c 2019 [Please consult the author] Notice Changes ... of Gamified Persuasive... · Green IS to induce desired behavioral patterns. Feldman and Dolan (2011) believe that gamification

Adoption of Gamified Persuasive Systems

Fortieth International Conference on Information Systems, Munich 2019 1

Adoption of Gamified Persuasive Systems to Encourage Sustainable Behaviors: Interplay

between Perceived Persuasiveness and Cognitive Absorption

Completed Research Paper

Nataliya Shevchuk University of Oulu

Pentti Kaiteran katu 1 Oulu 90014, Finland

[email protected]

Kenan Degirmenci Queensland University of Technology

2 George Street Brisbane QLD 4000, Australia [email protected]

Harri Oinas-Kukkonen

University of Oulu Pentti Kaiteran katu 1 Oulu 90014, Finland

[email protected]

Abstract

Gamification and persuasive technologies can change behaviors leading to desired outcomes, such as increasing consumer loyalty, raising health awareness, or developing eco-friendly mindsets. Since gamification triggers a state of flow enabling deep engagement with information systems, we consider the impact of perceived persuasiveness and cognitive absorption on the adoption of gamified persuasive systems aimed at encouraging sustainable behaviors. Having built a research model based on existing theories and frameworks, we survey 93 participants introduced to a gamified app with persuasive systems design features. The app is a social platform that rewards users for sustainable activities, like using renewable energy, recycling, and choosing pro-environmental modes of transportation, such as walking, cycling, public transport, or carsharing. Structural equation modeling shows significant influence of the persuasive design features with perceived persuasiveness exceeding the predictive value of cognitive absorption in explaining the intention to use the gamified app.

Keywords: Gamification, persuasive systems, mobile applications, environmental sustainability, behavior change, perceived persuasiveness, cognitive absorption, structural equation modeling

Introduction

Although new eco-friendly technologies are more widely available nowadays than in the past, they alone cannot guarantee a successful implementation of sustainable practices. Most of the time, the intensity of negative environmental impacts can be reduced, if not eliminated, by changing a root of environmental problems – human behavior (Steg and Vlek 2009). Human behavior remains the most important factor for achieving environmental sustainability (Osbaldiston and Schott 2012; Zhao et al. 2017). Furthermore, the role of human behavior has been an essential part in Green Information Systems (Green IS) (Elliot 2011), and advances in Green IS call for behavior changes because individuals need to accept, understand, buy,



and use innovative systems and technologies in proper ways. Therefore, it is crucial to look for ways to achieve pro-environmental or sustainable behaviors, i.e., behaviors that minimize environmental harm, or even create benefits for the environment (Steg and Vlek 2009). Frequently, this behavior is viewed as unenjoyable because it may result into loss of comfort, behavioral constraints, or other personal disadvantages (Midden and Ritsema 1983). Since sometimes changes in behavior may have to be enforced and are not usually enjoyable (Schoech et al. 2013), people show more resistance to sustainability concerns when they have to change their existing routines (Bengtsson and Ågerfalk 2011). Since many people are struggling with changing their behaviors, we suggest that gamification and Persuasive Systems Design (PSD) (Oinas-Kukkonen and Harjumaa 2009) are possible approaches that are likely to help people adopt Green IS to induce desired behavioral patterns. Feldman and Dolan (2011) believe that gamification can appeal to the users as well as boost the users’ motivation; they describe gamification techniques as straightforward, easy, and enjoyable tools for improving sustainable habits that incorporate habit-forming psychology, methods for self-evaluation, and virtual games. The use of social web and smartphone applications has substantially enlarged capabilities of employing multiuser, social, and real-time gamification techniques and persuasion strategies (Schoech et al. 2013).

Already now, both persuasive technologies and gamification have been employed for different purposes, such as marketing, changing attitude, and motivating (Hamari and Koivisto 2013). Gamification has been applied in different spheres, such as developing loyalty towards TV channels (GetGlue), improving health (Fitocracy), gamifying tracking personal goals (Mindbloom), and promoting greener energy consumption (Nissan Leaf) (Hamari 2015). Similarly, PSD has been introduced in several contexts, including health (Karppinen et al. 2014; Karppinen et al. 2016; Lehto et al. 2012a; Lehto et al. 2012b; Lehto and Oinas-Kukkonen 2015), well-being (Langrial et al. 2012a, Langrial et al. 2012b), social platforms (Stibe et al. 2011; Stibe and Oinas-Kukkonen 2014a; Stibe and Oinas-Kukkonen 2014b), and sustainability (Corbett 2013; Brauer et al. 2016). In terms of technology, mobile applications assist in monitoring behaviors and provide means to set and pursue individual and collective goals using various game-like approaches, such as visual aids, rank lists, competitions, etc. (Brauer et al. 2016). Numerous social mobile games have shown to improve users’ behaviors and behavior change outcomes in contexts focused on health, such as eating, fitness, controlling addictions, cognitive and physical therapies, cancer treatment adherence, combatting tobacco use, as well as self-help with diabetes and asthma (Oinas-Kukkonen 2013). Nevertheless, in the scope of a challenging area of behavior change, game development is in its early stages (Shoech et al. 2013).

Because both gamification and PSD prove to be effective at engaging people in certain activities, we aim to analyze how perceived persuasiveness and cognitive absorption influence intention to adopt a gamified persuasive system in the form of a mobile application aimed at increasing performance of sustainable behavior. Motivated by a need to understand users’ holistic experiences with IS, we focus on viewing perceived persuasiveness and cognitive absorption as important components required for desired behavioral outcomes. We propose to contribute to gamification research as well as gamified IS design by considering the use and impact of persuasive systems design on desired user experiences. In this study, we empirically evaluate the impact of persuasive systems design features on perceived persuasiveness and cognitive absorption using a gamified app that motivates users to act in a more environmentally sustainable manner. Therefore, the research question of the paper is the following:

RQ: How do perceived persuasiveness and cognitive absorption affect the intention to adopt a gamified persuasive system?

Background

Persuasive Systems Design

Fogg (2003) defined persuasive technologies as interactive information systems, for instance, the Internet, mobile and other ambient systems, aimed at modifying users’ behavior or attitudes. Behavior Change Support Systems (BCSSs) are essentially persuasive systems that create psychological and behavioral effects, described as sociotechnical IS configured to form, change or strengthen attitudes and short or long-term behaviors without deceiving or coercing the users (Oinas-Kukkonen 2013). In this regard, the Persuasive Systems Design (PSD) model is an instrument that has been created for assessment and design of BCSSs.



The first step of the PSD process involves addressing postulates related to the users’ perspectives, persuasion in general, and features of an information system (Oinas-Kukkonen and Harjumaa 2009). The following step in the PSD process is the analysis of the persuasion context. This step incorporates recognition of the intent of the persuasion, comprehension of the persuasion event, and definition of the strategies. After determining who the persuader is and whether persuasion will aim at attitude and/or behavior change, the problem domain, user and technology dependent features are considered. After that, the message and the route for its delivery are defined. The final stage of the PSD process is designing system features based on software principles classified in four categories. The design features of the primary task category, for example, reduction, tunneling, tailoring, personalization, simulation, self-monitoring, and rehearsal, aid the user to reaching primary goals. The features related to computer–human interaction, such as, praise, rewards, suggestions, similarity, reminders, social role, and liking, enable achieving established goals. Credibility support design features, such as trustworthiness, surface credibility, expertise, real-world feel, third-party endorsements, verifiability, and authority, make the system more persuasive and in turn more credible. Social support category introduces features that increase the users’ motivation by employing social influence techniques, namely, competition, recognition, cooperation, social learning, normative influence, social facilitation, and social comparison.

Kelders et al. (2012) suggest that persuasive system features increase involvement with the interventions. However, all available persuasive features are not required in a BCSS, because in some cases, additional persuasive features may cause a decrease in the overall persuasiveness (Oinas-Kukkonen 2013). Currently, promoting healthier lifestyle and habits has been the dominant application area of persuasive systems. PSD has already been used to evaluate and create systems supporting sustainable behavior (e.g. Corbett 2013; Brauer et al. 2016; Shevchuk and Oinas-Kukkonen 2016); however, this area of research remains less investigated compared to the others and requires more attention.

Gamification

Liu et al. (2017) noted several definitions of gamification in existing research: (1) using game components in non-game settings (Deterding et al. 2011), (2) a procedure of delivering game-like experiences to users, usually with the ultimate aim to influence the users’ behavior (Huotari and Hamari 2012, p.19), (3) including game elements (e.g. points, badges, leaderboards, etc.) into a system to improve the overall user experience and encourage the use of the system (Fitz-Walter et al. 2011), (4) promoting user engagement by using game elements in non-game settings (Mekler et al. 2013), (5) using game-based techniques including design, procedures, and logic in non-game settings in order to involve users, and encourage them to improve learning and problem-solving skills (Borges et al. 2014, p. 216). Liu et al. (2017) highlighted that gamification is integrated into real-world systems while their functionality is preserved; they identified two outcomes of a gamified system: experiential and instrumental outcomes. The duality of the gamification results into meaningful outcomes as the design provides not only enjoyable experiences but also enhances instrumental task outcomes.

Since gamification is able to impact the users’ psychological states, decision-making and behavior, it is related to persuasive technology and behavior change support systems (Fogg 2003; Oinas-Kukkonen and Harjumaa 2009; Oinas-Kukkonen 2013). The main difference is that gamification is more focused on invoking users’ intrinsic motivations than on the overall attitude or behavior change (Huotari and Hamari 2012). Being seen as a promising approach for evoking different behavioral changes, gamification techniques have been applied to facilitating sustainable practices, such as waste separation (Lessel et al. 2015) or reduction of energy use (Orland et al. 2014). Specifically, ranking lists allow performance comparison and contrast among users, creating a competitive environment to reduce the energy use or pursue sustainable modes of transportation (Loock et al. 2013); visual aids (a growing or dying tree) assist people realize the extent to which their mobility behavior affects environment (Froehlich et al. 2009). Although gamification techniques play an important role in persuasive systems, they constitute only a small part of potential persuasive design principles (Brauer et al. 2016).

Cognitive Absorption

Cognitive absorption is established on several connected concepts: the personality trait absorption (Tellegen and Atkinson 1974), flow (Csikszentmihaiyi 1975), and cognitive engagement (Webster and Ho 1997). Absorption is an individual disposition or trait defined as an intrinsic measurement of personality



that results into periods of undivided attention which occurs when all of the person’s attentional capacity is directed at a particular focus of interest (Tellegen and Atkinson 1974). It has the following four components: (1) readiness to experience deep immersion, (2) a sharp feeling of the realism of the focus of interest, (3) an unreceptiveness to actions that are usually distracting, and (4) an appraisal of information in idiosyncratic and unconventional ways (Roche and McConkey 1990, p. 91). Csikszentmihalyi (1975) described the flow as the complete involvement that creates a feeling of the comprehensive experience. Initially Csikszentmihalyi (1975) identified six components of it: merging action-awareness, centering of attention, loss of self-consciousness, feeling of control, coherent and noncontradictory demands, and autotelic nature. Webster and Ho (1997) state that cognitive engagement relates to playfulness, and that the state of playfulness is an equivalent to flow, only the following dimensions: intrinsic interest, curiosity, and attention focus, excluding the feeling of control. Flow is considered to incentivize skill development and individual progress as it promotes participation in complicated tasks (Engeser and Schiepe-Tiska 2012). Thus, it is insightful to study flow in the context of BCSSs, which also aim to extend an individual’s potential to change behavior.

Cognitive absorption consists of control, curiosity, and focused attention, and temporal dissociation (a notion of losing track of time) measures of flow (Csikszentmihaiyi 1975; Trevino and Webster 1992), the heightened enjoyment which is a combination of the intrinsic interest dimension of flow (Webster et al. 1993) and perceived enjoyment (Davis et al. 1992). Therefore, Agarwal and Karahanna (2000) describe cognitive absorption as a condition of intense engagement with the system consisting of the five dimensions: (1) temporal dissociation, or the failure to register the course of time during involvement in interaction; (2) focused immersion, or feeling a complete involvement and ignoring other possible foci of interest; (3) heightened enjoyment, or capturing the pleasant parts of the involvement; (4) control, or the user's impression of being in command of the involvement; and (5) curiosity, or arousal of a person's sensory and subconscious interest that stems from tapping into the experience. Liu et al. (2017) recognize cognitive absorption as a possible experiential outcome of engaging with a gamified system.

Research Model and Hypotheses

We examine how the PSD and cognitive absorption concepts influence perceived persuasiveness and intention to adopt a gamified app that motivates users to act in a more sustainable manner. Focusing on the persuasive design of the gamified app, we use constructs adapted from the PSD model: primary task support, dialog support, system credibility support, and social support (Oinas-Kukkonen and Harjumaa 2009).

Figure 1. Research Model

Social Support (SOCI)

Intention to Use

Application (INT_USE)


(COAB)

Perceived Persuasiveness

(PEPE)

Primary Task Support (PRIM)

Credibility Support (CRED)

Dialogue Support (DIAL)

Persuasive Systems Design Categories

Adoption of Gamified Persuasive System

H2c

H2d

H1a

H1b

H3a

H3b

H4a

H4b

H5a H2b

H2e

H6

H5b

H2a



Each of the constructs contains three respective items chosen based on the features present in the gamified app, namely, primary task support: reduction, self-monitoring, personalization; dialogue support: praise, rewards, social role; credibility support: trustworthiness, surface credibility, expertise; social support: social learning, social comparison, recognition. The proposed and examined research model is illustrated in Figure 1; the relationships among constructs and the derived hypotheses are presented below.

The primary task support helps the individual to complete the primary task. By increasing the user’s self-efficacy, the primary support features can reduce the cognitive load associated with the system usage (Nadkarni and Gupta 2007). Regarding sustainable behavior context, primary task support can assist with overcoming perceived burdens, such as perceived effort of engaging in sustainable behavior and other similar psychological obstacles (Lindenberg and Steg 2013). Thus, by reducing the effort of completing the primary task, it should be easier for the user to experience involvement with the gamified system and experience cognitive absorption. Self-monitoring helps the users observe their own behavior and form a stronger association with the own influence on the environment (Froehlich et al. 2009; Loock et al. 2013). Personalized systems create social, emotional, and cognitive impacts on the users (Blom and Monk 2003), and, therefore, are perceived as more convincing; thus, they increasingly contribute to desired outcomes (Berkovsky et al. 2015; Kurz 2002). Furthermore, primary task support promotes positive affect, which increases perceived persuasiveness of the source (Angst and Agarwal 2009; Derrick et al. 2011). Previous studies (Lehto et al. 2012a; Lehto et al. 2012b), which focused on different application domains, suggest that perceived persuasiveness is positively significantly affected by primary task support.

H1a: Primary task support positively impacts perceived persuasiveness.

H1b: Primary task support positively impacts cognitive absorption.

Dialogue support assists an individual with using the system and engaging in the target behavior by activating and motivating the user. Because IT artifacts are social actors, users perceive engagement with them as interpersonal, similar to social interactions (Al-Natour and Benbasat 2009; Fogg 2003; Fogg and Nass 1997). Dialogue support is implemented with various suggestions, reminders, and prompts (Oinas-Kukkonen and Harjumaa 2009), which provide appropriate argumentation and feedback to the users based on their performance. Compelling advice increases persuasion, whereas inadequate ones decreases it (Briñol et al. 2007). The cognitive fit theory (Vessey and Galletta 1991) states that the dialog between the system and the users both encourage the system use and motivate performance of intended sustainable actions. Existing research on Green IS and sustainable behavior suggests that praise and rewards given to the users for their accomplishments increase motivation to use Green IS and engage in pro-environmental behavior (Froehlich et al. 2010; Gifford 2011; Graml et al. 2011; Petkov et al. 2011). Moreover, the use of game-like elements, such as virtual rewards like badges, leaderboards, levels, ranking etc. implies involvement of gamification to increase people’s favorable impression of the systems. According to previous research (Lehto et al. 2012a; Lehto et al. 2012b), dialogue support influences perceived persuasiveness; however, none of the existing studies focused on Green IS, none of the studies investigated impact of dialogue support on social support and on cognitive absorption; thus, the following hypotheses are applied in a new context.

H2a: Dialogue support positively impacts perceived persuasiveness.

H2b: Dialogue support positively impacts cognitive absorption.

H2c: Dialogue support positively impacts primary task.

H2d: Dialogue support positively impacts credibility support.

H2e: Dialogue support positively impacts social support.

Credibility support strengthens persuasion outcomes by adding trustworthiness to the system (Oinas-Kukkonen and Harjumaa 2009). According to Van Vugt et al. (2006), users are able to assess the surface credibility of the system immediately after the first use. Hence, perceived credibility is not only greatly subjective but can also vary significantly among users. If users deem that evidence supporting sustainable initiatives and environmental outcomes suggested by the system lacks credibility (Flynn et al. 2009; Kollmuss and Agyeman 2002; Kurz 2002), they are likely not to interact with the system. However, in case the system is perceived as credible, the users become more comfortable using the system, consequentially reaching the state of cognitive absorption. To achieve perceived credibility, the system can provide



testimonials from well-known sources and authorities (e.g., privacy statements to ensure confidentiality, awards for recognizing the quality system, or endorsements by influential organizations). Usually, the more credible the source is perceived, the more persuasive it is, and vice versa (Pornpitakpan 2004). Often discussed in IS research, a closely related concept to credibility is trust (Lehto et al. 2012a). Everard and Galletta (2005, p. 59) list possible signs of computer credibility as follows: accepting the advice, trusting the information, and believing the output. Dwivedi et al. (2011) point out that credibility, trust, and belief can significantly impact the users’ engagement in the target behavior.

H3a: Credibility support positively impacts perceived persuasiveness.

H3b: Credibility support positively impacts cognitive absorption.

Social influence, a base for shaping pro-environmental mindsets and behaviors, is leveraged by the social support in order to motivate users (Gifford 2011; Stern et al. 1995). Having a feeling of necessity to join a community, some individuals will consider modifying their own behavior (Petkov et al. 2011). Opinions, beliefs and attitudes of family, peers, and friends are likely to influence a person’s outlook on sustainable behavior adoption (Gifford 2011). According to Lindenberg and Steg (2013), interaction with like-minded people, engaging in social activities with people with similar personal goals or interests can both increase the individuals’ favorable perception of the system and the individuals’ willingness to engage in sustainable behavior. Furthermore, adding gamified techniques, such as enabling cooperation and/or competition among the users, keeps the users involved and focused on performing the target behavior.

H4a: Social support positively impacts perceived persuasiveness.

H4b: Social support positively impacts cognitive absorption.

Cognitive absorption is relevant for researching technology use behavior, since it is an intrinsic motivation variable that shapes principal user attitudes towards IS. It was investigated in user behavior patterns, like usage intention and development of user beliefs, in work-related or organizational contexts involving utilitarian information systems (Reychav and Wu 2015). Since smartphone users tend to exhibit immersive behavior when engaging with applications, cognitive absorption appears to be relevant in the context of application usage. The related concept of flow was recognized to be useful in predicting behavioral intent both experimental and goal-oriented types of behaviors (Hoffman and Novak 1996). Additionally, because flow has an impact on attitude formation and consumer experience (Chen et al. 1999), it is likely to result into desired marketing outcomes, such as positive attitudes, purchase intentions, and purchase behaviors (Hoffman and Novak 2009). Flow was proven to have a positive impact on a variety of attitudinal outcomes, for instance, favorable attitudes towards online shopping (Korzaan 2003), web sites and brands (Mathwick and Rigdon 2004; Sicilia and Ruiz 2007), and playing games (Hsu and Lu 2004). These findings suggest that cognitive absorption is likely to positively influence both perceived persuasiveness and intention to use the gamified persuasive system.

H5a: Cognitive absorption positively impacts perceived persuasiveness.

H5b: Cognitive absorption positively impacts intention to use the gamified persuasive system.

Lehto et al. (2012a) defined perceived persuasiveness as an individual’s positive feeling about the system. Crano and Prislin (2006) suggest that the fundamental construct of attitude is crucial for understanding persuasion, because it embodies a critical combination of affects and cognitions. Based on the theory of reasoned action (TRA) (Fishbein and Ajzen 1975), the technology acceptance model (TAM) (Davis et al. 1989), and the theory of planned behavior (TPB) (Ajzen 1991), the attitude of an individual is crucial for determining intention, which is an essential component of engaging in a behavior. Ajzen (1991) defined attitude as an individual’s positive or negative feeling regarding performance of the target behavior. Successful persuasion occurs when the object of change (e.g., an attitude or a belief) is altered in the desired manner (Petty and Cacioppo 1986). Under appropriate conditions, the modified attitude may influence subsequent behavior (Crano and Prislin 2006). This indicates that people are likely to engage in sustainable behavior in the future if they develop or maintain an affirmative attitude towards the system through the persuasive experiences. TPB theorizes that when there is an opportunity to act, users’ intentions result in behaviors (Ajzen 1991). Among all other antecedents of behavior, behavioral intention is considered to be an immediate predictor and determinant of behavior and, consequentially, behavior change.

H6: Perceived persuasiveness positively impacts intention to use the gamified persuasive system.



Research Method

Instrument Design and Development

To test our hypotheses, we designed a task that involved using greenApes (https://www.greenapes.com), an app available for iOS and Android (Figure 2).

Figure 2. Screenshots of the Persuasive Gamified Application (greenApes)1

The app, implemented within the European Union’s Horizon 2020 research and innovation program, provides a digital platform in the form of a social network for users to engage in sustainable behaviors such as using renewable energy, recycling, or choosing sustainable modes of transportation, which include walking, cycling, public transport, or carsharing. The app includes gamification elements such as points and rewards. Points (called BankoNuts) are collected by connecting the app with third party apps for validation, for example, car2go for carsharing, or Apple Health/Google Fit for walking and cycling. Partners of the app offer rewards, e.g., grocery stores that offer discounts on a purchase, or cafes that give free smoothies for a certain amount of collected BankoNuts. Further gamification elements include badges (e.g., JungleSurfer) and leaderboards (“most inspiring apes”). We developed a survey tool to assess the constructs of our model. The latent variables were evaluated with the reflective multiple-item scales based on the pre-validated measures where possible (see Appendix Table A1). Additionally, we asked open-ended questions about examples of sustainable behaviors the respondents engaged in, and about the characteristics of the app the respondents liked or disliked. Finally, the survey included demographic-related questions about gender, age, employment, and education.

Data Collection Procedure and Sample Description

The data was collected through a paper-based survey distributed to students of a large university in Finland, and further respondents participated through an online version of the survey distributed via social media platforms. The same procedure was used in both cases: each participant freely explored the app for about 15-20 minutes and afterwards completed the survey. In total, 93 complete responses suitable for analysis were received (59 collected with the online survey, and 34 responses with the paper-based one). In order to control for potential bias, we (1) ex ante conducted the survey based on random sampling, and (2) ex post performed a correlation analysis and analyzed the influence of the demographics as control variables on intention to use. Results showed that all correlations between the demographics and the latent variables

1 https://play.google.com/store/apps/details?id=com.greenapes.horizontalandroid&hl=en_US



were lower than 0.325, indicating low correlations (Hinkle et al. 1988), and the influence of the demographics on intention to use was not significant with path coefficients lower than 0.056 and t-values lower than 0.725. In addition, the type of survey (paper-based/online) was also examined as a control for the latent variables, and again low correlations (lower than 0.379), path coefficients (lower than 0.046) and t-values (lower than 0.640) were identified. Furthermore, we compared results from the paper-based survey with the online survey, which were very similar to the overall results. The analysis of correlations and path coefficients suggests that the presence of the two samples creates no bias because the demographics had no confounding effects on the latent variables. Table 1 shows the combined descriptive statistics (i.e. both samples are considered together).

Table 1. Descriptive Statistics

Demographics Value Frequency Percent

Age 18 - 24

25 - 34

35 - 44

45 - 54

55 - 64

65 - 74

17

49

18

5

3

1

18%

53%

19%

5%

3%

1%

Gender Female

Male

Preferred not to answer

42

48

3

46%

52%

2%

Education High school

Bachelor's Degree

Master's Degree

Doctorate/Advanced Degree

16

50

22

5

17%

54%

24%

5%

Employment Employed full-time

Employed part-time

Self-employed

Student

45

9

4

35

48%

10%

4%

38%

Data Analysis and Results

For data analysis, we employed partial least squares structural equation modeling (PLS-SEM) because this technique works well for exploratory research (Gefen et al. 2011). Generally, PLS-SEM is used for predicting instead of testing established theory (Hair et al. 2011). Our sample size exceeds the PLS-SEM minimum requirement: being at least equal to ten times the largest number of structural paths directed at a particular latent construct in the structural model (Hair et al. 2011). Although PLS-SEM does not call for data normality (Hair et al. 2011), our data did not have any outliers. The PLS-SEM model testing is a two-step process: (1) assessment of the validity and reliability of the measurement model and (2) assessment of the structural model. The relationships among the constructs are reflected in the measurement model. To confirm validity and reliability of the measures of the constructs, we examined the discriminant and convergent validities of the utilized measures. Next, we assessed the structural model by inspecting relationships among the constructs. SmartPLS 3 was the software we used.

Measuring all variables with the same instrument creates a possible threat to validity of the findings known as common method bias (CMB), or common method variance (CMV). Confidentiality and anonymity of the study encouraged the respondents to give honest answers to minimize CMB ex ante. Additionally, we assessed the correlation matrix of the constructs as a possible ex post test and control for CMB. All correlations in the matrix were below 0.90, showing the evidence of absence of CMB (Pavlou et al. 2007).

All constructs were modeled as reflective and measured using multiple indicators (see Appendix for factor loadings). Cognitive absorption was modeled as a second-order variable (the higher-order component – HOC) composed of five first-order constructs: perceived enjoyment, temporal dissociation, focused



immersion, control, and curiosity (the lower-order components – LOCs). Establishing such higher-order model or hierarchical component model of reflective-reflective type allows to operationalize the construct of cognitive absorption at a high level of abstraction. The reflective-reflective type of the model indicates formative relationship between the LOCs and HOC, where each construct is measured by reflective indicators (Hair et al. 2011). To model hierarchical latent variables, the repeated indicator approach was used (Hair et al. 2011; Lohmöller 1989).

Assessment of Measurement Model

Item loadings, discriminant validity, and internal consistency are used to assess the properties of the scales. To be considered acceptable, item loadings and internal consistencies must exceed .70 (Fornell and Larcker 1981) (see Appendix Table A1 for factor loadings). Based on the composite reliability scores, ranging from .848 (PRIM) to .980 (INT_USE), the constructs in the model demonstrate good internal consistency. Additionally, adequate internal consistency is supported with the average variance extracted (AVE) values which are above the suggested minimum of .50 (Fornell and Larcker 1981) of all the constructs. Inspection of the latent variable correlations and square root of AVE demonstrates that all constructs share more variance with their own indicators than with other constructs (Table 2).

Table 2. Properties of Latent Variables

PRIM DIAL CRED SOCI PEPE PEEN TEDI FOIM CTRL CURI INT_USE

PRIM .808

DIAL .740 .837

CRED .542 .591 .810

SOCI .333 .429 .310 .824

PEPE .646 .768 .587 .356 .904

PEEN .622 .693 .641 .352 .715 .849

TEDI .546 .649 .567 .271 .682 .633 .900

FOIM .584 .675 .531 .297 .662 .602 .731 .880

CTRL .421 .477 .564 .394 .453 .585 .419 .394 .838

CURI .586 .718 .690 .367 .822 .797 .736 .644 .546 .948

INT_USE

.571 .653 .518 .229 .823 .771 .626 .586 .442 .830 .962

CA .731 .784 .736 .765 .926 .922 .882 .852 .856 .944 .973

CR .848 .875 .851 .864 .947 .939 .927 .911 .904 .964 .980

AVE .653 .700 .656 .679 .818 .722 .809 .775 .703 .899 .926

CA = Cronbach’s alpha; CR = Composite reliability; AVE = Average Variance Extracted; Bolded cells = Square root of AVE

Structural Model and Hypotheses Testing

In SmartPLS 3, the parametric bootstrapping technique using 5000 subsamples and the default settings (i.e. parallel processing, no sign changes) was used to assess the structural model. The path coefficients and explained variances (R²) were used to evaluate the model. While the path coefficients indicate how strong the relationships among independent and dependent variables are, R² corresponds to the proportion of the variance of dependent variables explained by independent variables (Hair et al. 2011). The results of the PLS analysis provide considerable support for the suggested research model since most of the hypotheses were supported (see Figure 3).



In the structural model, dialogue support related significantly to primary task support, to credibility support, and to social support explaining 54.7%, 35.0%, and 18.4% of the variance respectively. Primary task support, credibility support, and social support showed no significant relationship to perceived persuasiveness, while dialogue support and cognitive absorption did. Overall, 70.7% of the variance in perceived persuasiveness was explained. Cognitive absorption (HOC) showed strong relationships with its LOCs (perceived enjoyment, temporal dissociation, focused immersion, control and curiosity). Primary task support and social support showed no statistically significant relationship to cognitive absorption, while dialogue support and credibility support did, resulting in explaining 72.0% of the variance in cognitive absorption. Perceived persuasiveness and cognitive absorption explained 73.3% of the variance in intention to use gamified application.

Figure 3. Results of PLS-SEM analysis

Discussion

Our findings show significant support for the suggested theoretical model for studying intention to adopt gamified persuasive systems, because most of the relationships we hypothesized were supported. The results obtained from the data analysis have essential implications for both academics and practitioners. One of the core findings was discovering support for hypothesis about a positive influence of perceived persuasiveness and cognitive absorption on intention to adopt the gamified persuasive system aimed at increasing sustainable behavior. Nonetheless, since this study is an exploratory research, which considered only one gamified persuasive system, thus, further studies are needed for a continuing investigation of the proposed theoretical model.

As anticipated, perceived persuasiveness showed a statistically significant impact on intention to use the gamified application. This supports that the user’s favorable impression of the system assists in behavior change. More studies on relationship of perceived persuasiveness and intention to adopt a certain behavior will be valuable for further theory-building. Future research could look into comparing the impact of perceived persuasiveness and attitude on intention to adopt a behavior. From a practitioners’ standpoint, acknowledging significance of perceived persuasiveness can help enhancing BCSS development that will be beneficial for the increasing effectiveness of BCSS. When analyzing factors that impact perceived persuasiveness, some of the implications based on our findings differ from the one obtained in previous studies (Lehto et al. 2012a; Lehto et al. 2012b), and thus require further investigation. For example, out of all persuasive design principles only dialog support had a statistically significant impact on perceived

Social Support R2=.184


R2=.720

Perceived Persuasiveness

R2=.707

Intention to Use

Application R2=.733

Primary Task Support R2=.547

Credibility Support R2=.350

Dialogue Support

Persuasive Systems Design Categories

Adoption of Gamified Persuasive System

.740***

.591***

.437***

.063n.s.

.121n.s.

-.025n.s.

.382***

-.012n.s.

.057n.s.

Path Significance: * p < .05; ** p < .01; *** p < .001; n.s. = not significant, dashed line

Perceived Enjoyment

R2=.834

Temporal Dissociation

R2=.684

Focused Immersion

R2=.615

Control

R2=.479

Curiosity

R2=.826

Cognitive Absorption Dimensions

.321**

.429***

.913***

.542***

.496***

.404***

.909*** .827***

.784***

.74*

.692***



persuasiveness proving that the computer-human dialog facilitated by the system creates an overall favorable impression of the system. Since findings justified the hypothesized impact of dialogue support on other PSD categories, the dialogue support category was the only PSD category in this study which performed in line with the previous research findings. This might be due to the reason that we used a gamified application as the persuasive system, which differs from previous studies. Regarding primary task support, credibility support, and social support not showing a sufficient impact on perceived persuasiveness, such outcome could be attributed to the specific application used in the study. However, other reasons are possible. For example, considering primary task support, the high level of awareness of what the sustainable behavior is and actions contribute to it could have made the primary task support category less influential, since the users did not need to rely on the app to support carrying out the primary task. Statistical insignificance of the credibility support category could be explained by the lack of the users’ interest in scrutinizing to the credibility of the app, i.e. the users might be content with the credibility of the app as long as the app appears to be similar to other ones the users had used before. In such scenario, credibility support system neither contributes to nor undermines perceived persuasiveness of the system. Regarding social support being statistically irrelevant to perceived persuasiveness, the overwhelming presence of technologies and social media might have impacted the respondents in such a way that they do not associate communication with the other users of the app (provided with the social support features) with the overall favorable perception of the system. In any case, significant relationship between only one PSD category and perceived persuasiveness exemplifies the premise mentioned in theoretical background stating that a system does not have to include all available persuasive features, because adding persuasive principles does not guarantee an increase of the overall persuasiveness, and in some cases, it can even contribute to the decrease in the overall persuasiveness (Oinas-Kukkonen 2013). In order to check validity of these explanations and to find other possible reasons for the inconsistent performance of the relationship of the individual PSD categories across studies, further research with a wider range of systems and/or apps is required.

Examining performance of cognitive absorption in the research model, we found it to be one of the factors that influences perceived persuasiveness, as it was hypothesized. Similar to the relationship with perceived persuasiveness, dialogue support turned out to have a statistically significant impact on cognitive absorption. Another PSD category that performed as hypothesized in relation to cognitive absorption was credibility support. Apparently, the credible look and feel of the app contributes to the state of deep involvement with the app, but not necessarily to the overall favorable impression of the app (as suggested by the lack of statistically significant relationship between credibility support and perceived persuasiveness). Primary task support did not display a statistically significant relationship with cognitive absorption, suggesting that assistance with fulfilling the primary task does not create either immersive involvement with the app nor increases the favorable impression of the app as discussed earlier. Since to our knowledge, this is the first study that examined impact of the PSD categories on cognitive absorption, more research (involving different systems and/or apps) is required to improve the understanding of the relationship between PSD and cognitive absorption. Based on our results, we suggest that cognitive absorption of gamified app users is strongly influenced by dialog and credibility support, which provides useful implications for gamified app designers. Furthermore, we discovered that perceived persuasiveness outweighs the predictive value of cognitive absorption to explain intention to use a gamified persuasive system. Thus, relevant for both researchers and practitioners, this result support the premise that the PSD model should be considered for gamification design.

One of the limitations of our study is a short-term use of the gamified persuasive system. It might explain why some of the PSD categories did not show significant effects on perceived persuasiveness and on cognitive absorption. In the future, we suggest conducting an experiment able to capture long-term interaction of the users with the gamified persuasive system. For example, the users can be surveyed twice: after being introduced to the system for the first time and after several weeks of using the system. Additionally, our environmental awareness measure on a 7-point Likert scale has a mean of 5.65 (SD = 1.04), suggesting that the participants were highly environmentally aware. Thus, it might indicate a self-selection bias (Kankanhalli et al. 2005): participants who were more concerned about environmental sustainability might have been more likely to participate in this study. Therefore, further research could also look into differences between individuals with low and high environmental awareness. All in all, in spite of the limitations of the study, our findings provide a foundation for the future development of research on gamification and PSD regarding the adoption of sustainable behavior.



Conclusion

In this study, we gained insight into the influence of gamified persuasive systems on intention to adopt sustainable behavior. Based on the foundation of theories related to gamification and behavior change, we built and tested a research model focusing on the impact of perceived persuasiveness and cognitive absorption on intention to use a gamified persuasive application aimed at fostering a sustainable behavior. The results of our research suggest that perceived persuasiveness is exceeding the predictive value of cognitive absorption in explaining the intention to use the gamified app. Additionally, outcomes and implications of our study underline advantages of PSD and gamification techniques for encouraging sustainable behavior via mobile applications. We also highlighted implications of our findings that should be investigated more in-depth in the future. Altogether, knowledge gained in the course of our research augments existing academic work in the areas of Green IS, gamification, IS adoption, and behavior change as well as contributes to practice by analyzing features and techniques useful for designing persuasive gamified systems.

Acknowledgements

The authors would like to thank the associate editor and the three anonymous reviewers for their valuable comments and suggestions. Dr Degirmenci’s contributions were partially supported by a grant from the Australian Research Council (DP150100163).

References

Agarwal, R., and Karahanna, E. 2000. “Time Flies When You’re Having Fun: Cognitive Absorption and Beliefs About Information Technology Usage,” MIS Quarterly (24:4), pp. 665-694.

Ajzen, I. 1991. “The Theory of Planned Behavior,” Organizational Behavior and Human Decision Processes (50:2), pp. 179–211.

Al-Natour, S., and Benbasat, I. 2009. “The Adoption and Use of IT Artifacts: A New Interaction-Centric Model for the Study of User-Artifact Relationships,” Journal of the Association for Information Systems (10:9), pp. 661–685.

Angst, C. M., and Agarwal, R. 2009. “Adoption of Electronic Health Records in the Presence of Privacy Concerns: The Elaboration Likelihood Model and Individual Persuasion,” MIS Quarterly (33:2), pp. 339–37.

Bengtsson, F., and Ågerfalk, P. J. 2011. “Information Technology as a Change Actant in Sustainability Innovation: Insights from Uppsala,” Journal of Strategic Information Systems (20:1), pp. 96–112.

Berkovsky, S., Freyne, J., and Oinas-Kukkonen, H. 2015. “Influencing Individually: Fusing Personalization and Persuasion,” In Proceedings of the International Joint Conference on AI, pp. 4153–4157.

Blom, J. O., and Monk, A. F. 2003. “Theory of Personalization of Appearance: Why Users Personalize Their PCs and Mobile Phones,” Human-Computer Interaction (18:3), pp. 193–228.

Borges, S., Durelli, V. H. S., Reis, H. M., and Isotani, S. 2014. “A Systematic Mapping on Gamification Applied to Education,” In Proceedings of the Annual ACM Symposium on Applied Computing, pp. 216–222.

Brauer, B., Ebermann, C., and Kolbe, L. M. 2016. “An Acceptance Model for User-Centric Persuasive Environmental Sustainable IS,” In Proceedings of the International Conference on Information Systems, pp.1–22.

Briñol, P., Petty, R. E., Valle, C., Rucker, D. D., and Becerra, A. 2007. “The Effects of Message Recipients’ Power Before and After Persuasion: A Self-Validation Analysis,” Journal of Personality and Social Psychology (93:6), pp. 1040–1053.

Chen, H., Wigand, R. T., and Nilan, M. S. 1999. “Optimal Experience of Web Activities,” Computers in Human Behavior (15:5), pp. 585–608.

Cialdini, R. B., Petty, R. E., and Cacioppo, J. T. 1981. “Attitude and Attitude Change,” Annual Review of Psychology (32:1), pp. 357–404.

Corbett, J. 2013. “Designing and Using Carbon Management Systems to Promote Ecologically Responsible Behaviors,” Journal of the Association for Information Systems (14:7), pp. 339–378.

Crano, W. D., and Prislin, R. 2011. Attitudes and Attitude Change, Psychology Press. Csikszentmihalyi, M. 1975. “Beyond Boredom and Anxiety,” Book Reviews, pp. 703–707.



Davis, F. D., Bagozzi, R. P., and Warshaw, P. R. 1989. “User Acceptance of Computer Technology: A Comparison of Two Theoretical Models,” Management Science (35:8), pp. 982-1003.

Davis, F. D., Bagozzi, R. P., and Warshaw, P. R. 1992. “Extrinsic and Intrinsic Motivation to Use Computers in the Workplace,” Journal of Applied Social Psychology (22:14), pp. 1111–1132.

Derrick, D. C., Jenkins, J. L., and Nunamaker, J. F. J. 2011. “Design Principles for Special Purpose, Embodied, Conversational Intelligence with Environmental Sensors (SPECIES) Agents,” AIS Transactions on Human-Computer Interaction (3:2), pp. 62–81.

Deterding, S., Dixon, D., Khaled, R., and Nacke, L. 2011. “From Game Design Elements to Gamefulness: Defining Gamification,” In Proceedings of the International Academic MindTrek Conference, pp. 9–11.

Dwivedi, Y. K., Rana, N. P., Chen, H., and Williams, M. D. 2011. “A Meta-Analysis of the Unified Theory of Acceptance and Use of Technology (UTAUT),” In Proceedings of the International Working Conference on Governance and Sustainability in Information Systems, pp. 155–17.

Elliot, S. 2011. “Transdisciplinary Perspectives on Environmental Sustainability: A Resource Base and Framework for IT-Enabled Business Transformation,” MIS Quarterly (35:1), pp. 197-236.

Engeser, S., and Schiepe-Tiska, A. 2012. “Historical Lines and an Overview of Current Research on Flow,” Advances in Flow Research (Vol. 9781461423), pp. 1–22.

Everard, A., and Galletta, D. F. 2006. “How Presentation Flaws Affect Perceived Site Quality, Trust, and Intention to Purchase from an Online Store,” Journal of Management Information Systems (22:3), pp. 56–95.

Feldman, B., and Dolan, B.2011. Mobile Social and Fun Games For Health. Mobihealth News Research. Chester Street Publishing Inc.

Fishbein, M., and Ajzen, I. 1975. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley.

Fitz-Walter, Z., Tjondronegoro, D., and Wyeth, P. 2012. “A Gamified Mobile Application for Engaging New Students at University Orientation,” in Proceedings of the Australian Computer-Human Interaction Conference, pp. 138–141.

Flynn, R., Bellaby, P., and Ricci, M. 2009. “The ‘Value-Action Gap’ in Public Attitudes towards Sustainable Energy: The Case of Hydrogen Energy,” Sociological Review (57:SUPPL. 2), pp. 159–18.

Fogg, B. J. 2003. “Persuasive Technology: Using Computers to Change What We Think and Do,” Persuasive Technology: Using Computers to Change What We Think and Do.

Fogg, B. J., and Nass, C. 1997. “Silicon Sycophants: The Effects of Computers That Flatter,” International Journal of Human Computer Studies (46:5), pp. 551–561.

Fornell, C., and Larcker, D. 1981. “Evaluating Structural Equation Models with Unobservable Variables and Measurement Error,” Journal of Marketing Research (18:3), pp. 39–5.

Froehlich, J., Dillahunt, T., Klasnja, P., Mankoff, J., Consolvo, S., Harrison, B., and Landay, J. A. 2009. “UbiGreen: Investigating a Mobile Tool for Tracking and Supporting Green Transportation Habits,” In Proceedings Of The SIGCHI Conference On Human Factors In Computing Systems, pp. 1043–1052.

Gefen, D., Rigdon, E. E., and Straub, D. 2011. “Editor’s Comments: An Update and Extension to SEM Guidelines for Administrative and Social Science Research,” MIS Quarterly (35:2), III-XIV.

Gifford, R. 2011. “The Dragons of Inaction: Psychological Barriers That Limit Climate Change Mitigation and Adaptation,” American Psychologist (66:4), pp. 290–302.

Graml, T., Loock, C.-M., Baeriswyl, M., and Staake, T. 2011. “Improving Residential Energy Consumption At Large Using Persuasive Systems,” In Proceedings of the European Conference on Information Systems, pp. 1–15.

Hair, J. F., Ringle, C. M., and Sarstedt, M. 2011. “PLS-SEM: Indeed a Silver Bullet,” The Journal of Marketing Theory and Practice (19:2), pp. 139–152.

Hamari, J. 2015. “Gamification - Motivations & Effects,” Doctoral Dissertations 11/2015, Aalto University, Helsinki, Finland.

Hamari, J., and Koivisto, J. 2013. “Social Motivations to Use Gamification: An Empirical Study of Gamifying Exercise,” In Proceedings of the European Conference on Information Systems, pp. 1–12.

Hinkle, D. E., Wiersma, W., and Jurs, S. G. 1988. Applied Statistics for the Behavioral Sciences, (2nd ed.). Boston, MA: Houghton Mifflin Company.

Hoffman, D. L., and Novak, T. P. 1996. “Marketing in Hypermedia Environment Foundations: Conceptual Foundations,” Journal of Marketing (60:3), pp. 50–68.



Hoffman, D. L., and Novak, T. P. 2009. “Flow Online: Lessons Learned and Future Prospects,” Journal of Interactive Marketing (23:1), pp. 23–34.

Hsu, C. L., and Lu, H. P. 2004. “Why Do People Play on-Line Games? An Extended TAM with Social Influences and Flow Experience,” Information and Management (41:7), pp. 853–868.

Huotari, K., and Hamari, J. 2012. “Defining Gamification - A Service Marketing Perspective,” In Proceeding of the International Academic MindTrek Conference, p. 17.

Kankanhalli, A., Tan, B. C. Y., and Wei, K.-K. 2005. “Contributing Knowledge to Electronic Knowledge Repositories: An Empirical Investigation,” MIS Quarterly (29:1), pp. 113-143.

Karppinen, P., Alahaivala, T., Jokelainen, T., Keranen, A.-M., Salonurmi, T., and Oinas-Kukkonen, H. 2014. “Flow or No Flow? A Qualitative Study of Health Behavior Change Support System,” In Proceedings of the Hawaii International Conference on System Sciences, pp. 3044–3053.

Karppinen, P., Oinas-Kukkonen, H., Alahäivälä, T., Jokelainen, T., Keränen, A. M., Salonurmi, T., and Savolainen, M. 2016. “Persuasive User Experiences of a Health Behavior Change Support System: A 12-Month Study for Prevention of Metabolic Syndrome,” International Journal of Medical Informatics (96), pp. 51–61.

Kelders, S. M., Kok, R. N., Ossebaard, H. C., and Van Gemert-Pijnen, J. E. W. C. 2012. “Persuasive System Design Does Matter: A Systematic Review of Adherence to Web-Based Interventions,” Journal of Medical Internet Research, 14(6), pp. 2-25.

Kollmuss, A., and Agyeman, J. 2002. “Mind the Gap: Why Do People Act Environmentally and What Are the Barriers to pro-Environmental Behavior?” Environmental Education Research (8:3), pp. 239–260.

Korzaan, M. L. 2003. “Going With the Flow: Predicting Online Purchase Intentions,” Journal of Computer Information Systems (43:4), p. 25.

Kurz, T. 2002. “The Psychology of Environmentally Sustainable Behaviour: Fitting Together Pieces of the Puzzle,” Analyses of Social Issues and Public Policy (2:1), pp. 257–278.

Langrial, S., Oinas-Kukkonen, H., and Wang, S. 2012a. “Design of a Web-Based Information System for Sleep Deprivation - A Trial Study,” In Proceedings of the International Conference on Well-Being in the Information Society, pp. 41–51.

Langrial, S., Lehto, T., Oinas-Kukkonen, H., Harjumaa, M., and Karppinen, P. 2012b. “Native Mobile Applications For Personal Well- Being: A Persuasive Systems Design Evaluation,” In Proceedings of the Pacific Asia Conference on Information Systems, p.93.

Lehto, T., and Oinas-Kukkonen, H. 2015. “Explaining and Predicting Perceived Effectiveness and Use Continuance Intention of a Behaviour Change Support System for Weight Loss,” Behaviour and Information Technology (34:2), pp. 176–189.

Lehto, T., Oinas-Kukkonen, H., and Drozd, F. 2012a. “Factors Affecting Perceived Persuasiveness of a Behavior Change Support System,” In Proceedings of the International Conference on Information Systems, pp. 1–15.

Lehto, T., Oinas-Kukkonen, H., Pätiälä, T., and Saarelma, O. 2012b. “Consumer’s Perceptions of a Virtual Health Check: An Empirical Investigation,” In Proceedings of the European Conference on Information Systems, p. 154.

Lessel, P., Altmeyer, M., and Krüger, A. 2015. “Analysis of Recycling Capabilities of Individuals and Crowds to Encourage and Educate People to Separate Their Garbage Playfully,” In Proceedings of the Annual ACM Conference on Human Factors in Computing Systems, pp. 1095–1104.

Lindenberg, S., and Steg, L. 2013. “Goal-Framing Theory and Norm-Guided Environmental Behavior,” Encouraging Sustainable Behavior: Psychology and the Environment, van Trijp, H. C. M. (ed.). New York: Psychology Press, pp. 37–54.

Liu, D., Santhanam, R., and Webster, J. 2017. “Toward Meaningful Engagement: A Framework for Design and Research of Gamified Information Systems.,” MIS Quarterly (41:4), pp. 1011–1034.

Lohmöller, J.-B. 1989. “Latent Variable Path Modeling with Partial Least Squares,” Physica-Verlag Heidelberg.

Loock, C.-M., Staake, T., and Thiesse, F. 2013. “Motivating Energy-Efficient Behavior With Green Is: An Investigation of Goal Setting and the Role of Defaults.,” MIS Quarterly (37:4), pp. 1313-A5.

Lowry, P. B., Gaskin, J. E., Twyman, N. W., Hammer, B., and Roberts, T. L. 2012. “Taking ‘Fun and Games’ Seriously: Proposing the Hedonic-Motivation System Adoption Model (HMSAM),” Journal of the Association for Information Systems (14:11), pp. 617-671.

Mathwick, C., and Rigdon, E. 2004. “Play, Flow, and the Online Search Experience,” Journal of Consumer Research (31:2), pp. 324–332.



Mekler, E. D., Brühlmann, F., Opwis, K., and Tuch, A. N. 2013. “Disassembling Gamification: The Effects of Points and Meaning on User Motivation and Performance,” In Extended Abstracts on Human Factors in Computing Systems, p. 1137.

Midden, C. J. H., and Ritsema, B. S. M. 1983. “The Meaning of Normative Processes for Energy Conservation,” Journal of Economic Psychology (4:1–2), pp. 37–55.

Nadkarni, S., and Gupta, R. 2007. “A Task-Based Model of Perceived Website Complexity,” MIS Quarterly (31:3), pp. 501–524.

Oinas-Kukkonen, H. 2013. “A Foundation for the Study of Behavior Change Support Systems,” Personal and Ubiquitous Computing (17:6), pp. 1223–1235.

Oinas-Kukkonen, H., and Harjumaa, M. 2009. “Persuasive Systems Design: Key Issues, Process Model, and System Features,” Communications of the Association for Information Systems (24:1), pp. 485–500.

Orland, B., Ram, N., Lang, D., Houser, K., Kling, N., and Coccia, M. 2014. “Saving Energy in an Office Environment: A Serious Game Intervention,” Energy and Buildings (74), pp. 43–52.

Osbaldiston, R., and Schott, J. P. 2012. “Environmental Sustainability and Behavioral Science: Meta-Analysis of Proenvironmental Behavior Experiments,” Environment and Behavior (44:2), pp. 257–299.

Pavlou, P.A., Liang, H. and Xue, Y. 2007. “Understanding and Mitigating Uncertainty in Online Exchange Relationships: A Principal-Agent Perspective,” MIS Quarterly (31:1), pp. 105 -136.

Petkov, P., Köbler, F., Foth, M., and Krcmar, H. 2011. “Motivating Domestic Energy Conservation through Comparative, Community-Based Feedback in Mobile and Social Media,” in Proceedings of the International Conference on Communities and Technologies, p. 21.

Petty, R. E., and Cacioppo, J. T. 1986. “The Elaboration Likelihood Model of Persuasion,” Advances in Experimental Social Psychology (19:C), pp. 123–205.

Pornpitakpan, C. 2004. “The Persuasiveness of Source Credibility: A Critical Review of Five Decades’ Evidence,” Journal of Applied Social Psychology (34:2), pp. 243–281.

Reychav, I., and Wu, D. 2015. “Are Your Users Actively Involved? A Cognitive Absorption Perspective in Mobile Training,” Computers in Human Behavior (44), pp. 335–346.

Roche, S. M., and McConkey, K. M. 199. “Absorption: Nature, Assessment, and Correlates,” Journal of Personality and Social Psychology (59:1), pp. 91–101.

Schoech, D., Boyas, J. F., Black, B. M., and Elias-Lambert, N. 2013. “Gamification for Behavior Change: Lessons from Developing a Social, Multiuser, Web-Tablet Based Prevention Game for Youths,” Journal of Technology in Human Services (31:3), pp. 197–217.

Shevchuk, N., and Oinas-Kukkonen, H. 2016. “Exploring Green Information Systems and Technologies as Persuasive Systems : A Systematic Review of Applications in,” In Proceedings of the International Conference on Information Systems, pp. 1–11.

Sicilia, M., and Ruiz, S. 2007. “The Role of Flow in Web Site Effectiveness,” Journal of Interactive Advertising (8:1), pp. 1–15.

Steg, L., and Vlek, C. 2009. “Encouraging pro-Environmental Behaviour: An Integrative Review and Research Agenda,” Journal of Environmental Psychology (29:3), pp. 309–317.

Stern, P. C., Dietz, T., and Guagnano, G. A. 1995. “The New Ecological Paradigm in Social-Psychological Context,” Environment and Behavior (27:6), pp. 723–743.

Stibe, A., and Oinas-Kukkonen, H. 2014a. “Designing Persuasive Systems for User Engagement in Collaborative Interaction,” In Proceedings of the European Conference on Information Systems, pp. 1–17.

Stibe, A., and Oinas-Kukkonen, H. 2014b. “Using Social Influence for Motivating Customers to Generate and Share Feedback,” in Proceedings of the International Conference on Persuasive Technology, pp. 224–235.

Stibe, A., Oinas-Kukkonen, H., Bērziņa, I., and Pahnila, S. 2011. “Incremental Persuasion through Microblogging,” In Proceedings of the International Conference on Persuasive Technology, pp. 1–8.

Tellegen, A., and Atkinson, G. 1974. “Openness to Absorbing and Self-Altering Experiences (‘Absorption’), a Trait Related to Hypnotic Susceptibility,” Journal of Abnormal Psychology (83:3), pp. 268–277.

Trevino, L. K., and Webster, J. 1992. “Flow in Computer-Mediated Communication: Electronic Mail and Voice Mail Evaluation and Impacts,” Communication Research (19:5), pp. 539–573.

van Vugt, H. C., Hoorn, J. F., Konijn, E. A., and de Bie Dimitriadou, A. 2006. “Affective Affordances: Improving Interface Character Engagement through Interaction,” International Journal of Human Computer Studies (64:9), pp. 874–888.



Venkatesh, V., Thong, J. Y. L., and Xu, X. 2012. “Consumer Acceptance and Use of Information Technology: Extending the Unified Theory of Acceptance and Use of Technology,” MIS Quarterly (36:1), pp. 157-178.

Vessey, I., and Galletta, D. 1991. “Cognitive Fit: An Empirical Study of Information Acquisition,” Information Systems Research (2:1), pp. 63–84.

Webster, J., and Ho, H. 1997. “Audience Engagement in Multimedia Presentations,” ACM SIGMIS Database (28:2), pp. 63–77.

Webster, J., Trevino, L. K., and Ryan, L. 1993. “The Dimensionality and Correlates of Flow in Human-Computer Interactions,” Computers in Human Behavior (9:4), pp. 411–426.

Zhao, D., McCoy, A. P., Du, J., Agee, P., and Lu, Y. 2017. “Interaction Effects of Building Technology and Resident Behavior on Energy Consumption in Residential Buildings,” Energy and Buildings (134), pp. 223–233.

Appendix

Table A1. Survey Instrument and Factor Loadings

Construct Items L

Primary Task Support (PRIM) (Oinas-Kukkonen and Harjumaa 2009)

The app decreases complexity of sustainable behavior by breaking it into separate tasks. The app helps me track and monitor my behavior. The app provides me personalized content related to performing target behavior.

.690 .853 .869

Dialogue Support (DIAL) (Oinas-Kukkonen and Harjumaa 2009)

The app encourages me based on my behavior. The app rewards me for performing my target behavior. The app offers a virtual specialist to support me in performing my target behavior.

.916

.820

.768

Credibility Support (CRED) (Oinas-Kukkonen and Harjumaa 2009)

The app is truthful, fair, and unbiased. The app has a professional appearance. The app provides competent up-to-date information.

.744

.804

.876

Social Support (SOCI) (Oinas-Kukkonen and Harjumaa 2009)

The app allows me to observe actions and outcomes of other people. The app allows me to compare myself with others. The app enables receiving public recognition for performing my target behavior.

.856 .810 .807

Perceived Persuasiveness (PEPE) (Oinas-Kukkonen 2013; Lehto et al. 2012a)

The app has an influence on me. The app is personally relevant for me. The app makes me reconsider my habits. In my opinion, the app is convincing.

.919

.916

.907

.874

Cognitive Absorption (COAB) (Agarwal and Karahanna 2000; Lowry et al. 2012):

Perceived Enjoyment (PEEN)

I found using the app to be enjoyable. I had fun using the app. Using the app was boring. (reverse scaled) The app really annoyed me. (reverse scaled) The app experience was pleasurable. The app left me unsatisfied. (reverse scaled)

.887

.913

.882

.806

.787

.814

Temporal Dissociation (TEDI)

Time appeared to go by very quickly using the app. I lost track of time when I was using the app. Time “flew” when I used the app.

.890

.899

.909

Focused Immersion (FOIM)

I was able to block out most of distractions when I was using the app. I was absorbed in what I was doing when I used the app. I was immersed in the app.

.748 .957 .922 .471



I was distracted by other attentions very easily while using the app. (reverse scaled)

Control (CTRL)

I had a lot of control in the app. I could choose freely what I wanted to see or do in the app. I was in control in the app. I had no control over my interaction in the app. (reverse scaled) I was allowed to control my interaction in the app.

.878

.854

.891

.407

.719

Curiosity (CURI)

This app experience excited my curiosity.

This app experience made me curious.

This app experience aroused my imagination.

.960

.967

.917

Intention to Use (INT_USE) (Davis et al. 1989; Venkatesh et al. 2012; Lehto and Oinas-Kukkonen 2015)

I would use the app in the future.

I would be willing to try using the app in the future.

I would consider using the app in the future.

I can imagine myself using the app in the future

.962

.966

.962

.960

Environmental Awareness [control check] (Steg et al 2005)

Global warming is a problem for society. Energy savings help reduce global warming. The exhaustion of fossil fuels is a problem. The exhaustion of energy sources is a problem. Environmental quality will improve if we use less energy. It is not certain whether global warming is a real problem. (reverse scaled)

-

Note. Items in italics were deleted due to values of the outer loadings significantly below the critical value (.7)

Documents

c 2019 [Please consult the author] Notice Changes ... of Gamified Persuasive... · Green IS to induce desired behavioral patterns. Feldman and Dolan (2011) believe that gamification