rejected.

Construct 3

The final construct in the conceptual model concerns the effect of the attitude towards the destination on the behavioural intention to travel towards this destination. A regression analysis was executed, due to the fact that both variables were continuous. The model showed that the affect-based attitude had a significant effect on the behavioural intention to travel (β = 0.257, p = 0.01), opposed to cognitive-based attitude that had an insignificant effect on the intention to travel (β = 0.124, p > 0.05). Consequently we support hypothesis 7.

5.3 Supplementary mediation analysis

Supplementary analysis is executed in order to understand whether the dimensions of customer engagement have a mediating role in the relationship of information type and attitude. Only enthusiasm and absorption were analyzed of the dimensions of customer engagement, since these were the only variables that correlated with both the information type and the attitude.

(40)

40 The total effect of information type on the attitude types (table 4 and 5) is compared to the direct effect. Table 4 shows that the total effect of information type on the affect-based attitude is significant. This is not the case for the total effect of information type on the cognitive-based attitude (Table 5).

Affect-based attitude B SE t P LLCI ULCI

Total effect of information type on affect-based attitude

-0.17 0.07 -2.52 0.01 -0.3001 -0.0366

Table 5 Total effect affect-based attitude

Cognitive-based attitude B SE t P LLCI ULCI

Total effect of information type on cognitive-based attitude

-0.04 0.05 -0.83 0.41 -0.1372 -0.0562

Table 6 Total effect cognitive-based attitude

The PROCESS macro (Hayes, 2012) was used to test for mediation. Firstly, the dimension of enthusiasm was studied. The findings of the effect of information type on cognitive- and affect-based attitude, with enthusiasm as a possible mediator, are shown in Table 6 and 7. The total effect of information type on affect-based attitude (table 4) shows to be significant (B = -0.17, p < 0.05). This indicates that the information type has an effect on the affect-based attitude. Nevertheless, we have to look at the direct effect of information type on the affect-based attitude in order to find out whether enthusiasm has a mediating role (table 5). Full mediation if the direct effect is insignificant. This appeared to be the case (B = -0.12, p = 0.07) and provides support for the mediating role of enthusiasm between the information type and the affect-based attitude.

Secondly, the mediating role of enthusiasm on the cognitive-based attitude was analyzed. The total effect of information type on the cognitive-based attitude shows not to be significant (B = -0.04, p > 0.05). Therefore, no support can be provided for a mediating effect of enthusiasm between information type and the cognitive-based attitude.

The effect of engaging consumers through stories