The Effectiveness of Gamification of Mobile Apps
How can gamified mobile apps influence the customers on their path-to-purchase?
By:
Jérôme Faßhauer MSc. Marketing Management
University of Groningen Faculty of Economics and Business
The Effectiveness of Gamification of Mobile Apps
How can gamified mobile apps influence the customers on their path-to-purchase?
- Master Thesis -
Author:
Jérôme Faßhauer Peletierweg 32 51143 Cologne
Germany Tel: +31 6 39 68 21 61 jerome.fasshauer@googlemail.com
MSc Marketing Management Student Number: 3202984
1st Supervisor:
Dr. Hans Berger, Faculty of Economics and Business
2nd Supervisor
Dr. J.A. Voerman, Faculty of Economics and Business
Date of Submission: 15.01.2018
Management Summary
The ongoing advancement in technology entails various new opportunities for marketers to engage with customer, but at the same time also faces them with challenges. Due to the innovation of the smartphone and the subsequently associated rise of the digital channel, marketers need to find new cutting-edge ways, in order to grasp the attention of their
customers. To stay relevant in the highly competitive mobile application industry, companies need to anticipate new trends which deliver unique experiences to the customer. One of these trends relates to the implementation of game features into the development of mobile
applications, known as gamification.
Gamification has become a buzzword throughout the years and has drawn the attention of many academics. The main idea behind gamification is to transfer game elements, which exert motivational effects and foster certain behavior, and apply them towards a non-gaming related context.
Due to the relatively newness of the concept, the currently existing literature is still debating on the effectiveness of gamification. The aim of this paper is hence twofold. First it aims to investigate the effectiveness of specific game elements that were derived from the work of Shell (2008) in regards to the usage intention of mobile apps. Secondly, this research will try to contribute to the existing gap in literature referring to the effectiveness of gamified apps and the impact on consumer activities.
A self-administered survey was conducted in order to collect sufficient amounts of data. The questionnaire was distributed to 107 participants, of which 96 were eligible for analysis and asked them questions about their mobile phone usage and later about their impression of an application. Regarding the second part of the survey, participants were randomly assigned to one of two previously set up conditions, in which one evaluated a gaming application while the other group evaluated a gamified application.
With regard to the first goal of applying the elements from the Elemental Tetrad Model and test them for their effectiveness regarding the usage intention of a mobile application, none of the four elements could be reach significance. These findings suggest that there might be other game design models and elements, which can be used for successful application design.
Regarding the second goal of this paper, investigating the effectiveness of gamified
applications on consumer activities, the results demonstrated significance. All three measured variables - consideration, purchase intention and engagement - were found to be significant in relation to the usage intention of gamified mobile applications.
IV Based on the findings of this paper, managerial implications were discussed as well as
limitations and possibilities for future research.
Preface
This thesis has been written as a part of my masters programm in order to obtain the master’s degree in Marketing Management at the University of Groningen. The whole process of researching and writing this thesis took five months in total, starting during September 2017 till January 2018.
My motivation for this research stems from my own interest about the ongoing development of technology and how it affects values and behaviors of humankind. The advancement in technology has impacted our everyday lives and as a consequence we have to adopt to those changes that come along. A perfect example for this would be the introduction of the smartphone by Apple, releasing the iPhone in back in January 2007. These portable devices are nowaday accompanying us everywhere we go and throughout the years we got closely attached to them. These days it is hard to imagine that could ever renounce our smartphones due to the fact that they enable us to nearly do anything we want, anywhere and at any time.
The second aspect of this research considers the topic of gamification. This specific topic gained my interest due to the personal relevance I have towards the topic of gaming. Since I my youth, I have always spend a lot of my freetime playing videogames and based on the fact that I am studying Marketing, I was interested how videogames and more precise the effects of videogames on individuals are transferred into a business context.
Considering my field of study and the plan of starting my carrer as a marketer in the near future, I developed the above mentioned research question that combines the two topics of gamification and smartphones. Obtaining relevant literature for the research I was doing, was quite challenging due to the newness of the topic and futher because my topic tapped not only into the field of Marketing but also into fields like gamedesign, information technologies and psychology. Altough this whole process of researching a very specific topic and writing an empirical and academic paper was really challenging, it allowed me to use and further develop my skills that I acquired throughout my academic career.
It is a need for me to thank my supervior Dr. Hans Berger, who gave me guidance throughout the whole process of writing this thesis and always tried to help me whereever I had difficulties.
I would also like to express my appreciation to all the participants in my survery. Without them this master thesis would not have been possible.
VI Table of Contents
LIST OF FIGURES AND TABLES ... VIII
INTRODUCTION ... 1
THEORETICAL FRAMEWORK ... 4
LITERATURE REVIEW ... 4
Mobile shopping and Branded Mobile Applications ... 4
Gamification ... 4
Elemental Tetrad Model ... 5
Technology Acceptance Model (TAM) ... 6
CONCEPTUAL MODEL... 6
HYPOTHESES DEVELOPMENT ... 7
METHODOLOGY ... 11
EXPERIMENTAL DESIGN ... 11
QUESTIONNAIRE ... 12
Pre-questionnaire 1 ... 12
Pre-questionnaire 2 ... 12
Experiment ... 13
Consideration, Purchase Intention and Engagement ... 14
DATA COLLECTION ... 14
SAMPLE ... 14
ANALYSIS PLAN ... 15
ANALYSIS AND RESULTS ... 17
MEASUREMENT MODEL ... 17
STRUCTURAL MODEL ... 20
DISCUSSION ... 23
CONCLUSION & RECOMMENDATIONS ... 26
MANAGERIAL IMPLICATIONS ... 26
LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH ... 26
CONCLUSION ... 28
APPENDICES... 29
REFERENCES ... 31
List of abbreviations
app application
AVE average variance extracted
e.g. for example
MGA Multigroup Analysis
PLS Partial least squares
SEM Structural equation modeling
TAM Technology Acceptance Model
TPB Theory of Planned Behavior
TRA Theory of Reasoned Action
WoM Word-of-Mouth
VIII List of figures and tables
Figures
Figure 1: Conceptual Model ... 6
Tables Table 1: Sample characteristics ... 15
Table 2: Composite reliablilty, AVE and Cronbach alpha coefficient ... 18
Table 3: Discriminant validity of constructs ... 19
Table 4: Summary of test results for the structural model ... 21
Table 5: Results of multi-group analysis ... 22
Table 6: Composite reliablilty, AVE and Cronbach alpha coefficient for game application .. 30
Table 7: Composite reliablilty, AVE and Cronbach alpha coefficient for gamified application ... 30
Introduction
The field of marketing and more specifically the world of retailing underwent during the last decades several changes due to the integration of several new technologies into the business models. It all started with one single channel being the brick-and-mortar store and throughout the years the number of channels increased further and further. The first addition, the online channel, changed retailing enormously. Consumers’ shopping behavior changed due to the availability of information, which hence lead to customers’ empowerment. The integration of new technologies brings along several benefits for the customer, but at the same time entails various challenges for marketers (Ailawadi & Farris, 2017). With the ongoing digitalization and the integration of the mobile channels into the online and offline channels, the world of retailing further continued to change (Verhoef, Kannan & Inman, 2015). As Rigby (2011) suggested the retailing model once more needs to be adapted in order to incorporate the various different channels properly. Especially the integration of the mobile channel was considered to be a disruptive development (Rigby, 2011).
Previous to this development, retailers were mostly encountering the so-called webrooming, where customers search for information online but still make the purchase decision offline. But with the introduction of mobile devices another development called showrooming became important. Customers could now browse for information offline in store and simultaneously on their mobile devices (Verhoef, Neslin & Vroomen, 2007) in order to compare prices and search for discounts. The old traditional barriers that differentiated the channels from each other were vanishing and leading to the creation of a seamless experience across all different channels within the omni-retailing environment.
Lamberton & Stephen (2016) reflected in their article about how the digitalization affected marketing and the opportunities it entails. Further they point out the importance to deepen the understanding of consumers’ experience across the different channels due to the breaking down of the traditional barriers, how these should be designed and managed and additionally how the digital environment impacts consumer’s shopping behavior. According to Shankar et al. (2010) mobile marketing has the potential to change the paradigm of retailing. Based on the assumption that previously consumers had to enter the retailer’s environment, nowadays the retailer can enter the consumer’s environment due to the fact that mobile devices became constant companions of the consumer, and enable the retailer to continuously communicate. Due to the fast penetration of smartphones in our everyday lives, the mobile channel has become the third marketplace after online and offline and marketers shift their focus toward building and
As a marketer, it is important to deepen the understanding of how the usage of mobile devices and applications impacts the behavior and furthermore how these technologies can help to create and deliver valuable experiences to the customer. “The rapid adoption of smartphones and subsequent development of mobile applications (“app” or “apps” hereafter) have been changing the ways in which customers interact with a brand” (Kim, Wang & Malthouse, 2015, p. 28).
In 2016, consumers downloaded over 149.3 billion apps worldwide accounting for more than 88.3 billion dollars in revenue. Until 2021, the number of apps downloaded worldwide is projected to more than double to over 350 billion resulting in more than 188 billion dollars revenue (Statista, 2017a; Statista, 2017b). Based on the fact that the mobile application industry is rapidly growing and the level of competition is high, marketers are nowadays competing to be on the first page of the consumer’s mobile device with their mobile application. Such applications are designed to inform, entertain and communicate with customers and hence to attract new customers, increase the interest in the brand or company, creating deeper relationships with existing customers and increase their engagement, driving their behavior and delivering them an enhanced total experience (Alnawas & Aburub, 2016; Bellman et al., 2011;
Kim, Wang & Malthouse, 2015).
For this reason Shankar et al. (2016) examined the current state of mobile shopper marketing in their research paper and further proposed various important research avenues that are in need to be investigated. One of these research avenues proposed by Shanker et al. (2016) covers the topic of mobile applications and posits the question of:
“How can marketers optimize their mobile app design to best influence shoppers on their path to purchase?”
The purpose of this paper is to examine the above-mentioned research avenue, by elaborating on the impact of “Gamification” on mobile app design. Gamification refers to the implementation and “use of game-like elements in a non-game context” (Deterding et al., 2011). The concept of gamification is becoming more and more important in business practices and especially gained lots of popularity in the field of marketing (Yang, Asaad & Dwivedi, 2017).
According to Hofacker et al. (2016) a lot of apps that already implemented the gamification approach in their mobile app design do not live up to their full potential and hence the authors state that further research is needed in order to overcome the current inefficacy of gamified apps.
For this reason, the goal of this paper is to examine the effectiveness of gamified apps focusing on four different elements derived from the Elemental Tetrad Model by Shell (2008), and how these affect the customer’s behavior.
According, the research question of this paper was developed:
“How can gamified mobile apps influence the customers on their path-to-purchase?”
The remainder of this paper will be structured in different main parts, starting with the theoretical framework and methodology, followed by the analysis and results. The last part refers to the conclusion and recommendations.
Theoretical Framework
Literature review
Mobile shopping and Branded Mobile Applications
Previous research has attempted to explain the effects of mobile devices and applications on consumer’s behavior and its drivers (see e.g. Alnawas & Aburub, 2016; Bellman et al., 2011;
Hubert et al., 2017; Kim et al., 2017; Kim, Wang & Malthouse, 2015; Wang, Malthouse &
Krishnamurthi, 2015). For example, Wang, Malthouse & Krishnamurthi (2015) examine the impact of mobile shopping on customers purchase intention and furthermore come up with drivers that are enhancing mobile shopping. It is important as marketers to investigate the antecedents of mobile shopping in order to get a deeper understanding regarding the acceptance of smartphone-based mobile shopping. Hubert et al., (2017) identified drivers of mobile shopping acceptance and incorporated these in their study to investigate how they influence the development of mobile shopping applications.
Further researchers considered the effectiveness of branded mobile phone apps and how these affect consumer behaviors regarding brand attitudes and purchase intentions (Bellman et al., 2011; Kim, Wang & Malthouse, 2015). Branded mobile apps are defined as software downloadable to a mobile device that prominently displays a brand identity via the name of the app and the appearance of a brand logo or icon throughout the user experience (Bellman et al., 2011, p.191).
Additionally, regarding the functionally of branded mobile applications, Bellman et al. (2011) examined that apps which made use of an informational/user-centered style compared to an experimental game like app, were significantly more effective in shifting purchase intention, because they focus more on the attention on the user than on the phone.
Gamification
The term gamification became a buzzword and quite popular among academics in recent years (Yang, Asaad & Dwivedi, 2017) and since has emerged as a common business practice within several different fields (Deterding et al., 2011; Hamari, Koivisto & Sarsa, 2014; Werbach &
Hunter, 2012). Although the concept just gained lots of attention during recent years among academics, it cannot be considered as a new concept. Gamification originated from the digital media industry (Deterding et al., 2011), which underwent a rapid growth within the last 15 years (Seaborn & Fels, 2014). Videogames have been deeply integrated in our today’s culture and serve as a form of entertainment. Beyond its purpose of entertaining people, Anderson and
Dill (2000) stated that games in general have considerable effects on the ‘players’ behaviors and thought. Drawing from this, practitioners and business professionals are trying to transfer the elements out of games that have these motivational potentials, and apply them to different areas outside the game environment in order to influence and drive people behaviors and intentions.
Examples of these areas are education, information studies, health and marketing.
By definition, gamification is declared as “the use of game elements in non-game context”
(Deterding et al., 2011, p. 1) or as “a process of enhancing a service with affordance for gameful experiences in order to support user’s overall value creation” (Huotari & Hamari, 2012, p. 19).
Although the body of literature regarding gamification is growing, authors are still debating about its effectiveness regarding the change of people’s attitudes and further academic evidence for the effectiveness and benefits of gamification are still lacking. Hence this paper aims to shed more light onto this topic by incorporating the Elemental Tetrad Model by Shell (2008) and examining its effectiveness using the Technology Acceptance Model from Davis (1989).
Elemental Tetrad Model
In his book “The Art of Game Design. A book of Lenses” from 2008, Jesse Shell created a conceptual framework, which is needed in order to design games that deliver rich experiences to the consumer. His model consists of four basic elements, which cover and classify the elements that form a game. He calls these the “elemental tetrad” (Shell, 2008, p.41). The conceptual model is based on mechanics, aesthetic, story and technology.
Mechanics refer to the structural aspect of games. They describe how consumers can reach the set goal of the game and what happens when they do so. Aesthetics refer to the visual appearance of a game and are predominantly influencing the users’ perceived experience. The story gives the context to the game and can be seen as “a sequence of events that unfold”
throughout the game (Shell, 2008, p.41). Technology is defined as the medium that makes the game possible.
He points out that the elements are essential and interrelated, meaning that “none of the elements is more important than the others” (Shell, 2008, p.43). The framework is “arranged in a diamond shape not to show any relative importance, but only to help illustrate the “visibility gradient”; the fact that technological elements tend to be the least visible to the players, aesthetics are the most visible, and mechanics and story are somewhere in the middle”
(Shell, 2008, p.43)
Technology Acceptance Model (TAM)
The Technology Acceptance Model is one of the most well-known models referring to technology acceptance and hence became predominantly considered as the conceptual basis for many academics in the field of information systems literature. According to several studies (see Legris, Ingham & Collerette (2003); Ramayah & Jantan, 2003; Ramayah, Lam & Sarkawi, 2003) TAM has been successfully proven to be a theoretical model that helps to examine and predict certain user behavior of information technology. Drawing on the theory of reasoned action (TRA) (Fishbein & Ajzen, 1975) and theory of planned behavior (TPB), TAM proposed by Davis (1989) suggests that system acceptance is determined directly or indirectly by the user’s behavioral intentions, attitude, perceived usefulness and perceived ease of use of a system (Yang et al., 2017). Perceived usefulness is defined as “the degree to which a person believes that using a particular system would enhance his or her job performance” (Davis et al., 1989, p.985). Perceived ease of use is defined as the degree to which a person believes using an information system would be free of effort (Davis, 1989). According to Park (2009), the Technology Acceptance Model appears to be accountable for 40 to 50 percent of user acceptance.
Conceptual Model
Figure 1: Conceptual Model
For the purpose of this research, the employed approach is an adoption of the Technology Acceptance Model by Davis (1989). Additionally, this research is based on the conceptual framework that Hubert et al. (2017) derived for their study regarding the acceptance of smartphone-based mobile shopping. The conceptual model that is derived for the purpose of this study is going to be tested via an experiment.
Hypotheses Development
The embedded conceptual model of this paper integrates different sets of hypotheses. First assumptions about relationships between perceived ease of use, perceived usefulness and the behavioral intention to actual make use of a system, that are deduced in H2a-d, H3a-d. These hypotheses incorporate the four different elements derived from the Elemental Tetrad Model by Shell (2008). The second group of hypotheses (H1a-c) refers to the relationships between the usage intention and the three dependent variables consideration, purchase intention and engagement representing the path-to-purchase.
According to the Technology Acceptance Model by Davis (1989), it theorizes that the intention to use a system, in this case the mobile application, is determined by two beliefs:
(1) Perceived usefulness and (2) Perceived Ease of Use.
Perceived usefulness can be defined as the extent to which a person believes that using the a system, in this case the mobile application, enhances his or her performance regarding a certain task, while perceived ease of use refers to the extent that using the mobile app will be free of effort (Davis, 1989). In the case of this research, the author wants to examine how the attitude toward gamified mobile applications can influence customers along their path to purchase. The path to purchase encompasses the activities consumers undertake throughout their whole journey and within this paper is divided into three different outcomes being consideration, purchase intention, and engagement. In this research, consideration is defined as the degree to which a person would be likely to engage with the company and further the degree of the person to look for category-related products. Purchase intention is defined as the likelihood of making a purchase via the mobile application. The last outcome, engagement is conceptualized as the degree to which a person intends to generate positive Word-of-Mouth about the company and the likeliness to engage in co-creation. These concepts reflect the path to purchase by investigating the pre-purchase activities, the actual purchase intention of a consumer and the consumer activity after the purchase.
Against this background, it is assumed that usage intention of the mobile application has a direct effect on the three different path-to-purchase outcomes. The likelihood of actually using a system further increases the probability that the marketing outcomes are also affected.
Thus,
H1a: Usage intention is positively related to Consideration H1b: Usage intention is positively related to Purchase intention H1c: Usage intention is positively related to Engagement
Based on the main idea of this paper, to apply the concept of gamification on mobile app design taking the Elemental Tetrad Model into consideration, the four elements being Story, Mechanics, Aesthetics and Technology need to be examined in regards to their effectiveness measured by their perceptions of usefulness and ease of use.
As already depicted in this section, perceived usefulness and ease of use are predictors in order to determine the adoption of technology and hence the usage of a system.
In this research the element of Story is defined as the narrative format that provides context to a game and adds meaning to the whole consumption experience (Hofacker et al., 2016). A story is referred to as a sequence of events which unfolds (Shell, 2008) and that keeps the user or player engaged. A story gives a meaning to a game beyond the sole purpose of players trying to achieve goals in order to collect points, badges or achievements. It can enrich the users experience by inspiring and motivating the player. According to Hofacker et al. (2016) the element of story is an effective tool of persuasion, in order to change attitudes and beliefs.
Under the assumption that the element of story is perceived as useful and free of effort for making use of the mobile application the following hypotheses were derived,
H2a: The perceived usefulness of story is associated with usage intention H3a: The perceived ease of use of story is associated with usage intention
The second element derived from the proposed model by Shell is Mechanic. It refers to the rules and procedures of a game and further establishes a clear structure of how players or users are able to achieve certain goals and how they are going to be rewarded for achieving those (Shell, 2008). Most literature that deals with game design regarding mechanics
distinguishes between several crucial components that are important to consider (e.g. Reeves
& Read, 2009; Werbach & Hunter, 2012; Zichermann & Cunningham, 2011). Due to the fact
that mechanics give the structure of how players are able to achieve goals and how they are getting rewarded for doing so, academics mostly focus on the most relevant elements of (1) points, (2) badges and (3) leaderboards.
This research therefore proposes to test the relationship between the usefulness and the ease of use of mechanics regarding the usage intention of the mobile application.
Hence,
H2b: The perceived usefulness of mechanics is associated with usage intention H3b: The perceived ease of use of mechanics is associated with usage intention
Aesthetics relates to the overall visual appearance of a game. Visual imagery and presentation is important in order to create a unique, immersive and valuable experience for the consumer.
Aesthetics are one of the most important aspects of game design, due to the fact that they have the most direct relationship to a player’s experience (Shell, 2008, p.42). Further they are able to reinforce and strengthen the other elements to deliver the best experience possible. Under the assumption that aesthetics help to deliver those promised experiences, this research posits the following hypotheses that depict the relationships between the perceived usefulness and ease of use of aesthetics and the usage intention of the mobile application.
H2c: The perceived usefulness of aesthetics is associated with usage intention H3c: The perceived ease of use of aesthetics is associated with usage intention
Technology is the essential medium, through which the story is told, the mechanics operate and the aesthetics are taking place (Shell, 2008). It therefore shapes the whole experience.
Although technology might be the least visible of all four elements from the Elemental Tetrad Model, it does not mean that it is less important. According to Shell (2008) all four elements exhibit the same level of significance and hence this paper defines the hypotheses:
H2d: The perceived usefulness of technology is associated with usage intention H3d: The perceived ease of use of technology is associated with usage intention
Control variables
Demographic differences in the adoption and use of any new form of technology are important
gender, age and educational background. Due to the fact that this study investigates the perceptions of people regarding the usefulness and ease of use of the four different elements into a mobile application design, these perceptions may vary across these demographics.
According to various academics, variables like age and gender did not receive great attention within the information technology literature (Gefen & Straub, 1997; Sun & Zhang, 2006). On top of that, regarding to the research within the field of gamification, age and gender represent perspectives on games and gameplay wherein variation and preferences have been long disregarded (Greenberg et al., 2010; Griffiths et al., 2003; Williams et al., 2008). According to Venkatesh et al. (2003) the influence of age on technology adoption and use significantly differs in between younger and older users. This effect can be explained due to the fact that younger people have been exposed to digital technologies while growing up (Morris & Venkatesh, 2000) compared to older generations, that tend to be more technology anxiety and hence compared to younger people perceive their technological skills rather low (Chung et al., 2010; Czaja et al., 2006; Ellis & Allaire, 1999; Harrison & Rainer, 1992). In regard to gender, previous studies also found significant differences referring to the adoption and use of technology (Spence &
Helmreich, 1980; Venkatesh et al., 2000), indicating that men seem to be more affected by the usefulness of technology, while women seemed to be rather influence by ease of use (Venkatesh
& Morris, 2000). Furthermore, in respect to the purpose of this study, investigating the influence of game elements, previous academic literature on gender related gaming studies pointed out that males compared to women were more likely to be considered as gamers (Cassell & Jenkins, 1998; Gorriz & Medina, 2000). However, this notion changed throughout the years based on the fact that females now make up a huge percentage of all gamers (see Hamari & Järvinen, 2011; Paavilaien et al., 2013). The only difference between the genders indicated that, women have different preferences for their content, design and genre (Hartmann
& Klimmt, 2006; Kinzie & Joseph, 2008).
Being familiar or having experience with any kind of technology, plays an important role for adopting and using such technology. Several studies indicated that people who are more exposed to technology and hence more familiar, are less anxious and their perception of self- efficacy related to the use of technology rises (Bohlin & Hunt, 1995; Todman & Managhan, 1994). The more people are familiar with a technology, the more likely it will be that their attitudes towards the adoption and continued usage intention of technology will change (Brown
& Inouye, 1978; Davis, 1989; Venkatesh et al., 2003).
Methodology
Experimental design
In order to test the proposed relationships as presented in the conceptual model, an experimental study was conducted using an online survey. The benefit of using an online survey is that in a relatively short amount of time, sufficient amount of data for later analysis can be acquired.
The survey consisted of four different parts. The first part gathered information about the participants’ demographic background. The second part and simultaneously the first study of the survey acquired information about the participant’s mobile phone usage. During the third part and simultaneously second study of the survey respondents were asked question about their perceptions of a mobile application design. During this study, participants were randomly assigned to one of two conditions by the software: control group vs experimental group.
In the control group, participants were shown two images of a gaming application (Temple Run) (see Appendix B) and a brief description of the game and asked to give their opinions on several statements according to the four elements (Story, Mechanic, Aesthetic and Technology). In the experimental group, participants were confronted with two images of a gamified application (Nike Run) (see Appendix A) and also given a short description of the app. Both visualizations were created using the image editing software GIMP version 2.8.22.
Due to the statement of Anderson and Dill (2000), that games in general have considerable effects on the ‘players’ behaviors and thoughts it is important to investigate if these effects can also be evoked and transferred to a non-gaming related context. Therefore it is necessary to examine the different elements games are built on and study whether those are all relevant and applicable to a fully non-game context. The biggest difference between a gaming and a gamified app is that the gamified application just delivers a “game-like” experience by incorporating these elements. To further explain that difference, the experiment chose a fully-fledged game (Temple Run) where the player or user slips into the role of the fictional character called Guy Dangerous – the hero and main character of the game – who stole a valuable artefact out of a temple and has to escape on his way out monsters which are chasing him now. The aim of the game is to run away as long as possible in order to chase and aim for new high scores. The gamified application – in this research the example of “Nike Run” – makes use of certain elements like badges and leaderboards and incorporates these to foster a certain behavior like exercising. The research aims at finding the most relevant elements for app designers that actually result in a certain perceived usefulness that can further evoke and promote different
Furthermore, they were provided with information about the terminology of gamification to enhance their understanding regarding the given tasks.
After that the survey proceeded by asking the participants about their usage intention of either a gaming application (control group) or a gamified application (experimental group).
The last part of the survey asked the respondents to give their opinion regarding the three dependent variables, being Consideration, Purchase Intention and Engagement. In order to measure the outcomes of all the different variables depicted, previously developed and validated measurement scales were adapted and modified for the purpose of this research.
Questionnaire
The questionnaire that was specifically designed for the purpose of this research was created using the Qualtrics survey platform. It started with a preface introducing the participants to the topic and the purpose of the study. Further they were given an outline of the tasks they were about to engage in and an approximate time designation of the amount of time it would take to complete the whole survey. Additionally, the preface pointed out that all their data would be treated anonymously and confidentially and that they would solely be used for educational purposes.
Pre-questionnaire 1
After the preface, the survey started assessing relevant data regarding the participants demographic background, in which respondents were asked to indicate their age, gender, nationality, country of residence and additionally questions about their current situation (being a student or working) and about their highest completed education degree till that point in time.
Pre-questionnaire 2
The second part of the questionnaire and simultaneously the first study of the survey covered the topic of mobile phones usage. The purpose of this study is to investigate the technological savviness of the participants and furthermore about their attachment to their own mobile devices.
It started by asking the participant whether he or she owns a mobile phone, which is a set condition to continue with the next part of the survey. If the participants indicated they did not own a mobile phone, they were directly led to the end of the survey and were thanked for participation.
In order to determine the personal attachment of the participants to their mobile phones, they were asked to rate three statements e.g. “I can’t live without my mobile phone”, “I use my phone 24/7”, that were measured using a 7-point Likert scale (1=”strongly disagree”, 7=”strongly agree”). The measurement scales were adopted from the previous studies (see e.g.
Gao et al., 2013; Vincent, 2005, 2006; Wehmeyer, 2007) Furthermore, to assess data about their technological savviness, participants were asked to indicate for how long they have been owning a mobile phone and how much time they spend daily on their devices. Additionally, the questionnaire tried to examine the usage behavior of the participants by asking them about the mobile application they used most recently and which one they have been using the most. These questions helped to indicate if participants rather made use of utilitarian or hedonic applications.
Experiment
After completing the first study regarding mobile phone usage, the survey introduced the participants to the second study about mobile application design. The participants were asked to evaluate four different elements being Story, Mechanic, Aesthetic and Technology based on a mobile application. In order to deepen their understanding of these elements, respondents were provided with information about each of the four elements.
After providing the participants with the relevant information, they were randomly assigned to one of the two conditions, control group or experimental group. In order to assess the effectiveness of gamification, the control group was given the task to evaluate a gaming application while participants in the experimental group were confronted with a gamified application. In both conditions respondents were asked to rate statements that measured the perceived usefulness and perceived ease of use of the four different elements Perceived usefulness was measured with three items adapted and modified from previous studies (see Davis, 1989; Hubert et al., 2017; Koivisto & Hamari, 2014; Rodrigues et al., 2016; Yang et al., 2017), e.g. ‘I would find this mobile application useful’. Perceived ease of use was measured with three items adapted and modified from previous studies (see Davis, 1989; Hubert et al., 2017; Koivisto & Hamari, 2014; Rodrigues et al., 2016; Yang et al., 2017) e.g. ‘Learning to use this mobile application would be easy for me’. Both, perceived ease of use and perceived usefulness used a 7-point Likert scale anchored at “Strongly disagree” (1) to “Strongly agree”
(7).
Usage intention described the actual behavioral intention of the participant to use either a gaming application (control group) or a gamified application (experimental) group in the future.
The behavioral intention was measured using semantic scale that examined three items based on the study of Hubert et al. (2017).
Consideration, Purchase Intention and Engagement
The last part of the whole survey covered the measurement of the three dependent variables.
For Consideration, participants were asked to rate different statements. It was conceptualized as the degree to which a person would be likely to engage with the company and further the degree of the person to look for category-related products.
Consideration was measured using three items that have been developed for the purpose of this research e.g. ‘I would like to know more about the company that developed the application’, ‘I would consider to download applications that are similar’. These items used a 7-point Likert scale anchored at “Strongly disagree” (1) to “Strongly agree” (7).
Purchase intention assessed data about the likelihood of participants to make purchases through the application, measured with a Semantic Differential scale. It was captured with three items which were derived from the work of Spears & Singh (2004). The third and last dependent variable engagement assessed the intention of the respondents to potentially create and deepen a relationship to the company and the willingness to generate positive WoM and feedback. The items measured have been developed for the purpose of this research and were measured using a 7-point Likert scale anchored at “Strongly disagree” (1) to “Strongly agree” (7).
Data Collection
The survey was distributed among students, friends and family through personal messages containing the generic URL from Qualtrics. Additionally, data was gathered from people living in the Student Hotel Groningen and people visiting the University Library in the city center of Groningen. Recipients of the link were also kindly asked to further share the survey with their friends, family and colleagues in order to collect sufficient amount of data for later analysis.
Sample
In total 106 people participated in the survey. From the 106 participants, 97 participants completed the whole survey and 1 respondent had indicated that he or she did not own a mobile phone, which excluded this participant from the analysis. In total 96 responses were eligible for the analysis.
Out of these 96 respondents, 38 (39.58%) were identified as female and 58 as male. The youngest person that participated in the survey was 18 years old while the oldest was 62 years
old. The average age was 26.094. Regarding their educational background at the time of completing the survey, 23 (23.96%) respondents indicated that they completed their compulsory education, 45 (46.88%) respondents their Bachelor’s degree, 26 (27.08%) finished their Master’s degree and 2 (2.08%) indicated that they obtained another form of education that was not mentioned within the choices presented in the survey.
Additionally, 50 (52,08%) participants signaled to be full-time students, 10 (8,34%) participants to be studying and working and 38 (39,58%) to be full-time working.
Due to the fact that the survey was distributed among friends, family, students and residents of the Student Hotel Groningen, and people visiting the University Library in the city center of Groningen different nationalities (e.g. German, French, Spanish, Italian, Moroccan) took part in the survey.
Table 1: Sample characteristics
Gender Age Educational degree
Female 38 39.58% Min 18 Compulsory Education 23 23.96%
Male 58 60.42% Max 62 Bachelor 45 46.88%
Mean 26.094 Master 26 27.08%
Median 25 Other 2 2.08%
Analysis Plan
To test the proposed hypotheses from the conceptual model, this paper used structural equation modelling (SEM) based on partial least squares (PLS) using the SmartPLS software version 3.2.7. According to Richter et al. (2015) gained PLS-SEM lots of attention among management researchers and additionally is a well-established technique to test structural models.
Partial least squares (PLS) is a variance-based SEM and was chosen for this analysis due to the fact that it is an applicable method which works efficiently with a rather small sample sizes (n=96). It is a combination of factor analysis and multiple regression analysis and compared to covariance-based SEM’s, it allows to work with complex models and makes no assumption about the data (Cassel & Hackl & Westlund, 1999). Additionally, PLS-SEM is applicable for models using reflective and formative measurements and is applicable to constructs which are measured with single- and multi-item measures (Hair et al., 2016). Also it is highly robust to missing values as long as these do not reach a certain threshold (Hair et al., 2016).
PLS-SEM analysis is going to be done in two different stages (Hair et al., 2016). In the first
that represents the relationships between each latent construct and its associated indicators. The second stage examines the structural model, also known as the inner model that estimates the relationship between the latent constructs.
Before performing the actual analysis it is important to check if the amount of data is sufficient enough. The sample size of 96 is higher than the minimum required sample size. As suggested by Barclay & Higgins & Thompson (1995), the minimum sample size for a PLS study should be equal to the lager of the next two conditions “(1) ten times the amount of formative indicators used to measure a single construct” (in our model: 0 formative indicators, or “(2) ten times the largest number of structural paths directed at a particular construct in the structural model” (in our model: 8 structural paths) (Hair et al., 2016, p.24). Additionally it is necessary to check the data characteristics regarding missing values that may have been derived from the questionnaire and if any of the indicators exceeded the threshold level 15%. In this research, that was not the
case.
Analysis and Results
Measurement Model
The measurement model is developed form the combined data that was gathered with the help of the questionnaire. The data set (n=96) consists of the control group, representing the game application (n=58) and the experiment group (n=38), which represents the gamified application.
The first step when analyzing measurement models is to examine the convergent and discriminant validity of the latent constructs and the reliability of the indicators. Convergent validity is the extent to which a measure correlates positively with alternative measures of the same construct (Hair et al., 2016, p.112), whereas discriminant validity refers to the degree to which a construct is truly different from other constructs (Hair et al., 2016, p.115).
In PLS-SEM several rules of thumb need to be taken into consideration when evaluating reflective measurement models.
For the internal consistency reliability, all measures of the composite reliability should be higher than 0.70. The use of composite reliability is recommended over the use of Cronbach Alpha as it tends to give a more conservative measurement of reliability. Cronbach’s alpha can be seen as the lower bound of internal consistency reliability and on the other hand composite reliability as the upper bound of internal consistency reliability (Hair et al., 2016, p.122).
Furthermore to check for the indicator reliability, the indicator’s loadings should also exceed the minimum threshold of 0.7. Last, the average variance extracted (AVE) has to exceed the threshold level of 0.5. Discriminant validity is assessed by comparing the square roots of the average variance extracted (AVE) of each latent variable with the correlation coefficients of any other latent variable. Discriminant validity is given of the specific latent construct when the square root of the AVE of each latent variable is greater than the any of the correlations involved in that construct (Fornell & Larcker, 1981).
The findings in our study, as shown in Table 2, indicate that all the factor loadings are greater than 0.7 (p < 0.01) and AVE is greater than the threshold level of 0.5. Hence, supporting convergent validity. The results of the combined table (Table 2) also hold for the individual tables (see Appendix Table 5 and Table 6).
Further Table 2 and 3 are displaying all measures of composite reliability that are needed to assess the reliability of the latent constructs. The findings indicate that the value of composite reliability also exceeds the minimum value of 0.7.
Table 2: Composite reliability, AVE and Cronbach alpha coefficient
Construct Game and gamified application
Composite reliability AVE Cronbach alpha
Consideration 0.957 0.880 0.932
Ease of Use Aesthetics 0.974 0.925 0.959
Ease of Use Mechanics 0.964 0.901 0.945
Ease of Use Story 0.966 0.904 0.947
Ease of Use Technology 0.979 0.939 0.968
Engagement 0.950 0.862 0.920
Purchase Intention 0.969 0.913 0.952
Usage Intention 0.970 0.915 0.954
Usefulness Aesthetics 0.985 0.956 0.977
Usefulness Mechanics 0.963 0.897 0.942
Usefulness Story 0.962 0.895 0.941
Usefulness Technology 0.976 0.931 0.963
The findings regarding discriminant validity, as presented in table 4 and 5 indicate that all latent variables fulfill the requirement referring to the Fornell & Larcker criterion. Additionally, the inspection of the cross loadings further established that each indicator loads highest on the construct it is associated with.
The statistics depicted in the previous tables indicate that the measurement model is sufficient strong enough to further test the structural model.
Table 3: Discriminant validity of constructs
1 2 3 4 5 6 7 8 9 10 11 12
Consideration 0.938
Ease of Use Aesthetics 0.590 0.962
Ease of Use Mechanics 0.659 0.757 0.949
Ease of Use Story 0.448 0.518 0.595 0.951
Ease of Use Technology 0.616 0.811 0.757 0.434 0.969
Engagement 0.909 0.604 0.634 0.407 0.652 0.929
Purchase Intention 0.785 0.481 0.596 0.437 0.499 0.825 0.956
Usage Intention 0.788 0.523 0.539 0.468 0.491 0.749 0.729 0.957
Usefulness Aesthetics 0.732 0.812 0.652 0.404 0.698 0.761 0.601 0.597 0.978
Usefulness Mechanics 0.776 0.424 0.577 0.293 0.516 0.792 0.698 0.643 0.706 0.947
Usefulness Story 0.709 0.515 0.550 0.617 0.459 0.761 0.683 0.674 0.729 0.793 0.946
Usefulness Technology 0.766 0.681 0.668 0.318 0.812 0.814 0.650 0.586 0.839 0.808 0.681 0.965
Structural model
In the structural model or inner model, the author examines the relationships between the different latent constructs. In order to be able to assess the path significance of the structural model, and further to examine if there are significant differences between the gaming application and the gamified application regarding the relationships between the constructs, a multi-group analysis (MGA) was performed. It allows us to test whether the two groups have significant differences in their parameter estimates that are based on bootstrapping. For this research bootstrapping was performed with the recommended number of 5000 subsamples in order to ensure stability of the results.
The findings as shown in table 4 indicate that the data supported three hypotheses and did not support eight for the gamified application, while for the gaming application four hypotheses were supported and seven rejected. Comparing both types of application, the gaming application showed that the usefulness of mechanics is significant at the 0.10 level towards the usage intention (H2b: β=0.600, p < 0.1) which partially reflects the importance of the element
‘Mechanic’ from the Elemental Tetrad Model’ regarding games in general.
Due to the fact that this research focuses on the effectiveness of gamified mobile applications, the findings suggest that these types of applications exert an influence on consumer activities.
In line with the hypotheses H1a-c, it was observed that usage intention is positively associated with consideration (β=0.86, p < 0.01), purchase intention (β=0.789, p < 0.01) and engagement (β = 0.857, p < 0.01). This suggests that people using a gamified mobile app continuously are more likely to engage with the company, to make purchases with the mobile application and to generate positive word-of-mouth.
As already depicted, this research further aims at comparing the results of the gaming application to the outcomes of the gamified application. To assess whether the path coefficients differ significantly among those two kinds of applications a MGA was performed.
The findings as shown in Table 5 indicate that only one relationship between the constructs differ in both cases. The relationship between Usage Intention and Engagement (p-value = 0.095) reached significance at the 0.10 level.
Table 4: Summary of test results for the structural model Game application
Gamified application
Game application
Gamified application
Game application
Gamified application
Hypotheses Path Standardized Estimates p-value Hypothesis verification
H1a Usage Intention Consideration 0.732 0.860 < 0.01 < 0.01 Supported Supported
H1b
Usage Intention Purchase
Intention 0.685 0.789 < 0.01 < 0.01 Supported Supported
H1c Usage Intention Engagement 0.669 0.857 < 0.01 < 0.01 Supported Supported
H2a
Usefulness Story Usage
Intention 0.114 0.536 n.s. n.s. Rejected Rejected
H2b
Usefulness Mechanics Usage
Intention 0.600 0.107 n.s. n.s. Rejected Rejected
H2c
Usefulness Aesthetics Usage
Intention -0.260 -0.127 n.s. n.s. Rejected Rejected
H2d Usefulness Technology Usage
Intention -0.122 0.557 n.s. n.s. Rejected Rejected
H3a
Ease of Use Story Usage
Intention 0.204 -0.175 n.s. n.s. Rejected Rejected
H3b
Ease of Use Mechanics Usage
Intention -0.028 -0.065 < 0.1 n.s. Supported Rejected
H3c
Ease of Use Aesthetics Usage
Intention 0.468 0.268 n.s. n.s. Rejected Rejected
H3d
Ease of Use Technology Usage
Intention -0.021 -0.288 n.s. n.s. Rejected Rejected
Table 5: Results of multi-group analysis
Path Game application Gamified application Standardized Estimates
(of difference)
t-Value
(of difference) p-Value Standardized Estimates (Mean)
Ease of Use Aesthetics Usage Intention 0.456 0.315 0.200 0.322 0.748
Ease of Use Mechanics Usage Intention 0.001 0.040 0.036 0.092 0.927
Ease of Use Story Usage Intention 0.219 -0.175 0.379 1.133 0.260
Ease of Use Technology Usage Intention -0.043 -0.376 0.267 0.424 0.672
Usage Intention Consideration 0.733 0.866 0.128 1.168 0.246
Usage Intention Engagement 0.671 0.858 0.189 1.685 0.095
Usage Intention Purchase Intention 0.686 0.793 0.104 0.838 0.404
Usefulness Aesthetics Usage Intention -0.237 -0.059 0.133 0.220 0.827
Usefulness Mechanics Usage Intention 0.566 0.043 0.493 1.019 0.311
Usefulness Story Usage Intention 0.093 0.430 0.422 0.851 0.397
Usefulness Technology Usage Intention -0.108 0.581 0.679 0.952 0.344
Unfortunately, this research was not able to identify the key antecedents of that usage behavior.
The results were unexpected, as the four components of the Elemental Tetrad Model by Shell failed to indicate any significant impact referring to their usefulness and ease of use on the usage intention of the gamified mobile application. As already mentioned above, only within the control group the element of mechanic regarding the usefulness was significant on the 0.10 level for the usage intention.
Discussion
Although the body of literature regarding gamification is growing, authors are still debating about its effectiveness regarding the change of people’s attitudes and further academic evidence for the effectiveness and benefits of gamification are still lacking. This study aimed at contributing to this gap in the literature and further investigates the antecedents for an appropriate mobile application design.
The first goal was to investigate whether the four elements that were derived from the work of Shell (2008) can be transferred towards mobile application design. The second goal was to investigate the influence of gamified applications on consumer activities along path-to- purchase. Of the 11 in total proposed hypotheses, only three were confirmed. The remaining eight hypotheses were not supported with the model used in this research.
As the first goal of this research was to apply the four elements of the Elemental Tetrad Model by Shell, Story, Mechanics, Aesthetics and Technology and examine those as possible antecedents for future mobile application designs. This was done by using the Technology Acceptance Model that in past literature appeared to account for 40% - 50% of user acceptance (Park, 2009).
The first element Story, did not find any significant support in the applied model. Although story is an important element in regards to proper game design, it was not considered as an important element that could be transferred into a non-gaming context. Contrary to its function as to give meaning and relevance to the user’s whole experience, in this research it was not perceived as an element which would be rather useful nor easy to use. According to Hofacker et al. (2016) the story element in mobile gaming applications was seen as a possible tool of persuasion in order to influence certain consumer values and beliefs, due to the
assumption that a well-told story line will have the effect of keeping the user engaged and nearly getting him or her lost in the story, that might result in a temporary acceptance of
A possible explanation for the insignificant result could be that mobile applications are just used occasionally and mostly for short-term oriented enjoyment, where an engaging story line would be contradictory. Another reason could be, that a story is according to Shell (2008) described as an often pre-scripted sequence of events that might take away a certain degree of freedom when using the mobile application.
The next element that was considered and examined in this research was Mechanic. It refers to the rules and procedures of a game which determine how users are able to reach certain goals and how they are getting rewarded for doing so (Shell, 2008). It showed significance at the 0.10 level for the gaming application but exhibited no significance for gamified
application. In the current body of literature, the discussion and opinions about the importance and significance of mechanics is very much divided. On the one hand academics argued that mechanics exert a strong influence on the intrinsic and extrinsic motivations of people (see e.g. Werbach & Hunter, 2012; Zichermann & Cunningham, 2011), but on the other hand, this motivational power is challenged (see e.g. Lewis et al., 2016; Mekler et al., 2015). Another debate which needs to be addressed is the ongoing highlighting of mechanics such as points, badges and leaderboards, that are predominantly examined as the most important elements in regards to gamification. However game designer argue that there should not be such a focus only on these three elements due to the fact that they are not the most essential part of games.
The next element that this research examined was Aesthetic. Same as for the other two elements, it could not find any significant support regarding usage intention. This finding, based on the relative importance that was attributed by Shell, was quite surprising. Although it is associated with the most direct relationship to a user’s experience, the perception of people that participated in this study resulted in no significance regarding its perceived usefulness and ease of use. A possible explanation could be linked to the phenomenon of visual complexity (Sohn & Seegebarth & Moritz, 2017), which was studied as an indicator of how people process information (Titus & Everett, 1995). The study mentioned that although the improvements in technology offer more possibilities in regards to visual display, it can also result in increased complexity which ultimately would result in a less valuable
experience.
The last element of the four mentioned, Technology, could not reach significance either. This could be related towards its overall “visibility” assigned within the Elemental Tetrad Model.
Another possible explanation would be that, this research captured the perceptions of users
regarding the element. Users of the application seemed not to be interested in the technology that needs to be developed in order to make the application work. The assumption could be made that users do not seem to be affected by the technological aspect, as long as their whole experience is not harmed. According to the research done by Kim & Wang & Malthouse (2015), people delete mobile applications nearly immediately when they are experiencing issues regarding the functionality.
The second part, this research intended to investigate the effectiveness of gamified mobile applications on consumer activities. The three examined variables consideration, purchase intention and engagement found to be all significant. This is in line with the body of current literature (e.g. Alnawas & Aburub, 2016; Bellman et al., 2011; Kim & Wang & Malthouse, 2015; Yang & Asaad & Dwivedi, 2017;).
Further did the results regarding the effect of gamification towards purchase intention, challenge the findings of Bellman et al. (2011), who stated that game-like applications were less impactful in shifting purchase intention.
Conclusion & Recommendations
Managerial implications
Gamified mobile applications seem to provide several opportunities for marketers based on its examined effectiveness on the depicted consumer activities. Challenging the view of Bellman et al. (2011), managers should not be risk averse of implementing the concept of gamification within their mobile application design. They are further advised that gamification entails even more benefits regarding brand loyalty and awareness (Lucassen & Jansen, 2014), which indicates that the concept itself has a promising impact on all relevant marketing activities.
Due to the highly competitive nature of the mobile application industry, marketers can consider the gamification of mobile applications as a reasonable tool in order to gain a competitive advantage.
However, due to the fact, that this research could not find any significant effect for the four elements derived from the Elemental Tetrad Model and examined in regards to usage intention, it seems that other elements play a more important role. Therefore managers are advised to extend their research towards the antecedents of successful gamification regarding the design of mobile applications.
Limitations and Directions for Future Research
The present study has several implications which are going to be addressed in the following.
First it should be addressed that the acquired data and the resulting sample can be considered as non-representative due to the fact that the sample size was rather small and the average age of participants was 25. Moreover, the ratio between genders was with over 60% not typical.
Additionally the ratio between the control and experimental group within this research exhibited the same resulting controversy as depicted previously with gender.
The data furthermore included the responses of people from various different nationalities and cultures which could lead to a bias, regarding the relative small sample size obtained. Thus trying to draw conclusions for larger populations based on the results of this study could be misleading. Further, the results are built upon the perceptions of people regarding gamified and game applications, which could just be tested by providing them visual images of these applications, hence should future research conduct a more practical and realistic experiment, such as programming an application including the tested elements so that people can actually use it and get a real experience and make up their opinions.
