Gamified branded apps: the next level
loyalty program?
Assessing the effect of gamification elements on engagement intentions
towards loyalty programs within branded apps.
Master Thesis | By Aniek Groenwold MSc Marketing Management
Gamified branded apps: the next level
loyalty program?
Assessing the effect of gamification elements on engagement intentions
towards loyalty programs within branded apps.
University of Groningen Faculty of Economics and Business MSc Marketing
Master thesis September 2018
Akkerstraat 45 9717KG Groningen +31620574916
M.A.Groenwold@student.rug.nl Student number: 2339951
First supervisor: dr. J. Berger
Abstract
When consumers use the mobile application (app henceforth) of a brand, customer loyalty increases (Kim, Ling & Sung, 2013). A trend in the development of branded apps is
Preface
Before you lies the master thesis ‘Gamified branded apps: the next level loyalty program?’. Researching and writing this thesis took place between September 2018 and January 2019. The goal of this project is to graduate from the Master Marketing Management at the University of Groningen.
First of all, I want to thank my supervisor, dr. J. Berger, for his valuable insights and his guidance through the process of writing my thesis. Also, I want to thank my family, my friends and all respondents who supported me by sharing and filling in the survey. Without their cooperation, I would not have been able to conduct this research.
Enjoy reading; I hope you can take some interesting findings with you. Aniek Groenwold
Content
Preface 3 Content 4 List of abbreviations 5 Introduction 6 1.1. Academic relevance 7 1.2. Managerial relevance 7 2. Theoretical framework 8 2.1. Loyalty programs 8 2.2. Branded apps 9 2.3. Gamification 10 2.4. Rewards 11 2.4.1. Intrinsic motivation 11 2.4.2. Extrinsic motivation 13 2.5. Effort 13 2.5.1. App savviness 13 2.6. Moderators 14 2.6.1. Gender as moderator 14 2.7. Conceptual model 15 3. Method 16 3.1. Data gathering 16 3.2. Measures 163.3. Plan for analyses 18
4. Results 19
4.1. Sample statistics 19
4.2. Model specification 20
4.3. Measurement model assessment (‘outer model’) 21
4.3. Structural model assessment (‘inner model’) 25
5. Conclusion 27
5.1. Discussion 27
5.2. Theoretical implications 28
5.4. Limitations and recommendations for future research 29
Appendices 31
Appendix 1: Measurement scales 31
Appendix 2: PLS-SEM model 32
List of abbreviations
app Application
AVE Average variance extracted
CBS Centraal Bureau voor de Statistiek
DV Dependent variable
e.g. Abbreviation for exemplī grātiā (‘’for the sake of an example’’) i.e. Abbreviation for id est (‘’in other words’’)
IV Independent variable
IMI Intrinsic Motivation Inventory
PLS Partial least squares
1. Introduction
Mobile devices, ranging from smartphones to tablets, seem to have gradually become a necessity in today’s world. Mobile can be defined as a centrally connected portable device which delivers contextually relevant information (e.g. location, time, etc.) and assists users in various activities on the move (Shankar et al., 2016). Researchers from various scientific fields demonstrate how the adoption of mobile digital devices profoundly affects users’ social and individual behavior (McNaughton & Light, 2013). For example, research by Misra, Cheng,
Genevie & Yuan (2016), indicates that when people have conversations while a mobile device is noticeably present, lower levels of empathy are perceived. On the other hand, Andrews, Luo, Fang & Ghose (2015), who studied the behavior of mobile users under specific social
circumstances, concluded that people possibly evaluate their mobile phones as a “welcome relief”. In highly crowded spaces such as public transport during rush hour, people turn inward and thus become more susceptible to mobile ads. Speaking of internalizing, studies even suggest that users perceive their mobile devices as extensions of their identities (Walsh, White, Kox & Young, 2011).
From a marketing perspective, the enormous growth in mobile adoption offers new possibilities for brands to interact with customers. Through mobile marketing, marketers can view and hear users' voices, personalize marketing communications and target precisely (Lamberton &
Stephen, 2016). Also, as mobile is able to track users' transaction data, it becomes more easily and effective for firms to analyze and predict consumers' needs and wants (Shankar et al., 2016). From 2000 to 2015, a technological transformation took place through which digital, social media, and mobile have adopted an important position in the firm-consumer relationship (Lamberton & Stephen, 2016). Given that consumers stay attached to their devices more than ever, firms can easily enter consumer's personal environment and be there anywhere, anytime (Shankar, Venkatesh, Hofacker & Naik, 2010). Consequently, mobile and digital marketing play a vital role in shoppers’ decision journey. In fact, digital touch points seem to have replaced face-to-face interactions in some stages, as was found in a study conducted by McKinsey & Company (Banfi, Gbahoué & Schneider, 2013).
The interrelated developments in mobile marketing and shopper marketing have resulted in a domain where both area’s converge, often referred to as ‘mobile shopper marketing’. Mobile shopper marketing could be defined as ‘the planning and execution of mobile-based marketing activities that influence a shopper along and beyond the path-to-purchase: from a shopping trigger, to purchase, consumption, repurchase, and recommendation stages’ (Shankar et al., 2016). For instance, retailers can push sales promotions to consumers through the mobile channel to create an immediate shopping trigger (Shankar et al., 2010). Today’s proliferation in mobile technology and devices are expected to continue to rise and future mobile marketing will create endless possibilities (Fritz, Sohn & Seegebarth, 2017). Hence, creating a deeper
1.1. Academic relevance
This study aims to add meaningful insights to the existing theory on gamification in branded apps. Hofacker et al. (2016) argue that when gamification is applied to a mobile platform, it could influence various retailing outcomes and even improve in-store engagement. This implies that gamification could possibly affect various stages of the consumer decision process.
Although it currently attracts many researchers’ attention, limited empirical evidence is published for the effects of gamification on loyalty, customer engagement and motivation (Hamari, Koivisto & Sarsa, 2014). Studies on traditional loyalty programs show that both extrinsic and intrinsic motives increase consumer engagement (Bijmolt et al., 2018). As many companies nowadays replace their traditional loyalty programs with branded apps and thereby make use of
gamification elements, the question arises whether these findings also apply within a branded app context (Li, 2018). According to Blohm & Leimeister (2013) gamification elements (e.g., providing visual evidence of one's progress toward personal goals, social interaction in a community of peers and providing social recognition) offer the opportunity to serve as both intrinsic and extrinsic motivators. Hence, when brands replace their loyalty programs to branded apps, gamification appears to add significant value for the users. Research is needed to assess how the mix of intrinsic and extrinsic rewards affects the marketing effectiveness of mobile gamification (Hofacker et al., 2016; Li, 2018; Blom & Leimeister, 2013).
1.2. Managerial relevance
Given the central role of mobile applications in consumers everyday life, it is expected that marketers will continue to motivate consumer loyalty through the use branded apps. Insufficient research into the gamification of branded apps exists to draw any firm conclusions about the effect of gamification elements on consumer outcomes (Hofacker et al., 2016). Since
2. Theoretical framework
This chapter will firstly discuss relevant concepts, relations and insights on the subjects of loyalty programs, branded apps and gamification. The review of scholarly literature on these subjects then leads to a central research question. Subsequently, the factors mentioned in the research question will be discussed more profoundly.
2.1. Loyalty programs
Loyalty programs could be defined as ‘integrated systems of structured and customized marketing actions, designed to build customer loyalty among profitable customers’ (Bijmolt, Dorotic & Verhoef, 2011). A great deal of brands introduce loyalty programs with the goal of influencing buying behavior and, ultimately, increasing customer loyalty and retention (Demoulin & Zidda, 2009). Generally, consumers need to provide certain demographic information (e.g. name, gender, email address) in order to participate in a loyalty program (García Gómez, Gutiérrez Arranz, & Gutiérrez Cillán, 2012). Subsequently, details on the consumers’ buying behavior are saved in the company’s database and in exchange, the consumer receives a certain type of reward (García Gómez et al., 2012). Traditionally, firms use loyalty programs to reward those customers who frequently make purchases (Leenheer, van Heerde, Bijmolt & Smidts, 2007). However, loyalty programs might also be used to provide incentives for other behaviors which are beneficial for the firm, such as writing product reviews or using specific transaction channels (Bijmolt, Krafft, Sese & Viswanathan, 2018). With the goal of motivating certain behavior and creating loyalty, firms create various sorts of reward structures. They can vary them in multiple aspects, such as the rules for achievement and the balance between symbolic and monetary incentives (Hofacker et al., 2016).
Generally, when consumers evaluate the attractiveness of a loyalty program, two main
Nowadays, more and more firms replace their traditional loyalty programs with reward systems within their branded apps (Li, 2018). A recent study into coalition loyalty programs (e.g. an overlapping loyalty program which is used by two or more companies), shows that adding convenience by way of introducing a mobile app increases engagement and especially encourages participation among customers who usually did not redeem their rewards (Wang, Krishnamurthi & Malthouse, 2018). In the next paragraph, recent scientific literature on the subject of mobile applications will be discussed.
2.2. Branded apps
When Leeflang, Verhoef, Dahlström & Freundt (2014) asked management teams of large firms for their intended use of digital media over the next 4 years, social media and mobile apps represented the biggest growth areas. Since 82% of peoples’ digital media usage time is spend on mobile apps, many firms have already developed their own branded apps and it is expected that this trend will continue (Hsu & Lin, 2015). Branded apps should be notably distinguished from other categories, such as social networking apps (e.g. Facebook, Instagram) and utility apps (e.g. weather forecasting). Branded apps are typically downloadable for free and “prominently display a brand identity, often via the name of the app and the appearance of a brand logo or icon, throughout the user experience” (Bellman, Potter, Treleaven-Hassard, Robinson & Varan, 2011). Marketing through branded apps is found highly effective, as it enables consumers to access personalized content and services and interact with the brand irrespective of time and location (Alnawas & Aburub, 2016). Researchers found that branded apps are able to improve shoppers’ awareness of products and offers (Shankar et al., 2016). Also, a direct link was found between active engagement with branded apps and shoppers’ spending levels (Kim, Wang & Malthouse 2015). For this reason, it is important for marketers to monitor their branded app users’ engagement.
Importantly, the first condition for users to be able to engage with an app is to download it. Convincing consumers to actively decide to download an app might be an initial challenge for many marketeers (Wang, Kim, & Malthouse, 2016). Nowadays, Android users are able to choose between 3,8 million apps available for download and Apple's App Store offers 2 million unique apps (Statista, 2018). Due to this proliferation, it becomes highly essential for apps to align with its users. By using public data, Garg & Telang (2013) inferred the relationship between an app’s rank within Apple’s App store and the demand for an app. They concluded that the top ranked paid app for iPhone generates 150 times more downloads compared to the 200th ranked app. Given that these rankings are based on user evaluations, user relevance and ease of use seem to be important determinants for app success (Garg & Telang, 2013). As firms will continue to compete for mobile users to download their branded app, marketers should thus offer user-relevant solutions.
shoppers discontinue using the app (Kim, Wang & Malthouse, 2015). Given these points, once consumers have downloaded the app, the challenge for marketers remains to persuade app users to continue their usage (Wu, 2015). Researchers have suggested multiple factors through which marketers could increase engagement among mobile app users. Among these are perceived ubiquity, informativeness and personalization (Kim, Baek, Kim & Yoo, 2016),
employing unique mobile capabilities (e.g. personal history, geographic information) (Shankar et al., 2016) and creating interactive experiences which are perceived as personal and pleasurable (Kim, Jung & Malthouse, 2015). An approach in which multiple of these factors mentioned above come together is through gamification. The following chapter will clarify how gamification can be used as a method to induce app stickiness and drive customer engagement.
2.3. Gamification
Although the non-game use of game-like elements (e.g. collecting points, air miles, etc.) has been used for decades in traditional loyalty programs, gamification is a very recent trend in the field of branded apps (Blohm & Leimeister 2013). Gamification implies ‘’a process of enhancing a service with affordances for gameful experiences in order to support user’s overall value creation’’ (Huotari & Hamari, 2012: 20). Compared to traditional reward systems, gamification provides added social and motivational benefits through product usage rather than only
expenditures (Blohm & Leimeister, 2013). By providing fun and flow experiences, gamification is found capable of enhancing collaboration among users (Hsu & Lin, 2016). Gamification can also be used in a work context, applying game design to real-world tasks in order for employees to have fun in pursuing work-related goals (Jin, 2016). Although gamification has only been around shortly within the field of mobile marketing, it offers large future potential in this field, as one of its effects is to increase consumers’ retention of mobile apps (Hofacker et al., 2016). By inducing app stickiness, gamification is argued to be effective for driving customers’
engagement (Robson et al., 2015). Although measuring the actual engagement of app users might be challenging due to privacy concerns and self-selection bias (Shankar et al., 2016), user engagement intention is an effective predictor, since users' engagement motivation is a conscious decision choice (Davis, 1989).
Recently, Li (2018) found that gamification functions as a significant push factor for consumers to switch from traditional loyalty programs to loyalty programs within branded apps. A possible explanation for this effect is that engaging in game-like experiences deepens the
How does the junction of intrinsic rewards, specifically those facilitated by the use of gamification elements, extrinsic rewards and perceived effort affect the user engagement intention towards loyalty programs within branded apps?
The factors involved in this research question will be described in the following chapters, which will then lead to corresponding hypotheses. Two main subjects within this research paper are rewards and effort. When consumers evaluate the attractiveness of a loyalty program, the components taken into consideration are the rewards which can be earned and the required effort for receiving these rewards (Drèze & Hoch, 1998). In the following paragraphs, these two components will be discussed respectively.
2.4. Rewards
Motivation could be defined as one’s ‘willingness to apply time and energy to a task, determined by anticipatory expectations of outcomes and the resultant satisfaction of needs’ (Cooper & Jayatilaka, 2006). The self determination theory (Deci & Ryan, 1985) states that people’s motivation emanate from psychological needs and growth tendencies. Whereas extrinsic motivation arises from sources outside of the task itself, such as monetary rewards; intrinsic motivation arises from positive evaluations of a task itself (Lepper & Henderlong, 2000).
2.4.1. Intrinsic motivation
Traditional loyalty programs, which relied almost exclusively on economic benefits (e.g. extrinsic rewards) such as discounts and free goods, were evaluated to be very costly (Melnyk & van Osselaer, 2012). Moreover, whereas extrinsic motives often fail to increase behavioral motivation in the long run, intrinsic motives seem to have longer lasting effects (McGonigal, 2011). Therefore, it is not surprising that many companies nowadays turned to psychological rewards as a way to stimulate customer loyalty (Leenheer et al., 2007).
Rummel & Feinberg (1988) define intrinsic motivation as ‘the doing of an activity for its inherent satisfactions rather than for some separable consequence’. People act according to their intrinsic motivation when one engages in activities just by the mere pleasure of carrying them out (Deci & Ryan, 1985). Regardless of whether extrinsic rewards exist, intrinsic motivation is found to be a moderate to strong predictor of performance (Cerasoli, Nicklin & Ford, 2014). In the light of the current research, it is important to note that gamification elements are found to arouse the intrinsic motivation of consumers in a number of ways (Hofacker et al., 2016). Game-specific symbolic rewards, such as badges and status updates, provide visual evidence of consumers’ performance, document progress toward personal goals, facilitate social
interaction and function as an instrument of social recognition (Blohm & Leimeister, 2013). These intrinsic rewards are able to increase the engagement of mobile gamers (Hofacker et al., 2016).
include activities which are social in nature. Hence, exogenous use and gratifications are for relating an individual into social contexts and socializing with other people. In line with Gerlich et al. (2015), this research will also differentiate between endogenous motives and exogenous motives. In terms of exogenous motives, mobile gamification is found to create value by offering skill development, information acquisition and learning, which expands users' knowledge and expertise (Hofacker et al., 2016). According to cognitive evaluation theory, receiving feedback during a certain action can enhance intrinsic motivation for that action (Deci et al., 1999). Through the use of mobile technology, gamification offers the unique possibility to reward
consumers through self-learning. This refers to the process by which individuals gain knowledge about their abilities to perform specific activities or actions (Nambisan and Baron 2009). An effective way of reinforcing such learning is by explicitly tracking and reporting achievements (Blohm & Leimeister, 2013). Therefore, it is hypothesized that:
H1a: an increase in learning benefits created through gamification elements positively affect user engagement intentions
In terms of exogenous motives, gamification is found to offer certain social value. Van Lange, Bekkers, Schuyt & Vugt (2007) argue that, although gaming theories are often based on the assumption of rational self-interest, it is also able to provide social value to users.Gamification was found to offer users social value, both through social exchange and social recognition (Blohm & Leimeister, 2013). On the one hand, gamification is found to facilitate social exchange (Blohm & Leimeister, 2013), as mobile techniques may enable interactions with both the brand and other consumers. For example, gamification elements can create social value through interactions involving appreciation, compliments, and reciprocal exchange with others (Nambisan & Baron, 2009). Therefore, it is hypothesized that
H1b: social exchange created through gamification elements positively affect user engagement intentions
2.4.2. Extrinsic motivation
Extrinsic rewards refers to the monetary savings consumers receive from engaging in loyalty programs (Bijmolt, et al., 2018). According to Venkatesh et al. (2012), monetary value is a predictor of consumer intention with regard to technology usage. If consumers believed economic benefits were being provided by branded apps, their intention to switch from membership cards would be enhanced (Li, 2018). Hence, monetary rewards are expected to increase users’ engagement with a loyalty program. Therefore, it is hypothesized that
H2: Monetary rewards positively affect user engagement intentions
2.5. Effort
In the context of both loyalty programs and gamification, the balance between rewards and the required physical and mental effort seems to be highly important. Cognitive effort is typically seen as a type of cost that consumers seek to avoid or at least reduce (Gretzel & Fesenmaier, 2006). Requirements which are perceived as too difficult will cause consumers to abandon a program (Albuquerque & Nevskaya, 2015). On the other hand, however, was found that offering rewards at too low a level of difficulty will cause boredom (Csikszentmihalyi, 2014). Hence, when marketers make use of any kind of reward system, they should pay special attention to the reward–effort relationship (Hofacker et al., 2016). Kivetz & Simonson (2002) define the perceived effort of a loyalty program as ‘any inconvenience inherent in complying with the program requirements, such as making a special effort to buy at a particular store, purchasing more than one would have otherwise bought, or repeatedly engaging in a certain task (e.g., completing surveys or browsing websites)’.Participants of loyalty programs typically follow the principle of least effort (Wang, Krishnamurthi & Malthouse, 2018), which implies that they prefer greater absolute rewards and lower absolute efforts. Therefore, it is hypothesized that
H3a: An increase in perceived effort negatively affects user engagement intentions
2.5.1. App savviness
One factor that might influence consumers’ perceived effort with regards to loyalty programs within branded apps, is their personal convenience with those apps. Parasuraman & Grewal (2000) studied the impact of technology on drivers of customer loyalty and found that
consumers highly differ in the degree to which they embrace technologies to accomplish goals. Non-tech-savvy teachers are found to be less effective in the use of gamified education
H3b: An increase in app user savviness reduces the perceived effort of engaging with a gamified loyalty program within a branded app
2.6. Moderators
Consumer traits are likely to moderate how gamification design affects marketing effectiveness (Hofacker et al., 2016). The effects of technology design to enhance gamification is argued to depend on individual consumer experience, age, and gender (Venkatesh, Thong, and Xu 2012). Research into online gaming illustrates that, although the majority of consumers increase
participation in response to rewards, a minority, with stronger use habits and stronger intrinsic motivation, are found not to be affected by virtual reward offers (Nevskaya and Albuquerque 2015). Other research suggests that the attractiveness of gamification features will depend on consumers' existing game use (Hartmann, Jung, and Vorderer 2012). As the male–female dichotomy is found to be the most fundamental dichotomy in society, it affects both the social and cultural meanings associated with developing a marketing strategy (Prakash & Flores, 1985). Based on self-construal and gender scheme theories, gender also plays a crucial role in the engagement with the adoption of technologies (Okazaki, Navarro, & López-Nicolas, 2013). For this reason, the current study will specifically focus on the moderating effect of gender.
2.6.1. Gender as moderator
A factor which is expected to moderate the effect of intrinsic rewards on the consumer response to gamified branded apps is gender. Melnyk & van Osselaer (2012) have found that men and women differ in their response to psychological rewards in the context of loyalty programs. Therefore, assuming that men and women are intuitively aware of their response to
psychological rewards, one could expect that men and women differ in their attitude towards engaging with a gamified loyalty program, depending on the potential rewards the loyalty program offers. Melnyk & van Osselaer (2012) have found that men react more positively to gaining high status, especially when their status was highly visible to others. This implies that, for example, a gamified loyalty program which publicly track consumers’ achievement, will be evaluated as more attractive to men.
H4a: Being a male positively moderates the effect of social recognition rewards on attractiveness of engaging with the gamified loyalty program
On the other hand, females engage in more elaborate and motivated social perception and tend to be more interdependent and attentive to information that pertains to others (Meyers-Levy & Loken, 2015). This theory suggest that women will be triggered more by the social exchange value of psychological rewards
2.7. Conceptual model
The hypotheses mentioned above result in the following conceptual model:
3. Method
This study aims to test the aforementioned hypotheses by carrying out a statistical analysis with quantitative data. The following chapter will elaborate on the method that will be used for
obtaining results.
3.1. Data gathering
As secondary data was not available, this study will make use of primary data for testing the hypotheses. This implies that the data will be collected by the researcher itself (Malhotra, Hall, Shaw & Oppenheim, 2006). Due to time limitations, data will be gathered through an online survey, as this method offers several advantages in terms of flexibility, accuracy and efficiency (Tan & Teo, 2000). Respondents will be sampled based on relative ease of access. The survey will be spread among relatives of the researcher by means of social media (Facebook and LinkedIn) and WhatsApp. In order for results to be reliable, the study requires a large sample of respondents. Therefore, relatives will also be asked kindly to share the survey with their
networks. Hence, the sampling method of this study can be characterised as nonprobability sampling, specifically convenience and snowball sampling (Blumberg, Cooper & Schindler, 2011).
The survey will be hosted by the survey platform Qualtrics, as this software is highly specialized in facilitating surveys and has an official partnership with the University of Groningen. In the message for spreading the survey, a description of branded apps with gamified loyalty programs will be provided. Recipients will be asked if they currently use such a gamified loyalty program. If yes, the person is asked to start the survey, while keeping in mind the gamified branded app they used the last. If no, the person is kindly asked to not participate in the study.
The start of the survey includes a short preface which describes the purpose of the study and offers the guarantee that respondents’ personal data will be treated anonymously and
confidentially. After confirming that one has comprehended the preface, respondents are send to the main part of the survey. Within this part, respondents are asked various questions regarding their experience with gamified loyalty program within branded apps. With multiple validated measures (see 3.2.) adapted from prior research, respondents are questioned on learning benefits, social benefits, recognition benefits, monetary rewards, perceived effort, app savviness and their mobile engagement intention regarding the app they have in mind. Finally, the participants are asked to indicate their gender, age and education level. These descriptive statistics will be obtained to test whether the sample is diverse and representative. Various other app oriented studies have been using these demographic factors as control variables (Hsu & Lin, 2015; Hew, Lee, Ooi & Wei, 2015; Kim, Wang & Malthouse, 2015). Also, the data on gender will be used in the analysis for testing its’ moderating effect.
3.2. Measures
measured with a seven-point Likert scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). Namely, using equal scales will enable the author to draw parallels between items efficiently during subsequent analyses. All scales of the measures which are used and described below can be found in appendix 1.
User engagement intention (DV)
Kim, Kim & Wachter (2013) have proposed a mobile user engagement model by which one can explain mobile user engagement intention through user's motivations. Their scale, which will be adapted in this study, includes two items regarding user engagement intentions.
Intrinsic rewards
Alnawas & Aburub (2016) developed a scale to measure learning benefits (4 validated items) in the context of branded mobile apps.This scale will be adapted to measure to which degree respondents experience learning as an intrinsic reward for using the gamified loyalty program within a branded app.
A scale for perceived benefits of loyalty programs developed by Mimouni-Chaabane & Volle (2010) will be adapted to measure to which degree respondents experience social recognition (4 items) and social exchange (3 items) by using the gamified loyalty program within a branded app. More precisely, social recognition refers to which degree consumers perceive that the branded app publicly acknowledges their status. Social exchange, on the other hand, refers to which degree consumers perceive that the branded app facilitates interaction with the brand and with other consumers.
Monetary rewards
Mimouni-Chaabane & Volle (2010) also developed and validated 3 items by which one can measure perceived monetary savings through loyalty programs. This scale will be used for the current study to measure extrinsic, monetary rewards.
Effort
Respondents’ perceived effort will be measured through the Intrinsic Motivation Inventory (IMI) (Deci & Ryan, 2000). Their construct ‘Effort/Importance’ consists of 5 items. The three 3 items regarding effort are adapted to be used in the current study. The general term ‘activities’ is specified by ‘using this app’.
As it is expected that perceived effort greatly depends on the degree to which a respondent is experienced and keen in using apps, respondents’ mobile app savviness will be measured as well. More precisely, this will be measured by the 5 items regarding ‘ability’ imported from the tech-savviness scale (Shanahan & Hyman, 2010). The construct originally measures
consumers’ ability to adopt technologies, but in order to match the current research objective, the term ‘technology’ was changed to ‘apps’.
Construct: moderator
3.3. Plan for analyses
Ultimately, this study aims to find out the weighted effects of the independent variables (intrinsic and extrinsic rewards and perceived effort) on the dependent variable (mobile engagement intention) and to which degree app savviness predicts perceived effort. Furthermore, it aims to find out to which degree being female has a) a positively moderating effect on the relation between social benefits and mobile engagement and b) a negatively moderating effect on the relation between status benefits and mobile engagement. As the conceptual model includes more than one explanatory variable, the appropriate method for this research is a multiple regression analysis (Malhotra, Hall, Shaw & Oppenheim, 2006).
In the current study, latent variables are used to measure the various constructs. Latent variables are those which are not directly observed but are rather inferred from other variables that are directly measured (Hair, Sarstedt, Hopkins & Kuppelwieser, 2014). A method which is able to evaluate the measurement of latent variables, while also testing relationships between latent variables, is Structural Equation Modelling (SEM) (Babin, Hair & Boles, 2008). SEM is a combination of factor analysis and multiple regression analysis (Cassel, Hackl & Westlund, 1999).
A type of SEM which seems to have become a quasi-standard in marketing research lately, is Partial Least Squares Structural Equation Modelling (PLS-SEM)(Hair, Ringle & Sarstedt, 2011). According to Hair, Sarstedt, Hopkins and Kuppelwieser (2014), one argument for using
PLS-SEM is that it is applicable to small sample sizes. While smaller sample sizes highly decrease aspects like model fit and statistical power when using other types of SEM, PLS-SEM is still effective in these cases (Shah & Goldstein, 2006). Barclay, Higgins & Thompson (1995) suggested two requirements for the minimum sample size for a PLS model. These
4. Results
In the following chapter, the research sample will be discussed first. Afterwards, the PLS-SEM model for this study is specified. Subsequently, the measurement model and the structural model are analyzed. PLS-SEM uses a two-step approach, in which the first step is to determine the measurement model (often referred to as the ‘outer model’). When the measurement model shows sufficient validity, the structural model (often referred to as the ‘inner model’) can be analyzed (Henseler & Sarstedt, 2013).
4.1. Sample statistics
The survey launched at November 27th, 2018. After two weeks, the survey was taken down and the records were gathered. First of all, the dataset was prepared for further analyses. This included cleaning the dataset from missing values, converting textual results into numerical values and inverting the results of reverse scored items. Before the start of the data preparation process, 117 records were found. As it turned out that 8 respondents did not fully complete the survey, those records were deleted. Thus, the final sample consisted of 109 respondents (n=109).
A summary of the sample size characteristics can be found in the table below. Compared to data on the Dutch population according to Centraal Bureau voor Statistiek (CBS, 2018), female respondents were highly overrepresented within the final sample. Almost two thirds of the respondents is Dutch. As the researchers’ network was an important source for collecting respondents, students with various other nationalities (e.g. Bulgarian, Polish, Greek, Italian) took part in the survey. All of these students however, are currently living in the Netherlands. Concerning education, none of the respondents has indicated to have primary education as highest level of education.
Table 1: Gender, Nationality and educational level of the sample size Gender Nationality Educational level
Female 70,6%
Male 29,4% Dutch 69,7% Other 30,3% Secondary education 22% Community college (MBO) 18,4%
Polytechnic degree (HBO) 31,2% Bachelor's degree (WO) 12,8% Master's degree (WO) 15,6%
Table 2: Age groups of the sample compared to population
Age group Survey Dutch population of mobile users (CBS, 2018) 15-24 25-44 45-64 65+ 53,2% 37,6% 8,3% 0,9% 12,2% 24,2% 26,6% 12,3%
The demographic information on this sample size can largely be explained by the fact that the questionnaire was distributed within the authors’ network. This should be kept in mind when drawing conclusions. Overall, the sample seems to be a moderate representation of the mobile user population according to Dutch data (CBS, 2018).
4.2. Model specification
A PLS-SEM analysis starts with specifying the structural model and the measurement model (Hair, et al., 2014). Whereas the structural model displays relationships between the research variables, the measurement model displays relationships between the survey items (or
‘indicators’) and their corresponding variable (or ‘construct’) (Hair, Hult, Ringle & Sarstedt, 2016).
The structural model consists of a path model, which indicates the location of the variables and the relationships between them. As one can see in appendix 2, the structural model resembles the conceptual model of this research (which is shown in figure 1). Variables within the path model are typified as either exogenous or endogenous (Hair et al., 2014). Exogenous constructs are characterized by not having an arrow directed towards them (Hair, et al., 2016) (e.g.
‘Learning’, ‘Recognition’, ‘Social’, ‘Monetary rewards’ and ‘App savviness’). On the other hand, endogenous constructs have arrows directed towards them, which implies that they are
explained by other constructs (Hair, et al., 2016) (e.g. ‘Mobile user intention’ and ‘Effort’) . One might notice that within this model, ‘Effort’ is both an endogenous construct and an independent variable, as it is placed between two constructs.
Once the structural model is clear, the next step is to specify the measurement model. The specification of a measurement model is crucial for an effective analysis, ‘as the relationships hypothesized in the inner model are only as valid and reliable as the outer models’ (Hair et al., 2014: 111). As can be seen in appendix 2, the measurement model within this research includes the survey items. The arrows within the measurement model are directed from the survey items towards the constructs, which implies that all indicators are reflective (Hair et al., 2016).
4.3. Measurement model assessment (‘outer model’)
As mentioned before, the survey used for this research is based on measurement scales which were used in prior studies. Notably, only selections of items from these scales have been used and the items were adapted to fit the current study. As the formal procedure for PLS-SEM analysis dictates, the first step is to evaluate the measurement model (Hair et al., 2014). This evaluation will be based on the reliability and validity of the constructs, which are measured by running a factor analysis with the PLS algorithm.
Reliability
A first step in assessing the reflective constructs within the measurement model is to evaluate reliability. The internal consistency reliability of the constructs will be measured by means of composite reliability (Hair et al., 2014). The reason for using composite reliability instead of the frequently used Cronbach’s Alpha, is that PLS-SEM is able to accommodate differences in the indicator loadings. Also, one can avoid to underestimate internal consistency reliability based on the number of items in the scale, which bias is often associated with Cronbach’s alpha (Hair et al., 2016).
Table 3: Internal consistency reliability of the constructs.
Cronbach’s Alpha rho_A Composite Reliability
App Savviness 0.885 0.894 0.888 Effort 0.588 0.837 0.662 Learning 0.761 0.788 0.775 Mobile engagement 0.820 0.830 0.823 Monetary 0.871 0.896 0.873 Recognition 0.869 0.876 0.870 Social 0.887 0.891 0.888
A rule of thumb within PLS-SEM analyses prescribes that composite reliability should be
higher than 0.70 (Hair et al., 2014). Concluding from the values in the table above, all constructs except for ‘Effort’ are found to be internally consistent. As Effort is rather close to 0.70, it was concluded that no severe reliability issues were found in the measurement model.
Validity
The second step in assessing the reflective constructs within the measurement model is to evaluate validity. Firstly, convergent validity will be analyzed and afterwards discriminant validity.
As mentioned before, the current research only includes reflective indicators. Reflective indicators relate to a construct by means of loadings (Hair et al., 2014). A rule of thumb prescribes that convergent validity is arguably sufficient when each item has outer loadings above 0.70 (Hair et al., 2016). Hence, loadings lower than .7 are marked bold (see table 4). Table 4: Outer loadings of items
App Savviness
Effort Learning Mobile engagement
Again, one can conclude that the construct ‘Effort’ scores relatively low. This implies that the items Eff1 and Eff2 do not seem to be highly related to its’ construct. McAuley, Duncan, and Tammen (1987) examined the validity the IMI, among which the construct for effort, and found strong support for its’ validity. As Hulleman (1999) argues that only factor loadings which are smaller than 0.4 should be considered for deletion, it was decided to not delete any items. Average Variance Extracted (AVE)
Table 5: Average Variance Extracted (AVE) of constructs
Constructs AVE App Savviness 0.615 Effort 0.437 Learning 0.468 Mobile engagement 0.701 Monetary 0.700 Recognition 0.627 Social 0.725
A rule of thumb within PLS-SEM analyses prescribes that convergent validity is arguably sufficient when each construct’s average variance extracted (AVE) is 0.50 or higher (Hair et al., 2016). As one can conclude from the table above, although the AVE of the constructs Effort and Learning are relatively close, they do not meet the threshold of .5. This implies that the amount of variance which is captured by those constructs is relatively low compared to the amount of variance due to measurement error. Malhotra & Dash (2011) argue however, that AVE is a highly strict measure of convergent validity and that ‘On the basis of composite reliability alone, researchers may conclude that the convergent validity of the construct is adequate, even though more than 50% of the variance is due to error’ (Malhotra & Dash, 2011). For this reason, and for the fact that Effort and Learning are reasonably close to .5, composite reliability will be
considered as leading for now. This still implies however, that convergent validity is insufficient for the construct Effort. The measurement model evaluation is continued, however with caution for drawing any conclusions in which the construct of effort is involved.
Discriminant validity
al., 2016). This implies that the square root of the AVE must be greater than any of the interfactor correlations.
Table 6: Discriminant validity of constructs App
Savviness
Effort Learning Mobile engagement Monetary Rewards Recognition Social App savviness 0.784 Effort 0.508 0.661 Learning 0.532 0.515 0.684 Mobile engagement 0.290 0.090 0.352 0.837 Monetary Rewards 0.192 0.060 0.365 0.175 0.837 Recognition 0.542 0.340 0.618 0.113 0.453 0.792 Social 0.423 0.404 0.683 0.361 0.419 0.747 0.852
In the accompanying table, the square root of the AVE can be found on the diagonal. One can conclude that technically the Fornell and Larcker (1981) criterion was met. One should note however, that the constructs social and learning are found to have a relatively high interfactor correlation and are not highly distinct.
In terms of reliability and validity, six out of seven constructs were measured by effective indicators. As no strong conclusions could be drawn with regards to the construct Effort as it was measured initially, it was decided to perform a post-hoc analysis.
Post-hoc analysis
Taking a closer look at the measurement items for Effort, it appears that Eff1 is formulated as an expectation, while Eff2 and Eff3 are formulated as an actual experience. Perhaps this
observation could explain the relatively low outer loadings that were found for Eff1 and Eff2. Therefore, a post-hoc analysis was performed by repeating the tests for convergent validity while leaving out item Eff1. Consequently, increased loadings for Eff2 and Eff3 were found (see table 7) and the initial AVE value of 0.437 changed into 0.727, which meets the threshold of .5. Hence, the validity of the measurement construct Effort increases when item Eff1 is excluded. Table 7: Outer loadings after deleting Eff1
App Savviness
Effort Learning Mobile engagement
Monetary Recognition Social
Eff2 0.756
To conclude the first part of the two-step approach, the measurement model could be evaluated as moderately well. Hence, it is appropriate to continue with the second step: a structural model assessment.
4.3. Structural model assessment (‘inner model’)
In the following paragraph, a path analysis of the structural model will be discussed. For the final results of this analysis to be considered valuable, it is highly important that the independent variables are not correlated to each other, as this would imply that they measure the same construct. Multicollinearity occurs when predictor variables are highly correlated to each other, therefore predicting the variance of another variable, which ultimately leads to unreliable
regression coefficient estimates (Malhotra, 2010). Hence, a multicollinearity test was conducted by assessing the values of the inner variance inflation factors (VIFs) (see table 7).
Table 7: Inner Variance Inflation Factors of constructs
Effort Mobile engagement
App Savviness 1.000 Effort 1.417 Learning 2.305 Mobile engagement Monetary Rewards 1.320 Recognition 1.621 Social 2.410
Looking at the VIF scores, one can conclude that based on a cut-off score of 5 (Hair, Ringle & Sarstedt, 2011), no severe multicollinearity issues were found.
Subsequently, path coefficients were calculated by means of bootstrapping. In bootstrapping, subsamples are created with observations randomly drawn (with replacement) from the original set of data (Hair et al, 2014). To ensure stability of results, the number of subsamples should be large. As advised by SmartPLS, 5000 subsamples were created. As the study does not aim to highlight any indicator as particularly important or dominant, initial outer weights were left in their default setting. To test the hypothesized moderating effect of gender on ‘social benefits’ and ‘recognition benefits’, an interaction term was added to the inner model which was specified as a product indicator. This approach uses all possible pair combinations of the items that are used for the independent variable and the moderator variable. These combinations then serve as a new indicator ("product indicator") for the interaction effect.
In a post-hoc analysis it appeared that the validity of the measurement construct Effort
structural model seems to be fairly robust regardless the items that were included for the measurement of effort. The following results were found (see table 8).
Table 8: Path coefficients. ***p<0.01; **p<0.05; *p<0.1 Original Sample Sample Mean (M) Standard Deviation T STatistics P Values
App Savviness -> Effort 0.438 0.458 0.080 5.514 0.000***
Effort -> Mobile engagement -0.055 -0.043 0.126 0.441 0.660
Gender -> Mobile engagement -0.041 -0.034 0.097 0.426 0.670 Learning -> Mobile engagement 0.205 0.235 0.118 1.732 0.083* Monetary -> Mobile engagement 0.045 0.051 0.103 0.439 0.661 Recognition -> Mobile engagement -0.201 -0.155 0.184 1.088 0.276 Social -> Mobile engagement 0.330 0.276 0.139 2.381 0.017** Moderating effect recognition -> Mobile engagement -0.099 -0.111 0.141 0.701 0.483
Moderating effect Social ->
Mobile engagement -0.043 -0.044 0.121 0.375 0.721
5. Conclusion
In the following chapter, the results of the analyses will be discussed and theoretical and managerial implications will be drawn. Subsequently, limitations to the study and suggestions for future research will be described.
5.1. Discussion
The objective of this research is to explore how the mix of intrinsic and extrinsic rewards, and the perceived effort to participate in a gamified loyalty program affects the engagement with a branded app. By means of a PLS-SEM analysis, 8 hypotheses were tested. These hypotheses and corresponding findings are summarized in table 9, which will receive further elaboration in the following section.
Table 9: Results of hypotheses testing
Hypothesis Result
H1a: an increase in learning benefits created through gamification elements positively
affect user engagement intentions
Confirmed
H1b: social exchange created through gamification elements positively affect user
engagement intentions
Confirmed
H1c: social recognition created by gamification elements positively affect user
engagement intentions
Rejected
H2: Monetary rewards positively affect user engagement intentions Rejected
H3a: An increase in perceived effort negatively affects user engagement intentions Rejected
H3b: An increase in app user savviness reduces the perceived effort of engaging with a
gamified loyalty program within a branded app
Confirmed
H4a: Being a male positively moderates the effect of social recognition rewards on
attractiveness of engaging with the gamified loyalty program
Rejected
H4b: Being a female positively moderates the effect of social exchange rewards on
attractiveness of engaging with the gamified loyalty program
Rejected
Intrinsic rewards
Regarding the effect of intrinsic rewards on user engagement intentions, two out of three
suggested by Dowling and Uncles (1997), who argue that consumers might become frustrated by these ‘status benefits’, as they may induce feelings of inferiority.
Extrinsic rewards
The hypothesized positive effect of monetary rewards on user engagement intentions was rejected. This actually substantiates research by Lee and Cunningham (2001), who claim that consumer benefits which are not tied to financial incentives are the most important indicators for the success of a loyalty program. Another research which shares this idea is by Rosenbaum, Ostrom and Kuntze (2005), who found that consumers are more loyal and less predisposed to competitor switching when loyalty programs offer social benefits instead of simply using financial incentives.
Perceived effort
As participants of loyalty programs typically follow the principle of least effort (Wang, Krishnamurthi & Malthouse, 2018), it was hypothesized that perceived effort would have a negative effect on user engagement intentions. However, no significant evidence was found. When consulting literature, it appears that researchers in the past found similar results
regarding effort after using the Intrinsic Motivation Inventory for their study. Markland & Hardy (1997) explain this finding by arguing that effort is a motivational consequence rather than its’ predictor.
Moderating effect of gender
No evidence was found for the hypothesized moderating effects of gender. A likely explanation for this finding is that, rather than simply one’s gender, social and psychological mechanisms predict how individuals react towards social aspects of digital environments (Nysveen, Pedersen & Thorbjørnsen, 2005).
5.2. Theoretical implications
This study contributes to the literature on loyalty programs by acknowledging that many traditional loyalty programs are replaced to branded apps and therefore highlighting the
importance of identifying key factors for user engagement intentions towards these kind of apps. The results confirmed that learning benefits and opportunities for social exchange positively affect users’ engagement intentions towards a branded app.
5.3. Managerial implications
With respect to the hypotheses that were confirmed by this study, valuable information could be at hand for managers. The results confirmed the effect of social exchange benefits and learning benefits on mobile engagement intention. This implies that app functions which enable
that monetary value motivates consumers to participate (Venkatesh et al., 2012), the
hypothesized positive effect of monetary rewards on user engagement intentions was rejected. Hence, managers are advised to reconsider the use of financial incentives and further explore ways in which intrinsic motivational principles could be applied to functions of branded apps.
5.4. Limitations and recommendations for future research
Clearly, this research has a number of limitations. To develop a comprehensive understanding of the effect of gamification elements on loyalty programs within branded apps and consumers intentions to use these apps, further research must be done.
One of the limitations is the rather small sample size. Although PLS-SEM is found to be still effective with small sample sizes, increasing the sample size would improve the power of the hypothesis testing. Namely, the minimum sample size requirements by Barclay et al. (1995) fail to take into account the effect size, reliability, number of indicators and other factors that are known to affect power (Henseler, Ringle & Sinkovics, 2009). Also, within the sample, female respondents were highly overrepresented. This might be an explanatory factor for the insignificant effects that were found for gender as a moderator variable (e.g. H4a and H4b). Furthermore, the sample that was used for this study consists of people who currently use a loyalty program within a branded app. Hence, the arguments for consumers to not participate in such apps are unclear. Irrespective of the hypothesized effect of perceived effort, a significant path coefficient was found for the effect of app savviness on effort. It might be the case that brands exclude non-app savvy consumers when replacing their loyalty program to a branded app. Thus, the question arises to which degree the app savviness of a brand’s consumer base determines the success of a loyalty program within a branded app. Future research could address this topic.
Another limitation within the current research was the dubious function of effort. As concluded from the initial measurement model evaluation, no meaningful conclusion could be drawn on the construct of effort, due to low reliability and validity findings. Hence, a post-hoc analysis was performed. However, no differences were found in the path coefficients after adjusting the effort construct. As mentioned earlier, a possible explanation for this finding might be that effort is a motivational consequence rather than its’ predictor (Markland & Hardy, 1997). Researchers are encouraged to which degree this claim holds in the context of loyalty programs within branded apps.
Furthermore, the current research did not differentiate in types of gamification elements or degrees of gamification. One could argue that certain gamification elements are more effective than others, and high degrees of gamification might be distracting consumers from a core task or offer. Future research should also tackle the question whether customers become loyal to the rewards or the branded app. It is argued that in order for gamified rewards to enhance loyalty, they must be integrated with the firm's goods and services or real world currencies (Hofacker et al., 2016).
Appendices
Appendix 1: Measurement scales
Learning (Alnawas & Aburub, 2016)IRLearn1: The app helps me to obtain solutions to specific product-usage related problems IRLearn2:The app provides information that helps me make important decisions
IRLearn3: The app enhance my knowledge about the product and its usage IRLearn4: The app helps me better manage my money
Recognition (Mimouni-Chaabane & Volle, 2010) IRRecog1: The company takes better care of me IRRecog2: I'm treated better than other customers IRRecog3: I'm treated with more respect
IRRecog4: I feel I am more distinguished than other customers Social (Mimouni-Chaabane & Volle, 2010)
IRSocial1: I belong to a community of people who share the same values IRSocial2: I feel close to the brand
IRSocial3: I feel I share the same values as the brand Monetary rewards (Mimouni-Chaabane & Volle, 2010) By using this app...
ERMon1: I shop at a lower financial cost ERMon2: I spend less
ERMon3: I save money Effort (Deci & Ryan, 2000)
Eff1: I have to put a lot of effort into this.
Eff2: I do not try very hard to do well at this activity. (R) Eff3: I try very hard on this activity.
App Savviness (Shanahan & Hyman, 2010)
AppSav1: I keep up with the latest app developments. AppSav2: Other people come to me for advice on new apps. AppSav3: People think of me as an app expert.
AppSav4: I'm among the first to acquire new apps. AppSav5: I enjoy the challenge of figuring out apps. Mobile Engagement Intention (Kim, Kim & Wachter, 2013)
Sources
Alnawas, I., & Aburub, F. (2016). The effect of benefits generated from interacting with branded mobile apps on consumer satisfaction and purchase intentions. Journal of Retailing and
Consumer Services, 31, 313-322.
Andrews, M., Luo, X., Fang, Z., & Ghose, A. (2015). Mobile ad effectiveness: Hyper-contextual targeting with crowdedness. Marketing Science, 35(2), 218-233.
Babin, B. J., Hair, J. F., & Boles, J. S. (2008). Publishing research in marketing journals using structural equation modeling. Journal of Marketing Theory and Practice, 16(4), 279-286. Banfi, F., Gbahoué, B., & Schneider, J. (2013). Higher satisfaction at lower costs: Digitizing customer care. Recall, 22, 10-15.
Barclay, D., Higgins, C., & Thompson, R. (1995). The partial least squares approach to casual modeling: Personal Computer adoption and use as an Illustration. Technology Studies, 2, 285-309.
Bellman, S., Potter, R.F., Treleaven-Hassard, S., Robinson, J.A., & Varan, D., (2011), “The Effectiveness of Branded Mobile Phone Apps,” Journal of Interactive Marketing, 25 (4), 191–200.
Bijmolt, T. H., Dorotic, M., & Verhoef, P. C. (2011). Loyalty programs: Generalizations on their adoption, effectiveness and design. Foundations and Trends® in Marketing, 5(4), 197-258. Bijmolt, T. H., Krafft, M., Sese, F. J., & Viswanathan, V. (2018). Multi-tier loyalty programs to stimulate customer engagement. In Customer engagement marketing (pp. 119-139). Palgrave Macmillan, Cham.
Blohm, I., & Leimeister, J. M. (2013). Design of IT-Based Enhancing Services for Motivational Support and Behavioral Change.
Blumberg, B., Cooper, D. & Schindler, P. (2011). Business Research Methods, Berkshire, United Kingdom: McGraw-Hill Education.
Cassel, C., Hackl, P., & Westlund, A. H. (1999). Robustness of partial least-squares method for estimating latent variable quality structures. Journal of applied statistics, 26(4), 435-446.
Cerasoli, C. P., Nicklin, J. M., & Ford, M. T. (2014). Intrinsic motivation and extrinsic incentives jointly predict performance: A 40-year meta-analysis. Psychological bulletin, 140(4), 980. Cooper, R. B., & Jayatilaka, B. (2006). Group creativity: The effects of extrinsic, intrinsic, and obligation motivations. Creativity Research Journal, 18(2), 153-172.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS quarterly, 319-340.
Deci, E., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. Springer Science & Business Media.
Deci, E. L., & Ryan, R. M. (2000). The" what" and" why" of goal pursuits: Human needs and the self-determination of behavior. Psychological inquiry, 11(4), 227-268.
Demoulin, N. T., & Zidda, P. (2009). Drivers of customers’ adoption and adoption timing of a new loyalty card in the grocery retail market. Journal of Retailing, 85(3), 391-405.
Drèze, X., & Hoch, S. J. (1998). Exploiting the installed base using cross-merchandising and category destination programs. International Journal of Research in Marketing, 15(5), 459-471. Dreze, X., & Nunes, J. C. (2008). Feeling superior: The impact of loyalty program structure on consumers' perceptions of status. Journal of Consumer Research, 35(6), 890-905.
Fritz, W., Sohn, S., & Seegebarth, B. (2017). Broadening the perspective on mobile marketing: An introduction. Psychology & Marketing, 34(2), 113-118.
Furner, C. P., Racherla, P., & Babb, J. S. (2014). Mobile app stickiness (MASS) and mobile interactivity: a conceptual model. The Marketing Review, 14(2), 163-188.
Garg, R., & Telang, R. (2013). “Inferring App Demand from Publicly Available Data,” MIS Quarterly, 37(4), 1253–64.
García Gómez, B., Gutiérrez Arranz, A. M., & Gutiérrez Cillán, J. (2012). Drivers of customer likelihood to join grocery retail loyalty programs. An analysis of reward programs and
loyalty cards. Journal of Retailing and Consumer Services, 19(5), 492–500.
Gretzel, U., & Fesenmaier, D. R. (2006). Persuasion in recommender systems. International Journal of Electronic Commerce, 11(2), 81-100.
Gupta, S. (2013). For Mobile Devices, Think Apps, Not Ads. In Harvard Business Review. Retrieved October 1st, 2018, from:
Hair Jr, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2016). A primer on partial least squares structural equation modeling (PLS-SEM). Sage Publications.
Hair, J.F. Jr., Sarstedt, M., Hopkins, L., & Kuppelwieser, V.G. (2014). Partial least squares structural equation modeling (PLS-SEM) An emerging tool in business research. European Business Review, 26(2), 106-121.
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing theory and Practice, 19(2), 139-152.
Hamari, J., Koivisto, J., & Sarsa, H. (2014). Does gamification work?--a literature review of empirical studies on gamification. In 2014 47th Hawaii international conference on system sciences (HICSS) (pp. 3025-3034). IEEE.
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing. In New challenges to international marketing (pp. 277-319). Emerald Group Publishing Limited.
Henseler, J., & Sarstedt, M. (2013). Goodness-of-fit indices for partial least squares path modeling. Computational Statistics, 28(2), 565-580.
Hew, J. J., Lee, V. H., Ooi, K. B., & Wei, J. (2015). What catalyses mobile apps usage intention: an empirical analysis. Industrial Management & Data Systems, 115(7), 1269-1291.
Huotari, K., & Hamari, J. (2012). Defining gamification: a service marketing perspective. In Proceeding of the 16th international academic MindTrek conference (pp. 17-22). ACM. Hofacker, C. F., De Ruyter, K., Lurie, N. H., Manchanda, P., & Donaldson, J. (2016).
Gamification and mobile marketing effectiveness. Journal of Interactive Marketing, 34, 25-36. Hsu, C. L., & Lin, J. C. C. (2015). What drives purchase intention for paid mobile apps?–An expectation confirmation model with perceived value. Electronic Commerce Research and Applications, 14(1), 46-57.
Hsu, C. L., & Lin, J. C. C. (2016). Effect of perceived value and social influences on mobile app stickiness and in-app purchase intention. Technological Forecasting and Social Change, 108, 42-53.
Kim, Y. H., Kim, D. J., & Wachter, K. (2013). A study of mobile user engagement (MoEN): Engagement motivations, perceived value, satisfaction, and continued engagement intention. Decision Support Systems, 56, 361-370.
Kim, E., Lin, J. S., & Sung, Y. (2013). To app or not to app: Engaging consumers via branded mobile apps. Journal of Interactive Advertising, 13(1), 53-65.
Kim, S. J., Wang, R. J. H., & Malthouse, E. C. (2015). The effects of adopting and using a brand's mobile application on customers' subsequent purchase behavior. Journal of Interactive Marketing, 31, 28-41.
Kivetz, R., & Simonson, I. (2002). Earning the right to indulge: Effort as a determinant of customer preferences toward frequency program rewards. Journal of Marketing Research, 39(2), 155-170.
Lamberton, C., & Stephen, A. T. (2016). A thematic exploration of digital, social media, and mobile marketing: Research evolution from 2000 to 2015 and an agenda for future inquiry. Journal of Marketing, 80(6), 146-172.
Lee, M., & Cunningham, L. F. (2001). A cost/benefit approach to understanding service loyalty. Journal of services Marketing, 15(2), 113-130.
Leeflang, P. S., Verhoef, P. C., Dahlström, P., & Freundt, T. (2014). Challenges and solutions for marketing in a digital era. European management journal, 32(1), 1-12.
Leenheer, J., van Heerde, H. J., Bijmolt, T. H., & Smidts, A. (2007). Do loyalty programs really enhance behavioral loyalty? An empirical analysis accounting for self-selecting members. International Journal of Research in Marketing, 24(1), 31–47.
Lenhart, A., Madden, M., Smith, A., Purcell, K., Zickuhr, K., & Rainie, L. (2011). Teens,
Kindness and Cruelty on Social Network Sites: How American Teens Navigate the New World of" Digital Citizenship". Pew Internet & American Life Project.
Lepper, M. R., & Henderlong, J. (2000). Turning “play” into “work” and “work” into “play”: 25 years of research on intrinsic versus extrinsic motivation. In Intrinsic and extrinsic motivation(pp. 257-307).
Li, C.. 2018. Consumer behavior in switching between membership cards and mobile applications: The case of Starbucks. Computers in Human Behavior 84, 171-184.
Malhotra, N. K., Hall, J., Shaw, M., & Oppenheim, P. (2006). Marketing research: An applied application.
Markland, D., & Hardy, L. (1997). On the factorial and construct validity of the Intrinsic Motivation Inventory: Conceptual and operational concerns. Research Quarterly for exercise and Sport, 68(1), 20-32.
McAuley, E., Duncan, T., & Tammen, V. V. (1989). Psychometric properties of the Intrinsic Motivation Inventory in a competitive sport setting: A confirmatory factor analysis. Research quarterly for exercise and sport, 60(1), 48-58.
McGonigal, J. (2011) Reality is broken. Why games make us better and how they can change the world. Pinguin Press, New York.
McMillan, D. W., & Chavis, D. M. (1986). Sense of community: A definition and theory. Journal of community psychology, 14(1), 6-23.
Melnyk, V., & van Osselaer, S. M. (2012). Make me special: Gender differences in consumers’ responses to loyalty programs. Marketing Letters, 23(3), 545-559.
Mimouni-Chaabane, A., & Volle, P. (2010). Perceived benefits of loyalty programs: Scale development and implications for relational strategies. Journal of Business Research, 63(1), 32-37.
Misra, S., Cheng, L., Genevie, J., & Yuan, M. (2016). The iPhone effect: the quality of in-person social interactions in the presence of mobile devices. Environment and Behavior, 48(2),
275-298.
Nysveen, H., Pedersen, P. E., & Thorbjørnsen, H. (2005). Explaining intention to use mobile chat services: moderating effects of gender. Journal of consumer Marketing, 22(5), 247-256.
Okazaki, S., Navarro, A., & López-Nicolas, C. (2013). Assessing gender differences in ‘quick response’ code loyalty promotion acceptance. The Service Industries Journal, 33(12), 1165-1177.
McNaughton, D., & Light, J. (2013). The iPad and mobile technology revolution: Benefits and challenges for individuals who require augmentative and alternative communication.
Parasuraman, A., & Grewal, D. (2000). The impact of technology on the quality-value-loyalty chain: a research agenda. Journal of the academy of marketing science, 28(1), 168-174. Rosenbaum, M. S., Ostrom, A. L., & Kuntze, R. (2005). Loyalty programs and a sense of community. Journal of Services Marketing, 19(4), 222-233.
Rummel, A., & Feinberg, R. (1988). Cognitive evaluation theory: A meta-analytic review of the literature. Social Behavior and Personality: an international journal, 16(2), 147-164.
Salen, K., Tekinbaş, K. S., & Zimmerman, E. (2004). Rules of play: Game design fundamentals. MIT press.
Shah, R., & Goldstein, S. M. (2006). Use of structural equation modeling in operations
management research: Looking back and forward. Journal of Operations Management, 24(2), 148-169.
Shanahan, K. J., & Hyman, M. R. (2010). Motivators and enablers of SCOURing: A study of online piracy in the US and UK. Journal of Business Research, 63(9–10), 1095-1102.
Shankar, V., Venkatesh, A., Hofacker, C., & Naik, P. (2010). Mobile marketing in the retailing environment: current insights and future research avenues. Journal of interactive marketing, 24(2), 111-120.
Shankar, V., Kleijnen, M., Ramanathan, S., Rizley, R., Holland, S., & Morrissey, S. (2016). Mobile shopper marketing: Key issues, current insights, and future research avenues. Journal of Interactive Marketing, 34, 37-48.
Shen, L., Fishbach, A., & Hsee, C. K. (2014). The motivating-uncertainty effect: Uncertainty increases resource investment in the process of reward pursuit. Journal of Consumer Research, 41(5), 1301-1315.
Statista (2018), Number of apps available in leading app stores as of 1st quarter 2018. In Statista - The Statistics Portal. Retrieved October 1st, 2018, from:
https://www.statista.com/statistics/276623/number-of-apps-available-in-leading-app-stores/ Tan, M., & Teo, T. S. (2000). Factors influencing the adoption of Internet banking. Journal of the AIS, 1(1es), 5.
