Gamification Design for Educational Contexts: Theoretical and Practical Contributions 3031319486, 9783031319488

Citation preview

Armando Toda Alexandra I. Cristea Seiji Isotani Editors

Gamification Design for Educational Contexts Theoretical and Practical Contributions

Gamification Design for Educational Contexts

Armando Toda • Alexandra I. Cristea • Seiji Isotani Editors

Gamification Design for Educational Contexts Theoretical and Practical Contributions

Editors Armando Toda Universidade de São Paulo São Paulo, Brazil

Alexandra I. Cristea Durham University Durham, UK

Seiji Isotani Harvard University Cambridge, MA, USA

ISBN 978-3-031-31948-8 ISBN 978-3-031-31949-5 https://doi.org/10.1007/978-3-031-31949-5

(eBook)

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

During my academic career, I’ve been asked a lot of times “How can I gamify my classes? It seems pretty cool but I have no idea where to start”, and this led me to start my research on this topic. I’ve tried for a long time to use existing tools (available at the time) to help teachers and other education professionals to gamify whatever tasks they wanted, but failed miserably. Since then, I dedicated my research to help and support those same professionals, by creating tools, processes, and so on to ease their life. I don’t think I solved that problem during my PhD, but I believe to have created some tools that can be useful for teachers and other professionals, and help them to achieve what they couldn’t at some time ago. This book presents a compilation of research from me and my main co-authors regarding our contributions for gamification design in educational contexts, and we hope that this book can make other teachers’ lives easier when they want to use concepts related to gamification in their environments. São Paulo, Brazil January 2023

Armando Toda

v

Contents

Part I Fundamentals Gamification for Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Armando Toda, Alexandra I. Cristea, and Seiji Isotani

3

Gamification and Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Paula T. Palomino, Luiz Rodrigues, and Armando Toda

15

Ethical Challenges in Gamified Education Research and Development: An Umbrella Review and Potential Directions. . . . . . . . . . . . . . . Ana Carolina Tomé Klock, Brenda Salenave Santana, and Juho Hamari Theories Around Gamification in Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Paula T. Palomino

37 49

Part II Methods and Tools TGEEE: Analysis and Suggestions for Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Armando Toda and Alexandra I. Cristea Using Participatory Design to Design Gamified Interventions in Educational Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Armando Toda, Elad Yacobson, Giora Alexandron, Paula T. Palomino, Mauricio Souza, Elian Santos, Alinne Corrêa, Rodrigo Lisboa, Thiago Damasceno Cordeiro, and Alexandra I. Cristea Data Mining in Gamified Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luiz Rodrigues and Armando Toda

69

85

97

Part III Miscellaneous The Dark Aspects of Gamification in Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Armando Toda Evaluation of TGEEE by Education Professionals . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Armando Toda, Marcia Almeida, and Luis Salinas vii

Contributors

Giora Alexandron Weizmenn Institute, Rehovot, Israel Marcia Almeida University of Sao Paulo, Sao Paulo, Brazil Alinne Corrêa Federal University of Technology - Parana, Curitiba, Brazil Thiago Damasceno Cordeiro Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil Juho Hamari Gamification Group, Tampere Univeristy, Tampere, Finland Ana Carolina Tomé Klock Gamification Group, Tampere Univeristy, Tampere, Finland Rodrigo Lisboa Federal Rural University of the Amazon, Pará, Brazil Paula T. Palomino Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil Luiz Rodrigues Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil Luis Salinas University of Sao Paulo, Sao Paulo, Brazil Brenda Salenave Santana Federal University of Rio Grande do Sul, Porto Alegre, Brazil Elian Santos Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil Mauricio Souza Lavras Federal University, Lavras, Brazil Elad Yacobson Weizmenn Institute, Rehovot, Israel

ix

Acronyms

ADDIE AR CPR CRISP-DM CSV DDGD EA FAIR GA GDPR HCI ITS LGPD LX MDA MR OSFA PBL RPG SDT STEM TGEEE UX VR

Analyse, design, develop, implement, and evaluate Augmented reality Cardiopulmonary resuscitation Cross Industry Standard Process for Data Mining Comma-separated value Data-driven gamification design Educational actors Findability, accessibility, interoperability, and reusability Gamification actor General Data Protection Regulation Human-computer interaction Intelligent tutoring systems General Law of Data Protection (Translated) Learning experience Mechanics, dynamics, and aesthetics Mixed reality One-size-fits-all Points-badges-leaderboard Role-playing game Self-determination theory Science, technology, engineering, and mathematics Taxonomy of gamification elements for educational environments User experience Virtual reality

xi

Part I

Fundamentals

In this part we describe fundamental concepts to understand gamification, gamification design, personalisation in gamification design, and what to consider before applying gamification in educational contexts. In the first chapter, we deal with some basic concepts of games and gamification, expliciting some differences and current trends, as well as the delineation of the book. The second chapter deals with psychology theories related to gamification and why they are used, we present a brief description of the most used theories in gamification literature. Following, we present a study that deals with ethics in gamification in education literature, presenting an umbrella review that summarises some ethical issues and how to deal with them in educational contexts. Finally, this part present some general theories about gamification by discussing and presenting how gamification and educational concepts can be tied through frameworks to support teachers and instructors.

Gamification for Education Armando Toda, Alexandra I. Cristea, and Seiji Isotani

1 Games and Gamification Gamification has gained some notoriety in recent years. This happens due its cost to create gameful experiences focused on increasing users’ engagement and motivation. However, it is difficult to explain its success without talking about games, the main concept where gamification derives from [16]. Games are a part of different societies since ancient times. These societies have used playful activities as ways to engage with others, and build relationships [10, 19]. According to Huizinga [10], games have always been a part of our history and culture; Huizinga calls them “a magic circle”. In this concept, the “game” is a voluntary task that happens within a limited time and space, with predetermined and obligatory rules. Tackling the task is accompanied by a feeling that is different from what the “player” is used to in their daily routine (Fig. 1). In Huizinga, the author explains that the magic circle is a fictional space that is delimited by its rules, narrative, and immersion. While on this circle, the individual becomes the player, and the rules of this circle creates a narrative that leads the player to immersion. An example would be the action of kicking a ball through a net, in the real world, this is just a meaningless action, however in the magic circle of “football”, kicking the ball through a net achieves the player a goal.

A. Toda () University of Sao Paulo, Institute of Mathematics and Computer Science, Sao Paulo, Brazil A. I. Cristea Durham University, Department of Computer Science, Durham, UK e-mail: [email protected] S. Isotani Harvard University, Cambridge, MA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_1

3

4

A. Toda et al.

Magic Circle Real World

Rules Narrative Player

Immersion

Individual

Fig. 1 Magic circle

Caillois expanded these definitions, by mapping different types of games that are independent from their social, material, and temporal context, creating the first typology of games that would be used and expanded across the years [4]. According to Caillois, a “game” is the opposite of laborious work. However, the author also recognises the potential that games have to influence other cultural and daily aspects of society, and can be used to understand those same cultures and societal pillars that sustain that society. In summary, for a long time, games may have been seen as anti-productive tasks that only hindered the economical aspects of a given industry [18]. Those same games, however, were also part of a culture that pushed these industries as a form of entertainment. These concepts were the basis for many of the definitions of games and digital games that are used nowadays [11]. This societal economy membership is repeated in the digital age, from the emergence of the first digital games and consoles in the 1970s, where video games started to gain a major space in general media, and their marketing growth took off exponentially [33]. Both of these concepts defined in Huizinga and Caillois [4, 10] are important to understand the actual concepts of digital games that have been used and diffused in the literature. According to Salen and Zimmerman [25], a game can be defined as a virtual system that engages the players in a conflict that is limited by its own rules with a predetermined outcome. This concept is further expanded by Koster [17] which also adds that this system has a continuous and instant feedback that evaluates the players’ actions, and the outcome of these actions produce an emotional response.

Gamification for Education

5

More recently, McGonigal [19] proposes a set of characteristics that make a game (both digital and non-digital), which are: a set of objectives that gives purpose to players; a set of rules that set limitations on the ways in which the players can achieve their objectives; the instant feedback that evaluates the players’ actions; and voluntary, “autonomous” participation, meaning that players have total autonomy to accept or decline any of the rules or objectives that are imposed. Due to their increasing popularity, researchers tried to understand how they could use these same systems, that once were called out for being anti-productive, to improve their employers’ productivity and, consequently, overall productivity of a given industry. This led to many failed attempts such as the “Fun-at-work” movement in the 1990s, and “Funfication”, because only using games without actually improving working conditions was not enough to have a drastic influence over the employers productivity [21]. The reader might ask themselves, ‘why is this important to Gamification?’. In the early 2000s’, Nick Pelling, an Amazon employee, gave a talk about how Amazon used game-like elements to create almost “gameful” experiences on the Amazon website [20]. Since then, many applications have emerged that also used game-like elements, to create “gameful” experiences to change behaviours. This led to the first formal definition of “gamification”, coined by Deterding et al. [7], which is “the use of game-like elements outside of a game, aiming to create gameful experiences”. To better contextualise this definition, Deterding et al. describe the concept stating the major differences (Fig. 2) between a game and a gamified application, with respect to the preponderance of game-like elements in that application—thus ranging from the whole (full fledged games) and parts of this whole (gamification). They also differentiate between “gameful” and playful, where playful is derived from toys (whole) and playful interactions (parts of). Since then, gamification has been promised as an universal panacea for many existing problems related to productivity. Many approaches to “gamify your business” emerged and, consequently, many claims of what gamification could improve [27]. When considering the first concept defined in Huizinga [10], we can observe that gamification could be considered a middle point between the magic circle and the real world (Fig. 3). Where the individual is not a player, but also interacts with the rules and narratives from the circle (in the real world), which may cause immersion. In parallel, many research on gamification has been conducted in the past 10 years, and here, our focus will be mainly on the education field which has been accumulated the majority of studies related to gamification [3, 8, 15], since educational technologies were often not sufficient to engage students in finishing their learning tasks. These studies provided many insights on how gamification can affect (positively and negatively) students’ engagement, motivation, and consequently their performance [1, 8]. Since then, gamification has become a major ally in the field of education, with research spanning over many fields, from artificial intelligence (e.g. adaptive gamification [15]), to sociology (e.g. ethics and gamification [14]).

6

A. Toda et al.

Gaming

Full-fledged games

Gamification

Whole

Parts

Playful design

Toys

Playing Fig. 2 Differences between games, playful interactions, and gamification, according to Deterding et al. [7] Fig. 3 Gamification and the magic circle

Magic Circle

Real World

Gamification

After this brief contextual introduction, the remainder of this book is thus dedicated to gamification in education and its applications. We next provide basic definitions regarding the concept as well as tools that can be used by practitioners (e.g. teachers, professors, instructional designers, etc.) to help them “gamify” learning contexts. We them discuss the application of these tools and what could be improved in practical implementations, as well in-depth analysis of the positive and negative points regarding gamification, to guide practitioners.

2 Gamification in Education Gamification in education is the use of these game-like mechanics to improve students’ engagement and motivation, as well as improve training processes [13]. The gamification concept commenced drawing the attention of educational technologies

Gamification for Education

7

researchers and practitioners alike, due to its low-cost benefit, and because it also has been shown to have a positive impact on students’ engagement and motivation [1, 24]. This can be justified by psychological theories, such as the Flow theory [5] and the Self-Determination theory [6] (as is further explained in detail in the next chapters). According to research, both engagement and motivation have a major impact on students’ performance in learning environments [9, 26]. Since gamification has a direct impact on students’ motivation, it also needs to be carefully applied, otherwise it can lead to some negative effects. According to the literature [28], if not well designed, gamification can hinder the learning process (which is further detailed in the next chapters, including methodological advice on how the negative impact can be avoided). Firstly, it is important to understand that there are two types of gamification [13] in the field of education: structural- and content gamification. Structural gamification is the application of the game elements in a macroenvironment that is applied across all content, and to all learners, but does not interact with the learning content within it. Usually this type of gamification is focused on the high-level learning process, through tasks and objectives to be achieved. One example of structural gamification in real life is found in the Duolingo1 learning system (Fig. 4). The application contains a series of game elements focused on completing tasks (e.g. levels, competition, progression, etc.), but the content itself is not gamified. As we can see, Duolingo2 presents many competitive elements and a progression-based gamification, which consists in telling the learner their current progress, as a constant feedback loop. This type of gamification is usually “cheaper”, since the gamification designer only needs to care about the environment and not the content itself. Content gamification, on the other hand, is the use of game-like elements in the learning content itself, as a way to make it more attractive and engaging to the learner. In this type of gamification, the objective is to make the content be more “game-like”, possibly by using fictional elements, such as narrative and storytelling. Using this kind of gamification requires more effort from the gamification designers, since they need to align the possible game elements with learning objectives, whilst also taking into account the subject being taught. In virtual environments, this could be the content presented in a visual-novel style3 and activities could be presented as badge-based quizzes (e.g. the learner wins a badge when providing the right answer, via a character they play in the story). In non-virtual environments, the instructional designer can use storytelling in an RPG4 -like manner and provide challenges to the students, while (sometimes indirectly) explaining the content.

1 https://duolingo.com/. 2 https://press.duolingo.com/#press-kit. 3 Visual novel refers to a game genre based on storytelling, which consists in the player interacting with a plot through text and images. 4 Role Playing Game, a game genre focused on storytelling and acting.

8

A. Toda et al.

Fig. 4 Duolingo screenshot seen in their press kit

An example of content gamification in education is Lifesaver app,5 which is an interactive application that focus on teaching basics of cardiopulmonary resuscitation through interactive storytelling. In this app, the student is exposed to real-life situations where they must take action to save people using CPR. Game Elements For both these two major types of gamification, at a finer granularity level, the decision needs to take place on what ‘game elements’ to use. For instance, a progression bar can be used in structural gamification as a way to provide the students’ progress through a content, and this same progression bar can also be used within a content gamification as a way to present the remaining content that is needed to be seen or watched by the student (in this case, within the content). Seaborn and Fels [27] reported in their study, together with and other authors in the literature, that there are many (possibly, too many) game element classifications [30]. In this book, we adopt our previous work, where we summarised existing game elements into only 21 constructs (Fig. 5), for which we showed that their respective synonyms can be used to “translate” to-from other classifications. Importantly, we use the TGEEE classification (more details on Chap. 5), as it was designed to focus on educational environments and how these elements could affect students’ engagement and motivation.

5 https://life-saver.org.uk/.

Gamification for Education

9

Fig. 5 Taxonomy of gamification elements for education contexts [30]

Besides these game elements, it is well known that gamification is contextdependent [15, 27], which means that each instance of a gamified strategy will have different outcomes, depending on the contextual and cultural factors of a given environment. In educational environments, beside these factors, the different students of a classroom come also with different personalities and traits. This variational problem led researchers to study how to provide the most suitable game elements to attend students’ needs, focused on their characteristics, starting a new field on personalised gamification [15, 24]. Personalised gamification in education consists in analysing students’ profiles and matching their characteristics with gamification elements, to be used in content—and/or structural gamification. These profiles can be either behavioural, psychological, or even based on their demographic characteristics [15]. Considering existing behavioural profiles, we can cite player types, which are a set of traits based on the preferred gameplay strategy of the student. Player types emerged with the study of Bartle [2], who classified 4 original types: Explorer, Achiever, Killer, and Socialiser. Since then, this classification has expanded to different types and to different genres of games. However, according to Bartle, his classification is not

10

A. Toda et al.

Table 1 HEXAD player type model for gamification definitions, as seen in Tondello Player type Philanthropists

Driven by Purpose, willing to give without expecting rewards Suggested elements: Cooperation and reputation, e.g. providing items that can be traded between students; administrative roles through reputation Socialisers Relatedness, will to connect to others Suggested elements: All the Social elements Achievers Competence, willing to progress by proving themselves Suggested elements: All the Performance elements Free Spirits Autonomy, willing to act without external control Suggested elements: All the Fiction and Personal elements Players Rewards, willing to get every type of reward in the system Suggested elements: All the Performance and Ecological elements Disruptors Change, willing to change things within the system (sometimes for good, sometimes for bad) Suggested elements: Narrative, Imposed choice, and Personal elements

suitable to be used in gamified environments.6 This motivated other researchers to design and develop different player types focused on gamification; one of the most famous models being the HEXAD, designed by Tondello et al. [32] The HEXAD model7 is based on 6 types, each one driven by a given motivational construct, e.g. the socialisers are motivated by relatedness, where they want to create connections with other users. A complete explanation of each of the gamer types can be seen in Table 1, alongside their suggested game elements, translated from our classification in Fig. 5. Besides the player profiles, we also have profiles based on psychology, which are broader and encompass a series of different characteristics and needs. In the study conducted by Palomino [22], the author proposed an initial classification based on a mapping of characteristics seen in Jung’s archetypes [12]. These archetypes were mapped using semiotic approaches. Table 2 contains each one of these archetypes, with their own concepts and their related characteristics [23].

6 https://www.gamedeveloper.com/design/an-interview-with-richard-bartle-about-gamesgamification. 7 https://hcigames.com/user-types-hexad/.

Gamification for Education

11

Table 2 Palomino’s mapping of concepts and characteristics onto Jung’s archetypes [23] Archetype The Innocent

Concepts related to the archetype (Firstness) Freedom, happiness, and naiveness

The Sage

Wisdom, Intelligence, and Meticulous

The Explorer

Autonomy, ambition, and inner emptiness

The Outlaw

Outrageousness, idealism, and radical freedom

The Magician

Make things happen, manipulation, and determination

The Hero

Competence, courage, and arrogance

The Lover

Passion, gratitude, commitment, and weak identity

The Jester

Joy, frivolity, and playfulness

The Everyman

Realism, empathy, and lack of pretense

The Caregiver

Compassion, generosity, and martyrdom Responsibility, leadership, and authoritarianism Creativity, imagination, and perfectionism

The Ruler The Creator

Characteristics related to the archetype (Secondness) Aims to do things right and dislikes doing things wrong Aims to find the truth and fears being misled Aims to experience a fulfilling life and fears conformity Aims to overturn what is not working and fears being powerless Aims to understand the laws of the universe and fears negative consequences Aims to achieve expert mastery in a way that improves the world and fears weakness Aims to be in a good relationship and fears being alone or unwanted Aims to have a great time and fears being bored Aims to belong and fears to be left out Aims to help others and fears ingratitude and selfishness Aims to create a prosperous community and fears chaos Aims to realise a vision and fears mediocre execution

Finally, within personalised gamification, recent studies also tried to identify patterns of game elements’ preference related to students’ demographics. Up to now, no study found any statistically relevant relation between the students’ gender and possible gamification elements [24, 29]. However, students’ age and other characteristics, like time they play games, and favourite game genre do influence their preferences, and consequently the most suitable strategies that could be applied to them [31]. It is important to understand that knowing all students’ characteristics is complex, and personalisation models tends to focus on specific traits [15, 24]. Furthermore, as previously noted, it is important to consider that in practice other external factors might influence gamification acceptance by students like, context and culture.

12

A. Toda et al.

3 For Practitioners This book ties in and presents many concepts related to how you can design and apply gamification in a given learning environment. We provide both higher level concepts, as well as concrete tools that can be used and present some examples on how they can be used. However, it is very important for practitioners that the gamified strategies that are created are tied in their respective contexts; e.g., if the practitioner does not have access or resources to apply player type questionnaires, they can design a one-size-fits-all approach, based on other characteristics of their group of students. Another example, if the practitioner does not have access to ways to implement certain elements virtually, they can try to work with their existing resources. The intention of this book is not to oblige other practitioners to use specific gamification strategies, but to present ways on how to create them, and present examples that had positive outcomes in their specific contexts. In the next chapters we will explore these elements in detail. In this first part, we will present some concepts related to gamification, gamification design in education, psychology and learning theories that are and can be related to gamification, as well as ethics, diversity, and fairness related to gamification in educational contexts. Following, in the second part, we present some tools on how we can use gamification in learning environments, by showing in depth our taxonomy, alongside its elements, and how they can be used individually, or in groups. We also present a way on how to generate those strategies using data-driven algorithms and tools to support those who are not familiarised within these concepts. Furthermore, we also present case studies on how we can use those elements present in the taxonomy to co-design gamification with teachers and virtual learning environments designers. In the last part, we present our final thoughts on our taxonomy of gamification elements, as well as discuss beneficial and harmful aspects of gamification in learning. We close this book by proposing future research directions for educational practitioners and opportunities in this exciting field that is expanding every year.

References 1. Bai, S., Hew, K.F., Huang, B.: Does gamification improve student learning outcome? Evidence from a meta-analysis and synthesis of qualitative data in educational contexts. Educ. Res. Rev. 30, 100322 (2020) 2. Bartle, R.: Hearts, clubs, diamonds, spades: players who suit MUDs (1996). http://mud.co.uk/ richard/hcds.htm 3. Borges, S.d.S., Durelli, V.H.S., Reis, H.M., Isotani, S.: A systematic mapping on gamification applied to education. In: Proceedings of the 29th annual ACM symposium on applied computing - SAC ’14, Icmc, pp. 216–222 (2014). https://doi.org/10.1145/2554850.2554956. http://dl.acm.org/citation.cfm?doid=2554850.2554956 4. Caillois, R., Barash, M.: Man, Play, and Games. University of Illinois Press (1961)

Gamification for Education

13

5. Csikszentmihalyi, M.: Flow: The Psychology of Optimal Experience. Harper & Row (1975). http://www.amazon.com/Flow-The-Psychology-Optimal-Experience/dp/0061339202 6. Deci, E.L., Ryan, R.M.: Intrinsic Motivation and Self-determination in Human Behavior. Springer US, Boston, MA (1985). https://doi.org/10.1007/978-1-4899-2271-7. http://link. springer.com/10.1007/978-1-4899-2271-7 7. Deterding, S., Dixon, D., Khaled, R., Nacke, L.: From game design elements to gamefulness: defining “gamification”. In: Proceedings of the 15th International Academic MindTrek Conference: Envisioning Future Media Environments, pp. 9–15 (2011) 8. Dichev, C., Dicheva, D.: Gamifying education: what is known, what is believed and what remains uncertain: a critical review (2017). https://doi.org/10.1186/s41239-017-0042-5. http:// educationaltechnologyjournal.springeropen.com/articles/10.1186/s41239-017-0042-5 9. Hanson-Smith, E.: Games, gaming, and gamification: some aspects of motivation. TESOL J. 7(1), 227–232 (2016). https://doi.org/10.1002/tesj.233 10. Huizinga, J.: Homo ludens. Editora Perspectiva SA (2020) 11. Huotari, K., Hamari, J.: Defining gamification: a service marketing perspective. In: Proceeding of the 16th International Academic MindTrek Conference, pp. 17–22 (2012). http://dl.acm.org/ citation.cfm?id=2393137 12. Jung, C.G.: The Archetypes and the Collective Unconscious. Routledge (2014) 13. Kapp, K.M.: The gamification of learning and instruction: game-based methods and strategies for training and education (2012). http://dl.acm.org/citation.cfm?id=2378737 14. Kim, T.W., Werbach, K.: More than just a game: ethical issues in gamification. Ethics Inf. Technol. 18(2), 157–173 (2016). https://doi.org/10.1007/s10676-016-9401-5. http://link. springer.com/10.1007/s10676-016-9401-5 15. Klock, A.C.T., Gasparini, I., Pimenta, M.S., Hamari, J.: Tailored gamification: a review of literature. Int. J. Human-Comput. Stud., 102495 (2020). https://doi.org/10.1016/J.IJHCS.2020. 102495. https://www.sciencedirect.com/science/article/pii/S1071581920300975 16. Koivisto, J., Hamari, J.: The rise of motivational information systems: a review of gamification research (2019). https://doi.org/10.1016/j.ijinfomgt.2018.10.013. https://www.sciencedirect. com/science/article/pii/S0268401217305169 17. Koster, R., Wright, W.: A theory of fun for game design (2004). http://dl.acm.org/citation.cfm? id=1207478 18. McDonald, P.D.: The principle of division in Roger Caillois’s man, play and games. Games Cult. 15(8), 855–873 (2020). https://doi.org/10.1177/1555412019853080 19. McGonigal, J.: Reality is broken: why games make us better and how they can change the world. Deterding. Penguin Group. (2011). http://dl.acm.org/citation.cfm?id=2207854. http:// dl.acm.org/citation.cfm?id=1972527 20. Mora, A., Riera, D., Gonzalez, C., Arnedo-Moreno, J.: A literature review of gamification design frameworks. In: 2015 7th International Conference on Games and Virtual Worlds for Serious Applications (VS-Games), pp. 1–8. IEEE (2015). https://doi.org/10.1109/ VS-GAMES.2015.7295760. http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber= 7295760 21. Nelson, M.J.M.: Soviet and American precursors to the gamification of work. In: Proceeding of the 16th International Academic MindTrek Conference (pp. 23–26) pp. 23–26 (2012). https:// doi.org/10.1145/2393132.2393138. http://dl.acm.org/citation.cfm?id=2393138 22. Palomino, P.T.: Gamification of virtual learning environments: a narrative and user experience approach. Ph.D. thesis, Universidade de São Paulo, São Carlos (2022). https://doi.org/ 10.11606/T.55.2022.TDE-26072022-105616. https://www.teses.usp.br/teses/disponiveis/55/ 55134/tde-26072022-105616/ 23. Palomino, P.T., Toda, A.M., Oliveira, W., Rodrigues, L., Isotani, S.: Gamification journey: a novel approach for classifying gamer types for gamified educational systems. In: Simpósio Brasileiro de Jogos e Entretenimento Digital 2019 (SBGames 2019) (2019) 24. Rodrigues, L., Toda, A.M., Palomino, P.T., Oliveira, W., Isotani, S.: Personalized gamification: a literature review of outcomes, experiments, and approaches. In: ACM International Conference Proceeding Series, pp. 699–706 (2020). https://doi.org/10.1145/3434780.3436665

14

A. Toda et al.

25. Salen, K., Zimmerman, E.: Rules of Play: Game Design Fundamentals (2003). http://dl.acm. org/citation.cfm?id=1215723 26. Sanmugam, M., Zaid, N.M., Mohamed, H., Abdullah, Z., Aris, B., Suhadi, S.M.: Gamification as an educational technology tool in engaging and motivating students; an analyses review. Adv. Sci. Lett. 21(10), 3337–3341 (2015). https://doi.org/10.1166/asl.2015.6489 27. Seaborn, K., Fels, D.I.: Gamification in theory and action: a survey. Int. J. Hum. Comput. Stud. 74, 14–31 (2014). https://doi.org/10.1016/j.ijhcs.2014.09.006 28. Toda, A.M., Valle, P.H.D., Isotani, S.: The dark side of gamification: an overview of negative effects of gamification in education. In: Communications in Computer and Information Science, vol. 832, pp. 143–156. Springer, Cham (2018). https://doi.org/10.1007/978-3-31997934-2_9 29. Toda, A., Palomino, P.T., Rodrigues, L., Oliveira, W., Shi, L., Isotani, S., Cristea, A.I.: Validating the effectiveness of data-driven gamification recommendations: an exploratory study. In: Proceedings of the SBIE 2019, vol. 30, pp. 763–772. Brazilian Computer Society (Sociedade Brasileira de Computação - SBC), Brasilia (2019). https://doi.org/10.5753/cbie. sbie.2019.763. https://br-ie.org/pub/index.php/sbie/article/view/8804 30. Toda, A.M., Klock, A.C.T., Oliveira, W., Palomino, P.T., Rodrigues, L.L., Shi, L., Bittencourt, I., Gasparini, I., Isotani, S., Cristea, A.I.: Analysing gamification elements in educational environments – using an existing Gamification Taxonomy. Smart Learn. Environ. 6(1), 16 (2019). https://doi.org/10.1186/s40561-019-0106-1. https://slejournal.springeropen.com/ articles/10.1186/s40561-019-0106-1 31. Toda, A., Pereira, F.D., Klock, A.C.T., Rodrigues, L., Palomino, P., Oliveira, W., Oliveira, E.H.T., Gasparini, I., Cristea, A.I., Isotani, S.: For whom should we gamify? Insights on the users intentions and context towards gamification in education (Cbie), pp. 471–480 (2020). https://doi.org/10.5753/cbie.sbie.2020.471 32. Tondello, G.F., Wehbe, R.R., Diamond, L., Busch, M., Marczewski, A., Nacke, L.E.: The gamification user types hexad scale. In: Proceedings of the 2016 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’16, pp. 229–243. ACM Press, New York, USA (2016). https://doi.org/10.1145/2967934.2968082. http://dl.acm.org/citation.cfm?doid= 2967934.2968082 33. Wolf, M.J.P.: The Video Game Explosion: A History from PONG to Playstation and Beyond. ABC-CLIO (2008)

Gamification and Motivation Paula T. Palomino, Luiz Rodrigues, and Armando Toda

1 Introduction Gamification in education is the use of game design elements and game mechanics in non-game contexts to engage and motivate students to achieve their goals, as well as making the learning process more engaging [10]. Since those elements tackle directly into students’ motivation, it is important to understand which motivational theories exist and how it affects motivation. Motivational theories are psychological theories that attempt to explain why people behave the way they do and what drives their behavior. These theories can be used to understand and predict the motivation of individuals and groups in different contexts, including in the context of gamification. In education, understanding motivation also help us to understand how to improve learning methods to foster students’ performance [4]. In this chapter, we present some of the most used motivational theories that were mapped in gamification studies in the field of education [31]. We present some basic concepts regard the given theory, as well as some studies that have used or explored these theories.

P. T. Palomino () · L. Rodrigues Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil e-mail: [email protected]; [email protected] A. Toda University of Sao Paulo, Institute of Mathematics and Computer Science, Sao Paulo, Brazil e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_2

15

16

P. T. Palomino et al.

2 Self-determination Theory The Self-Determination Theory (SDT) is a theory about motivation widely used in several areas, including education. Motivation is the energy directed toward achieving a specific goal, and it is essential in the educational field as it can directly influence the learning process. According to the SDT, any person has three psychological needs that constitute motivation: Autonomy, that is, the feeling of choice and freedom that a person has and that supports their behaviour, whereas the opposite experience makes the person feel compelled or controlled in their behaviour; competence, that is, the experience of mastering and being effective in an activity; relatedness, or a refers to the need to feel connected and a sense of belonging with others. Figure 1 shows this relationship [21]. The social environment can promote or hinder people’s efforts to the extent that it meets their basic psychological needs. Autonomy is supported by understanding and acknowledging the person’s desires, preferences, and perspectives, conveying a sense of their point of view, providing justification for engaging in a behaviour, and providing a choice of how to behave in a certain way. One of the ways to support someone’s autonomy is to refrain from trying to control or pressure that person to act in a particular manner. Competency is supported by providing the person with optimal challenges and opportunities, encouraging their sense of initiative, and giving structure to mobilise and organise behaviour and provide relevant feedback. Finally, the relatedness is supported when other people are involved, show interest in the person’s activities, are empathetic in responding to their feelings, and convey that the person is meaningful, cared for, and loved. When these needs are optimally met, evidence suggests that people are more autonomous in their behaviours, are more likely to persist in them, and generally feel better.

Competence The experience of mastery and being effective in one's activity

Autonomy

Relatedness

The feeling one has choice and willingly endorseing ones' behaviour

The need to feel connected and belongingness with others

Motivation

Fig. 1 SDT and the psychological needs [21]

Gamification and Motivation

17

2.1 Dissecting Self-determination Theory The SDT derives from five other theories; each developed to explain a set of underlying motivational phenomena, addressing a facet of personality motivation or functioning. Cognitive Evaluation Theory (CET) concerns intrinsic motivation (i.e., motivation based on the satisfaction of behaving “for its own sake”). Examples of intrinsic motivation are children’s exploration and play, but intrinsic motivation is a lifetime creative source. CET explicitly addresses the effects of social contexts on intrinsic motivation and how factors such as rewards, interpersonal controls, and ego affect intrinsic motivation. CET describes the critical roles played by competence and autonomy supports in promoting intrinsic motivation, which is crucial in education. Organismic Integration Theory (OIT) addresses the issue of extrinsic motivation in various forms, with its properties, determinants, and consequences. Generally speaking, extrinsic motivation is instrumental behaviour that aims at outcomes extrinsic to the behaviour itself. However, there are distinct instrumentality forms, including external regulation, introjection, identification, and integration. These extrinsic motivation sub-types are viewed as falling along a continuum of internalisation. The more internalised the extrinsic motivation, the more autonomous the person will perform the behaviours. The OIT is also concerned with the social contexts that enhance or impede internalisation—what causes people to resist, partially adopt, or deeply internalise values, goals, or belief systems. The OIT particularly highlights support for autonomy and relatedness as critical to internalisation. Relationships Motivation Theory (RMT) is concerned with social relationships. It postulates that a certain amount of such interactions are not only desirable for most people but are, in fact, essential for their adjustment and well-being. However, research shows that it is not only the need for relatedness satisfied in high-quality relationships but also the need for autonomy and, to a lesser extent, the need for competence. As such, the highest quality personal relationships are those in which each partner supports the other’s autonomy, competence, and relatedness needs. Orientation Theory of Causality (COT) describes ways in which people’s tendencies are different toward environments and how they regulate behaviour. The COT describes and evaluates three types of causality orientations: • The autonomy orientation in which people act out of interest and appreciation of what is happening. • The control orientation where the focus is on rewards, earnings, and approval. • The impersonal or amotivated orientation characterised by competence anxiety. Basic Psychological Needs Theory (BPNT) elaborates on how psychological needs evolve and their relationship to psychological health and well-being. The BPNT argues that psychological well-being and optimal functioning are based on autonomy, competence, and relatedness. Therefore, contexts that support versus

18

P. T. Palomino et al.

Fig. 2 SDT diagram and the different types of motivation [21]

frustrate these needs must invariably impact well-being. For this theory, the three needs are crucial and if any are frustrated there will be distinct functional costs. As basic needs are universal aspects of functioning, the BPNT analyses cross-cultural and cross-cultural development settings for validation and refinement. Goal Contents Theory (GCT), arises from how intrinsic and extrinsic goals differentiate between themselves, considering their impact on motivation and wellbeing. Goals are seen as differentially providing satisfactions of basic needs. Extrinsic goals were specifically contrasted with intrinsic goals, with the former likely to be associated with lower well-being and higher distress [21]. Excessive focus on rewards significantly reduces intrinsic motivation as it shifts the perceived locus of causality from internal to external. The autonomy continuum of intrinsic and extrinsic motivation extends from amotivation, a state of apathy/indifference because the skill is lacking or the value of the activity is not understood, to extrinsic motivation, such as external regulation, introjection (ego involvement, self-esteem), identification (valuing the result and therefore voluntarily engaged) and integration (activity aligned with the individual’s value system) to intrinsic motivation (inherent joy in doing the activity) as shown in Fig. 2. Amotivation, external regulation, and introjection generate the lowest motivational quality, where the perceived locus of causality is external. At the same time, identification, integration, and intrinsic motivation generate the highest motivational quality, where the perceived locus of causation is internal. External motivation doesn’t last long after the stimulus that drives it is removed (rewards or punishments), so it’s an unstable form of motivation. Identification, integration, and intrinsic motivations are stable forms of motivation.

2.2 Application Self-Determination Theory has being applied in education and specifically in gamification for education frameworks and experiments. Marczewski proposed six

Gamification and Motivation

19

types of users that differ in the degree to which they can be motivated by intrinsics or extrinsic motivational factors. Instead of basing the model on observations behaviour, user types are personifications of intrinsic and extrinsic motivations, as defined by the SDT [18]. Other research investigates students’ intrinsic motivation when designing and implementing an eClass through the practical use of gamification based on core components of SDT to improve students’ engagement in classes [11]. Van Roy and Zaman created nine gamification heuristics based on gamification from the perspective of self-determination theory, considering the educational domain. The authors describe how various types of motivation guide people’s behaviour differently and point to the importance of basic psychological need satisfaction [30].

2.3 Summary This section detailed in a broad way Self-Determination Theory, one of the most famous theories used in gamification recently. Motivation is crucial for teachers as they try to motivate themselves or their students to learn. People are often motivated by external factors such as rewards or the opinions of others. However, they can also be driven by internal factors such as interests or values. These intrinsic motivations are not always externally rewarded but can drive passions and creativity. The focus of Self-Determination Theory is the relationship between external factors and intrinsic motivations.

3 Flow Theory The flow theory, first proposed by psychologist Mihaly Csikszentmihalyi, describes a state of complete immersion and engagement in an activity. Flow is often associated with intense focus and optimal performance during a task. When considering tasks, two factors are important: the level of difficulty and the individual’s skill level. When a task is challenging, and the individual’s skills are low, it may lead to feelings of anxiety. However, when a task is easy and the individual’s skills are high, it may lead to relaxation or boredom, as shown in Fig. 3. The best chance for achieving flow is when the task difficulty is slightly greater or equal to the individual’s skills. With consistent feedback and minimal distractions, one can enter a flow state. Additionally, it’s important to continue challenging oneself by increasing task difficulty and improving skills, as high task difficulty and high skills are linked to flow. In contrast, low task difficulty and low skills lead to apathy.

P. T. Palomino et al.

High

20

Challenge level

Anxiety

Arousal

Worry

Flow

Control

Boredom

Relaxation

Low

Apathy

Low

Skill level

High

Fig. 3 Flow theory diagram [5]

3.1 Dissecting Flow Theory Csikszentmihalyi and his team identified nine psychological characteristics commonly associated with the optimal experience during a flow state in their research. These characteristics were reported by a diverse group of participants including artists, athletes, academics, and everyday workers. Despite their varied professions, they all described their flow experiences in similar terms. 1. 2. 3. 4. 5. 6. 7. 8. 9.

Complete concentration on the task; Clarity of goals and reward in mind and immediate feedback; Transformation of time (speeding up/slowing down); The experience is intrinsically rewarding; Effortlessness and ease; There is a balance between challenge and skills; Actions and awareness are merged, losing self-conscious rumination; There is a feeling of control over the task; Activity becomes autotelic.

It’s worth noting that the ability to experience flow can vary among individuals. Research suggests that those with an autotelic personality, or those who engage in activities for their own sake, are more likely to experience flow. This personality

Gamification and Motivation

21

type is characterised by a strong interest in life, persistence, and a lack of selfcenteredness. Most importantly, peak experience is characterised by fully immersing in the task for its own sake. The experience becomes self-rewarding, and all thoughts of success or failure are absent. Individuals with an autotelic personality, who engage in activities for their own sake, have no ulterior motives such as seeking money, status, recognition, or applause. It is the joy of the task that drives them. The motivation comes from within rather than being externally driven. A recent study examined the relationship between flow and five personality traits and found a negative correlation between flow and neuroticism and a positive correlation between flow and conscientiousness [28]. By the same logic, neurotic individuals may be more susceptible to anxiety and self-doubt, which can hinder the ability to enter a flow state. Finally, conscientious individuals are more likely to invest time in mastering challenging tasks, which is an essential aspect of the flow experience, particularly in the workplace.

3.2 Application Flow theory is another popular theory in Gamification for Education. Huang et al. examined whether gamification could enhance student engagement in a flipped course, conducting a comparison study involving two classes of undergraduate students in an Information Management course, based on this theory [8]. Jagust et al. examined competitive, collaborative and adaptive gamification in students learning math, also using flow theory [9].

3.3 Summary Flow is a state of mind that combines cognitive, physiological, and emotional aspects. It is an optimal state of being, defined by feelings of being “in the zone” or “in the groove.” Flow often leads to peak performance and can be used with gamification to enhance students’ focus and concentration in learning and therefore is extensively used in gamified educational strategies.

4 Cognitive Load Theory Cognitive load, in Cognitive Load Theory, is defined as the amount of working memory used by the brain, similar to the Random Access Memory (RAM) in a computer. When the cognitive load is high, learning becomes less efficient, similar to a computer slowing down when memory usage is high. Cognitive load can be

22

P. T. Palomino et al.

classified in three types: intrinsic, germane, and extraneous. Intrinsic load is the inherent difficulty of the subject matter and cannot be changed by instructional designers. Germane load is the portion of memory used for integrating new information and should be increased in instructional design. Extraneous load is memory used for processes unrelated to learning and should be reduced as much as possible [24].

4.1 Dissecting Cognitive Load Theory There are two main types of knowledge: innate or inborn, and acquired. Innate knowledge, also known as biologically primary knowledge, is information that humans have evolved to acquire over thousands of generations. It is obtained unconsciously and without instruction, because it is essential for human survival and the functioning of human societies. Examples of innate knowledge include basic problem-solving and thinking skills, such as learning to speak and listen in a native language at a young age, generalising, transferring, and performing basic social skills like recognising faces. Since this knowledge is acquired automatically, it does not need to be taught and does not place a heavy cognitive load on the individual. However, acquired knowledge also referred to as biologically secondary knowledge, is information that needs to be explicitly taught and not left for students to discover. It requires conscious effort, and most of the subjects taught in formal education fall under this category. Examples include reading, writing, mathematics, history, science, and other subjects traditionally taught in schools and universities. Since acquiring this knowledge requires conscious effort, it places a cognitive load on the individual [24]. Sweller argues that most primary knowledge results in generic-cognitive skills, basic skills that can be applied across different domains. On the other hand, most of what is taught in schools consist of domain-specific skills. These biologically secondary skills require explicit instruction because, without them, students may have to rely on trial and error, which can impose a heavy cognitive load. Therefore, domain-specific skills should be explicitly taught to reduce cognitive load [24]. Cognitive load theory describes the process of acquiring new knowledge and how it is stored in memory. It explains how working and long-term memory work together to aid learning. Figure 4 illustrates the different components of the human memory system, which are used to describe the distinct processes related to working memory and long-term memory. The memory system, as shown in Fig. 4, is not a single entity but rather two interconnected components. These components are working memory and long-term memory. Working memory is the conscious aspect of memory, where new information is temporarily stored and manipulated for reasoning, learning, and comprehension. The working memory must first process information to give it meaning before transferring it to long-term memory in a learning process. Long-term memory, on the other hand, is the unconscious aspect of memory where information is stored indefinitely in knowledge structures called

Gamification and Motivation

23 Individual Learn

Environment Short-term memory

Long-term memory

Attention Remember

Forget

Fig. 4 Cognitive load theory diagram [24]

schemas. These schemas are activated when one encounters familiar information and are automatically and easily transferred from long-term memory to working memory for a specific purpose. A more knowledgeable person in a particular domain can quickly transfer large amounts of organised schematic information from longterm memory to working memory to assist with complex problem-solving tasks due to their more sophisticated schemas in magnitude, complexity, and refinement compared to those of less knowledgeable individuals [29].

4.2 Application This theory was used in gamification for education to enhance instructional outcomes and students’ engagement. Landers and Armstrong examined the effectiveness of the Technology-Enhanced Training Effectiveness Model (TETEM) in a gamification setting by randomly assigning participants to read either scenarios of gamified instruction or traditional instruction using PowerPoint, and measuring their level of engagement, using Cognitive Load Theory [13]. This research investigated the impact of gamification strategies on students’ cognitive load and academic performance and also evaluated students’ perspectives on gamification [27].

4.3 Summary The concept of load in Cognitive Load Theory stances for anything that takes from people’s working memory capacity. As such, the theory’s main directive is to reduce cognitive load to increase learning. Strategies for reducing cognitive load include

24

P. T. Palomino et al.

sequencing, chunking, reflection, interleaving, and worked examples. Additionally, instructional designers can reduce the extraneous load by keeping students in the gamified educational system, for example, setting links to open in a new tab, and keeping course navigation clean and simple.

5 Goal-Setting Theory The Goal-setting Theory [16, 17] has been widely used to explain how to motivate people to perform better. Its backgrounded on the idea that conscious goal affect action, considering goal as the object/aim of an action. An example of goal is to finish reading all chapters of this book. Additionally, this theory considers goals have two main attributes. Content concerns the result/object being sought, while intensity involves three aspects: (1) effort required to set the goal, (2) the goal’s position in one’s goal hierarchy; and (3) a person’s degree of commitment to attainment. Furthermore, two main empirical findings, extracted based on hundreds of studies, were core drivers in the Goal-setting Theory’s development. The first concerns evidence that the degree of a goal difficulty has a linear relationship with one’s performance. The second concerns evidence that specific, difficulty goals are associated with higher performance compared to no goals or abstract ones (e.g., do your best). Moreover, the research that originated these core findings also informed another two key points of Goal-setting theory: how those goals increase performance (i.e., mechanisms) and factors that might improve or complicate the relationship between goal and performance (e.g., moderators). Next, we elaborate of these mechanisms and moderators.

5.1 Dissecting Goal-Setting Theory Prior research [16, 17] discusses four main mechanisms through which a specific, high goal leads to improved performance: 1. Orientation: It (1) guides people’s effort and attention toward activities relevant toward to the goals, (2) put individuals away from irrelevant activities, and (3) activates one’s skills and knowledge required to attain it. 2. Effort: Also known as a mediator, the effort one puts vary depending on the demands to attaining the goal. 3. Persistence: Concerns how long one takes to attain the goal (e.g., time spent seeking it). 4. Knowledge or Task Strategy: A more cognitive perspective compared to the other mechanisms, this relates to using one’s knowledge and skills to seek a goal when there is the need to go beyond persistence and effort (e.g., to perform a specific procedure, such as driving or cutting logs).

Gamification and Motivation

25

Additionally, the literature [16, 17] have raised several moderators of the relationship between specific, high goals and performance. Some examples are: • Ability plays a role in one’s choice of a goal because of possibly lacking the knowledge/skill to reach the expected performance level. • Performance feedback helps with self-regulation and, consequently, is assumed to play a positive role on people’s performance. • Goal commitment concerns one’s determination or attachment to a given goal. Research suggests higher commitment levels are associated with better improvements on performance. • Task complexity seems to maximize the goal-setting-performance relationship when complexity is straightforward for people (i.e., when people have the knowledge and skills required to perform well). • Situational constraints/resources might jeopardize goal attainment due to, for instance, the lack of resources such as supplies, task information, and materials. Besides those mechanisms and moderators, previous research has indicated other relevant components. For instance, studies have raised other key findings in terms of setting multiple goals, how assigned, participative and self-set goals compare, the relationship between goals and affect, and the role of self-efficacy and personal goals. For the interested reader, please refer to [16, 17].

5.2 Applications Although research on gamification applied to education has widely focused on SDT, the Goal-setting Theory has also been present [26, 31]. In [2], for instance, the authors conducted a synthesis of qualitative findings on gamification in educational contexts, which resulted in data from 32 studies. Among those, 11 studies reported that one of gamification’s main benefits is promoting goal setting. An example is [7], wherein the authors discuss that most of the study participants felt motivated to complete more tasks and set higher goals due to gamification. Overall, researchers have also explored how the Goal-setting Theory to inform gamification design. In [14], for instance, the authors experimented with different leaderboard designs to understand the most effective approach. Based on an experimental study, the authors compared four goal types (do-your best, easy, difficulty, and impossible) to a leaderboard condition, in the context of a brainstorming task, based on participant performance (i.e., number of ideas generated). The authors additionally captured participant goal commitment and analyzed it as a moderator when comparing the different goal types. Mainly, their findings suggested that leaderboard played a role similar to that of difficulty and impossible goals and that goal commitment behaved similarly to its behavior in the context of traditional goals. Hence, providing empirical evidence on the usefulness of Goalsetting Theory to the design of leaderboards in the context of brainstorming, a

26

P. T. Palomino et al.

task of highest complexity according to Bloom’s Taxonomy of cognitive learning objectives [3].

5.3 Summary The Goal-setting Theory was conceived based on consistent empirical evidence demonstrating that (1) there is a linear relationship between goal difficulty and performance and that (2) some goal types are more effective than others. The mechanisms through which that relationship happens involves goal orientation and people’s effort, persistence, knowledge, and skills related to a given goal. Nevertheless, there are several factors, such as goal commitment and task complexity, that might either maximize or attenuate the goal-performance relationship. Gamification studies has explored the Goal-setting Theory, both in terms of empirical evaluations as well as theoretical conceptualizations, providing valuable directions to inform practical usage and future research efforts.

6 Theory of Gamified Learning While gamification research has heavily relied on well-established theories, such as SDT and Flow, researchers also have proposed specific frameworks to understand how it works. The Theory of Gamified Learning [12] is one example, which has been considered suitable to plan and test gamification applied to educational scenarios [22]. One of the main reasons for its value is acknowledging the several process through which gamification might act to reach the desired outcome. Consider the SDT, for instance. Gamification studies using it as the theoretical background assume game elements will motivate users—often based on autonomy, competence, and relatedness—and, hence, lead to a behavioral outcome, such as increased class attendance (e.g., [8, 11, 30]). In contrast, the Theory of Gamified Learning proposes game elements might affect learning through two processes: mediation and moderation [12]. Next, we follow [12] to introduce the five propositions composing the Theory of Gamified Learning (ToGL hereafter), which is summarized in Fig. 5. Subsequently, we discuss gamification studies providing empirical evidence that supports the ToGL.

6.1 Dissecting the Theory First, the ToGL proposes that instructional content influences learning outcomes and behaviors (connections A to B and A to C). Intuitively, one might expect that the better the instructional content, the better the learning outcomes. Additionally,

Gamification and Motivation

Game Elements (D)

Instructional Content (A)

27

Behavior / Attitude (C)

Learning Outcomes (B)

Fig. 5 Representation of the Theory of Gamified Learning. Adapted from [12]

such quality is likely to play a role on learner behaviors and attitudes. For instance, if instruction is poor, one might become demotivated to engage and dedicate time to it, and vice-versa. Importantly, the ToGL emphasizes that the goal of gamifying learning is to improve, rather than replace instruction. Accordingly, if the instructional content fails to help students learn, gamification is not to blame. Second, the ToGL proposes that behaviors and attitudes influence learning (connection C to B). With this regard, the author discuss a number of behaviors and attitudes that, according to educational theory, have been connected to learning outcomes. One can, for instance, draw a parallel between this connection and Flow Theory, given that flow relates to concentration. Hence, if the student is more concentrated (behavior), it is likely they will achieve better learning outcomes. Similarly, if the student does not like the subject (attitude), this might jeopardize their learning experience. Third, the ToGL proposes that game elements might change behaviors and/or attitudes (connection D to C). For example, if we consider the TGEE, its game elements are grouped into five broad categories (see Chapter 5). Each of those is likely to affect users differently (e.g., fictional compared to social game elements). Similarly, different people might have distinct preferences and attitudes toward the same game elements. Consequently, it is imperative to select appropriate game elements aligned to users and target behavior/attitudes. Fourth, the ToGL proposes that behaviors and attitudes moderate instruction effectiveness (C in-between connection A to B). To exemplify this point, assume young students consider math boring and, importantly, the math instruction in question is effective (as discussed in proposition one). The instructor then could transform the act of solving assignments into a competition of who solves more assignments with the goal of making students perceive it as a fun activity. In that case, the student attitude (fun perception) could then improve instruction effectiveness because they completed more assignments motivated by the competition game element (as in the third proposition). Note, however, that the opposite is possible for students that, for instance, are not motivated by that game element. That is, such moderation might either maximize or minimize instruction effectiveness.

28

P. T. Palomino et al.

Lastly, the ToGL proposes that behaviors and attitudes mediate the relationship between game elements and learning outcomes (connection D to C to B). Fundamentally, this proposition states that for gamification for affect learning outcomes, it must influence behaviors or attitudes related to such learning outcomes. For instance, if gamification targets information retention (learning outcome), the game elements must be designed so that they motivate behaviors known to enhance information recall (e.g., solving quizzes [20]). On the other hand, if the game elements are designed to reward getting to class in time, the expected learning outcome (information retention) might not be affected. Yet, one cannot say gamification did not work if such rewards successfully motivated students to get to class in time. Gamification worked, but it was not properly designed in light of the target learning outcomes. Thereby, gamification will only affect learning outcomes by firstly influencing attitudes and behaviors, which must be aligned to the expected outcomes.

6.2 Applications In [6], the authors present an experimental study that supports the ToGL based on its mediation process. Figure 6 demonstrates the ToGL model for this study, indicating the implementation of each element. In summary, [6] provides empirical evidence on gamification’s effect on exam scores mediated by self-testing. The authors conducted a controlled experiment comparing the effect of two game elements: points and badges, which would be used in isolation or combined. The experimental task concerned creating and answering multiple-choice questions on an online website, which generated the study behavior measure. Then, the final learning outcome was measures based on student exam scores. As a result, they found evidence supporting the ToGL’s fifth proposition: student behavior (i.e., self-testing) mediates the relationship between game elements (i.e., badges) and learning outcomes (i.e., exam scores).

Game Element: Badges

Behavior: Self-testing

Instructional Content: Quizzes Fig. 6 The ToGL model as implemented in [6]

Learning Outcome: Exam Score

Gamification and Motivation

Game Element: Competition

29

Attitude: Intrinsic Motivation

Instructional Content: Brainstorming

Learning Outcome: Performance

Fig. 7 The ToGL model as implemented in [1]

In another experimental study [1], the authors present support for the ToGL based on its moderation process. Figure 7 demonstrates the ToGL model for this is study along with indications of how each component was implemented. In summary, [1] provides empirical evidence concerning gamification’s role on user perception regarding the instructional content. The authors conducted an experimental study wherein participants would either interact with standard instructional content or a gamified version of the same instruction. In this case, they used fictional game elements to transform the instructional content into a story. Based on self-reported satisfaction, the authors found gamification enhanced participants’ attitudes towards the content. However, they also found participants of the gamified version performed worst than those of the standard one in terms of procedural knowledge. Thereby, this study provides empirical support for gamification’s potential to enhance instruction in terms of participant attitude and demonstrates the importance of behaviors/attitudes aligned to the target learning outcome.

6.3 Summary The ToGL is one of the first models specifically thought to model how gamification applied to education works. It assumes game elements either moderate instruction effectiveness or have its effect mediated by attitudes and behaviors. Experimental research has provided empirical evidence supporting this model, posing it as a viable alternative for gamification studies.

7 Gamification Science Despite the recognition from the literature [22], the Theory of Gamified Learning misses two key entities: the user and the context. Gamification research has long

30

P. T. Palomino et al.

Game elements (Predictors)

L1 Design-relevant moderators

L2

Psychological States (Mediators)

L4

L7

L3 L6

Design-irrelevant moderators

Distal outcome (Criteria)

L8 Behavior (Mediator)

L5

Fig. 8 Theoretical relationships between Gamification Science’s constructs. Adapted based on [15]

emphasized the importance of considering the user as well as the usage context in gamification research (see Chapter 4). However, the ToGL does not take these entities into account when modeling the process through which gamification works within the educational domain. Additionally, the Theory of Gamified Learning models behaviors and attitudes as a single component. In contrast, the SDT understands one’s motivation (which might be seen as an attitudes) as their driver for action/behavior. That context opened the possibility of extending the Theory of Gamified Learning, which led its author to introduce the Gamification Science Framework [15]. Gamification Science [15] addresses those gaps as follows. First, it acknowledges that psychological states (which is the term used instead of attitudes) and behavior influence the distal outcome (e.g., learning) through different mechanisms. Second, it assumes those effects are likely to be moderated by user and contextual factors. It specifically explores theories behind how gamification designs work, along with presenting a model to empirically validating those. Therefore, this research will rely on it. Next, we introduce how Gamification Science works, which is summarized in Fig. 8, describing it according to [15].

Gamification and Motivation

31

7.1 Dissecting Gamification Science Gamification Science was built upon four main constructs: predictors, mediators, moderators, and criteria. Predictors are the game elements, such Points, Cooperation or Storytelling [25]. Criteria is the distal outcome to be affected (e.g., students’ knowledge retention). Mediators are users’ psychological states (e.g., intrinsic motivation) and behaviors (e.g., completing quizzes). Lastly, moderators are independent factors that might either improve or decrease predictors’ or mediators’ effects, which are defined as design-relevant and design-irrelevant, respectively [15]. Examples of those are the user’s gender and instruction quality, respectively. The starting point of the relationships between those four constructs is the predictors. Predictors are designed to affect psychological states (Link 1; L1), and this effect is likely to be moderated by design-relevant moderators (L2). For instance, one might expect to improve users’ motivation (psychological state) to perform a task by giving them a badge (predictor) as a reward to do so. It might be that being acknowledged is less relevant for older users and, therefore, age moderates badge’s effects. The third link is that psychological states affect behaviors (L3). It has been discussed that intrinsic motivation is the best regulation to drive behaviors (e.g., [30]). Consequently, as one’s intrinsic motivation to perform a task increases, their behavior (e.g., spending more time-on-task) will likely change the most. Both psychological states (L4) and behaviors (L5) are expected to affect the criteria. Concretely, one might expect this to happen as, for example, students completing more quizzes (behavior), as well as those with higher motivation (psychological state), are likely to learn (criteria) the most. Lastly, there the moderator effects of design-irrelevant moderators (L6, L7, and L8) on L3, L4, and L5. To exemplify, this might happen for both L4 and L5 when students were motivated (L7) and completed quizzes (L8) but quizzes’ quality was poor, moderating how much they learned. One should note that, in practice, it is possible to find a relationship between the predictors and the behavior when, for instance, the behavior is being influenced by psychological states other than those measured. For example, one is measuring only intrinsic motivation but the behavior is being driven by external motivation.

7.2 Applications In [19], the authors present an experimental study that partially supports Gamification Science. Overall, they study the mediation role of intrinsic motivation on game elements’ effect on learning gains, as well as the moderator effect of contextual factors. Figure 9 demonstrates the research model based on Gamification Science’s connections.

32

P. T. Palomino et al.

Game elements: Badges, Objectives, Cooperation Design-relevant moderators: - Intervention duration - Familiarity with the subject Psychological State: Intrinsic Motivation Distal Outcome: Learning Gains

Behavior: Completing Quizzes

Fig. 9 Research model from [19] based on the Gamification Science Framework

Specifically, [19] reports a six-week experimental study with two factors: gamification (yes or no)—implemented with Badges, Objectives, and Cooperation— usage time (1–6; in weeks). The experimental task was completing out-of-class assignments of an introductory programming class, and the study measures captured student intrinsic motivation (psychological state), number of completed quizzes (behavior), and learning gains—the difference between pre and posttests (distal outcomes). Additionally, the authors captured participants familiarity with the subject (i.e., introductory programming) at the start of the intervention. Based on that, the authors tested five out of the eight connections proposed by Gamification Science. As a result, they found indication that intrinsic motivation mediated game elements’ effect on learning gains, that completing quizzes also affected the distal outcome, and that intervention duration and familiarity with the subject moderated gamification’s effect on intrinsic motivation. On the other hand, they found no support for intrinsic motivation’s effect on the completing quizzes behavior. Hence, the study provides partial empirical support Gamification Science Framework. Another study that provides partial support for Gamification Science is [23]. In this research, the authors present an experimental study on gamification’s effect on behavior and test scores, along with the moderator role of student grade. Figure 10 demonstrate the study research model according to that of Gamification Science. Similar to the other example, [23] reports a 16-week experimental study with gamification (yes or no) as the main factor. The experimental task also was completing quizzes. However, this study concerned students enrolled in an introductory psychology course. The study measures were the number of completed quizzes (behavior) and test scores (distal outcome). Behavioral measures were captured in three occasions: five, eight and 13 weeks after the term started. Accordingly, the authors analyzed how gamification’s effect changed over time based on the different time points in which students completed the tests. Additionally, they analyzed the

Gamification and Motivation

33

Game elements: Progress Bar, Wager option, Encouragement

Design-irrelevant moderators: Course Grade

Behavior: Quizzes Completed

Distal Outcome: Test Score

Fig. 10 Research model from [23] based on the Gamification Science

moderator role of students final grade on gamification’s effect. Unlike the other example, this study’s moderator analysis tested final grade as both design-relevant and irrelevant. That is, the possibility of moderating gamification’s direct effect as well as behavior’s effect on the distal outcome. Hence, they tested four out of the eight connections proposed by Gamification Science. As a result, they found empirical support for the effect of the of quizzes completed on test score. Also, they found support for the design-relevant moderation of final grade on gamification’s effect on test score. However, they found no support for the final grade designirrelevant moderation. An important note is that the relationship between quizzes completed and test score was stable throughout the three measures. In contrast, the relationship between gamification and quizzes completed was statistically significant only at the first measure. Thereby, these results similarly provide partial support for Gamification Science.

7.3 Summary The Gamification Science framework is an evolution of the widely researched Theory of Gamified Learning. Mainly, Gamification Science differs by disentangling how game elements affect psychological states (referred to as attitudes in the previous version) and behaviors differently, as well as acknowledging that there are both design-relevant and design-irrelevant moderators to be considered when

34

P. T. Palomino et al.

modeling how gamified learning works. Empirical studies have taken steps towards empirically validating this evolved model. However, in a smaller extent if compared to research on the Theory of Gamified Learning, probably due to its increase complexity, which complicates planning and conducting such experimental studies.

8 Concluding Remarks In this chapter we presented motivational theories related to gamification studies. These theories were selected based on existing literature reviews and based on their relevancy to the field. Through this chapter, we can provide an explanation and examples on how these theories were covered in the literature, so other practitioners can understand and choose the best one to suit their practice.

References 1. Armstrong, M.B., Landers, R.N.: An evaluation of gamified training: using narrative to improve reactions and learning. Simul. Gaming 48(4), 513–538 (2017) 2. Bai, S., Hew, K.F., Huang, B.: Does gamification improve student learning outcome? Evidence from a meta-analysis and synthesis of qualitative data in educational contexts. Educ. Res. Rev. 30, 100322 (2020) 3. Bloom, B.S.: Taxonomy of Educational Objectives: The Classification of Educational Goals. Longman, New York (1956) 4. Boruchovitch, E.: A motivação do aluno: contribuição da psicologia contemporânea. Vozes (2001) 5. Csikszentmihalyi, M.: Toward a psychology of optimal experience. In: Flow and the Foundations of Positive Psychology, pp. 209–226. Springer (2014) 6. Denny, P., McDonald, F., Empson, R., Kelly, P., Petersen, A.: Empirical support for a causal relationship between gamification and learning outcomes. In: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, pp. 1–13 (2018) 7. Huang, B., Hew, K.F.: Implementing a theory-driven gamification model in higher education flipped courses: Effects on out-of-class activity completion and quality of artifacts. Comput. Educ. 125, 254–272 (2018) 8. Huang, B., Hew, K.F., Lo, C.K.: Investigating the effects of gamification-enhanced flipped learning on undergraduate students’ behavioral and cognitive engagement. Interactive Learn. Environ. 27(8), 1106–1126 (2019) 9. Jagušt, T., Botiˇcki, I., So, H.J.: Examining competitive, collaborative and adaptive gamification in young learners’ math learning. Comput. Educ. 125, 444–457 (2018) 10. Kapp, K.M.: The Gamification of Learning and Instruction. Wiley San Francisco (2012) 11. Lamprinou, D., Paraskeva, F.: Gamification design framework based on SDT for student motivation. In: 2015 International Conference on Interactive Mobile Communication Technologies and Learning (IMCL), pp. 406–410. IEEE (2015) 12. Landers, R.N.: Developing a theory of gamified learning: linking serious games and gamification of learning. Simul. Gaming 45(6), 752–768 (2014) 13. Landers, R.N., Armstrong, M.B.: Enhancing instructional outcomes with gamification: an empirical test of the technology-enhanced training effectiveness model. Comput. Hum. Behav. 71, 499–507 (2017)

Gamification and Motivation

35

14. Landers, R.N., Bauer, K.N., Callan, R.C.: Gamification of task performance with leaderboards: a goal setting experiment. Comput. Hum. Behav. 71, 508–515 (2017) 15. Landers, R.N., Auer, E.M., Collmus, A.B., Armstrong, M.B.: Gamification science, its history and future: definitions and a research agenda. Simul. Gaming 49(3), 315–337 (2018) 16. Locke, E.A., Latham, G.P.: Building a practically useful theory of goal setting and task motivation: a 35-year odyssey. Am. Psychol. 57(9), 705 (2002) 17. Locke, E.A., & Latham, G.P. (Eds.).: New Developments in Goal Setting and Task Performance (1st ed.). Routledge (2012). https://doi.org/10.4324/9780203082744 18. Marczewski, A.: Even Ninja Monkeys like to Play. Blurb Inc, London (2015) 19. Rodrigues, L., Pereira, F.D., Toda, A.M., Palomino, P.T., Pessoa, M., Carvalho, L.S.G., Fernandes, D., Oliveira, E.H., Cristea, A.I., Isotani, S.: Gamification suffers from the novelty effect but benefits from the familiarization effect: findings from a longitudinal study. Int. J. Educ. Technol. High. Educ. 19(1), 1–25 (2022) 20. Rowland, C.A.: The effect of testing versus restudy on retention: a meta-analytic review of the testing effect. Psychol. Bull. 140(6), 1432 (2014) 21. Ryan, R.M., Deci, E.L.: Self-determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness. Guilford Publications (2017) 22. Sailer, M., Homner, L.: The gamification of learning: a meta-analysis. Educ. Psychol. Rev. 32, 77–112 (2020). https://doi.org/10.1007/s10648-019-09498-w 23. Sanchez, D.R., Langer, M., Kaur, R.: Gamification in the classroom: examining the impact of gamified quizzes on student learning. Comput. Educ. 144, 103666 (2020). https:// doi.org/10.1016/j.compedu.2019.103666. http://www.sciencedirect.com/science/article/pii/ S0360131519302192 24. Sweller, J.: Cognitive load theory. In: Psychology of Learning and Motivation, vol. 55, pp. 37–76. Elsevier (2011) 25. Toda, A.M., Oliveira, W., Klock, A.C., Palomino, P.T., Pimenta, M., Gasparini, I., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.I.: A taxonomy of game elements for gamification in educational contexts: Proposal and evaluation. In: 2019 IEEE 19th International Conference on Advanced Learning Technologies (ICALT), vol. 2161-377X, pp. 84–88. IEEE (2019). https:// doi.org/10.1109/ICALT.2019.00028 26. Tondello, G.F., Premsukh, H., Nacke, L.E.: A theory of gamification principles through goalsetting theory. In: HICSS (2018) 27. Turan, Z., Avinc, Z., Kara, K., Goktas, Y.: Gamification and education: achievements, cognitive loads, and views of students. Int. J. Emerg. Technol. Learn. 11(7), 64–69 (2016) 28. Ullén, F., de Manzano, Ö., Almeida, R., Magnusson, P.K., Pedersen, N.L., Nakamura, J., Csíkszentmihályi, M., Madison, G.: Proneness for psychological flow in everyday life: associations with personality and intelligence. Personal. Individual Differences 52(2), 167– 172 (2012) 29. Van Merriënboer, J.J., Sweller, J.: Cognitive load theory in health professional education: design principles and strategies. Med. Educ. 44(1), 85–93 (2010) 30. van Roy, R., Zaman, B.: Why gamification fails in education and how to make it successful: introducing nine gamification heuristics based on self-determination theory. In: Serious Games and Edutainment Applications, pp. 485–509. Springer (2017) 31. Zainuddin, Z., Chu, S.K.W., Shujahat, M., Perera, C.J.: The impact of gamification on learning and instruction: a systematic review of empirical evidence. Educ. Res. Rev. 30, 100326 (2020)

Ethical Challenges in Gamified Education Research and Development: An Umbrella Review and Potential Directions Ana Carolina Tomé Klock, Brenda Salenave Santana, and Juho Hamari

1 Introduction Using game design elements to promote game-like experiences throughout many daily tasks—namely gamification—has garnered growing attention among scholars and practitioners over recent years [1]. Gamification is an emergent phenomenon in multiple domains, having a meaningful role in upholding many Sustainable Development Goals (SDG), such as good health and well-being (SDG3) [2], decent work and economic growth (SDG8) [3], and climate action (SDG13) [4]. Quality education (SDG4) is prominent for gamification among this variety of possibilities [5], since it offers a gameful way to engage and inspire students during the teachinglearning process. Consequently, there is a continuously increasing interest in both uses and implications of gamified education [6]. Nevertheless, despite contributing towards several improvements to the educational field, gamification also introduces adverse effects when not suitably applied [7]. For instance, while it aims to promote more appealing and rewarding learning, gamification affects diverse students in distinct ways [8]. Such individual differences raise questions on how scholars and practitioners should consider and handle data regarding personal characteristics and preferences—questions that are ethical by nature. Thus, understanding and promoting ways to ethically research and develop gamified applications in all fields, but especially in education, is an essential question yet to be addressed.

A. C. T. Klock () · J. Hamari Gamification Group, Tampere University, Tampere, Finland e-mail: [email protected]; [email protected] B. S. Santana Federal University of Rio Grande do Sul, Porto Alegre, Brazil e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_3

37

38

A. C. T. Klock et al.

Towards an answer to such a question, this chapter investigates and discusses ethical challenges of gamification research methodology and application development in teaching-learning processes. This study is organised as follows: Sect. 2 provides an overview of normative ethics and its philosophical theories, as well as their relation to gamification research and development, and further decisionmaking processes regarding planning, conducting and communicating gamification outcomes. Section 3 describes the methodology, research questions, inclusion criteria, search process, screening procedure and data extraction plan. Section 4 details bibliometric information of the secondary studies, while elaborating on how to make ethical gamification and how to make gamification ethical. Finally, Sect. 5 concludes this chapter by presenting the final remarks and limitations.

2 Background Ethics is an extensive philosophy branch that analyses and conceptualises moral behaviours and determines right from wrong through multiple perspectives (e.g., meta-ethics, normative ethics, and applied ethics) [9]. From these sub-branches, this work focuses on the normative ethics perspective by seeking to establish standards of conduct in a practical manner to promote better gamification research and development for educational settings. Normative ethics is a broad term that describes the moral reasoning (i.e., norms of conduct of what is acceptable or not) from multiple philosophical theories (e.g., consequentialism, deontological, and virtue ethics) and decision-making processes (e.g., planning, conducting and communicating) [10]. Briefly describing some of these philosophical theories, the consequentialism perspective follows a utilitarian rationale, prioritising societal over individual interests (i.e., favouring actions that benefit the majority of people), regardless of potential harms. In opposition, the deontological perspective follows standards based on universal moral principles and duties to others (e.g., what would happen if everyone adhered to this standard?), disregarding how these codes vary according to their context and their unexpected results. At last, virtue ethics perspective follows behaviours towards living a virtuous life by practising good traits (e.g., honesty, integrity) and emphasising the interdependency of human beings, which creates a moral obligation to care for dependent groups (e.g., children, older people) and to use emotional virtues (e.g., sensitivity, responsiveness) when interacting with people. Gamification scholars and practitioners usually focus on enhancing teaching-learning processes with motivational affordances to invoke psychological and behavioural outcomes, in which game elements are justified by their utility towards the so-called common good (i.e., consequentialism) [1]. At the same time, multiple studies discuss the so-called right way to design gamification, following a set of rules defined in a framework (i.e., deontology) [11]. Relatively newer in the gamification field, the virtue ethics viewpoint invites scholars and practitioners to move from coercion to facilitating the best life and from instrumental perfection

Ethical Challenges in Gamified Education Research and Development

39

to critical transformation (i.e., “a critical, transformative, socio-technical systems design practice for motivational affordances in the service of human flourishing”— eudaemonist virtue [12]). In this sense, scholars and practitioners must ensure that gamification in educational settings focuses on students’ flourishing, such as by providing a fulfilling and meaningful gameful learning experience, through a well-thought decision-making process throughout its research and development. For instance, in the planning phase, those researching or developing gamification for educational domains must evaluate their competence in terms of skills and expertise (or collaborating with those with the necessary complementary abilities), as well as be familiar with relevant ethical guidelines for technology-assisted education through the lenses of cultural relativism and applicable legislation [13]. As another example, the conduction phase should follow principles of fairness, accountability, transparency and ethics (FATE) to avoid gamification designs that are exploitative or addictive [14], while ensuring that data is findable, accessible, interoperable and reusable (FAIR) [15]. As gamification scholars and practitioners, communicating the results of implementing such technology in education should be clear and comprehensive and respect students’ privacy, whose data should be anonymised since the earlier stages of the gamified project [16]. Yet, a better understanding of the unethical issues in gamified education and a comprehensive set of guidelines to address them is still needed towards making ethical gamification development and making gamification research ethical.

3 Methodology An umbrella review aims to summarise evidence from multiple research syntheses to provide an overall picture of findings for particular questions or phenomena [17]. According to this goal, the umbrella review might also include studies regarding different conditions and populations [18]. Furthermore, this methodology comprises a protocol detailing investigated research questions, inclusion criteria, search process, screening procedure and data extraction plan [18]. Regarding this study, despite the existing multiple secondary studies on the intersection of gamification and ethics, there is still no comprehensive understanding of how to address and overcome unethical issues in the research and development of gamification in education. Towards this goal, this chapter describes an umbrella review that complies with the following:

3.1 Research Questions Towards understanding and finding ways to mitigate the ethical challenges of gamification when researching and developing such technologies, this chapter

40

A. C. T. Klock et al.

focuses on two main research questions: 1. How to make ethical gamification? addressing how to design, implement and evaluate the effects of gamified educational applications towards living a virtuous life 2. How to make gamification ethical? addressing how to plan, conduct and communicate the outcomes of gamification in educational settings

3.2 Inclusion Criteria Based on these research questions for a high-quality and assertive outcome, this umbrella review adopted the following inclusion criteria: • Language: Studies need to be written in English; • Venue: Studies need to be published as Journal articles, Conference papers or Book chapters; • Methodology: Studies need to conduct a secondary study; • Intervention: Studies need to investigate gamification research, design or implementation; and • Outcome: Studies need to tackle any ethical issues emerging from gamification.

3.3 Search Process The search string was set following the PICOC method, which defines the Population, Intervention, Comparison, Outcome and Context of the desired studies [19]. In this chapter, the Population includes any secondary study, using gamification as the Intervention, and focusing on ethics as the main Outcome. Based on the research questions, there is no Comparison to be made among the studies. No limitations were defined based on the Context of these studies, as investigating ethical aspects of gamification in a broader sense also contributes to a deeper understanding and anticipation of potential issues towards their mitigation in the educational domain, especially given that it also employs, reuses and benefits from overall research methodology and development. Therefore, the search was conducted on Scopus, which indexes many of the literature databases available, and considered studies that meet gamification AND ethic* AND (review OR systematic) in their title, abstract, or keywords. The search was conducted in September 2022 and returned a total of 34 works.

Ethical Challenges in Gamified Education Research and Development

41

3.4 Screening Procedure Based on the selection criteria (described in Sect. 3.2), the authors excluded studies based on their language (.n = 1), venue (.n = 7), methodology (.n = 6), intervention (.n = 7), and outcome (.n = 8). Accordingly, a total of five secondary studies on the ethical issues and potential harms of gamification were included in this umbrella review [20–24].

3.5 Data Extraction Plan Following the guidelines proposed by Aromataris et al. [17], this umbrella review extracted bibliometric information (i.e., citation details, objectives of the included reviews, review type, the context of the application, number of databases searched, publication range, number of studies included, and country of origin), and outcomes and implications reported that are related to the aforementioned research questions.

4 Results In terms of bibliometric information, Arora and Razavian [20] conducted a systematic review to understand the ethical issues in the existing empirical work on the effects of gamification in health tracking. For this, the authors analysed 23 studies ranging between 2012 and 2021 from six different search engines (i.e., ACM Digital Library, IEEE Xplore, PhilPapers, PubMed, Scopus, and Web of Science). The second study, from Benner, Schöbel and Janson [21], conducted a systematic review to understand the current ethical considerations in persuasive system design. The authors included 17 studies published between 2011 and 2020 from eight search engines (i.e., ACM Digital Library, AISel, Emerald, IEEE Xplore, JSTOR, PubMed, ScienceDirect and SpringerLink). Next, Hassan and Hamari [22] conducted a systematic review to summarise what has been carried out on gamified e-participation. In this study, a total of 66 papers from Scopus were included, which the publication range is between 2012 and 2018. The fourth study, from Humlung and Haddara [23], was a systematic review of how to apply gamification in business as a means to create an innovative environment. While searching in Google Scholar, 19 studies published between 2013 and 2019 were analysed. The last included study, from Hyrynsalmi, Smed and Kimppa [24], conducted a systematic review to understand the perceived negative side effects of applying gamification in a more general context. A total of 22 studies published between 2013 and 2016 were included in this study based on six search engines (i.e., ACM digital library, AISel, IEEE Xplore, ScienceDirect and Wiley Online Library). All these five systematic reviews were written by authors affiliated with European Universities (i.e., Finland

42

A. C. T. Klock et al.

[22, 24], The Netherlands [20], Germany and Switzerland [21], and Norway [23]). Details on their outcomes and implications are detailed below while addressing our research questions.

4.1 How to Make Ethical Gamification? (RQ1) Towards achieving such critical and transformative gamification as aimed by virtue ethics, the current means to design, implement, and evaluate it must be rethought. In this sense, the ethical challenges found in the included studies were: Power Dynamics and Paternalism Shaping or reinforcing behaviours through persuasive technologies, such as gamification, without the proper communication of the intentions behind gamification aligns with the consequentialism rationale, in which the benefits for the general audience are perceived to outweigh the potential harms [10, 23]. More than a lack of communication, gamification has paternalistic characteristics that limit autonomy and freedom of choice by positioning scholars and practitioners as authorities of the “correct” behaviour from the deontology rationale, while fostering the stigmatisation of those who are not able to meet these desired behaviours from gamification goals [20, 21]. Thus, despite being generally used to foster “good” habits, gamification design, implementation and evaluation usually focus on top-down approaches that patronise individuals instead of promoting autonomous and voluntary engagement [22]. While gamification scholars and practitioners are all susceptible to their own implicit biases, consequentialism may take over again when the interest of a third party (e.g., companies, schools) contradicts and overcomes the original intention of helping people to achieve their own goals (e.g., dissonance, imbalance, conflict of interest), which now instead focuses on exploiting their real-world vulnerabilities [20, 21]. To address these challenges, any (digital) nudge should be disclosed to preserve people’ autonomy and freedom, despite potential undesirable outcomes, given that those aware of this persuasion might react differently [21]. On top of that, paternalistic characteristics can be avoided by educating people on nudging, so that they would be aware of potential issues by themselves [24]. Furthermore, gamification scholars and practitioners should include diverse stakeholders to account for multiple user voices during gamification design, implementation and evaluation processes [22], while ensuring that its persuasive effects are not misused to exploit people physically, financially, emotionally, or psychologically [21]. In educational settings, gamification could also benefit from being based and aligned with transformative learning theories that allow a non-hierarchical dialogue among students and educators, so that individual needs are considered and learning becomes a more autonomous process where knowledge is promoted as collective construct [25]. Lack of Voluntarity Following power dynamics from gamification being unfair to one party, another ethical issue relates to people feeling obliged to use gamified systems. For instance, gamification might be deeply rooted in educational settings

Ethical Challenges in Gamified Education Research and Development

43

as an efficiency metric, while not being translated as beneficial for the students [20, 21]. Gamification not being entirely voluntary supports the power imbalance by intentionally or accidentally sugar-coating coercive practices (i.e., by questionable means) and the reinforcement of desirable outcomes (i.e., for questionable purposes) [22]. Examples of this can be seen, for instance, in a classroom where the educator gamifies a specific task that is more aligned with school needs (e.g., for external evaluation purposes) than individual learning meaningfulness, and the assessment of students is conditional on their participation, regardless if they have the means or desire to perform this task in a gamified way. To avoid such misconduct, gamification should include an opt-in design and proper anonymisation of people’s information [21]. Confidentiality Issues Another issue is the interaction between users and gamification providers. Mismanagement of the necessary communication between the parties can inflict fundamental ethical issues related to confidentiality. Anonymisation and providing information on what data and why they are collected while asking for explicit permission would allow information security and data privacy in gamified systems [20]. This would prevent dark patterns in the interaction design, such as cookies and consent default options favouring the gamification provider [21], intentionally luring people to share personal data through game elements, sharing or even selling personal data with third parties, and making people uncomfortable, anxious or any other psychological and emotional harm with their data being tracked or shared [20]. In this sense, special attention should also be given when designing, implementing and evaluating gamification that might not ensure students’ privacy—such as avatars, challenges and competitions, which recognise and record information on students’ characteristics, performance, and opponents [26]. Cognitive Manipulation Gamification can also be a means to inhibit autonomy and undermine self-reflection in unjustifiable ways (e.g., distraction, addiction). For instance, gamification requiring instant reaction in some works and job positions, such as medics and firefighters, add unnecessary steps and distractions that might cause dangers, remarkable losses, and threatening situations—overall physical and psychological harms [20]. At the same time, the potential moral, ethical and legal challenges of gamblification require further investigation [21], while the overall addiction might have detrimental effects on people, such as obsessively relying on game incentives or choosing goals that are potentially harmful (e.g., dieting based on what other people consider healthy) [20]. Thus, gamification should provide safe restrictions or warnings against cognitive effects, allow autonomy and personalisation, and focus on facilitating internal motivation (e.g., self-determination and self-reward) for a sustainable behaviour change [20]. Furthermore, scholars and practitioners need to consider the context and target audience, such as adding gamification in products and services marketed for children or considering their impact on people with a history or tendency towards addiction [24]. This, however, should not be a way to justify hyper-focus on specific target audiences, in which some people would have privilege over others, nor a means to reinforce stereotypes,

44

A. C. T. Klock et al.

in which different genders would have different colours or game elements in a reductionist way [20]. Thus, gamification should be deeply aligned with intended learning outcomes to ensure it is not distracting students from the educational content. At the same time, scholars and practitioners should understand the interaction of the students with the learning environment and its game elements to avoid reward dependency in the teaching-learning process [27]. Social Comparison By drawing inspiration from games, gamification also allows competition and rivalry, potentially giving a sense of social overload and straining people [21]. This might lead to a loss of motivation and a feeling of segregation for those who systematically perform worse than their counterparts and, furthermore, might lead to cheating. Thus, gamification scholars and practitioners should avoid giving people a sense of defeat while also transferring the responsibility to them by allowing cheating to some extent—with more autonomy in defining their own tasks, gamification supports a tolerant community of individual choice-making and acknowledges individual differences [20]. In educational settings, while considering that students have different learning styles, a tailor-made gamification might be a good alternative to allow everyone to have fun and play the game even though the rules are not the same for everybody [28].

4.2 How to Make Gamification Ethical? (RQ2) More than rethinking what aspects of gamified educational applications can be designed, implemented and evaluated in an unethical manner (RQ1), scholars and practitioners should follow ethical principles to promote ethical research and development of gamification (RQ2). Thus, gamification research and development are indivisible from ethics, as scholars’ and practitioners’ choices during the planning, conduction and communication have inherently ethical aspects [10]. For instance, whenever planning gamification research and development in educational settings, scholars and practitioners must consider which values and interests are promoted by the research questions and by the purpose of gamification in the educational application. Given supervisors’ individual interests and business’ own agendas, gamification investigation and execution might be loaded with contradictory assumptions (i.e., conflict of interest) [10], which very much relate with power dynamics and paternalism discussed in RQ1. Thus, scholars and practitioners must reflect on the gamification implications in educational settings before starting the project to ensure the research and development integrity [29], especially regarding sampling and data collection methods through ethical lenses. Since the gamified education target audience tend to be quite broad and focuses on human beings, scholars and practitioners should clearly define whom the subjects are (i.e., who is included, but especially who is excluded and why), while avoiding presenting people in stereotyped ways (e.g., tailored gamification that generalises preferences based on a single characteristic). More than understanding

Ethical Challenges in Gamified Education Research and Development

45

ethical principles when involving people in research and development, scholars and practitioners should provide informed consent that ensures transparency and voluntarity, clearly communicates potential risks and benefits of the participation, guarantees confidentiality, privacy and anonymisation, prevents social comparison, as well as affords special protections against addiction, distraction and further needs for targeted populations (e.g., children, minority and elderly) [29]. Regarding conducting gamification research and development in educational settings, gamification research and development should ensure integrity and quality regardless of the chosen methodologies. For instance, involving participants from co-participatory approaches beyond the design phase allows a further co-production that promotes the inclusion of participants and their social worlds in the analysis (e.g., giving them a voice to agree or disagree with scholars and practitioners’ understanding of their data, as opposing to power dynamics and paternalism) and dissemination of the results (e.g., giving them credits for the co-production, while guaranteeing confidentiality) [10]. Overall, reliability and validity are some examples of measurements that ensure research quality in quantitative methods, while transparency and data triangulation are examples of means to ensure quality when applying qualitative methods [10]. Nevertheless, doing ethical gamification research and development to the highest possible standard is not only a need but also a moral obligation. Scholars and practitioners should commit to following upto-date ethic codes (e.g., APA Ethical Principles and Code of Conduct1 ) to promote fairness [15] and avoid misinterpretation or misrepresentation of their gamification outcomes in educational settings. Moreover, gamification outcomes need to be communicated to relevant audiences, such as academia, industry and to the general public. Before that happens, it is essential to have an upfront discussion about the authorship with scholars and practitioners involved in the research and development of gamification [10]. While communicating, misconduct in all of its forms should be prevented: scholars and practitioners must not alter data (i.e., falsification), nor publish data that were not actually collected (i.e., fabrication), and mostly not steal other’s ideas, methods or data without proper attribution (i.e., plagiarism) [30]. Finally, communication should be complete (i.e., avoid fragmentary publication) and comprehensive (i.e., with a sufficient description of methods, corrections, and retractions) [29].

5 Final Remarks This chapter addressed a series of ethical issues in gamification research and development, with a particular focus on the educational field. From the analysis of secondary studies, this umbrella review explored many ethical challenges

1 https://www.apa.org/ethics/code.

46

A. C. T. Klock et al.

of gamified educational applications and proposed potential solutions for future research and development. Towards designing, implementing and evaluating the effects of gamified educational applications towards living a virtuous life, we elaborated on how to make ethical gamification (RQ1). Based on major unethical outcomes reported by the secondary studies, we propose that scholars and practitioners should ensure that digital nudging is properly disclosed to preserve people’s autonomy and freedom, while educating people on nudging to raise awareness on potential issues from one’s perspective. Gamification research and development should also involve diverse stakeholders to account for multiple user voices to further avoid power dynamics and paternalism. As a mean to avoid coercive practices in face of the lack of voluntarity from people in using gamification, gamification should include an optin design and proper anonymisation of user’s information. Data anonymisation is also a good strategy to prevent confidentiality issues, while providing information on what data and why they are collected when asking for explicit permission from the students. Gamification scholars and practitioners need also to be careful when designing and implementing some game elements (e.g., avatars, challenges and competitions) that might not ensure students’ privacy. Gamification should provide safe restrictions or warnings against cognitive manipulation, allowing autonomy and personalisation, and focusing on facilitating internal motivation. Towards preventing distractions and addiction, gamification needs to be aligned with the intended learning outcomes and understand students’ interaction with the learning environment to avoid reward dependency. The autonomy is also important to stop social comparison, by allowing students to define their own learning path or even promoting an autonomous tailoring system that understand students’ individual preferences and needs. Towards planning, conducting and communicating the outcomes of gamification in educational settings, we elaborated on how to make gamification ethical (RQ2), especially regarding conflict of interests, sampling and data collection methods, ensuring integrity and quality regardless of the chosen methodologies, misinterpretation and misrepresentation, authorship agreement, and misconduct through falsification, fabrication, and plagiarism. However, this study has some limitations. As with any umbrella review, it was not possible to ensure the quality assessment of the included primary works, which may vary from one secondary study to another. As with any systematic study, our work and the secondary ones may have issues with the definition of the string, search engines and selection criteria, which may not have retrieved relevant papers during the search. Finally, since the screening of the study was conducted by one senior researcher in systematic reviews and gamification, but yet only one, we may have some false negative papers along the way. Because of that, our results could be slightly different from similar analysis approaches. Still, the ethical challenges in gamified educational research and development, as well as potential directions for future works, might be useful for scholars and practitioners as a first step towards promoting critical transformation of gamified educational applications and making them a tool to facilitate the best life.

Ethical Challenges in Gamified Education Research and Development

47

Acknowledgments This work was supported by the Academy of Finland Flagship Programme [grant No 337653, Forest-Human-Machine Interplay (UNITE)] and the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie [grant No 101029543, GamInclusive].

References 1. Hamari, J., Koivisto, J., Sarsa, H.: Does gamification work? – a literature review of empirical studies on gamification. In: Proceedings of the 47th Hawaii International Conference on System Sciences, pp. 3025–3034 (2014) 2. Johnson, D., Deterding, S., Kuhn, K.A., Staneva, A., Stoyanov, S., Hides, L.: Gamification for health and well-being: a systematic review of the literature. Internet Interv. 6, 89–106 (2016) 3. Warmelink, H., Koivisto, J., Mayer, I., Vesa, M., Hamari, J.: Gamification of production and logistics operations: status quo and future directions. J. Bus. Res. 106, 331–340 (2020) 4. Galeote, D.F., Rajanen, M., Rajanen, D., Legaki, N.Z., Langley, D.J., Hamari, J.: Gamification for climate change engagement: review of corpus and future agenda. Environ. Res. Lett. 16(6), 063004 (2021) 5. Koivisto, J., Hamari, J.: The rise of motivational information systems: a review of gamification research. Int. J. Inf. Manage. 45, 191–210 (2019) 6. Sousa Borges, S., Durelli, V., Reis, H., Isotani, S.: A systematic mapping on gamification applied to education. In: Proceedings of the 29th Annual ACM Symposium on Applied Computing, pp. 216–222 (2014) 7. Toda, A.M., Valle, P.H., Isotani, S.: The dark side of gamification: an overview of negative effects of gamification in education. In: Researcher Links Workshop: Higher Education for All, pp. 143–156. Springer, Berlin (2017) 8. Smiderle, R., Rigo, S., Marques, L., Miranda Coelho, J., Jaques, P.: The impact of gamification on students’ learning, engagement and behavior based on their personality traits. Smart Learn. Environ. 7, 3 (2020) 9. Smith, C., Babich, C.: Theories of Individual and Collective Learning. eCampusOntario (2021) 10. Farrimond, H.: Doing Ethical Research. Bloomsbury Publishing, London (2012) 11. Mora, A., Riera, D., González, C., Arnedo-Moreno, J.: Gamification: a systematic review of design frameworks. J. Comput. High. Educ. 29(3), 516–548 (2017) 12. Deterding, S.: Eudaimonic design, or: six invitations to rethink gamification. In: Rethinking Gamification, pp. 305–331. Meson Press, Lüneburg (2014) 13. Aguinis, H., Henle, C.A.: Ethics in research. In: Handbook of Research Methods in Industrial and Organizational Psychology, pp. 34–56. Blackwell Publishing, Hoboken (2004) 14. Kim, T.W.: Gamification ethics: exploitation and manipulation. In: Proceedings of ACM SIGCHI Gamifying Research Workshop, pp. 1–5 (2015) 15. Wilkinson, M., Dumontier, M., Aalbersberg, I., Appleton, G., Axton, M., Baak, A., Blomberg, N., Boiten, J., Silva Santos, L., Bourne, P., Bouwman, J., Brookes, A., Clark, T., Crosas, M., Dillo, I., Dumon, O., Edmunds, S., Evelo, C., Finkers, R., Gonzalez-Beltran, A., Gray, A., Groth, P., Goble, C., Grethe, J., Heringa, J., Hoen, P., Hooft, R., Kuhn, T., Kok, R., Kok, J., Lusher, S., Martone, M., Mons, A., Packer, A., Persson, B., Rocca-Serra, P., Roos, M., Schaik, R., Sansone, S., Schultes, E., Sengstag, T., Slater, T., Strawn, G., Swertz, M., Thompson, M., Lei, J., Mulligen, E., Velterop, J., Waagmeester, A., Wittenburg, P., Wolstencroft, K., Zhao, J., Mons, B.: The FAIR Guiding Principles for scientific data management and stewardship. Sci. Data 3, 160018 (2016) 16. Goswami, P., Madan, S.: Privacy-preserving data publishing and data anonymization approaches: a review. In: Proceedings of the 2017 International Conference on Computing, Communication and Automation, pp. 139–142. IEEE (2017)

48

A. C. T. Klock et al.

17. Aromataris, E., Fernandez, R., Godfrey, C.M., Holly, C., Khalil, H., Tungpunkom, P.: Summarizing systematic reviews: methodological development, conduct and reporting of an umbrella review approach. JBI Evid. Implement. 13(3), 132–140 (2015) 18. Becker, L., Oxman, A.: Overviews of reviews. In: Cochrane Handbook for Systematic Reviews of Interventions. The Cochrane Collaboration, London (2011) 19. Wohlin, C., Runeson, P., Höst, M., Ohlsson, M.C., Regnell, B., Wesslén, A.: Systematic literature reviews. In: Experimentation in Software Engineering, pp. 45–54. Springer, Berlin (2012) 20. Arora, C., Razavian, M.: Ethics of gamification in health and fitness-tracking. Int. J. Environ. Res. Public Health 18(21), 11052 (2021) 21. Benner, D., Schöbel, S., Janson, A.: It is only for your own good, or is it? Ethical considerations for designing ethically conscious persuasive information systems. In: Americas Conference on Information Systems (2021) 22. Hassan, L., Hamari, J.: Gameful civic engagement: a review of the literature on gamification of e-participation. Gov. Inf. Q. 37(3), 101461 (2020) 23. Humlung, O., Haddara, M.: The hero’s journey to innovation: gamification in enterprise systems. Procedia Comput. Sci. 164, 86–95 (2019) 24. Hyrynsalmi, S., Smed, J., Kimppa, K.: The dark side of gamification: how we should stop worrying and study also the negative impacts of bringing game design elements to everywhere. In: GamiFIN, pp. 96–104 (2017) 25. Freire, P.: Pedagogy of the Oppressed. Continuum, New York (2000) 26. Mavroeidi, A.G., Kitsiou, A., Kalloniatis, C.: The role of gamification in privacy protection and user engagement. In: Security and Privacy From a Legal, Ethical, and Technical Perspective, IntechOpen (2020) 27. Andrade, F.R., Mizoguchi, R., Isotani, S.: The bright and dark sides of gamification. In: Proceedings of the 13rd International Conference on Intelligent Tutoring Systems, pp. 176– 186. Springer, Berlin (2016) 28. Klock, A.C.T., Gasparini, I., Pimenta, M.S., Hamari, J.: Tailored gamification: a review of literature. Int. J. Hum. Comput. Stud. 144, 102495 (2020) 29. Pimple, K.D.: Six domains of research ethics. Sci. Eng. Ethics 8(2), 191–205 (2002) 30. Zigmond, M.J., Fischer, B.A.: Beyond fabrication and plagiarism: the little murders of everyday science. Sci. Eng. Ethics 8(2), 229–234 (2002)

Theories Around Gamification in Education Paula T. Palomino

1 Introduction Gamification in Education is based on motivational, behavioural and pedagogical theories, as saw in chapter “Gamification and Motivation”. However, for gamified strategies to be successfully applied, they must be: (i) grounded by solid theories and models and (ii) systematised by frameworks or guidelines, to facilitate their application in classrooms or gamified educational systems. But after all, what is the difference between theory, model, framework and guideline? The theory is characterised as a generalised statement of abstractions, ideas, concepts and definitions that affirm, seek to explain or predict a certain meaning of relations, nature or connections between phenomena within a specific context, considering the limits of critical assumptions. A model already involves a deliberate simplification of a phenomenon or specific aspect of a phenomenon and is often confused with theory. They need not be accurate representations of reality to be of value and can be described as theories with a narrower scope. A model is descriptive, whereas a theory can be both explanatory and descriptive [29]. Frameworks, on the other hand, denote a consistent structure, system or plan of various categories, concepts, constructs or variables, and the relationships between them that are presumed to refer to a given phenomenon. A framework will represent and describe ways to apply or represent something. The guidelines, on the other hand, are a simplification of frameworks and constitute a step by step to be followed in order to reach a certain objective [29]. In this sense, there are several theories, from Human Computer Interaction, User Experience and Narrative that can be used when designing and thinking about

P. T. Palomino () Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_4

49

50

P. T. Palomino

Gamification Strategies. The next section will cover some of them and how they can relate to gamification in education.

2 Gamification Theories for Education Meta-analysis of the effect of gamification within the educational context show its effectiveness by demonstrating that gamification can affect different learning outcomes (i.e., behavioural and psychological), with its effect depending on several moderators [2, 40]. However, unlike in other areas where gamification is applied, it is necessary to consider the desired behaviour and learning gain in education. Thus, too much engagement in parallel with the educational content can hinder the learning process, even if it keeps the student immersed in the environment [20, 22]. There are cases where gamification may be associated with adverse effects [44], which further support the need for such balance. Many researchers assign these limitations to bad designs [24, 46], often arguing that using the same game elements and strategies to everyone (i.e., the one-size-fits-all—OSFA—approach). But how do we do better gamification design? First, we need to understand the context in which Gamification is usually applied. In short, if it is applied in a system, it will necessarily be related to Human-Computer Interaction (HCI) [4] and User Experience (UX) [13, 37]. If it is unplugged, it still can benefit from engagement and motivation theories, such as Flow Theory [8] and narrative and storytelling concepts [34]. Also, to grasp gamification as a whole, it is important to understand the gameful system behind it [23]. And finally, to consider the specificities of gamification for education, the understanding of popular instructional design frameworks such as ADDIE and educational taxonomies such as Bloom’s Taxonomy [21], and how they can be related to gamification might be of great use. Next, we present some of these theories and frameworks that research demonstrated to be good companions to gamification strategies.

2.1 Human-Computer Interaction and User Experience According to Barbosa [4], a system is the interface set plus the application itself, the interface being The entire portion of the system with which the user maintains physical (motor or perceptual) or conceptual contact during interaction. [...] The vast majority of users believe that it is the system. [...]. We can consider user-system interaction as being a process of manipulation, communication, conversation, exchange, influence, and so on. (Author’s translation from [4].)

Theories Around Gamification in Education

51

SYSTEM ACTION

INTERFACE

USER

APPLICATION

INTERPRETATION

Fig. 1 The process of human-computer interaction—adapted from [4]

Barbosa further defines that “a set of operations that can be performed with the interactive system, as well as [...] the ways to accomplish them by manipulating the interface elements” are called affordances. It is possible to visualise the basic scheme of the human-computer interaction process in Fig. 1. Much like games studies, HCI is a multidisciplinary science that encompasses several areas such as Computer Science, Cognitive Psychology, Organizational and Social Psychology, Ergonomics and Human Factors, Engineering, Design, Anthropology, Sociology, Philosophy, Linguistics and Intelligence Artificial. The set of all these areas aims to systematically support several steps and procedures of the Human-Computer Interaction methodologies since this is an area that works with aspects of science, technology, engineering, and mathematics (STEM) as with the physical, psychological and behavioural aspects, also considering the social context in which a particular system is inserted for its users [37]. To measure the quality of a particular software, including digital games, HCI has some criteria and goals that can be analyzed, cataloged and used to validate the resulting interface and interaction, such as Usability and User Experience, Accessibility and Communicability [4]. Usability is related to the ease of learning, use of the interface and user satisfaction due to this use [28]. In this sense, this research is highly based on the hedonic qualities of usability (novelty, stimulation, and attractiveness) [13]. User Experience, commonly known by the acronym UX, measures the quality related to the feelings and emotions of the users during the interaction [37]. The user experience field also studies the practical, experiential, affective, meaningful and valuable aspects of Human-Computer Interaction (HCI) [30] and how it can influence the system’s design. Accessibility is related to the removal of barriers that prevent more users from being able to access and interact with the system interface and can be defined as the flexibility provided for access to information and interaction so that users with different needs can access and use these systems [4]. Finally, communicability deals with the designer’s responsibility in communicating to the user their design intentions and the logic that governs the behaviour of the interface.

52

P. T. Palomino An interactive system is the result of a design process in which a designer establishes a vision (interpretation) about the users, their objectives, the domain and the context of use and makes decisions about how to support them. For the user to enjoy better computational support, it is desirable for the designer to remove interface barriers that prevent the user from interacting (accessibility), making it user-friendly, and communicating to the user their conceptions and intentions when designing the system. [4].

This process can be better represented by relating the designer’s mental models (design model, which is the image of the system from the perspective of the designer, that is, as the designer imagines the system to be), the user (system image from the perspective of the user, that is, as the user imagines that the system will attend) and the system itself (the image of the system, which constitutes the physical materialisation of the two mental models) [30]. Ideally, the System Image must be identical to the design model and user model, as shown in Fig. 2 below. It is desirable that the system’s image faithfully convey the experience it was intended to offer. To achieve this goal, one must explore how the user perceives and interacts with an application and the satisfaction and emotions brought about through this interaction. In this sense, the concepts of gamification become highly relevant to the user experience itself [15] and should be considered in gamification for education strategies.

2.2 Gameful Experiences, Gameful Systems, Gameful Design Based on the recent studies by Hodent [14], the part of UX that is most relevant for a user to want to use a system and to continue using it (i.e., its retention rate) is the “engagement capacity,” or “game flow,” a well-known term that, despite

Fig. 2 Three aspects of mental models [30]

Theories Around Gamification in Education

53

Fig. 3 Flow Theory [8]

representing some aspects of engagement, does not encompass everything essential to engage these users (or players) and keep them in that state. For the author, the term expresses the ability of an application or game to engage its user. The variables behind usability and engagement ability are not working in silos; they interact with each other [14]. She also describes the capacity for engagement divided into three aspects: motivation, which is the origin of all human behaviour, the emotions that serve the motivation in the sense that they help to choose the correct behaviour (for example, to escape when one is afraid) and the game flow, a state of deep focus and immersion, which is a concept derived from Theory of the Flow [8], shown in Fig. 3, and which functions as a connection between all other aspects. According to Fig. 3, depending on the balancing of the task that is being performed, the user enters the ’flow,’ where their productivity and engagement are at an ideal point. When the difficulty to perform a task is superior to the user’s abilities, the user may feel frustrated. Likewise, when their skills are superior to those required for the task, they may feel bored [8]. Also, according to Pink’s motivational theory [36], the motivation of the individual is influenced by three predominant factors: Autonomy, Mastery and Purpose, the first being the control of the individual about their actions, the second dealing with their abilities that are improved, and the latter is represented by the intrinsic need of the individual to perform a particular action. These three pillars can be observed in digital games since the player is able to control their actions (autonomy), hone their skills (mastery) and do it voluntarily (purpose). Recent research sought to refine these concepts to help design these game-like systems. According to [23], gamefulness is commonly cited as the primary goal of gamification, encompassing engagement, motivation and behaviour changes; however, the term is not well defined across the literature, hampering the definition of what to expect from gamification strategies. Therefore, the authors presented

54

P. T. Palomino

a theory that divided gamefulness into three different constructs: (i) gameful experience, a psychological state resulting from three characteristics: having nontrivial and attainable goals to pursue; being motivated to pursue them according to a set of rules and being willing to accept those rules because they make such activity possible; (ii) gameful system, as any system that provides for its users a gameful experience and (iii) gameful design as a process to design systems with characteristics that can provide their users’ gameful experiences. This process, according to [43], consists of a series of events that results in a gamified strategy, that is, a specific action related to game elements. The success of a gamification strategy can be hindered by the absence or inadequate implementation of a gameful design. For example, an inadequate gamification strategy in education can lead to the loss of motivation and engagement to the impairment of the learning process [10, 44]. However, according to the literature, frameworks can support the design process, ensuring that all steps and elements of the gameful design are present [1]. The user is considered a key agent in the process, and their experience is paramount.

2.3 Gamification Frameworks As stated before frameworks can be defined as a set of steps and tools aimed at specific results (in this case, a gameful design). According to recent research, the number of frameworks for gamification significantly increased in the last few years [26]. These frameworks can be categorised in structural frameworks (based on structural game elements such as ranking, points, badges and leaderboards); and content frameworks (based on subjective game elements applied directly to the content such as narrative, storytelling and sensation) [18]. Currently, one of the most accepted frameworks for the digital game and gamification design by academia and the market is Mechanics-Dynamics-Aesthetics, or MDA [17]. Mechanics are understood to be base components of the game—its rules, each main action that the player can take in the game, algorithms and data structures of the game engine; Dynamics is the run-time behaviour of mechanics that acts on the player’s input and cooperates with other mechanics, and aesthetics are the emotional responses evoked in the player. Figures 4 and 5 illustrate how the MDA formalises the consumption structure of a game and makes the parallel with its parts in the design. For the authors, the development of digital games is a “double-way,” where the player and the designer observe the artefact (digital game) from different

Fig. 4 Consumption structure of a game [17]

Theories Around Gamification in Education

55

Fig. 5 Parallel with its parts in design [17]

Fig. 6 Diagram explaining the perspective of player and designer observation [17]

perspectives, and it is possible to draw a parallel with the three mental models of Norman [30], as shown in Fig. 6. According to Hamari et al. [12], most of the frameworks for gamification developed are structural (i.e. aggregating scoring systems, ranking), with few content frameworks (where you apply elements and concepts of games in the content itself to be explained, to make it more like a game). This view is confirmed by the recent systematic review by Mora et al. [26], where 40 unique works that addressed the development of frameworks for gamification were studied and catalogued. However, most of these frameworks were developed based on MDA, and very few have a content approach. The majority of these gamification frameworks, however, are generalists. They can be used for education, but they are not concerned with learning specificities. However, a couple of education frameworks are being used, together with gamification frameworks to create gamification strategies for education, such as Bloom’s Taxonomy [21] and ADDIE’s Instructional Design Framework [27].

2.4 Learning Objectives and Learning Activities Types (LATs) Bloom’s original research, published in 1956, presented a framework to be used by teachers to support the instructional design of their classes [7]. In 2001 this framework was revised, focusing on a more dynamic iteration [21] and it is composed of the statement of a learning objective, where the verb (and the action associated with) refers to the cognitive process, and the object (usually a noun) refers to the knowledge expected the students to acquire. As such, the authors refer to two dimensions: the cognitive process one, categorised in six hierarchical stages (i.e., Remembering, Understanding, Applying, Analysing, Evaluating, Creating); and the Knowledge Dimension, categorised in factual, conceptual, procedural and meta-cognitive, as shown in the examples from Table 1.

Cognitive process dimension Remembering: Relevant knowledge from long-term memory Understanding: Construction of meaning through instructional messages Applying: Application of a procedure in a given situation Analyzing: Distinguish information between different parts Evaluating: Judging based on criteria and standards Creating: Join or organize elements in a new form, pattern or coherent structure

Conceptual: The relationships between the basic knowledge that allows them to make sense together Recognize

Classify

Provide

Differentiate

Determine

Assemble

Knowledge dimension Factual: The basis that the student must have acquired with a subject List

Summarize

Respond

Select

Select

Generate

Table 1 Revised Bloom’s taxonomy learning objectives example [21]

Design

Judge

Integrate

Carry out

Clarify

Procedural: How to apply knowledge, methods, skills and techniques Recall

Create

Reflect

Deconstruct

Use

Predict

Meta-cognitive: Knowledge in its broadest form, awareness of the existence of this knowledge Identify

56 P. T. Palomino

Theories Around Gamification in Education

57

Bloom’s taxonomy of learning objectives was already used in gamification, matching the learning activities gamification designs to a cognitive taxonomy [3] and to map which gamification design users consider the most suitable to help them in performing a particular learning activity [38].

2.5 ADDIE Instructional Design Framework Instructional frameworks provide a structure of components adaptable to work with different teaching styles, content areas and students’ needs. They are designed to provide a step-by-step for teachers to create their learning content with confidence and method. One of the most famous instructional design frameworks is ADDIE, which stands for Analyse, Design, Develop, Implement, and Evaluate [27] as seen in Fig. 7. This sequence, however, does not impose a strictly linear progression through the steps. Teachers, instructional designers, and training developers consider this a practical approach because having clearly defined stages facilitates effective learning content implementation.

Fig. 7 ADDIE model [27]

58

P. T. Palomino

But how to incorporate these two instructional design frameworks into gamification strategies for education. The secret is to add context, to make the student want to learn something because this something is related to other thing that makes sense to them. And how do we achieve that? Using Narrative and Storytelling game elements.

2.6 Narrative Concepts For Salen et al. [41], there are two distinct narrative lines in a game: the embedded narrative and the emergent narrative, the first containing the predefined story of the work, that is, the script, and consisting of two narrative axes (the first focusing exclusively on the logical chain of events portrayed in the story, comparable to the narrative found in literary works) and the axis of sensory experiences (which extends the previous axis, bringing all other sensory elements such as image and sound to the narrative, being comparable to the narratives of cinema). Figure 8 illustrates it. On the other hand, the emergent narrative, also called ludonarrative by other authors [6], has only one narrative axis, that of interactivity, dealing with narrative development directly related to the actions performed by the player throughout the game. It can be said then that the emergent narrative is the story that each player builds throughout the game, using the available gameplay features [39]. Figure 9 shows this relationship (Figs. 10 and 11). The narrative using these features (gameplay) is responsible for engaging and motivating the player, providing a decent and acceptable user experience that keeps

Fig. 8 Embedded narrative [41]

Fig. 9 Emergent narrative [41]

Theories Around Gamification in Education

59

Fig. 10 Narrative axis [41]

Fig. 11 Dimensions of the 5W2H framework [19]

them in the game. Figure 12 shows how game narrative incorporates these two types of narrative and their three axes. Although much is discussed about narrative in other media, this does not happen in gamification, where users of gamified environments may benefit far more by using this concept, specially in educational contexts. Next we present great Gamification for Education Frameworks, including one, the Gamification Journey, that is based on several of the theories presented in this section, and related Narrative and Storytelling to gameful and learning experiences.

60

P. T. Palomino

Fig. 12 Framework GAMIFY-SN [45]

3 Gamification Frameworks and Guidelines for Education In the last decade, several research tackled gamification frameworks specifically for education, some examples are: Klock’s et al. gamification conceptual model is a great example in engaging students in e-learning systems [19], where they deal with adaptive gamification that also influences in the user experience. They also consider the narrative game element, defining it mainly in its correlation with story, as in embedded narrative [41], that is a pregenerated narrative content that exists prior to a player’s interaction with the game, such as cut scenes and back story. However, they only addressed a part of the full concept of narrative. Marczewski’s “Gamification Periodic Table of Gamification Elements” also describes narrative as the element that deals with stories and plot lines, again including only one of the concepts related to the narrative definition [50]. Toda et al. created an approach for planning and deploying gamification concepts with social networks within educational contexts [45]. In this approach the storytelling is considered as an element option for teachers to choose when planning their solution. However, the description utilised, as in the context created by the designer [26] is too broad and confusing with the narrative concepts, which the paper also address as what occurs when the designer create a background to contextualise the user outside its reality [11], being examples of its uses: (i) Create a storyline to join multiple levels [5] and (ii) Create a storyline that allows the students to form groups and discuss with their peers their next actions [42]. As for the user experience, they define it as per Mora et al.’s definition, aka everything within the gamified practice that can be interacted by the participant [26].

Theories Around Gamification in Education

61

Later on Toda et al. adopted a data-driven approach to provide a set of game element recommendations, based on user preferences, that could be used by teachers and instructors to gamify learning activities. For that they applied a survey to 733 people and collected information regarding user preferences on game elements [49]. This survey however lacked the existence of narrative and storytelling game elements, leading to an updated research where a new survey was elaborated and first validated with specialists from the areas of gamification and game development and/or research [48]. This validation led to the inclusion of the two elements on the updated survey, so that the research could be re-ran. Other recent studies have proposed approaches (e.g., architectures, frameworks, methodologies) for tailor gamified educational systems or even serious games [25, 32, 33]. However, again, such studies propose the personalisation in terms of specific game elements (i.e., points, trophies, ranking, and others), but do not consider the narrative, leaving open such challenges. Demerval also proposed a process to aid teachers to develop gamified ITS. However his process is centred in the teacher/instructor and not the students. Also, he does not address the concept of narrative in his work [9]. To the best of our knowledge, there are no other studies that aim at providing gamification recommendations in educational scenarios based on narrative approaches and being user-centred. Finally, Palomino et al. [35] proposed a narrative gamification framework for education as seen in Fig. 13, considering both learning aspects and the students experience and freedom of choice in how to progress in what the authors called The Gamification Journey. The experience on this journey is represented by the user experience supported by Narrative and the learning experience supported by Storytelling. Therefore, the full Gamification Learning Journey uses a summarised adaptation of the four quadrants, as follows: 1. Call to Action: the first quadrant is related to the UX, and the students first contact with the system and their motivation to start learning. 2. Trials: in the second act, the student is in their special world and the start of their learning experience. The stages of Remembering, Understanding and Applying from Bloom’s Taxonomy better reflect the pedagogical content they should be studying. 3. Transformation: in the third act, the student is already used to the common learning challenges and should be able to transfer their newly acquired knowledge. Bloom’s Analysing, Evaluating and Creating steps are more suitable for this arc’s pedagogical contents. 4. Results: the last quadrant is responsible for evaluating the student’s whole journey, what they learned, what they felt and how they changed in the process, and therefore is related to their user experience. The framework works iteratively and incrementally, i.e., the educational gamification strategy can be implemented by blocks (or modules) in several different

62

P. T. Palomino

Fig. 13 Narrative gamification framework for education [35]

cycles, or it can be considered a journey for an entire subject, implementing the content progressively accordingly [35].

4 Gamification as User Experience (UX) Based on prior research presented on this chapter, narrative in gamification for education is defined as “the process in which the users build their own experience through a given content, exercising their freedom of choice in a given space and time, bounded by the system’s logic” [34] and is also known as karma system and implicit decisions. This intrinsic concept is the order of events in the game through the user experience. This experience is influenced by implicit choices made

Theories Around Gamification in Education

63

by the user. Examples of this are providing the content in different ways for the learner to choose by themselves, creating a branch and consequently a different user experience [47]. We can affirm that by these definitions, the student must have several options to do content, but the final goal remains the same: to learn that content. Considering User Experience (UX) as the set of elements and factors related to the user’s interaction with a product, system or service whose result generates a positive or negative perception [31], it is possible to use certain UX techniques, such as mapping users’ journeys to map narrative, creating milestones for particular tasks where the student can choose one path or another and predicting their behaviour when interacting with the system to present the next step. In parallel to this, Storytelling deals with “how the context is presented and the plot developed (the story is told) in a particular environment, which can be through text, voice or even sensory” [47], and Learning Experience (LX) deals with methods that focus on user learning, making the student protagonist of this process.

5 Concluding Remarks This chapter presented theories, models, frameworks and guidelines that could be used to design gamification strategies for education. This research field is increasing each year, and research are converging to consider gamification not only as ‘the use of game elements outside its context [18]’ but more like the use of strategies to improve gameful experiences, bringing positive gaming sensations to contexts outside games [35]. Gamification is much more than applying points, badges and leaderboards to people. It is about bringing the reason why people play games, why people enter the magic circle [16] in the first place, to environments that originally are not games themselves, and to give the students in the educational context, a ludic reason to why they should strive to learn something.

References 1. An, Y., Zhu, M., Bonk, C.J., Lin, L.: Exploring instructors’ perspectives, practices, and perceived support needs and barriers related to the gamification of MOOCs. J. Comput. High. Educ. 33(1), 64–84 (2021) 2. Bai, S., Hew, K.F., Huang, B.: Is gamification “bullshit”? Evidence from a meta-analysis and synthesis of qualitative data in educational contexts. Educ. Res. Rev. 30, 100322 (2020) 3. Baldeón, J., Rodríguez, I., Puig, A.: Lega: a learner-centered gamification design framework. In: Proceedings of the XVII International Conference on Human Computer Interaction, Interacción ’16, pp. 45:1–45:8. ACM, New York (2016). http://doi.acm.org/10.1145/2998626. 2998673 4. Barbosa, S., Silva, B.: Interação humano-computador. Elsevier, Brasil (2010)

64

P. T. Palomino

5. Bell, J., Sheth, S., Kaiser, G.: Increasing student engagement in software engineering with gamification. In: Proceedings of the 4th International Workshop on Social Software Engineering - SSE ’11 pp. 1–2 (2011). https://doi.org/10.1145/2024645.2024657. http://dl.acm.org/ citation.cfm?doid=2024645.2024657 6. Bissell, T.: Extra lives: Why video games matter. Vintage (2011) 7. Bloom, B.S.: Taxonomy of Educational Objectives: The Classification of Educational Goals. Longman, New York (1956) 8. Csikszentmihalyi, M.: Toward a psychology of optimal experience. In: Flow and the Foundations of Positive Psychology, pp. 209–226. Springer, Berlin (2014) 9. Dermeval, D., Albuquerque, J., Bittencourt, I.I., Vassileva, J., Lemos, W., da Silva, A.P., Paiva, R.: Amplifying teachers intelligence in the design of gamified intelligent tutoring systems. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10948, LNAI, pp. 68–73. Springer, Cham (2018). http://link.springer.com/10.1007/978-3-319-93846-2_13 10. Dichev, C., Dicheva, D.: Gamifying education: what is known, what is believed and what remains uncertain: a critical review (2017). http://educationaltechnologyjournal.springeropen. com/articles/10.1186/s41239-017-0042-5 11. Dignan, A.: Game Frame. Free Press, New York (2011) 12. Hamari, J., Koivisto, J.: Measuring flow in gamification: dispositional Flow Scale-2. Comput. Hum. Behav. 40, 133–143 (2014). https://doi.org/10.1016/j.chb.2014.07.048 13. Hassenzahl, M.: The thing and i: understanding the relationship between user and product. In: Funology 2, pp. 301–313. Springer, Berlin (2018) 14. Hodent, C.: The Gamer’s Brain: How Neuroscience and UX Can Impact Video Game Design. CRC Press, Boca Raton (2017) 15. Hsu, C.L., Chen, M.C.: How does gamification improve user experience? An empirical investigation on the antecedences and consequences of user experience and its mediating role. Tech. Forcasting Soc. Chang. 132, 118–129 (2018) 16. Huizinga, J.: Homo Ludens. Routledge & K. Paul, London (1949) 17. Hunicke, R., LeBlanc, M., Zubek, R.: Mda: a formal approach to game design and game research. In: Proceedings of the AAAI Workshop on Challenges in Game AI, vol. 4, p. 1722 (2004) 18. Kapp, K.M.: The Gamification of Learning and Instruction: Game-Based Methods and Strategies for Training and Education. Wiley, New York (2012) 19. Klock, A.C.T., Da Cunha, L.F., de Carvalho, M.F., Rosa, B.E., Anton, A.J., Gasparini, I.: Gamification in e-learning systems: a conceptual model to engage students and its application in an adaptive e-learning system. In: International Conference on Learning and Collaboration Technologies, pp. 595–607. Springer, Berlin (2015) 20. Koivisto, J., Hamari, J.: The rise of motivational information systems: a review of gamification research. Int. J. Inf. Manage. 45, 191–210 (2019). https://doi.org/10.1016/j.ijinfomgt.2018.10. 013 21. Krathwohl, D.R.: A revision of bloom’s taxonomy: an overview. Theory Pract. 41(4), 212–218 (2002). https://doi.org/10.1207/s15430421tip4104_2 22. Landers, R.N., Auer, E.M., Collmus, A.B., Armstrong, M.B.: Gamification science, its history and future: definitions and a research agenda. Simul. Gaming 49(3), 315–337 (2018) 23. Landers, R.N., Tondello, G.F., Kappen, D.L., Collmus, A.B., Mekler, E.D., Nacke, L.E.: Defining gameful experience as a psychological state caused by gameplay: replacing the term ‘gamefulness’ with three distinct constructs. Int. J. Hum. Comput. Stud. 127, 81–94 (2019) 24. Loughrey, K., Broin, D.: Are we having fun yet? Misapplying motivation to gamification. In: 2018 IEEE Games, Entertainment, Media Conference (GEM), pp. 1–9. IEEE (2018). https:// doi.org/10.1109/GEM.2018.8516535 25. Monterrat, B., Lavoué, É., George, S.: A framework to adapt gamification in learning environments. In: European Conference on Technology Enhanced Learning, pp. 578–579. Springer, Berlin (2014)

Theories Around Gamification in Education

65

26. Mora, A., Riera, D., González, C., Arnedo-Moreno, J.: Gamification: a systematic review of design frameworks. J. Comput. High. Educ. (2017). http://link.springer.com/10.1007/s12528017-9150-4 27. Morrison, G.R., Ross, S.J., Morrison, J.R., Kalman, H.K.: Designing Effective Instruction. Wiley, New York (2019) 28. Nielsen, J.: Usability inspection methods. In: Conference Companion on Human Factors in Computing Systems, pp. 413–414. ACM, New York (1994) 29. Nilsen, P.: Making sense of implementation theories, models, and frameworks. In: Implementation Science 3.0, pp. 53–79. Springer, Berlin (2020) 30. Norman, D.A.: Emotional Design: Why We Love (Or Hate) Everyday Things. Basic Civitas Books, New York (2004) 31. Norman, D.: The Design of Everyday Things: Revised and Expanded Edition. Basic Books, New York (2013) 32. Orji, R., Mandryk, R.L., Vassileva, J., Gerling, K.M.: Tailoring persuasive health games to gamer type. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 2467–2476. ACM, New York (2013) 33. Orji, R., Vassileva, J., Mandryk, R.L.: Modeling the efficacy of persuasive strategies for different gamer types in serious games for health. User Model. User Adap. Inter. 24(5), 453– 498 (2014) 34. Palomino, P.T., Toda, A.M., Oliveira, W., Cristea, A.I., Isotani, S.: Narrative for gamification in education: why should you care ? In: International Conference of Advanced Learning Techniques - Icalt 2019 (2019) 35. Palomino, P.T., Isotani, S., Nacke, L.P.E.: Gamification of Virtual Learning Environments: a Narrative and User Experience Approach. University of São Paulo (USP) (2022), Doctoral Thesis,) 36. Pink, D.H.: Drive: The Surprising Truth About What Motivates Us. Penguin, London (2011) 37. Preece, J., Rogers, Y., Sharp, H.: Interaction Design: Beyond Human-Computer Interaction. Wiley, New York (2015) 38. Rodrigues, L., Toda, A.M., Oliveira, W., Palomino, P.T., Vassileva, J., Isotani, S.: Automating gamification personalization: to the user and beyond (2021). arXiv preprint arXiv:2101.05718 39. Rogers, S.: Level Up! The Guide to Great Video Game Design. Wiley, New York (2014) 40. Sailer, M., Homner, L.: The gamification of learning: a meta-analysis. Educ. Psychol. Rev. (2019). https://doi.org/10.1007/s10648-019-09498-w 41. Salen, K., Tekinba¸s, K.S., Zimmerman, E.: Rules of Play: Game Design Fundamentals. MIT Press, Cambridge (2004) 42. Sung, H.Y., Hwang, G.J.: A collaborative game-based learning approach to improving students’ learning performance in science courses. Comput. Educ. 63, 43–51 (2013) 43. Toda, A.M., Toda, A.M., da Silva, A.P., Isotani, S.: Desafios para o Planejamento e Implantação da Gamificação no Contexto Educacional. RENOTE 15(2) (2017). https://doi.org/10.22456/ 1679-1916.79263. http://seer.ufrgs.br/index.php/renote/article/view/79263 44. Toda, A.M., Valle, P.H.D., Isotani, S.: The dark side of gamification: an overview of negative effects of gamification in education. In: Cristea, A.I., Bittencourt, I.I., Lima, F. (eds.) Higher Education for All. From Challenges to Novel Technology-Enhanced Solutions, pp. 143–156. Springer International Publishing, Cham (2018). https://doi.org/10.1007/978-3-319-97934-2_ 9 45. Toda, A.M., do Carmo, R.M.C., da Silva, A.P., Bittencourt, I.I., Isotani, S.: An approach for planning and deploying gamification concepts with social networks within educational contexts. Int. J. Inf. Manage. (2018). https://doi.org/10.1016/J.IJINFOMGT.2018.10.001. https://www.sciencedirect.com/science/article/pii/S0268401218304614 46. Toda, A., Oliveira, W., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.: Planning gamification strategies based on user characteristics and DM: a gender-based case study. In: 12th International Conference on Educational Data Mining, pp. 438–443 (2019)

66

P. T. Palomino

47. Toda, A.M., Klock, A.C., Oliveira, W., Palomino, P.T., Rodrigues, L., Shi, L., Bittencourt, I., Gasparini, I., Isotani, S., Cristea, A.I.: Analysing gamification elements in educational environments using an existing gamification taxonomy. Smart Learn. Environ. 6(1), 16 (2019). https://doi.org/10.1186/s40561-019-0106-1 48. Toda, A.M., Oliveira, W., Klock, A.C., Palomino, P.T., Pimenta, M., Gasparini, I., Cristea, A.: I.: a taxonomy of game elements for gamification in educational contexts: proposal and evaluation. In 2019 IEEE 19th International Conference on Advanced Learning Technologies (ICALT), vol. 2161, pp. 84–88. IEEE (2019, July) 49. Toda, A.M., Oliveira, W., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.: Planning gamification strategies based on user characteristics and DM : a gender-based case study. In: Proceedings of the Educational Data Mining 2019 Conference, I. Montréal (2019) 50. Tondello, G.F., Wehbe, R.R., Diamond, L., Busch, M., Marczewski, A., Nacke, L.E.: The gamification user types Hexad scale. In: Proceedings of the 2016 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’16, pp. 229–243. ACM Press, New York (2016). https://doi.org/10.1145/2967934.2968082. http://dl.acm.org/citation.cfm?doid= 2967934.2968082

Part II

Methods and Tools

In this part we describe how the Taxonomy of Gamification Elements for Educational Environments (TGEEE) tool works, by delving deeply into its elements and presenting some examples on how they can be used in educational environments. Following, we present two case studies in using participatory design to plan and implement gamified strategies in educational contexts. Finally, we present a datadriven framework using the TGEEE as part of the pipeline, to generate strategies using data mining and recommendation systems, to support gamification planning.

TGEEE: Analysis and Suggestions for Use Armando Toda and Alexandra I. Cristea

1 Introduction One significant problem in gamification applied to education, according to literature, is the lack of definitions [27, 34]. This lack of definitions is defined by the absence or little attention to definitions that are used, which might cause confusion in the reader [33]. One classical example is the definition of “badges”, which is a kind of reward that is given to the user when they conduct a certain behaviour in the system. These badges can also be represented as “Achievements” in other works [9], or “Medals” [8], or even “Trophies” [29], these are different words but semantically are the same concept [16]. Some might say that gamification frameworks are a solution for this problem, however not even these frameworks manage to assess the different elements and how to deal with them properly, specially in educational contexts [21, 33]. As an example, one given broad framework like 6 steps to gamification [36] provides 30 elements divided in 3 categories, while another analysis framework like Octalysis [2] provide the user with 77 elements, divided in 8 categories. The first framework provide a set of elements that is most focused on extrinsic rewards which might not address the complexity of what game elements might be (e.g. Storytelling), as for the second framework, it addresses many elements that can be described as system functions and not necessarily all of them are even encountered in educational contexts. This is not exclusive to broad frameworks, but domain specific frameworks as well, most of existing frameworks present different words describing what seems A. Toda () Institute of Mathematics and Computer Science, University of Sao Paulo, Sao Paulo, Brazil e-mail: [email protected] A. I. Cristea Durham University, Durham, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_5

69

70

A. Toda and A. I. Cristea

like the same concept [33]. Based on this problem, we proposed in our previous work, the Taxonomy of Gamification Elements for Educational Environments (TGEEE), which consists of a classification of gamification elements that can be used in educational contexts [34].

2 Taxonomy of Gamification Elements for Educational Environments The TGEEE was designed based on systematic literature reviews and existing frameworks, by extracting and analysing the explicit game elements that were used. Then, we compared those elements with a set of game elements used on the development of Behavioral Games [7]. The framework Game Frame was used as a basis because it was one of the first frameworks that dealt with gamification (without being called gamification at the time), and provided an initial set of game elements that could be used to design Behavioral Games. Based on these elements, we analysed the semantics of each element of the frameworks and tied to a concept defined in the base framework. This mapping process was conducted by two experts on games and gamification. After the mapping process, we designed a questionnaire to evaluate the initial classification, focused on five aspects: • Comprehensibility: Evaluate the comprehension of the concept that we want to define a group of the same element (e.g. Acknowledgment is the common denominator of Badges, Achievements, Trophies, and Medals); • Description: Evaluate the definition for the concept, the description needed to be clear and comprehensive, as well as specific enough to not create a confusion in the reader (broad enough to include the semantics of all synonyms); • Examples: Evaluate the practical examples on how the given concept could be used to represent it (e.g. Points could be presented as score points in a shooter game, or experience points in a RPG game); • Coverage: Evaluate if the initial proposed elements were adequate to represent all types of gamification needed in an educational environment; • Comments: Included additional comments or observations that the evaluator noted regarding the classification. The questionnaire was applied to 19 researchers on gamification, where 17 were also teachers, and the Comprehensibility, Description, Examples, and Coverage were evaluated through a Likert Scale [17] from 1 to 5, where 1 represented Totally Disagree and 5 represented Totally Agree. In summary, most of the concepts obtained a good rate of response (4–5 in the Likert Scale, with more than 65% if the evaluators’ response) some of the evaluators suggested the inclusion of Narrative and Storytelling which were included in the final version. After defining these initial 21 elements, we extended the original definitions while creating groups (dimensions) for each set of gamification elements, and

TGEEE: Analysis and Suggestions for Use

71

presenting examples on how these groups could be used to analyse gamified applications [33]. In this work, we present a detailed explanation of each dimension as well as its elements, and how these elements can be used in educational environments.

2.1 Performance/Measurement Dimension This dimension presents elements related to the environment that can be used to give feedback to the student. This response can be used also as a kind of evaluation of the students’ actions and behaviors. Lack of this dimension might influence the way the students orient themselves, and might think their actions does not impact on anything. For educational purposes, it is important that students have some kind of feedback to check on their learning progress, so using at least one of these elements is advised.

2.1.1

Acknowledgment

Can be also known as badges, medals, trophies, achievements. It praises the students’ specific set of actions in a learning environment, e.g. their diploma or certificate is a kind of achievement [28]; a trophy for completing the max number of tests [8]; a badge for helping other students [12]; an achievement for participating in classes [30], etc. This type of reward must be related to meaningful actions in the learning environment, it can also be used to stimulate positive actions such as helping others. The lack of Acknowledgment may influence in how the student may perceive their actions, where they might find they are not that significant [33]. Besides, Acknowledgment can also be used as an element of analysis, e.g. by defining an achievement for a certain behaviour, the teacher or instructional designer can verify the number of students that attained that achievement and generate insights on certain desired learning behaviours in both virtual and nonvirtual environments.

2.1.2

Level

Also known as skill level, character level, or any kind of structure that provides feedback as a hierarchy to the student. This hierarchy might provide some advantages as they interact with the gamified environment, e.g. students can accumulate levels and convert this level into grades [31]; students have access to different skills when they increase a level [5]. This element is usually tied to Points or other kind of elements that provides accumulation [6], it can also sometimes be confused with Progression, but different from the later, Level would be tied to the students advances while Progression, in our definition, is tied to the global progression of the environment.

72

A. Toda and A. I. Cristea

The lack of Level can make the students feel that they are not advancing or having any kind of gains while they progress in the learning environment.

2.1.3

Progression

Also seen as progression bars, steps, maps, or any kind of structure that can represent the students’ progress in a gamified environment, offering guidance. Progression can be implemented through the usual progress bars [37] or trough interactive maps that can be accessed by the students while they interact [13]. Lack of Progression can make the students feel they are not advancing, similar to levels [33]. It is recommended that either Progression or Level should be used to represent the advance of the students.

2.1.4

Point

Can also be known as score, experience points, skill points, and so on. It is a simple feedback to the users’ actions, can either sum or subtract from the students’ score based on their behaviour in the system. It is already implemented in the traditional learning process as the students’ grades [28], it can also be used as an agreement score in a social interaction, as up-voting a students’ commentary in a virtual system [19]. It is considered a basic concept found in almost all gamified applications as a form of feedback [16]. Points, as Acknowledgment, are feedback to the students’ direct action with the system, and the lack, just as achievements, may lead them to feel that their actions are meaningless [33].

2.1.5

Stats

Also known as information screen, results, Head Up Display (HUD), data, etc. It is related to the visual information presented to the student, by the environment. It can also be represented as a dashboard or profile screen [3]. This element is important because it can provide an overview of the students’ current progression, levels, points, achievements, and other types of elements that are inserted within the environment. The lack of Stats can make the students’ feel disoriented or that their actions have no meaning, similar to the previous elements in this group [33]. In summary, Performance elements will probably be the main type of feedback in a basic gamified environment. These elements can also be used together, e.g. students can accumulate Points, that will give them Levels, which might unlock Acknowledgments and Progression in the environment, and everything can be seen in a Stats screen (dashboard). Points tied to Levels are probably one of the most common elements found in many gamified environments [16], Acknowledgments and Progression are also present significantly in many studies regarding gamification [35], Stats on the other hand is not always treated as one game element, but it

TGEEE: Analysis and Suggestions for Use

73

is a significant part of games and are usually intrinsically present in any gamified system [2]. As said before, try to always consider at least one Performance element when gamifying a learning environment.

2.2 Ecological Dimension This dimension is related to the environment itself that is being gamified. These elements are properties of the environment that will help us to model some strategies alongside the other dimensions. The elements of this dimension are focused on influencing the students’ autonomy and consequently their interactions within the environment. Different from the Performance dimension, these elements are not strongly mandatory but they can influence (positively or negatively) their learning experience. Lack of this dimension might minimise the interactions within the environment.

2.2.1

Chance

This represents randomness related elements, like luck, fortune or probability. It is related to the random property of an event or outcome, creating a certain unpredictability in the environment. In a learning environment, teachers and instructional designers can have a bank of questions and randomly select one and assign it to students, another possible use is to have a bank of tips during exams where the lecturer might randomly select one and give it to the class [30]. The lack of this element might create a predictable experience to the students, however, depending on the length of the gamified solution. In short gamified interventions (e.g. less than 20 hours course for example), the Chance can be implemented as a prize draw for the participants, in long gamified interventions (e.g. more than a month at least), Chance can provide some unpredictability as a drive to students, e.g. draw a question from a bank of questions.

2.2.2

Imposed Choice

This element represent choices within the environment, can be known as choice, judgment, paths, and so on. It occurs when students are faced with an explicit choice and this influence the outcome of the task, e.g. the students can choose the next content of the lecture [22]; or students can choose the date of an exam [30]. The lack of this element might influence the way students feel about their autonomy in the environment. This element is also influenced by the length of the gamified intervention, if it is a short gamified intervention it might be difficult to implement the choices the students are about to make, however in long gamified interventions, the students have more time to think about how their choices influenced in the final

74

A. Toda and A. I. Cristea

outcome of a course or lecture. Imposed Choice might provide some strong sense of autonomy (both singular or of a community). 2.2.3

Economy

This element represent any kind of transaction in the environment, also known as trades, market, exchanges, etc. Here, students can spend some kind of currency to acquire advantages in the environment, e.g. spend points to buy tips during exams, or change the date of an exam [30]. Economy can give the student a sense of autonomy or possession, and also stimulate social interactions among other students if tied to other elements (e.g. Cooperation). Again, in short gamified interventions, Economy might be a difficult element to implement, however the instructor can offer a store to spend points or other type of currencies available during the gamified intervention. In long gamified interventions, the instructor can offer more advantages related to the student, groups of students, or the entire classroom. 2.2.4

Rarity

This element is associated with something that might be difficult to obtain, also known as limited items, collections, exclusivity, etc. It usually is represented by limited resources within the environment that can stimulate engagement to obtain these resources. As an example, students can obtain a rare achievement for doing a specific action within the environment [20], they can also obtain a certain prize after a Competition [10], students can also obtain different grades of achievements, where some grades are rarer than others, and so on. This element has a lot of potential to drive competitions in short gamified interventions or long-term prizes in long gamified interventions. Rarity can incite some sense of possession in the students that are participating in the environment. 2.2.5

Time Pressure

This element is associated with time itself, and its influence over the students’ tasks, it is also known as countdown timers or clocks. In this concept, time is used as a drive that exerts pressure in the learners’ actions, in learning environments it is represented as deadlines [11]. It is a very difficult element to use, also it is not wellaccepted by students but it can drive some behaviours if tied with other elements (e.g. an achievement related to time completion). It is a good element to ensure tasks are done in a short gamified intervention, and can provide the students some guidance (e.g. deadlines) in long gamified interventions. Even though Time Pressure can assure some tasks are due on time, it can also cause the students to feel less motivated. In summary, the elements of this dimension act more as properties of a given task in a gamified environment. Again, they are not obligatory and can also be used with

TGEEE: Analysis and Suggestions for Use

75

other elements. Most of these elements can work very well when tied to elements from the Performance dimension (e.g. limited badges (Rarity), score for positive choices (Imposed choice), etc.).

2.3 Social Dimension This dimension contains the elements that are responsible for the interactions between students in the gamified environment. These elements can be useful when dealing with a large group of students (e.g. Cooperation can create sub-groups of students to facilitate the instructor management), and the lack of these elements might cause the students to feel isolated since they won’t be able to interact with other students. It is desirable to have at least one of these elements since they are directed related to the “Relatedness” concept seem in Self-Determination Theory and other motivational theories that implies that social behaviour is innate to us as members of a society [4].

2.3.1

Competition

This element represents conflict, also known as scoreboards, player vs player, leader boards, and so on. This element is responsible for creating dispute between two or more students. It influences a sense of comparison in the students (which also can impact in the Social Pressure as will be seen soon). This element is relatively easy to implement since it can be used with a Performance element to create scoreboards focused on points or badges. The most common representation of this element is the use of leader boards [15] which creates an internal competition in learning systems. Designers must be cautious when using this element since too much competition can lead to undesired behaviours such as cheating or gaming the system [32]. It is recommended to create healthy competitions which might stimulate positive actions (e.g., students who helps more students than others, or that have positive contributions during lectures or group tasks). In short gamified interventions, a small scale competition can be created to drive a given behaviour and award the winner with a limited prize (paired with Rarity). In long gamified interventions, a large scale competition between groups (paired with Cooperation) can be used to drive long term behaviours in the students as scoring (paired with Points) the most constructive comments during a lesson.

2.3.2

Cooperation

This element represents forms of cooperating between students, it is also known as teamwork, guilds, co-op, groups, and so on. Cooperation relates to the collaboration between users to achieve a common goal. This element can be used to create a

76

A. Toda and A. I. Cristea

sense of unison between students, instructors and teachers can create groups and give tasks to be divided among the members of each group [14], another possible strategy is students can praise other students tasks or contributions with feedback [14]. Cooperation can also be used to stimulate collaborative learning scenarios [1]. In both short or long gamified interventions, Cooperation can be implemented through groups of students that must collaborate to solve a certain task (paired with Objectives or Puzzles).

2.3.3

Reputation

This element is related to social status, also known as titles, classification, and so on. It is related to the social status a student can obtain in the gamified environment, this status could be affected by their level, achievements, completed tasks, and any other metric or element chosen by the designer. Students can gain a title based on their position of the leader board, or by achieving a certain goal (like help other 5 students). Reputation can, sometimes, be confused or associated with Acknowledgments, however the first is related to a social status achieved for certain tasks and the other is related to a reward for doing a pre-determined task. Both elements can also be used simultaneously, where the student can get a title that will be shown to other, by completing an achievement [9]. In short gamified interventions, Reputation can be used in dynamics where one person manages others (paired with Cooperation), as for long gamified interventions, students can accumulate functions based on their tasks or interactions in class.

2.3.4

Social Pressure

This element represents the social interactions that exert pressure on the student, can also be known as peer pressure, co-op missions, praising actions, and so on. This is an element that will be probably be present intrinsically when using score/leader boards (paired with Competition) or in groups (paired with Collaboration). Extrinsically, this element can be represented with praising system (where other students can praise others’ actions) or even paired with Imposed Choice in a voting system (where one student might choose what happens to another). Social Pressure can be used to understand certain situations within a group but, as Competition, should be used with caution, since too many pressure can lead students to a loss of motivation. In summary, elements from this dimension are versatile to use with other elements from the previous explained dimensions, and can add a layer of interactions between students and not just the students and the environment.

TGEEE: Analysis and Suggestions for Use

77

2.4 Personal Dimension This dimension is related to the students’ interactions in the gamified environment, it contain more abstract concepts that can be translated in many functions provided in the same environment. The lack of these elements might cause the student to feel less motivated since the environment is not meaningful to the student. It is recommended to use at least one of these elements alongside one from the Performance dimension, as a basis for the gamified environment.

2.4.1

Novelty

This element is related to the content inserted in the gamified environment, it is also known as updates, surprises, changes, etc. It represents the improvements of a content in the environment, it could be either the learning content (e.g. unlocking new content) or the gamification content (e.g. new elements). It is a good element to keep students engaged, and it is also essential in long gamified interventions otherwise, just like in games, students can feel bored for interacting with the same thing over and over for a long time. In short gamified interventions, Novelty can act as an element of surprise, it is also something intrinsically to new gamified interventions which impacts directly in the so called “Novelty Effect” [26].1 Lack of Novelty can lead to boredom, as stated before, students might get bored due to the lack of new content. One important aspect to consider in Novelty is that the gamified environment might need changes over time, so it is important to assess those environments in order to insert new changes (e.g., students are not liking the way the gamified intervention was implemented).

2.4.2

Objectives

This element is related to the goals of the environment, aligned with the learning objectives, it is also known as missions, side-quests, milestones, etc. This element provides the purpose to the environment. Every gamified environment has a purpose, usually related to increase the students motivation and/or engagement, which is tied to the learning goals (e.g., learn a specific content) [31]. Then, the Objectives can be represented through a list of missions that will tell the student what they need to achieve that course/lecture. It can describe missions telling how the gamified interventions will occur. In short gamified interventions, the Objectives can be implemented through a list of tasks that the students need to do to finish a task, and in a long gamified intervention, the instructor/designer can provide a list of long term goals that the student needs to complete the course. 1 It is related to the acceptance or interest in new technologies that an individual is initially exposed to.

78

2.4.3

A. Toda and A. I. Cristea

Puzzle

This element is related to the challenges faced by the student, it can also be called challenges, cognitive tasks, quizzes, etc. This concept is intrinsically related to learning activities within a learning environment, since they are focused on cognitive tasks. It is very difficult for a learning environment to not have some instance of this concept, specially in the assessment of the student. In virtual environments, Puzzles are usually represented as quizzes, but can be translated into other type of interactive activities as simulations. For non-virtual environments, Puzzles can be used in a dynamic of an escape room, where students needs to interact with the environment in order to find clues to solve problems [18]. This element is also viable to be used in both short and long gamified interventions.

2.4.4

Renovation

This one is related to a new chance provided to the student, it is also known as boosts, extra life, renewal, re-do, etc. Usually it is also intrinsically tied to learning activities, e.g., when the activity allow the student to re-do the task to increase their grade. This is one of the crucial elements to be present in educational environments since it is one of the main characteristics that makes games fun [28]. This element can also be easily implemented in both short and long gamified interventions, since it is related to this re-do property of a given task, e.g., students can also have extra chances to answer a given problem.

2.4.5

Sensation

This element is related to sensory stimuli, e.g. visual and/or sounds. It is intrinsically related to the learners’ experience when stimulated through the visual (e.g. gamelike interfaces) or sounds (e.g. game soundtracks). It can appeal to the students’ affective memory when using elements from different games to create a connection [25]. This elements can also be presented with different technologies as Mixed Reality (MR), Augmented Reality (AR), or Virtual Reality (VR). Considering its implementation, it can be used in both short and long gamified interventions as well as the other elements in this group, e.g., teachers can use visual images of students’ favourite games to create an immersive experience appealing to their affective memory [25]. In summary, these elements are responsible to create meaning in the environment, so the student can relate. Some elements here can be tied also to students’ affective memory through nostalgia which can engage them even more in the environment.

TGEEE: Analysis and Suggestions for Use

79

2.5 Fictional Dimension This dimension relates to both student (Narrative) and the environment (Storytelling), tying the students’ experience with the context of the gamified environment. The elements of this dimension are very important for the students’ experience however they are not so trivial to implement. They can be used alongside the elements from previous dimensions, specially progression which is usually the most common association [22]. The lack of Fiction can lead to a loss of context of the gamified environment.

2.5.1

Narrative

This element is intrinsically related to the students’ experience in the gamified environment, it can be represented through karma system, implicit decisions, and so on. It can be defined as the order of events as they happen in the environment, that affects the students’ experience [24]. Their experience is influenced by implicit choices they make, e.g. by selecting a given number of choices, the student is conducted to an alternative content. Independent from the length of the intervention, Narrative is not a trivial element to implement since it requires some knowledge about the students and their possible interactions in the learning environment [23]. Virtually, Narrative can be used alongside recommendation systems to guide the students’ experience to a more tailored content, while in non-virtual environments, Narrative can be used alongside Puzzles and create an “Escape Room” experience.

2.5.2

Storytelling

This element represents a fictional context inserted in the learning environment, can be represented through audio queues, text stories, etc. It is the way the script of the story is told, it can support Narrative by creating an immersive experience [24]. The difficulty to implement Storytelling varies with the creativity of the teacher or instructor that is using it. There are some frameworks that can help in developing the story as the Hero’s Journey. In a virtual environment, it is common to see this element in “visual novel2 ” like applications, in non-virtual environments, Storytelling can contextualise the students in a fictional campaign that they need to overcome to make it to the end of the learning activity.

2A

game genre focused on stories.

80

A. Toda and A. I. Cristea

3 Discussions Limitations of Our TGEEE In this chapter we presented TGEEE, although being the first classification of gamification elements for educational environments, it is important to discuss how it can be used. Initially, it was designed to be used as a dictionary to teachers and other education professionals, to differentiate and explain the concepts of gamification elements that are most commonly found. The focus is the education professionals since most of these same professional, specially teachers, does not have time nor resources to study about the different concepts regarding gamification. The TGEEE was also designed to support the analysis of existing gamified applications or interventions, since it provides the descriptions and possible synonyms of each element [33]. By analysing these elements, it is possible to identify how the gamified strategies works and what is lacking in the environment. The classification can also be used alongside other existing frameworks, since it can fit almost every framework related to education. Another possible use is to use it alone as a way to think and design gamified strategies on demand. One important note about the TGEEEE, It can be useful for the purpose it was designed, but we cannot assure these strategies will efficiently work since studies on the field of gamified education often lacks to isolate gamification related variables. It is important to consider context and culture, as stated in chapter “Gamification for Education”, before just using the same strategy in a certain learning activity. It is of utmost importance to note that this classification was created as a proposal, so it can be altered to the user needs and context. Another limitation is the coverage of the gamification elements, in the first study [34] we evaluated the TGEEE with a number of gamification and education experts but we understand that this population might not be representative to attest the entire coverage, so it would be interesting in the future to validate this coverage of gamification elements with more experts. Up to now, many studies have used the TGEEE as it is, without any major modifications.

4 Concluding Remarks In this chapter we discussed the Taxonomy of Gamification Elements for Educational Environments, presenting some steps regarding its creation and evaluation, and expanding existing concepts of gamification elements by providing some examples on how they can be used based on the length of the gamified intervention (long or short) and environment (virtual or non-virtual). We then, present some known limitations of the TGEEE while discussing some potential uses with existent frameworks.

TGEEE: Analysis and Suggestions for Use

81

References 1. Challco, G.C., Mizoguchi, R., Isotani, S.: An ontology framework to apply gamification in CSCL scenarios as persuasive technology. Revista Brasileira de Informática na Educação XX(02), 67 (2016). https://doi.org/10.5753/RBIE.2016.24.02.67. http://br-ie.org/pub/index. php/rbie/article/view/6436 2. Chou, Y.K.: Actionable gamification: beyond points, badges, and leaderboards (2012). https:// doi.org/10.1017/CBO9781107415324.004 3. de Freitas, S., Gibson, D., Alvarez, V., Irving, L., Star, K., Charleer, S., Verbert, K.: How to use gamified dashboards and learning analytics for providing immediate student feedback and performance tracking in higher education. In: Proceedings of the 26th International Conference on World Wide Web Companion, pp. 429–434 (2017) 4. Deci, E.L., Ryan, R.M.: Intrinsic Motivation and Self-determination in Human Behavior. Springer US, Boston (1985). https://doi.org/10.1007/978-1-4899-2271-7. http://link.springer. com/10.1007/978-1-4899-2271-7 5. Decker, A., Lawley, E.L.: Life’s a game and the game of life: how making a game out of it can change student behavior. In: Proceeding of the 44th ACM Technical Symposium . . . , pp. 233–237 (2013). http://dl.acm.org/citation.cfm?id=2445269. http://www.scopus.com/inward/ record.url?eid=2-s2.0-84876276011&partnerID=tZOtx3y1 6. Dichev, C., Dicheva, D.: Gamifying education: what is known, what is believed and what remains uncertain: a critical review (2017). https://doi.org/10.1186/s41239-017-0042-5. http:// educationaltechnologyjournal.springeropen.com/articles/10.1186/s41239-017-0042-5 7. Dignan, A.: Game Frame. Free Press, New York (2011) 8. Domínguez, A., Saenz-de Navarrete, J., De-Marcos, L., Fernández-Sanz, L., Pagés, C., Martínez-Herráiz, J.J.: Gamifying learning experiences: practical implications and outcomes. Comput. Educ. 63(Null), 380–392 (2013). https://doi.org/10.1016/j.compedu.2012.12.020 9. Grant, S., Betts, B.: Encouraging user behaviour with achievements: an empirical study. In: Proceedings of the Tenth International Workshop on . . . , pp. 65–68 (2013) 10. Hew, K.F., Huang, B., Chu, K.W.S., Chiu, D.K.: Engaging Asian students through game mechanics: findings from two experiment studies. Comput. Educ. 92–93, 221– 236 (2016). https://doi.org/10.1016/j.compedu.2015.10.010. http://www.sciencedirect.com/ science/article/pii/S0360131515300592. http://www.scopus.com/inward/record.url?eid=2-s2. 0-84945976671&partnerID=tZOtx3y1 11. Jang, J., Park, J.J., Yi, M.Y.: Gamification of online learning. In: International Conference on Artificial Intelligence in Education, pp. 646–649. Springer, Berlin (2015) 12. Knutas, A., van Roy, R., Hynninen, T., Granato, M., Kasurinen, J., Ikonen, J.: A process for designing algorithm-based personalized gamification. Multimed. Tools Appl. 78(10), 13593– 13612 (2019). https://doi.org/10.1007/S11042-018-6913-5/FIGURES/1. https://link.springer. com/article/10.1007/s11042-018-6913-5 13. Konstantinov, O., Kovatcheva, E., Palikova, N.: Gamification in cultural and historical heritage education. In: Proc. 12th Int. Technol., Educ. Develop. Conf.(INTED), pp. 8443–8451 (2018) 14. Kuo, M.S.S., Chuang, T.Y.Y.: How gamification motivates visits and engagement for online academic dissemination - an empirical study. Comput. Hum. Behav. 55, 16– 27 (2016). https://doi.org/10.1016/j.chb.2015.08.025. http://www.sciencedirect.com/science/ article/pii/S0747563215301011 15. Landers, R.N., Landers, A.K.: An empirical test of the theory of gamified learning: the effect of leaderboards on time-on-task and academic performance. Simul. Games 45(6), 769–785 (2015). https://doi.org/10.1177/1046878114563662. http://sag.sagepub.com/content/45/6/769. abstract 16. Lewis, Z.H., Swartz, M.C., Lyons, E.J.: What’s the point?: A review of reward systems implemented in gamification interventions. Games Health J. 5(2), 93–99 (2016). https://doi. org/10.1089/g4h.2015.0078

82

A. Toda and A. I. Cristea

17. Likert, R.: A technique for the measurement of attitudes. Arch. Psychol. 140(140), 44–53 (1932) 18. López-Belmonte, J., Segura-Robles, A., Fuentes-Cabrera, A., Parra-González, M.E.: Evaluating activation and absence of negative effect: gamification and escape rooms for learning. Int. J. Environ. Res. Public Health 17(7), 2224 (2020) 19. Martínez-Núñez, M., Fidalgo-Blanco, Á., Borrás-Gené, O.: New challenges for the motivation and learning in engineering education using gamification in MOOC. Int. J. Eng. Educ. 32(1B), 501–512 (2016) 20. McDaniel, R., Fanfarelli, J.: Building better digital badges: pairing completion logic with psychological factors. Simul. Games 47(1), 73–102 (2016). https://doi.org/10.1177/ 1046878115627138. http://sag.sagepub.com/content/47/1/73.abstract 21. Mora, A., Riera, D., González, C., Arnedo-Moreno, J.: Gamification: a systematic review of design frameworks. J. Comput. High. Educ. (2017). https://doi.org/10.1007/s12528-017-91504 22. Palomino, P., Toda, A., Oliveira, W., Rodrigues, L., Cristea, A., Isotani, S.: Exploring content game elements to support gamification design in educational systems: narrative and storytelling. In: Brazilian Symposium on Computers in Education (Simpósio brasileiro de informática na educação-SBIE), vol. 30, p. 773 (2019) 23. Palomino, P.T., Toda, A.M., Oliveira, W., Rodrigues, L., Isotani, S.: Gamification journey : a novel approach for classifying gamer types for gamified educational systems. In: Proceedings of SBGames 2019, Rio de Janeiro (2019) 24. Palomino, P.T., Toda, A.M., dos Santos, W.O., Cristea, A.I., Isotani, S.: Narrative for gamification in education: why should you care? In: Proceedings of the 19th IEEE International Conference on Advanced Learning Technologies (2019) 25. Rodrigues, L., Arndt, D., Palomino, P., Toda, A., Klock, A.C.T., Avila-Santos, A., Isotani, S.: Affective memory in gamified learning: a usability study. In: Anais do XXXIII Simpósio Brasileiro de Informática na Educação, pp. 585–596. SBC (2022) 26. Rodrigues, L., Pereira, F.D., Toda, A.M., Palomino, P.T., Pessoa, M., Carvalho, L.S.G., Fernandes, D., Oliveira, E.H., Cristea, A.I., Isotani, S.: Gamification suffers from the novelty effect but benefits from the familiarization effect: findings from a longitudinal study. Int. J. Educ. Technol. High. Educ. 19(1), 1–25 (2022) 27. Seaborn, K., Fels, D.I.: Gamification in theory and action: a survey. Int. J. Hum. Comput. Stud. 74, 14–31 (2014). https://doi.org/10.1016/j.ijhcs.2014.09.006 28. Smith-Robbins, S.: This game sucks: how to improve the gamification of education. Educ. Rev. 46(1), 58–59 (2011). https://doi.org/10.1109/IEMBS.2005.1616259 29. Tenório, K., Dermeval, D., Monteiro, M., Peixoto, A., Pedro, A.: Raising teachers empowerment in gamification design of adaptive learning systems: a qualitative research. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12163 LNAI, pp. 524–536 (2020). https://doi.org/10. 1007/978-3-030-52237-7_42/FIGURES/2 30. Toda, A., Silva, Y., Cruz, W., Xavier, L., Isotani, S., Rafael, Y., Cruz, W., Xavier, L., Isotani, S.: Um processo de Gamificação para o ensino superior: Experiências em um módulo de Bioquímica. In: Anais do Workshop de Informática na Escola, vol. 22, p. 495 (2016). https://doi.org/10.5753/cbie.wie.2016.495. http://www.br-ie.org/pub/index.php/wie/ article/view/6856 31. Toda, A.M., do Carmo, R.M., da Silva, A.P., Bittencourt, I.I., Isotani, S.: An approach for planning and deploying gamification concepts with social networks within educational contexts. Int. J. Inf. Manage. (2018). https://doi.org/10.1016/J.IJINFOMGT.2018.10.001. https://www.sciencedirect.com/science/article/pii/S0268401218304614 32. Toda, A.M., Valle, P.H.D., Isotani, S.: The dark side of gamification: an overview of negative effects of gamification in education. In: Communications in Computer and Information Science, vol. 832, pp. 143–156. Springer, Cham (2018). http://link.springer.com/10.1007/9783-319-97934-2_9

TGEEE: Analysis and Suggestions for Use

83

33. Toda, A.M., Klock, A.C.T., Oliveira, W., Palomino, P.T., Rodrigues, L.L., Shi, L., Bittencourt, I., Gasparini, I., Isotani, S., Cristea, A.I.: Analysing gamification elements in educational environments – using an existing gamification taxonomy. Smart Learn. Environ. 6(1), 16 (2019). https://doi.org/10.1186/s40561-019-0106-1 34. Toda, A.M., Oliveira, W., Klock, A.C., Palomino, P.T., Pimenta, M., Gasparini, I., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.I., Shi, L., Gasparini, I., Isotani, S., Cristea, A.I., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.I.: A taxonomy of game elements for gamification in educational contexts: proposal and evaluation. In: IEEE 19th International Conference on Advanced Learning Technologies (ICALT), pp. 84–88 (2019). https://doi.org/10.1109/icalt. 2019.00028 35. Toda, A.M., Oliveira, W., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.: Planning gamification strategies based on user characteristics and DM: a gender-based case study. In: Proceedings of the Educational Data Mining 2019 Conference, i, Montréal, pp. 438–443 (2019). http://arxiv. org/abs/1905.09146 36. Werbach, K., Hunter, D.: For the Win: How Game Thinking Can Revolutionize Your Business. Wharton Digital Press, Philadelphia (2012) 37. Wongso, O., Rosmansyah, Y., Bandung, Y.: Gamification framework model, based on social engagement in e-learning 2.0. In: Technology, Informatics, Management, Engineering, and Environment (TIME-E), 2014 2nd International Conference, pp. 10–14 (2014). https://doi.org/ 10.1109/TIME-E.2014.7011583. www.files/22712/Wongso2014.pdf

Using Participatory Design to Design Gamified Interventions in Educational Environments Armando Toda, Elad Yacobson, Giora Alexandron, Paula T. Palomino, Mauricio Souza, Elian Santos, Alinne Corrêa, Rodrigo Lisboa, Thiago Damasceno Cordeiro, and Alexandra I. Cristea

1 Introduction Designing gamified strategies sometimes requires complex knowledge on profiling, when considering personalisation techniques to specialise those strategies to a specific public [7]. When considering personalisation in real environments (e.g.,

A. Toda () University of Sao Paulo, Institute of Mathematics and Computer Science, Sao Paulo, Brazil e-mail: [email protected] E. Yacobson · G. Alexandron Weizmann Institute of Science, Rehovot, Israel e-mail: [email protected]; [email protected] P. T. Palomino · T. D. Cordeiro Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil e-mail: [email protected]; [email protected] M. Souza Lavras Federal University, Lavras, Brazil e-mail: [email protected] E. Santos Education Center, Federal University of Alagoas, Maceió, Brazil e-mail: [email protected] A. Corrêa Federal University of Technology - Parana, Curitiba, Brazil e-mail: [email protected] R. Lisboa Federal Rural University of the Amazon, Pará, Brazil A. I. Cristea Durham University, Durham, UK © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_6

85

86

A. Toda et al.

classrooms), it can become too complex based on the number of students and due to the already existent workload of many teachers and instructors [11]. One practical example, imagine having a classroom with more than 20 students, and the teacher needs to collect all of their students’ player profiles to design a possible gamified strategy to please the majority of these profiles. In this example, one possible solution would be: the teacher apply questionnaires to collect students’ player profiles; next, the teacher needs to understand the division of these profiles within their class; following, the teacher needs to think of a strategy to please all of the profiles or a group of these profiles. Using this solution for the problem above, the results are most likely to appeal only to a few students, and not the entire class, which might lead to negative impacts over time due to the lack of proper design [15]. In this sense, personalisation of a gamified environment not only can make the teacher spend lots of time to design those strategies, but also not be accepted by their students. Even though literature has demonstrated that personalised gamified environments tends to have a positive impact on students [2], it restricts its efficiency to virtual environments. Furthermore, most of these personalised interventions does not consider students’ context and culture, which might lead to the same design being applied to different classes and leading to different results [11]. In non-virtual environments, where the collection and division of game elements among students is not something automatically, it falls on the already overloaded teacher or instructor to do this work. In another practical example, imagine hypothetically designing a gamified learning system1 focus solely on literature based strategies. Again, one gamified strategy that might work in a certain context might not work the same way in another context, this is important to consider in the design of these gamified learning systems and often is not considered in approaches focused on this development [12]. This might lead to a system that is not adequate to the teachers’ or institutions’ needs and lead to a bad acceptance overall. In summary, personalisation is important but also needs to consider the actual users’ (in this case, students, teachers, and institutions) needs, which leads us to their actual context and culture at the time the gamified intervention will be applied [7]. To address this problem, one possible solution is to involve the students and also teachers in the design process of the gamified strategies [14]. Based on these premises, in this chapter we tackle the following research problem “How to involve students and teachers in the design of gamified strategies in educational contexts?”. Through two case studies, we present how we can involve teachers in the design process of gamified strategies through the use of participatory design in educational environments.

1 A gamified learning system is a learning system with game elements that were pre-implemented to address specific educational needs.

Using Participatory Design to Design Gamified Interventions in Educational. . .

87

2 Methods To tackle our research problem, we developed an approach to create gamified strategies based on participatory design. Participatory design, according to Muller and Kuhn [9] is a democratic process for design of systems that involves the users and stakeholders equally. In other words, both users and stakeholders choices have the same weight as of the designers that are involved in the process. According to Rosenzweig et al. [13] participatory design can be used from different approaches such as brainstorming, co-creation, and creation of task scenarios by the users of a given system. We opted to use participatory design because its characteristic of involving the users’ (in our case, students and teachers) insights into the early design phase. To materialise our approach, we devised a process that can be adapted to create both virtual and non-virtual gamified strategies (Fig. 1). In this process, there are two groups of actors, the education and gamification. Education actors (EAs) are the public of the gamified strategy, they can be students or teachers and instructors. They are responsible to discuss the game elements that are going to be used. The Gamification actors (GAs) are the responsible for conducting the design of the gamified strategy, as well as its implementation and evaluation. As we can observe in Fig. 1, it is an iterative process, which means that it can happen more than once in the design of the gamified strategies. It starts with the

Fig. 1 Our participatory approach

88

A. Toda et al.

“Define learning goals” step, where the EAs explicit what are their objectives with the gamified strategies. In the case of the students being the EAs, the teacher or instructor that is responsible for their learning is the one that explicit these objectives. These objectives can be aligned with instructional design practices, and define a content or subject to be taught, e.g., To teach basic programming concepts. Following, the next step is to “Present Gamification”. This task is conducted by the GA, where they are responsible to explain basic concepts related to gamification to situate the other actors involved. This can be done through an informal discussion or either a formal presentation, the importance of this step is to even the knowledge of actors that are present and will be responsible for the design phase. It is imperative that the GA use concepts and definitions that are based on the existent literature of gamification. Next, both actors began the step “Define elements”, where they will discuss the best elements to use. This can be done through brainstorming sessions and cocreation. In these sessions it is important to define: Which gamification elements will be used? How they will be related (if more than one is selected)? What are the rules of these elements in the educational environment? For how long these elements will be used? This is the most important part of the process, and the inputs of the EAs are of extremely importance, the decisions should be democratic and have a full consensus of both parts, both the EAs and GAs. Considering a case of students as EAs and teachers as GAs, this second step might require some previous knowledge on gamification on the teachers’ side, so they can explain the concepts to the students. Since the GAs does not need to be a single person, but a group, the teacher can also recruit a gamification designer or expert to support in this step. Based on the definition of gamification elements, the teacher can also explicit what can be doable to implement or not in their lectures, and discuss a consensus with the students. Furthermore, the “Implement” and “Evaluate” steps are related to the GA, where they will be responsible to put the gamified strategies in practice, with everything that was discussed and agreed previously with the EAs, and evaluate it. The implementation in virtual environments (e.g., a Learning Management System) is usually done by an development team and the GA might act alongside a software engineer in that process. In non-virtual environments, the GA can deploy the game elements and its rules the way they discussed with their EAs. As for the evaluation process, it depends on the objectives defined between the EAs and GAs. The GA might be able to propose instruments and forms of evaluation, e.g., if increasing students’ motivation was defined as an objective, the GA can propose an instrument or questionnaire to measure that construct. Finally, the “Analyse” step is conducted by both actors, where they will analyse and make the report regarding the use of the gamified strategies. Depending on the time the strategy was implemented, they can make improvements in a posterior cycle.

Using Participatory Design to Design Gamified Interventions in Educational. . .

89

3 Case Studies To demonstrate the participatory design process, we will present two case studies. The first is focused on implementing gamified strategies, in a virtual platform, to increase teachers’ engagement [20]; the second is focused on presenting a case where we create gamified strategies for a higher education platform. The second case is focused on the process of generating the strategies up to the “Implement” step.

3.1 Gamification of PeTeL PeTeL is an open educational resources repository that is used by teachers to find and share educational resources [19]. However, stakeholders of the system reported that the engagement of teachers was too low, teachers were not using the system nor its functionalities. This posed as an opportunity to use gamification to increase the teachers’ engagement. In the context of PeTeL, the teachers’ engagement is measured through the system logs that contains information regarding the resources these teachers interacted with. The system also contains some social network functions to improve teachers’ interactions. In this context, our EAs are the physics teachers, we opted to begin this participatory design with physics teachers due to convenience sampling, since these teachers where more accessible to speak to about the system. The GAs were three researchers that have worked with gamification previously, one of them were mediating the meetings and discussions with the teachers, while the other two gave feedback or proposals based on the EAs’ feedback. We presented them the concepts related to gamification, and presented the taxonomy designed by Toda et al. [18] which covered 21 gamification elements focused on educational environments (Step 1: Present gamification). This presentation and dynamic was conducted through a workshop that occurred in 2021, at the annual PeTeL conference. During this session, 17 teachers were present five different mock-ups with different gamified strategies. Then, we asked these teachers to rank up these elements based on their utility to engage them. Alongside the mock-ups that were presented, we also held a group discussion so that the teachers could share their opinions towards the gamification strategy that was being implemented. In summary, teachers reported that they have a need to feel that their recommendations in the system are meaningful and taken seriously, that they want to actively contribute to their community. They also stated that social recognition by their peers were also important for that aspect (Step 2: Define elements). Based on the EA feedback, the strategy that was designed consisted of a social panel based on social recognition, where each interaction (e.g., recommendation)

90

A. Toda et al.

Fig. 2 PeTeL social panel

by the teachers would be seen by all of the community. This was designed based on the Cooperation element presented in the taxonomy that was used, where the users interact together to overcome a given challenge, as well as Reputation, where the teachers’ interactions are seen by every other teacher using the platform in realtime. The implementation (Step 3: Implement) was conducted by a software development team, monitored by one of the GAs. The software team implemented the strategy as it was designed (Fig. 2). After implementing the panel, teachers were monitored during one year to understand their engagement after the gamified strategy. In the previous year, the teachers used around 2.372 learning resources in the platform, and reviewed 61 of them, which reached 2.6% response rate. After 7 months of the implemented gamified strategies, the number of interactions almost tripled (increase in x2.7 related to the previous year), with 114 reviews. The authors also verified the response rate (e.g., reviews that were given to a learning resource in the platform), in which they found out an increase of x.23 related to the previous year, since after the gamified strategies, the teachers used 1.888 learning resources and received 114 reviews, resulting a response rate of 6%. Following, this gamified strategy was extended to chemistry teachers, to verify if the same kind of strategy (designed alongside physics teachers) would affect positively in the interactions of chemistry teachers. These sample of teachers were also chosen due to convenience sampling [1], since they had availability to participate in the following study. In this study, the authors conducted an A/B testing where they wanted to measure the efficiency of the gamified strategies and the non-gamified strategies. They created a control group (N = 59) that used no gamification, and the experimental group (N = 59) that was exposed to the gamified strategies.

Using Participatory Design to Design Gamified Interventions in Educational. . .

91

In the first group, 34 out of 59 teachers performed 64 interactions (reviews), the response rate of this group was 15%. As for the second group, 24 out of 59 teachers filled in 61 reviews, and their response rate was 16.3%. Using a onetailored proportion test showed that the differences between the two groups were not statistically significant (.p 0.05). After an interview with both teacher groups, the authors could identify that the Physics teachers were more likely to externalise a stronger feeling of belonging within their community due to the gamified strategies that were implemented, differently from the Chemistry teachers, since these group of teachers did not participated in the design phase.

3.2 Gamification of a Higher Education Virtual Course Following, we conduct a second study where we used participatory design to create gamified strategies. In this case, the context is the development of a virtual learning platform focused on adult training. In this sense, during the design and implementation of an educational system, there usually are a pedagogical team and software engineers that are involved where the first acts as the stakeholder and the second as the medium for a team of development [3]. Usually, the pedagogical team is responsible for the requirements related to the learning activities and other pedagogical aspects of a given platform (e.g., instructional designers delivering content). However, when considering gamified learning environments, few approaches consider the pedagogical aspects, software requirements needs, and gamified interventions within the environment. In literature, most of these approaches only consider the gamification designer [8]. Based on this premise, this second study focuses on tackling the following research problem: “How can we involve instructional designers and software engineers in the gamification design process of gamified learning environments?”. We conducted an action research, with insights regarding the gamification design process involving both pedagogical and software engineering teams. We opted to conduct an action research method, where the researcher participates collaboratively with the participants to change their practices and understanding of working situations [5]. In the action research approach, a joint intervention between the researcher and the participants of the study occurs, aiming at improving existing practices [4]. In the participatory design, the participants have agency to change and control the design process, rather than just be consulted to provide data (Sanders, 2002). Through the participatory design approach, the instructors and engineers are also part of the process of designing the gamified strategies that will be implemented. Following the types of gamification defined in Kapp [6], we defined a gamified structural design, and a content design. According to [6], a structural design is focused on the frame where the learning content is inserted (in this case, the gamified educational system), while the content design is focused on how the content is

92

A. Toda et al.

presented to the students. Since this study is focused on the design of gamified strategy focused on an educational system, we defined some roles that would be important in the gamification design process as: the gamification designer; the teachers and/or instructors; the requirements engineer; and the stakeholders of the system. The gamification designer (GA) is responsible to propose, design, and validate the gamification strategies. In the context of this work, a gamification strategy is an educational task which contains the learning objectives and the interactions between the game elements [16]. The teacher and/or instructor is the practitioner, responsible for designing and developing the learning materials that would be inserted in the learning platform. As for the requirement engineer, representing the software engineering team, they are responsible to translate the gamified tasks into requirements that will be used by the software team that will develop the educational gamified system. The stakeholders were responsible to validate the gamified strategies as well as to provide materials to the instructional designers to develop the learning content. In this case, the instructional designer, requirements engineer, and stakeholders are the EAs group. In this study, we explain how the roles of teacher/instructor and requirements engineer can impact in the gamified strategies that are designed, and how participatory design can benefit the design of these strategies, not focusing on a single role to conduct this task. Initially, the instructional designer produces the content that will be used in the platform, similar to the original approach. Following, the GA can use the content as an input to design the gamified strategies. The instructional designer participates in the process alongside the GA. Based on the gamified strategies that are designed, the instructional designer can provide inputs that are concerned with the Instructional Design theories and change or adapt what is necessary until they reach an agreement on the final gamified strategy. Then, the requirements engineer is responsible to translate the gamified strategy to system requirements to be implemented, which also differs from the original proposal of the aforementioned approach. The GA is responsible to work with the requirements engineer to create the requirements to be implemented in the final version of the system. After updating the requirements engineering documentation with the gamified strategies, the stakeholder is responsible to validate the entire process and return to whichever step that might need some rework. Finally, after the approval of the stakeholder, the gamified strategies can be passed to the software team in the Implementation phase. The team of GAs was composed of 2 gamification designers, with more than 4 years of experience working on gamification design and more than 5 years working in the field of education. One of the designers also had some experience (more than 5 years) working with software development. The EA team of teaching was composed of 3 higher education instructors with more than 5 years of experience working in the design and development of learning materials using Instructional Design. The requirements team was composed of 2 software engineers with more than 3 years experience working on the field of software development.

Using Participatory Design to Design Gamified Interventions in Educational. . .

93

The structural design is focused on presenting information to the students of the system. It is important for this kind of public to be recognised by their academic achievements, so we opted to use the following elements: Progression, Acknowledgments, and Objectives (concepts that were retrieved from the Taxonomy created by Toda et al. [17]. Progression, in this sense, is used so the student can visualise their advance in the content, it is represented through the use of progression bars that track their progress in the content, as well as their medals (how many medals they have obtained within a given learning content); Acknowledgment is the present through the medals the student might acquire by interacting and progressing with the learning content. In the structural design, Acknowledgment is represented through a medal board, where the student can track which medals they have achieved in the content; Finally, the Objectives are shown through a list of missions where the student has access to the tasks they are completing that are referred to the content. This list of missions is always accessible to the student so they can check which missions they still have to complete. For the content design, we focused on aligning the activities that were planned within the content to use game-like elements. There were two types of activities: (a) quizzes activities and; (b) simulation activities. Quizzes consisted of questions that would appear to the student while interacting with the content. For these activities, we chose the set of elements consisting of Storytelling, and Renovation. The Storytelling element is presented through the insertion of fictional characters in the learning environment, that would guide the learning content and present some questions to the students, based on the content they were studying. Renovation is presented through the chances that are given to the student to choose the correct answer in the quizzes. If a student picks a wrong answer, they will have the opportunity to re-do that question to achieve a better Progression. Storytelling is implied to induce immersion in the learning content while Renovation may ease the feeling of the student that they are not allowed to fail in a learning environment. Both of these elements are aligned with the intentions of the ID involved in the designed learning content. For simulation activities, which consisted in a series of choices that would be affected by the previous ones, to show the student the consequences of their actions in applying the content that they were exposed to. For these activities, we used Imposed Choice, Points, Stats, Renovation, and also Storytelling. Imposed Choice is presented through the choices the student is given in each situation. To advance to the next situation, the student must pick a choice amongst the ones that are available. This choice influences the next situation which the student is exposed to, which is aligned with the immersion and practical tasks that the student is needed to be able to perform after interacting with the content. Each situation has some choices, one that is correct, one that is half-correct, and one or many that are wrong. For each correct or half-correct answer, the student gains stars (representation of Points in this context). At the end of the simulation, the student receives a report of their performance based on their choices, and how many stars they accumulated (representation of Stats). Sometimes, the student might choose too many incorrect

94

A. Toda et al.

answers, and this may lead to a premature ending, closing the simulation and presenting their feedback. However, similar to the quiz, the Renovation element guarantees that the student has another chance to take on the simulation, knowing their wrong choices. The Storytelling is presented in the same manner as the first task. These elements also influence the systems’ structural design elements as Objectives, Acknowledgments, and Progression, since interacting with those activities earn the students a medal, are aligned with their objectives, and also counts towards their progression. These elements were chosen based on the instructional objectives defined by the instructor, and also aligned with the set of elements present in the work of Palomino et al. [10]. All these strategies were designed through sessions between the EAs. The instructional designers assured the gamification elements would be aligned with the instructional objectives of each one of the learning content that was designed, while the requirement engineers could translate those strategies into requirements to be validated with the stakeholders. The validation occurred in two steps, first we conducted an internal validation with the team of EAs in weekly meetings. The GAs were responsible to draft the initial strategies, then present these strategies to the group and receive the feedback from both fronts. The instructional designers were responsible to assure that the instructional objectives would be inherent in the strategies. After this step, the strategy would be translated by the requirements engineers to requirements that would then be passed and discussed with the user experience (UX) team to design the UI prototypes that, along with the requisites would be used by the implementation team. The requirements engineers also helped in designing the strategies so that they could be aligned with the implementation team capabilities and skills (otherwise, the strategies could become too complex to implement). After the two groups reached an agreement through the weekly sessions, another session with the stakeholders would be scheduled to present the results. For each of the strategies that were developed (3) we had at least one meeting with the stakeholders. The strategies were presented as they were planned and the stakeholders could give their feedback. For the gamified structural design, the stakeholder had a positive reception towards how the gamified ‘frame’ of the platform was designed, however they were reluctant due to the lack of visualisation on how those game elements would be presented. Based on this initial feedback, the other gamified strategies that were part of the gamified content design would be accompanied by a simple wire frame. Considering the content design, we presented the first strategy focused on Storytelling and Progression, which was praised by the stakeholder, stating that it would be perfectly aligned with the objective of engaging the learners into the content. The stakeholder had some observations regarding the content, but not the gamification itself. As for the second design, based on Imposed Choice, Points, Stats, and Renovation, we presented an animated demonstration on how a simulation would occur. Again, this design was praised by the stakeholders, stating that these elements, aligned with the content that was created, were adequate for the objectives of the learning environment that was being developed. Based on the feedback given

Using Participatory Design to Design Gamified Interventions in Educational. . .

95

by the stakeholder, the requirements engineer could pass on the requirements to the development team and thus began the implementation phase of the gamified strategies. It is important to note that this process only covers the gamification design process of the whole software development cycle in a specific context of a real educational system development. This process does not cover other processes that are aligned with the entire software development (e.g. the development of the database, or the knowledge structure of an educational system).

4 Concluding Remarks In this chapter we presented some concepts related to participatory design, presenting some cases on how we can use this type of approach to support the gamification design process. Two case studies were presented alongside a discussion proposing guidelines on how this can be used in other types of learning environment. It is important to consider the students and other education actors as part of the gamification design process to achieve better involvement of these same users in the gamified interventions.

References 1. Alkassim, R.S., Tran, X., Rivera, J.D., Etikan, I., Abubakar Musa, S., Sunusi Alkassim, R.: Comparison of Convenience Sampling and Purposive Sampling. Am. J. Theor. Appl. Stat. 5(1), 1–4 (2016). https://doi.org/10.11648/j.ajtas.20160501.11 2. Bai, S., Hew, K.F., Huang, B.: Does gamification improve student learning outcome? Evidence from a meta-analysis and synthesis of qualitative data in educational contexts. Edu. Res. Rev. 30, 100322 (2020) 3. Bittencourt, I.I., Costa, E., Silva, M., Soares, E.: A computational model for developing semantic web-based educational systems. Knowl. Based Syst. 22(4), 302–315 (2009) 4. Filippo, D., Roque, G., Pedrosa, S.: Pesquisa-ação: possibilidades para a informática educativa. Metodologia de Pesquisa Científica em Informática na Educação: Abordagem qualitativa de Pesquisa 3 (2018) 5. Hammond, M., Wellington, J.: Research Methods: The Key Concepts. Routledge, Oxfordshire (2012) 6. Kapp, K.M.: The Gamification of Learning and Instruction: Game-Based Methods and Strategies for Training and Education (2012). ACM, New York. http://dl.acm.org/citation.cfm? id=2378737 7. Klock, A.C.T., Gasparini, I., Pimenta, M.S., Hamari, J.: Tailored gamification: a review of literature. Int. J. Hum.-Comput. Stud. 144, 102495 (2020). https://doi.org/10.1016/J.IJHCS. 2020.102495 8. Mora, A., Riera, D., González, C., Arnedo-Moreno, J.: Gamification: a systematic review of design frameworks. J. Comput. Higher Edu. 29, 516–548 (2017). https://doi.org/10.1007/ s12528-017-9150-4 9. Muller, M.J., Kuhn, S.: Participatory design. Commun. ACM 36(6), 24–28 (1993)

96

A. Toda et al.

10. Palomino, P.T., Toda, A.M., Oliveira, W., Cristea, A.I., Isotani, S.: Narrative for gamification in education: why should you care? In: 2019 IEEE 19th International Conference on Advanced Learning Technologies (ICALT), vol. 2161, pp. 97–99. IEEE (2019) 11. Rodrigues, L., Toda, A.M., Palomino, P.T., Oliveira, W., Isotani, S.: Personalized gamification: a literature review of outcomes, experiments, and approaches. In: Eighth International Conference on Technological Ecosystems for Enhancing Multiculturality. ACM International Conference Proceeding Series pp. 699–706 (2020). https://doi.org/10.1145/3434780.3436665 12. Rodrigues, L., Toda, A.M., Oliveira, W., Palomino, P.T., Vassileva, J., Isotani, S.: Automating gamification personalization to the user and beyond. IEEE Trans. Learn. Technol. 15(2), 199– 212 (2022) 13. Rosenzweig, E.: Successful User Experience: Strategies and Roadmaps. Morgan Kaufmann, Burlington (2015) 14. Tenório, K., Dermeval, D., Monteiro, M., Peixoto, A., Silva, A.P.D.: Exploring design concepts to enable teachers to monitor and adapt gamification in adaptive learning systems: a qualitative research approach. Int. J. Artif. Intell. Edu. 32(4), 867–891 (2022) 15. Toda, A.M., Valle, P.H.D., Isotani, S.: The dark side of gamification: an overview of negative effects of gamification in education. In: Communications in Computer and Information Science, vol. 832, pp. 143–156. Springer, Cham (2018). https://doi.org/10.1007/978-3-31997934-2_9 16. Toda, A.M., do Carmo, R.M., da Silva, A.P., Bittencourt, I.I., Isotani, S.: An approach for planning and deploying gamification concepts with social networks within educational contexts. Int. J. Inf. Manag. 46, 294–303 (2019). https://doi.org/10.1016/J.IJINFOMGT.2018. 10.001 17. Toda, A.M., Klock, A.C.T., Oliveira, W., Palomino, P.T., Rodrigues, L.L., Shi, L., Bittencourt, I., Gasparini, I., Isotani, S., Cristea, A.I.: Analysing gamification elements in educational environments – using an existing gamification taxonomy. Smart Learn. Environ. 6(1), 16 (2019). https://doi.org/10.1186/s40561-019-0106-1 18. Toda, A.M., Oliveira, W., Klock, A.C., Palomino, P.T., Pimenta, M., Gasparini, I., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.I., Shi, L., Gasparini, I., Isotani, S., Cristea, A.I., Shi, L., Bittencourt, I., Isotani, S., Cristea, A.I.: A taxonomy of game elements for gamification in educational contexts: proposal and evaluation. In: IEEE 19th International Conference on Advanced Learning Technologies (ICALT), pp. 84–88 (2019). https://doi.org/10.1109/icalt. 2019.00028 19. Yacobson, E., Toda, A., Cristea, A.I., Alexandron, G.: Encouraging teacher-sourcing of social recommendations through participatory gamification design. In: International Conference on Intelligent Tutoring Systems, pp. 418–429. Springer, Berlin (2021) 20. Yacobson, E., Toda, A.M., Cristea, A.I., Alexandron, G.: Assisting teachers in finding online learning resources: the value of social recommendations. In: International Conference on Artificial Intelligence in Education, pp. 391–395. Springer, Berlin (2022)

Data Mining in Gamified Learning Luiz Rodrigues and Armando Toda

1 Introduction The use of gamification elements in learning environments have increased in the past decade due to the positive tendency they can achieve in those same environments [3]. This led studies to focus on the design step of gamification, where researchers and designers are focused on understand the best ways to design gamified strategies,1 such as frameworks [24]. On one hand, these frameworks can support the design process by presenting concepts that are tied to the gamification requirements to be effective on students’ engagement and motivation [26]. On the other hand, these frameworks do not always provide support to teachers and other education professionals when considering the creation of gamified strategies, e.g. how to select the best elements to a given context [32]. This led to the emergence of a field called Data-Driven Gamification Design (or DDGD), which is the use of data-driven algorithms to support the gamification design process [23]. In this sense, data mining algorithms can be used to explore and understand the best combination of gamification elements to be applied in a given

1 In

this sense, a gamified strategy is a task tied with gamification elements [40]

L. Rodrigues () Center for Excellence in Social Technologies (NEES), Federal University of Alagoas, Maceió, Brazil e-mail: [email protected] A. Toda University of Sao Paulo, Institute of Mathematics and Computer Science, Sao Paulo, Brazil e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. Toda et al. (eds.), Gamification Design for Educational Contexts, https://doi.org/10.1007/978-3-031-31949-5_7

97

98

L. Rodrigues and A. Toda

context (e.g., [35]). In addition, allied with recommender systems,2 these can be powerful tools to support the gamification design process by teachers and education professionals, by providing ”ready-to-use” gamified strategies that can be deployed in many environments [43]. Considering the exposed, this chapter presents a broad overview of DDGD ranging from a data mining project’s planning to directions for future research. Specifically, Sect. 2 demonstrates how to plan and conduct, as well as important information to report on a data mining project. Then, Sect. 3 introduces helpful tools for conducting such projects, including alternatives for those with coding skills and those without. Next, Sect. 4 discusses two DDGD studies along with code examples to replace them. Based on that context, Sect. 5 provides practical recommendations for DDGD projects and Sect. 6 discusses directions for future research. Lastly, Sect. 7 draws our final considerations.

2 Data Mining Project The CRoss Industry Standard Process for Data Mining (CRISP-DM) is widely used for data science and is specially recommended for goal-oriented projects [47]. Because gamification must be designed with a specific goal in mind, we consider CRISP-DM suitable to frame data mining projects in the context of gamification applied to education. According to [47], CRISP-DM is a cyclic, hierarchical process that involves four levels of abstraction: phases, generic tasks, specialized tasks, and process instances. To provide an overview of how CRISP-DM helps framing data mining projects, the remainder of section introduces its six phases, their respective generic tasks (Sect. 2.1), and associated techniques (2.2). See Sect. 4 for hands-on examples of specialized tasks and instances applied in practice.

2.1 Planning and Reporting CRISP-DM’s first phase is business understanding. It concerns understanding the project’s objective and requirements, from a business perspective, to convert them into the the project’s problem and preliminary plan. This phase’s generic tasks are (i) determining business objectives, (ii) assessing the situation (e.g., resources and risks), (iii) defining the data mining goals, and (iv) designing the project plan. The second phase is data understanding. It ranges from collecting to discovering the first insights into the data. This phase is closely connected to the first one as it provides prominent help to establishing the data mining problem and project plan. Specifically, this phase’s generic tasks are collecting, describing, exploring, and verifying the quality of the data available.

2 Systems

that recommend things based on pre-determined analysis.

Data Mining in Gamified Learning

99

The subsequent phase is data preparation. It concerns all activities related to constructing the final dataset, which will be the next phase’s input. Data preparation’s generic tasks are data selection, cleaning, construction, integration, and formatting. CRISP-DM’s fourth phase is modeling. It concerns selecting and applying techniques, such as classification and regression, as well as optimizing their parameters, to find a model that successfully solves the data mining problem. The generic tasks of this phase are selecting the modeling techniques, generating the test design, building the model(s), and assessing the model(s) based on pre-defined criteria, domain knowledge, and test design. The fifth phase is evaluation. Unlike phase four, it concerns evaluating the models from a business perspective. Also, this phase involves reviewing the previous steps to ensure the final model meets the business requirements. This phase’s generic steps are checking whether business success criteria were met, reviewing the process, and defining the next steps (e.g., move to deployment or start new iteration?). Finally deployment is the last phase. Mainly, it concerns transforming the knowledge gained from the final model in a way it is useful to customers/stakeholders. For instance, it might range from writing a simple report to creating a recommender system. Accordingly, deployment’s generic tasks are planning the deployment along with its monitoring and maintenance, producing the final report, and conducting a retrospective analysis of the project’s ups and downs.

2.2 Executing This section briefly introduces techniques for the different phases of CRISP-DM.

2.2.1

Data Understanding

Data understanding tasks concern collecting, describing, exploring, and verifying the quality of the data available. For data collection, surveys provide cheap alternative to achieving large amounts of data, but might lack information from true usage (e.g., capturing potential instead of real experiences). System logs might address this limitation, besides allowing the generation of fine-grained data concerning user behaviors over time. Nevertheless, capturing logs demands a running system with active users. For describing data, examining the document is a valuable starting point to understand properties like number of records, attributes available and their formatting, whether the data is structured, characters encoding, and so on. For data exploration, basic statistics (e.g., counts, mean, standard deviation, and confidence intervals) offer a simple yet reliable description. Next, plotting individual attributes with, for instance, histograms is helpful to further understand their variance and distribution. Other simple plots, such as Boxplots, Scatter Plots, and Barplot are

100

L. Rodrigues and A. Toda

valuable to identify relationships between attributes. Furthermore, line plots and word-clouds are examples that can be used to understand time series and textual data, respectively. For verifying data quality, important steps are looking for missing data, errors that might have occurred while storing the data (e.g., typos or technical issues), coding inconsistencies (e.g., using different codes for the same information), measurement errors (e.g., constructs measured using wrong or unreliable scales).

2.2.2

Data Preparation

Data preparation tasks concern selection, cleaning, construction, integration, and formatting. Data selection refers to choosing valid data sets (as previously checked), as well as documenting the reasons why it enables achieving business objectives. Data cleasing is one of the longest and most important steps. In case of missing data, one might remove them, which would reduce the sample size/dimensionality if rows/columns are dropped. Alternatively, there is data imputation, which can be as simple as using the attribute’s mean value or based on Neighbours similarity and related techniques [14]. In the case of having incorrect values, one needs to similarly deal with them by either correction or removal. Dealing with outliers is another important point. Those should be inspected to ensure they are not products of, for instance, measure or coding error. Then, data could be analysed with and without them to understand their role [1]. For data construction, a classic example is combining height and weight to create a body mass index. This approach can be extended to other domains, such as extracting code patterns in computer science education (e.g., [27]). Accordingly, a subject matter expert is probably the most important asset for this task [47]. There also are purely data-driven techniques for feature construction, or extraction, such as Principal Component Analysis [48], but describing this and similar approaches is outside this chapter’s scope. Data integration can be achieved by combining data from multiple sources. While finding multiple datasets aligned to the business problem, as well as compatible among themselves, is challenging, this offers invaluable benefits by increasing data representativeness. Finally, data formatting will likely be necessary, especially when working with data from multiple sources. This might involve standardising attributes names, transforming numerical values stored as strings to numbers, converting categorical values to numbers (i.e., hot encoding) so modelling algorithms can deal with them, and expanding nominal attributes into boolean ones (i.e., dummy coding) to avoid implicit ordering (e.g., teaching an algorithm that males—if represented by 1—are higher than females—if represented by 0).

2.2.3

Data Modelling

This phase concerns selecting the modelling techniques, generating the test design, building the model(s), and assessing them. Selecting the modelling technique mainly depends on the data mining task, which often is classification (target variable is

Data Mining in Gamified Learning

101

discrete), regression (target variable is continuous), or clustering (target variable are clusters/groups yet unknown). For classification and regression tasks, Linear (logistic) regression is valuable because it is highly interpretable, but cannot model non-linear problems. Decision trees overcome this limitation, but are still limited in terms of predictive power. Ensemble techniques, such as Random Forests, rely on several small decision trees and offer substantial improvements in predictive power. They offer interpretability by indicating each feature’s importance, but how features relate to each other, and how this affects predictions, is complicated by the several decision trees. Artificial Neural Networks provide state of the art predictive power. However, they are often considered black boxes, meaning that interpreting the patterns they find is highly challenging. For clustering, K-means and Hierarchical Clustering are widely used. Their main limitation is that the researcher must define the number of clusters to be found as it is commonly unknown. One can use techniques such as the knee-point to address this limitation. Then, a common practice is to compare clusters based on their averages and/or use Association Rules to identify the main factors leading a sample to be part of a cluster. For further information on such algorithms, please refer to [37]. The main test design are holdout and cross-validation [9]. The former concerns saving part of the dataset to test the model on unknown data. The latter splits the dataset into k folds, creates a model using .k − 1 of them, tests the model in the remaining fold, and then repeats this process until each fold was used for testing. In practice, combining both is the most effective approach: saving part of the dataset to test the model that performed the best on cross-validation. Building the model is straightforward as it refers to, for instance, training a random forest with the parameters that yielded the highest performance in cross-validation. Finally, assessing the models concerns comparing their performance on cross-validation and the testing set, contrasting it to the project’s goals, conducting error analysis (e.g., categories with better/worst performances), interpreting whether attributes importances are coherent with the problem (when applicable), and so on. For an overview on those approaches, please refer to [9].

2.2.4

Evaluation

Evaluation concerns comparing the modelling results to success criteria, reviewing the process, and defining the next steps. To compare the models, one should prioritise performance on unknown data (e.g., test set) when checking if success criteria were met as it demonstrates if the models found generalisable patterns [9]. In reviewing the process, one must ensure all steps were properly executed, summarise the main results, and correct anything needed. Finally, the next step should be move to deploying if there were no errors and the objectives were met, iterate in case reviewing/improvements are needed, or start a new project in case they consider a dead end was reached [47].

102

2.2.5

L. Rodrigues and A. Toda

Deploying

Deploying concerns making the model useful to its stakeholders, producing the final report, and reviewing the whole project. To make the model useful, one might create a presentation (e.g., with an accurate decision tree), design a website for users to input data and visualise the predictions, embed the model in an API (Application Programming Interface) developers can query to get predictions, and so on, depending on the business objectives. In producing the final report, it is prominent to detail all steps followed throughout the projects, the challenges associated with each phase and how they were addressed, and the main data mining results. Finally, the project review should involve a retrospective analysis to identify lessons learned, the positives and negatives, and recommendations to be used in future projects [47].

3 EDM Tools This section introduces four tools that might be used to conduct projects for DDGD from different perspectives, namely, Weka, Orange, R, and Python.

3.1 Weka Weka is an open source collection of several machine learning algorithms for data mining tasks [7]. It is a desktop software compatible with Windows, MacOS, and Linux. Weka offers an interactive user interface (i.e., the explorer), which enables conducting data mining projects regardless of coding knowledge. Nevertheless, it is possible to use it based on a simple command line as well. Figure 1 presents Weka’s explorer, which provides an overview of the several tasks it enables, ranging from preprocessing to visualisation. With Weka, one might work on several phases of a data mining projects. Despite Weka’s native data storage format is ARFF, the explorer is able to read Comma-Separated Value (CSV) spreadsheets. This makes it usable for many data sources as most database programs, such as Excel and Database Management System, often allow exporting data into CSV format. After the data is loaded, the explorer enables selecting the attributes one needs to inspect and, according to its type (e.g., ordinal or numeric), presents descriptive statistics and graphical visualisations. Additionally, the explorer has other data preparation features, such as data filtering (e.g., adding or removing attributes) and transformation (e.g., changing and converting values). Furthermore, Weka offers several algorithms for modelling, such as Trees and Bayesian classifiers, Neural Networks, clustering algorithms, and tools for association rule mining. Moreover, Weka has options to optimising modelling, such as attribute selection algorithms and experimental functionalities.

Data Mining in Gamified Learning

103

Fig. 1 Weka explorer interface. Source: [7]

In summary, Weka is a tool that, once data is collected, offers several functionalities to conducting data mining projects in terms of data understanding and preparation, modelling, and evaluation. It is free to use and features a graphical user interface, besides both simple and advanced algorithms. Hence, it is a valuable alternative for DDGD, especially for those who do not want to have to code.

3.2 Orange Orange is another open source tool, which is focused on machine learning and data visualisation [5]. It features a large, diverse toolbox that enables building workflows visually. Orange is compatible with Windows, MacOS, and Linux, might be installed as a standalone application or based on alternatives such as Anaconda and Python. Additionally, Orange features a portable version that removes the need for installing it. Figure 2 presents a Orange workflow joining two files (File 1 and File 2), which is accomplished using the concatenate widget (i.e., building blocks, such as functions, that enable creating data analysis workflows), to create a new dataset (data table).

104

L. Rodrigues and A. Toda

Fig. 2 Orange visual workflow demonstrating the joining of two datasets. Source: orangedatamining.com/screenshots/

Based on a wide catalogue of widgets, Orange might be used to work on core phases of a data mining project. Concerning data understanding, for instance, Orange features widgets for importing CSV files and SQL databases, checking data information and attributes statistics, and saving data, besides featuring visualisation tools that create boxplots, scatterplots, linear projections, and others. Concerning data preparation, among its several widgets, Orange enables taking random samples from a dataset, filtering data based on attributes or rows, merging and concatenating data, aggregating and grouping values, making attributes discrete, and imputing new values. Concerning modelling, Orange can deal with classification, regression, clustering, and association rule mining. It features several algorithms, such as decision trees, ensembles, support vector machine, and neural networks. Additionally, it features widgets for text mining, survival analysis, bioinformatics, time series analysis, and image analytics, among others. Lastly, Orange also features widgets concerning the evaluation phase, such as those to make predictions, generate confusion matrices, and test modelling performance. In summary, Orange is another valuable tool to conducting data mining projects, especially those with a focus on machine learning. It also is free to use and features several features for the different phases of a data mining project. Furthermore, it is based on visual workflows that do not require coding. The user only needs to select the widgets to use and connect them. Hence, those interested in working with DDGD based on machine learning, who do not know or want to code, will likely benefit from Orange.

Data Mining in Gamified Learning

105

3.3 R R is a software environment aimed for statistical computing and graphics [28]. Similar to the other alternatives, R is free, open source, and might be downloaded and installed on Windows, MacOS, and various UNIX platforms. By default, R is to be used based on scripts and command line. Nevertheless, R Studio is a widely used IDE (Integrated Development Environment) that features several tools aimed to help increasing productivity with R [38]. Unlike Weka and Orange, however, R is not based visual development. As Fig. 3 shows, RStudio provides a single workspace where one can write (top-left) and run (bottom-left) their scripts, inspect the working environment (top-right), and visualise generated plots, among other information (bottom-right). R offers numerous functionalities for conducting data mining projects. By using R’s standard base package, one can already perform several data mining tasks, such as reading CSV files, inspecting descriptive statistics, generating visualisations (e.g., boxplots and scatterplots), transforming data (e.g., filtering and converting), fitting

Fig. 3 RStudio screenshot demonstrating example code for the iris dataset. Code source: stat.ethz.ch/R-manual/R-devel/library/datasets/html/sleep.html

106

L. Rodrigues and A. Toda

models (e.g., linear regressions), and evaluating models (e.g., checking predictions’ accuracy/correlation). Furthermore, R has an active community that constantly contributes packages to it. To cite a few of the several alternatives available, DBI might be used to connect R to relational database management systems, tidyverse and dplyr are widely used for data preparation, ggplot2 is a famous option for data visualisation, randomForest might be used to machine learning-based modelling, multcomp is valuable for models comparison, and shiny and R Markdown facilitate data reporting. Because of this wide, active community, one will likely find a package for any data mining task on R’s repository.3 In summary, R is a free-to-use tool especially valuable for statistics and data visualisation. Compared to the previously discussed alternatives, R differs because it is based on coding instead of a graphical user interface. While this limits its usage for those not familiar with coding, R provides increased options for controlling and exploring each step of a data mining project. Furthermore, due to its wide and active community of contributors, one will likely be able to run any type of data mining task with R. Hence, DDGD researchers and practitioners familiar with or interested in learning to code will greatly benefit from the varied opportunities R offers.

3.4 Python Python is an open source programming language aimed to be friendly and easy to learn [44]. Accordingly, it demands coding skills to be used, similar to R and in contrast to Weka and Orange. Python is also compatible with Windows, MacOS, and Linux, and might be used to create scripts as well as object-oriented programs. To run Python code, one might use its command line as well as several IDEs, such as PyCharm4 and Spyder.5 In the case of Spyder, as Fig. 4 demonstrates, one can simultaneously check the file directory (left), the python scripts (center), the programming environment (top-right), and generated plots, among other information (bottom-right). As a programming language, one might use Python for all tasks of a data mining project. Nevertheless, a valuable asset for Python users is Anaconda [2], a widely used open source distribution platform. Among its several features, Anaconda offers (i) a repository with over 8000 packages for data science and machine learning, (ii) a package and environment manager (i.e., Conda), (iii) a navigador that facilitates managing and integrating applications, packages, and environments through a graphic user interface, and (iv) a cloud environment for backup. To cite a few of the numerous packages that contribute to Python being one of the most used tools for data mining projects, there are Pandas and Numpy for data manipulation, matplotlib

3 https://cran.r-project.org/web/packages/available_packages_by_name.html. 4 https://www.jetbrains.com/pycharm/. 5 https://www.spyder-ide.org/.

Data Mining in Gamified Learning

107

Fig. 4 Screenshot of Spyder, an IDE for the Python programming language. Source: https://www. spyder-ide.org/

for generating visualisations, scikit-learn for machine learning, and TensorFlow and PyTorch especially for deep neural networks. In summary, as a general-purpose programming language, Python is by default a viable option for any data mining task. However, its popularity in this domain is mainly supported by the several tools and packages it features, with most of those being easily accessible from Anaconda. On the other hand, it demands coding skills to be used. Hence, for those interested in working with DDGD, Python will be a valuable tool for the varied range of data mining tasks available as long as you are familiar with coding or willing to learn it.

4 Hands-on Data-Driven Gamification This section describes two projects using data mining to design gamification and present. For each, we discuss task alternatives, how they relate to CRISP-DM’s phases, and demonstrate sample codes on how specifics tasks might be coded.

108

L. Rodrigues and A. Toda

4.1 GARFIELD: A Regression-Based Recommender System GARFIELD is a recommender system that helps personalising gamified learning by suggesting which game elements to use for a given user to achieve a desired level of intrinsic motivation [35]. It was originally developed following CRISP-DM and using R.6 Next, we briefly describe each CRISP-DM phase with code examples of how to implement then using R. Please refer to [35] for details. The business understanding phase does not involve any coding. In this one, the authors defined the project goal (i.e., create a model to personalise gamification based on learners’ intrinsic motivations captured after using a gamified educational systems) and requirements (i.e., the model should (i) consider learners’ characteristics and (ii) be interactive). In the data understanding phase, the authors used a dataset publicly available. Next, the authors discussed the overall distribution of the dataset’s attributes. In R, one can replicate these steps with the following code. l i b r a r y ( r e a d x l ) # imports l i b r a r y to read x l s x f i l e s d s