Content uploaded by Antti Knutas
Author content
All content in this area was uploaded by Antti Knutas on Apr 25, 2018
Content may be subject to copyright.
Creating Student Interaction Profiles for
Adaptive Collaboration Gamification Design
Antti Knutas, Lappeenranta University of Technology, Lappeenranta, Finland
Jouni Ikonen, Lappeenranta University of Technology, Lappeenranta, Finland
Dario Maggiorini, Università di Milano, Italy
Laura Ripamonti, Università di Milano, Milan, Italy
Jari Porras, Lappeenranta University of Technology, Lappeenranta, Finland
ABSTRACT
Benefits of collaborative learning are established and gamification methods have been used to
motivate students towards achieving course goals in educational settings. However, different
users prefer different game elements and rewarding approaches and static gamification
approaches can be inefficient. We present an evidence-based method and a case study where
interaction analysis and k-means clustering are used to create gamification preference profiles.
These profiles can be used to create adaptive gamification approaches for online learning or
collaborative learning environments, improving on static gamification designs. Furthermore, we
discuss possibilities for using our approach in collaborative online learning environments.
Keywords: Collaborative learning, interaction analysis, profiling, k-means clustering,
gamification, adaptive gamification, software engineering education
INTRODUCTION
Collaborative learning, or the cooperative activity of students working together towards a
specific learning goal with the teacher as a facilitator (Bruffee, 1995; Dillenbourg, 1999b), has
become an increasingly topic important in education (Okamoto, 2004). This collaborative
approach to education has been shown to develop critical thinking, deepen the level of
understanding and increase the shared understanding of the material (Gokhale, 1995; D. W.
Johnson & Johnson, 1999; R. Johnson & Johnson, 1994). Computer-supported collaborative
learning (CSCL) extends and facilitates this cooperation by using electronic communication
tools (Dillenbourg, 1999a). CSCL has several benefits, including wider participation for
knowledge building, and improved student productivity and satisfaction (Resta & Laferrière,
2007).
Computer supported collaboration is also essential in software engineering education, because
working and efficiently collaborating teams is at the basis of software engineering industry
(Coccoli, Stanganelli, & Maresca, 2011). The impact of collaboration has been studied in both
physical classrooms (Alaoutinen, Heikkinen, & Porras, 2012) and in online environments
(Dewiyanti, Brand-Gruwel, Jochems, & Broers, 2007) with positive course outcomes. However,
the people who benefit most from this collaboration do not always interact (Knutas, Ikonen, &
Porras, 2013).
In recent studies it has been shown that students can be guided towards educational goals like
collaboration by using gamification (Glover, 2013), which is the application of game-like
elements to non-game environments (Deterding, Dixon, Khaled, & Nacke, 2011; Groh, 2012).
Approaches that use some elements of gamification have been shown to increase student
collaboration (Moccozet, Tardy, Opprecht, & Léonard, 2013) and the motivation towards
achieving course goals (Sheth, Bell, & Kaiser, 2012) in educational settings.
Although we instinctively recognize than games and fun are tightly related, both concept and
their interrelation are quite slippery to define (Caillois, 1961; Crawford, 1984; Huizinga, 1950;
Juul, 2003; Rollings & Adams, 2003; Salen & Zimmerman, 2005). The investigation of these
issues has led neuroscientists and cognitive psychologists to examine how playing a game and
learning are connected (S. Johnson, 2004; Miller, 1956). The basic observation is that humans
have always used games as playgrounds for learning and exercising safely specific skills. During
this process, human brain secretes endorphins (which makes a game an enthralling and fun
activity), is highly focused on recognizing recurring patterns in problems, and on creating
appropriate neural routines to deal with them. Once the pattern is fully caught by the player, the
game becomes boring, but the skill has been accurately acquired. In a certain sense, we could say
that “Fun is the emotional response to learning” (Crawford, 2003) and that the first and main
reason for a (video) game to exist is to provide fun to its players (Koster, 2005), that is achieved
not only through alluring game mechanics, but also by providing an environment that fosters
immersivity (Csikszentmihalyi, 1991; Fullerton, 2008; Salen & Zimmerman, 2004).
Our proposed solution is to use an evidence-based method for deciding which gamification
elements to apply and how to apply them. In this method we build collaborative behavior profiles
for students by using interaction analysis and teamwork profiling surveys. These profiles and the
collected profiles of interactions can be used to model how different students react to
gamification elements and the available goals in order to create or improve adaptive gamification
systems.
In this paper we detail our profiling method and present a case study where we profile the
collaborative behavior patterns of students who participated in a software engineering course.
We also present our plan of how to use these profiles to create custom user-centric gamification
approaches for a gamification system, with an ultimate goal of using them to improve
collaboration in CSCL environments. Specifically, our research questions in this study are:
1. What kind of collaborative interactions are present on a collaborative software engineering
course?
2. Do these interactions have repetitive patterns that can be used for profiling?
3. Which team worker roles and gameplay styles the profiled students prefer?
4. How can these profiles be used to improve gamification systems?
In the next section we review previous approaches to gamification in education. After that we
detail our research setup, methods of analysis and the research results. In discussion we consider
the implication of these results and how these results can be applied to a gamification CSCL
system. The paper finishes with conclusions.
GAMIFICATION APPROACHES IN SOFTWARE ENGINEERING EDUCATION
Gamification, at its core, is the use of game design elements in a non-game environment
(Deterding et al., 2011). However, effective gamification is about using these game elements
along three important principles, in order to satisfy users’ three innate needs for intrinsic
motivation (Groh, 2012), which were originally adapted by Deterding (2011) and Schell (2011)
from Deci and Ryan’s (1985) self-determination theory. The three principles are (Groh, 2012):
Relatedness: The universal need to interact and be connected with others.
Competence: The universal need to be effective and master a problem in a given
environment.
Autonomy: The universal need to control one’s own life.
These elements have been shown to be satisfied by gamification elements by Rigby and Ryan
(2011), where they also demonstrate the connection between the self-determination theory and
gamification.
Gamification is an upcoming method in learning, with ongoing research in its use as
motivation or engagement method (Barata, Gama, Jorge, & Goncalves, 2013; Glover, 2013;
Herranz et al., 2015; Sheth et al., 2012), for fostering collaboration (Moccozet et al., 2013;
Romero, Usart, Ott, & Earp, 2012; Leif Singer & Schneider, 2011; L. Singer & Schneider, 2012)
or engaging by competition (Dubois & Tamburrelli, 2013; Yamawaki, 2013). It has also been
used as an engagement method in the increasingly popular Massive Online Open Courses
(MOOCs) (Romero & Usart, 2013; Tan, 2013). However, the successful application to
gamification methods is not necessarily a straightforward affair, with the application of those
methods sometimes negatively affecting the course (Berkling & Thomas, 2013). The Table 1
summarizes the gamification methods applied in each of the previously mentioned studies.
Table 1. Recent gamification approaches used in education studies
Study
Scope
Methods
Conclusions
Barata et al., 2013.
“Engaging Engineering
Students with
Gamification”
2-year long
study with an
engineering
course
Added experience
points, levels, challenges
and a leaderboard for
progressing through the
course.
Participation and
proactivity increased in
the course. Students
were more engaged.
Berkling and Thomas,
2013. “Gamification of
a Software Engineering
Course and a detailed
analysis of the factors
that lead to it’s failure”
1-year long
study with a
software
engineering
course
Added study paths,
project unlocking, levels,
points and marketplace-
style collaboration.
The game platform
collided with a formal
culture. The students
looked for more strict
structure and did not
always answer to the
gamification-based
challenges.
Dubois and Tamburrelli,
2013. “Understanding
Gamification
Mechanisms for
Software Development”
A software
development
course
Added a competitive
scoring system based on
assignment scores that
was publicized to a
leaderboard.
Results are not
conclusive, but they
suggest that scoring
and leaderboards
increase motivation by
competition.
Moccozet et al., 2013.
“Gamification-based
assessment of group
work”
One year long
bachelor course
Adding rewards for
interaction and sharing
resources: intra-group,
inter-group or at global
class levels. Rewarding
the users for contributing
to the global platform
knowledge.
The system encourages
collaborators to
contribute and interact.
Herranz et al., 2015.
“Gamiware: A
Gamification Platform
for Software Process
Improvement”
Bachelor level
course in
software
development
and
management
Introducing a role-based
gamification system that
has competition,
cooperation and
challenge –based
mechanics.
The introduced
gamification system
increased motivation
and student satisfaction
in a pilot study.
Sheth et al., 2012.
“Increasing Student
Engagement in
Software Engineering
with Gamification”
A software
testing course
Using operant
conditioning by
providing simple
rewards for completing
tasks (experience,
currency and
achievement titles).
Added a MMORPG-like
structure.
Preliminary data shows
indications that
gamification increases
student motivation.
Studies in the field indicate that gamification methods are successful in fostering
collaboration. This applies when some gamification aspects relate to the internal motivational
elements of relatedness, competence or autonomy. This is especially evident in studies by
Moccozet et al. (2013) and Dubois & Tamburrelli (2013), where collaboration was rewarded or
the course participants were able to publish their competence and compare the results to their
peers. Herranz et al. (2015) further propose that when gamifying collaboration it is essential to
have supporting teamwork structures in place in order to have a positive impact on motivation.
The elements of success in these studies are the fact that they connect the users’ achievements to
a meaningful community who share some of the user’s personal goals. Additionally, by
providing variance the approaches in these studies allow the user to retain their personal
autonomy, by allowing the users to choose how exactly to achieve their goal. At the same time
individual elements of gamification have been studied and recent research concludes that simply
applying a single outward aspect of gamification, like badges (Falkner & Falkner, 2014), does
not work and instead gamification has to consider the motivation and goals of the course as a
complete system.
RESEARCH METHOD
In this study we observed 17 students over a five day long intensive format and collaborative
software engineering course. The course had 10 hours of lecturing and 40 to 64 hours (depending
on the student team) of collaborative teamwork around a set task. The topic for the course was to
develop a new mobile or tablet application before Friday’s deadline and the students had no
other courses concurrently. The students spent their time in the same computer classroom, with
each student team sitting at their own table group.
All student interactions that occurred in the classroom were recorded. The video and audio
recordings were combined into multi-angle and surround sound videos that allowed the
researchers reviewing the video to hear several concurrent interactions. This resulted in 40 hours
of video, from which 3366 interactions were recorded for analysis. To gain additional
information about preferred teamwork roles, the students were interviewed and were asked to fill
surveys about their teamwork methods. All 17 students participated in the interviews and 15
students participated in the surveys.
Analysis Process
The main source of data for statistical and network analysis was interaction analysis based on the
classroom interactions. The problem of analyzing classroom interactions into quantitative data
has been approached with interaction analysis and Social Network Analysis (SNA). While SNA
is concerned about who communicates with whom (Otte & Rousseau, 2002), interaction analysis
inspects how people communicate with each other (Jordan & Henderson, 1995). More
specifically, interaction analysis is an interdisciplinary method for investigation of interactions of
human beings (Jordan & Henderson, 1995). Interaction analysis has also been used to examine
student collaboration behavior by identifying common interaction patterns with k-means
clustering and correlation analysis (Serçe et al., 2011). Our research approach combines
interaction analysis and social network analysis by first using interaction analysis to gain
information about communication activities and then using social network analysis to
quantitatively model the collected data as a graph, from which further statistics can be generated.
The activities that were detected by observing interactions were classified according to
activity types defined by Vivian et al. (2013). In their study they also introduced an approach for
coding collaborative interactions in CSCL using team collaboration analysis roles originally
defined by Dickinson and McIntyre (1997). To summarize, these interaction types are: Team
leadership, which involves providing direction, structure and support for other team members.
Team orientation, which refers to attitudes that members have towards one another and the team
task. Social, non-professional communications were included under this category. Monitoring,
which is observing other team members’ performance or activities. Coordination, which
involves process reporting and goal setting. Profession or learning related communications,
which involves the exchange of information in a prescribed manner and by using proper
terminology. Additional behaviors introduced by Vivian et al. (2013) were also used: Seeking,
receiving or giving feedback about performance and seeking help (seeker) or receiving help
(supporter).
The list of analyzed interactions was collated, resulting in a frequency distribution of
interactions by type and by interaction target for each person. These lists give individual
communication profiles between the students, but it is not immediately apparent from these
individual profiles if there are repeating patterns in the student interactions. To find these
patterns k-means clustering was used, which is a statistical analysis method for automatically
partitioning a dataset into a specified amount of groups (Wagstaff, Cardie, Rogers, & Schrödl,
2001).
Finally, in order to gain further insight of which kind of teamwork and gameplay the students
would prefer, two different profiling methods were applied: Belbin’s team work inventory
(2010) and Bartle’s (2004) classification of player types. Belbin’s team role inventory divides the
participants to three major categories (action, people, cerebral) and each of the major categories
into three subcategories. Bartle’s player type classification divides the player types into two
separate axes: Whether the player prefers to act or interact, and whether the player prefers to
interact with other players with the world. Players who act towards other players are called clubs,
players who act towards the world are named diamonds. Players who interact with the world are
named spades and players who interact with other players are hearts. The division of player
types is visualized in the Figure 1, where the player types are divided by the interaction target
and interaction type axes.
Figure 1. The two axes of Bartle player types, adapted to learning environments
Analysis Results
3366 interactions were analyzed based on the student communication interactions, of which
81.79% were internal team interactions and 18.21% were to communications to outside the team
(external). The most common internal interaction type was communication (42.93%) and the
most common external interaction type was team orientation (7.90%).
K-means clustering analysis with Pearson’s correlation coefficient as a distance measure
resulted in four clusters of student profiles that share same communication behaviors. The
average silhouette coefficient for the resulting clusters is 0.64, which means that the data points
group well and are distinct from each other, but not perfectly (Kauffman & Rousseeuw, 1990).
These clusters are detailed in the Table 1, which lists the clusters CL1 to CL4, their members and
the most commonly occurring profiles in these clusters. Individual team members are first
labelled by their group alphabet and then their number within the group. Additionally, each
cluster has been assigned a nickname that characterizes roles the cluster’s profiles tend to take in
collaborative interactions.
Table 2. Student profile clusters
Cluster ID;
(Nodes);
Nickname
Belbin Profiles
Bartle
Profiles
Most Common
Internal Actions
Most Common
External Actions
Members
CL1
(A4, B4)
“Cooperative
workers”
Resource
Investigator;
Coordinator
Heart;
Club
Communication
(33.05%);
Team
Orientation
(8.91%);
Coordination
(8.05%)
Communication
(19.54%);
Team
Orientation
(9.48%);
Monitoring
(7.66%)
2
CL2
(A1, A3, C1,
D1, D2)
“Social team
members”
Coordinator;
Resource
Investigator;
Team Worker
Heart;
Club;
Spade
Communication
(32.26%);
Monitoring
(16.04%); Team
Orientation
(13.18%)
Team
Orientation
(11.61%);
Monitoring
(7.37%)
5
CL3
(A2, B3, D5)
“Achievement-
oriented
leaders”
Implementor;
Monitor
Evaluator;
Resource
Investigator
Diamond;
Club
Communication
(41.74%);
Supporter
(12.26%);
Team Leadership
(11.72%)
Communication
(7.64%);
Team
Orientation
(4.44%)
3
CL4
(B1, B2, C2,
C3, C4, D3,
D4)
“Task-oriented
worker”
Complete
Finisher;
Implementor;
Plant; Team
Worker
Diamond;
Spade
Communication
(54.65%);
Team
Orientation
(10.77%);
Coordination
(7.62%)
Team
Orientation
(6.01%)
7
At a first glance the most common internal and external actions in these profile clusters
appear similar. However, when one looks at the distribution of all tasks, presented in the Figure
2, the specialties and differences in the clusters become more apparent. For example, profiles in
Cluster 2 (“task-oriented workers”) communicate mostly internally and about the task at hand.
Compared to cluster 2, profiles in Cluster 1 (“cooperative workers”) are much more prone to
discuss the task with other teams and monitor what they are doing. While profiles in Cluster 3
(“achievement-oriented leaders”) also engage in professional communication, they are the only
profile type who exhibit significant levels of team leadership or the support type of
communication.
Figure 2. Profile cluster interaction distributions
The communication patterns inside the classroom can also be modelled visually as a graph by
using social network analysis (Bastian, Heymann, & Jacomy, 2009; Scott, 2012), as presented in
the Appendix A. Both the clustering and the graph visualization are based on the same dataset
and the visualization presents the structure and patterns of team communication in a more
interpretable manner. In the visualization each node represents a student and each line
communication between the students. Each node is also tagged with their profile cluster type.
The presented lines, or edges, of communication are unidirectional, with the direction of
communication progressing clockwise. If the communication is bidirectional, there are two edges
connecting the nodes, each circling in a clockwise direction. Each edge is tagged with the most
common type of communication between the connected nodes. For clarity internal team task-
related communication and external task orientation types have been removed from the graph,
because all team members frequently communicated with each other and team orientation does
not relate to the learning or team tasks.
From the social network analysis graph in Appendix A it can be seen that all teams
communicated with each other, with team members communicating about the task at hand or
technical issues, and monitored each other to see possible solutions or how the other teams were
progressing at problem solving. One notable fact is also that the students of cluster type three
(“achievement-oriented leaders”) are distributed evenly, with a maximum of one leader emerging
per team. Another notable feature in the communication patterns is that students do not only seek
help from inside their own team. Several nodes can be seen establishing seeking for help and
support/lead pairs with nodes external to their own teams. This suggests towards the fact that the
right kind of profile types can be encouraged to cooperate and support students outside their
immediate group.
Describing the Detected Profile Clusters
Cluster 1: “Cooperative workers”
Profiles in Cluster 1 concentrated most on professional communications, which essentially
means that they mostly communicated about software engineering tasks at hand. The difference
to the other worker-type profile cluster, the Cluster 4, is clear: The more collaborative Cluster 1
members identify themselves as social (resource investigator) or coordinator type of team
workers. The second distinctive feature when compared to the other work-oriented profile cluster
is that the profiles of Cluster 1 also cooperated outside their own team and examined
(“monitored”) often what other teams were doing.
Cluster 2: “Social team members”
The second most distinct cluster is the Cluster 2. These team members identify themselves as
seekers of new information (resource investigator) or team organizers (coordinator). Their
Bartle profile, hearts, describes their activity preferences well. They socialize and watch more
than the more directly productive activities of reporting and organizing. However, each team
needs to stop, reflect and examine their goals and the value of these kind of profiles should not
be underestimated. When considering applicable gamification approaches, these types of profiles
could be motivated to use their social skills more constructively and then report the solutions
forward from other groups to members of the Cluster 3, so that the shared knowledge would
spread further.
Cluster 3: “Achievement-oriented leaders”
Perhaps the most interesting case are the members of the Cluster 3. These students identify
themselves as people who get things done (implementor), or critical, logical thinkers (monitor
evaluator). However, they exhibited most leadership and supporting actions. They also listened
least to feedback and requested the least amount of help. In short, these people could be
characterized as people who want to get things done and have seized the opportunity to lead
people towards practical goals. Their weakness is getting little input and advice from others. If
there was a reward system for seeking more new information and group feedback, it could be
targeted towards the members of this cluster group, encouraging them to be more inclusive in
decision-making.
Cluster 4: “Task-oriented workers”
Profiles in Cluster 4 concentrated most on professional communications, which essentially
means that they mostly communicated about software engineering tasks at hand. The difference
to Cluster 1 that the members of the first cluster also collaborated with other team members and
members of the Cluster 4 concentrated almost solely on working with their own team members.
The difference between the cluster teamwork profiles is also clear: The more collaborative
Cluster 1 members identify themselves as social (resource investigator) or coordinator type of
team workers, while the members of the Cluster 4 are more goal-oriented (complete finisher,
implementor). This Cluster 4 is also the largest cluster, containing 41% of the participants. A
possible gamification approach for these profile types would be to encourage reflection and goal
coordination at key points in the project instead of directly proceeding to work.
DISCUSSING THE UTILIZATION POSSIBILITIES OF ADAPTIVE GAMIFICATION
PROFILES IN ONLINE LEARNING ENVIRONMENTS
In the initial research four distinct clusters were detected based on student interaction patterns.
The cluster members were also polled on their teamwork preferences, but these factors were not
taken into account in initial clustering, because the initial sorting is based only on objective
measurements. Despite being based on the students’ subjective views, in many of the cases the
students’ Belbin and Bartle preferences match the observed actions.
Benefits of Adaptive Gamification
Usually in gamification abstract points or achievement levels are used as rewards, but simply
applying points or badges does to make an effective game design (Falkner & Falkner, 2014;
Groh, 2012). Game design can go beyond that. According to Bartle (2004) there are four major
player types that enjoy different activities and also different kinds of rewards in online or
multiplayer games. This leads to the fact that different users enjoy different gamification
approaches. Often games or gamification systems are designed so that each type of player can
find something to interest them. However, a more advanced approach would be to try to profile
the student and then try to present gamification approaches and rewards most suitable to the
student. For example, for an explorer (spade) type of player providing more areas to explore can
be a reward or having other team members benefit from the players’ exploration.
This mirroring of preferences and rewards is of great interest and importance in the
gamification process. The core idea of our research is to borrow approaches, methodologies and
techniques from the field of (video) game design to create - and test – an engaging learning
environment, able to foster cooperation among students, to promote positive behaviors and to
impact on their overall learning performances. Beside the “traditional” way to convey teaching,
learning patterns have changed radically (Zyda, 2005): new generations are experiencing new
forms of computer and video game entertainment and this has shaped their preferences and
abilities, while offering an enormous potential for their learning (Prensky, 2006).
The usefulness of games as learning tools is a well-known phenomenon, especially in the first
years of our life (Din & Calao, 2001; Durik & Harackiewicz, 2003; Ritterfeld & Weber, 2006),
and it is demonstrated that games are able to guarantee high learning effectiveness in quite short
time (Koster, 2005; Dario Maggiorini, Previti, Ripamonti, & Trubian, 2013; Laura Anna
Ripamonti & Peraboni, 2010; Squire & Jenkins, 2003; Susi, Johannesson, & Backlund, 2007).
Moreover, games are able to train cognitive functions – such as memory (Squire & Jenkins,
2003) –, and to teach how to exert control over emotions (Michael & Chen, 2005; Mitchell &
Savill-Smith, 2004; Laura A. Ripamonti & Maggiorini, 2011; Wong & Looi, 2011).
Design Plan for an Adaptive Online Learning System
To exploit positive traits of games, we are planning to combine in different ways the building
blocks used by game designers (summarized in Table 3, and by Fullerton (2008)) to design and
deploy one – or more – “gamified” learning environment(s) for students. The learning
environment(s) will be built according to the following guidelines:
The “game” should be alluring for different types of Bartle’s player at the same time.
We must take care that the “pattern” learned by the students-players will enforce their
willingness to collaborate and their teamwork skills, independently from their Bartle
type.
Table 3. Building blocks of games
Formal elements
Dramatic elements
Other relevant elements
Players (interaction)
Objectives
Procedures
Rules
Resources
Conflict
Boundaries
Outcomes
Challenge
Play
Premise
Character
Story
Dynamics
Feedback
Rewarding
Core mechanics
Immersivity/flow
By considering the Bartle player types and which kind of game elements the defined player types
find most interesting, we can ensure that the gamified collaborative learning environment has
game elements that are attractive to all player types. The profile clusters, the clusters’ detected
Bartle player types and the most attractive game elements to each player type evaluated using
Bartle’s theory (2004) are listed in the Table 4. This table can be used to help in the requirements
design for a gamification system in order to ensure that the design has elements that are attractive
to the profiles and their player types. An even more advanced design would use an adaptive
approach and use interaction analysis or an introduction questionnaire to target game elements to
those player types who find the elements most interesting.
Table 4. Most attractive elements for profile cluster types
Profile cluster
Exhibited Bartle
player types
Most applicable game elements
CL1: “Cooperative
workers”
Heart, club
Formal elements: Player interaction, rules,
conflict, outcomes.
Dramatic elements: Character, challenge, play.
CL2: “Social team
members”
Heart, club,
spade
Formal elements: Player interaction, rules,
resources, boundaries, outcomes.
Dramatic elements: Premise, story, character,
challenge, play.
CL3: “Achievement-
oriented leaders”
Diamond, club
Formal elements: Player interaction, rules,
procedures, conflict, boundaries, outcomes.
Dramatic elements: Challenge, play.
CL4: “Task-oriented
workers”
Diamond, club,
spade
Formal elements: Rules, conflict, resources,
outcomes.
Dramatic elements: Premise, challenge, play.
Adaptive functionality would enable the next step in gamification by allowing the system to
customize and target the game elements to each player type. For example, a social team member
might not be interested in a scoreboard and the possibility to challenge other teams, but friendly
competition would be attractive to an achievement-oriented leader. An adaptive, or a profile-
aware system, could also provide source material for peer matching tools or automatic group
composition (Rubens, Vilenius, & Okamoto, 2009; van Tienen Marijn & Rothkrantz, 2015).
Research into adaptive gamification systems (Codish & Ravid, 2014) is ongoing, with
Monterrat et al. (2015; 2015) presenting a player model and a design for a generic, pluggable
gamification system design for learning environments. They use predefined player types and
present a player adaptation model, where player preferences and elements are evaluated in a
matrix to target gamification elements. Their evaluation model could also be applied to
collaborative profiles we present in this paper, though it would need to be extended to consider
interaction targets further.
Future Research
Once the gamified learning environment is in place, we are planning to run several activities to
collect data in order to dis/prove our thesis. In particular, on one hand we will run experiments
with small groups of real students in the area of Computer Science. As a side effect, this will
offer us the possibility to verify to what extent their cultural background has an impact on the
perception of the playing experience. Moreover, we will develop and test a model mimicking the
relationships among the students in the learning environment. This will be the basis to build a
large-scale simulation to verify the effects on the composition of the student population (in terms
of Bartle’s types) deriving from variations into the structure of the gamified learning
environment. The simulation will be run both from the perspective of achieving a higher degree
of satisfaction for students and from that of providing leverages to the teachers, useful for
affecting students behaviors (see e.g. Maggiorini, Nigro et al. (2012) for similar approaches).
CONCLUSION
In this case study we studied software engineering student collaboration behavior, sorted them
into profile clusters with the k-means algorithm and found common behaviors among them. We
found similarities in the profile clusters to Bartle player types and presented ways of how to
apply gamification approaches for the defined profile clusters. We also discussed adaptive
gamification approaches for a collaborative learning system that would utilize the presented
profiles to target each user with most attractive game elements and goals.
We defined four profile types using interaction analysis data, determined Bartle player types
for them and found interaction patterns that could be improved using targeted gamification
approaches. Existing literature has established adaptive gamification approaches for the
gamification of learning (Monterrat, Lavoué, et al., 2015). With our design we extend the current
state of research by proposing that adaptive gamification can be applied to collaborative learning
using our profiling method. Furthermore, we present a method where new profiles can be created
with interaction analysis and clustering to improve on predefined player types.
These results cannot be generalized yet because of the limited sample size, but the analysis
method itself is generally applicable and be repeated with the steps detailed in this study in order
to create profiles based for other learning environments. Repeating the analysis on more
extensive datasets would allow defining more general profile types. For future research there are
also behavior studies that use similar teamwork roles (Serçe et al., 2011; Vivian et al., 2013),
which could be used to gain additional and comparative material for the created profiles.
REFERENCES
Alaoutinen, S., Heikkinen, K., & Porras, J. (2012). Experiences of learning styles in an intensive
collaborative course. International Journal of Technology and Design Education, 22(1),
25–49.
Barata, G., Gama, S., Jorge, J., & Goncalves, D. (2013). Engaging engeneering students with
gamification. In Games and Virtual Worlds for Serious Applications (VS-GAMES), 2013
5th International Conference on (pp. 1–8).
Bartle, R. A. (2004). Designing virtual worlds. New Riders.
Bastian, M., Heymann, S., & Jacomy, M. (2009). Gephi: An open source software for exploring
and manipulating networks. In International AAAI conference on weblogs and social
media (Vol. 2).
Belbin, R. M. (2010). Team roles at work. Routledge.
Berkling, K., & Thomas, C. (2013). Gamification of a Software Engineering course and a
detailed analysis of the factors that lead to it’s failure. In Interactive Collaborative
Learning (ICL), 2013 International Conference on (pp. 525–530).
Bruffee, K. A. (1995). Sharing Our Toys: Cooperative Learning Versus Collaborative Learning.
Change: The Magazine of Higher Learning, 27(1), 12–18.
Caillois, R. (1961). Man, play, and games. Gelncoe, New York: The Free Press.
Coccoli, M., Stanganelli, L., & Maresca, P. (2011). Computer Supported Collaborative Learning
in software engineering. In 2011 IEEE Global Engineering Education Conference
(EDUCON) (pp. 990–995).
Codish, D., & Ravid, G. (2014). Adaptive Approach for Gamification Optimization. In
Proceedings of the 2014 IEEE/ACM 7th International Conference on Utility and Cloud
Computing (pp. 609–610). Washington, DC, USA: IEEE Computer Society.
Crawford, C. (1984). The art of computer game design. Retrieved from
http://www.vic20.vaxxine.com/wiki/images/9/96/Art_of_Game_Design.pdf
Crawford, C. (2003). Chris Crawford on game design. New Riders.
Csikszentmihalyi, M. (1991). Flow: The Psychology of Optimal Experience. New Riders.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. Plenum, New York.
Deterding, S., Dixon, D., Khaled, R., & Nacke, L. (2011). From game design elements to
gamefulness: defining gamification. In Proceedings of the 15th International Academic
MindTrek Conference: Envisioning Future Media Environments (pp. 9–15). ACM.
Dewiyanti, S., Brand-Gruwel, S., Jochems, W., & Broers, N. J. (2007). Students’ experiences
with collaborative learning in asynchronous Computer-Supported Collaborative Learning
environments. Computers in Human Behavior, 23(1), 496–514.
Dickinson, T. L., & McIntyre, R. M. (1997). A conceptual framework for teamwork
measurement. Team Performance Assessment and Measurement, 19–43.
Dillenbourg, P. (1999a). Collaborative-learning: Cognitive and computational approaches.
Elsevier.
Dillenbourg, P. (1999b). What do you mean by collaborative learning? Collaborative-Learning:
Cognitive and Computational Approaches., 1–19.
Din, F. S., & Calao, J. (2001). The Effects of Playing Educational Video Games on Kindergarten
Achievement. Child Study Journal, 31(2), 95–102.
Dubois, D. J., & Tamburrelli, G. (2013). Understanding gamification mechanisms for software
development. In Proceedings of the 2013 9th Joint Meeting on Foundations of Software
Engineering (pp. 659–662).
Durik, A. M., & Harackiewicz, J. M. (2003). Achievement goals and intrinsic motivation:
Coherence, concordance, and achievement orientation. Journal of Experimental Social
Psychology, 39(4), 378–385.
Falkner, N. J. G., & Falkner, K. E. (2014). “Whither, badges?” or “wither, badges!”: a metastudy
of badges in computer science education to clarify effects, significance and influence (pp.
127–135). ACM Press.
Fullerton, T. (2008). Game design workshop: a playcentric approach to creating innovative
games. CRC Press.
Glover, I. (2013). Play As You Learn: Gamification as a Technique for Motivating Learners. In
World Conference on Educational Multimedia, Hypermedia and Telecommunications
(Vol. 2013, pp. 1999–2008).
Gokhale, A. A. (1995). Collaborative Learning Enhances Critical Thinking. Journal of
Technology Education, 7(1).
Groh, F. (2012). Gamification: State of the art definition and utilization. In Proceedings of the
4th seminar on Research Trends in Media Informatics (pp. 39–46).
Herranz, E., Colomo-Palacios, R., & Seco, A. de A. (2015). Gamiware: A Gamification Platform
for Software Process Improvement. In Systems, Software and Services Process
Improvement (pp. 127–139). Springer International Publishing.
Huizinga, J. (1950). Homo ludens. Boston: The Beacon Press.
Johnson, D. W., & Johnson, R. T. (1999). Making cooperative learning work. Theory Into
Practice, 38(2), 67–73.
Johnson, R., & Johnson, D. (1994). An Overview of Cooperative Learning. In J. Thousand, R.
Villa, & A. Nevin (Eds.), Creativity and Collaborative Learning. Brookes Press.
Johnson, S. (2004). Mind wide open: Your brain and the neuroscience of everyday life. Simon
and Schuster.
Jordan, B., & Henderson, A. (1995). Interaction analysis: Foundations and practice. The Journal
of the Learning Sciences, 4(1), 39–103.
Juul, J. (2003). The game, the player, the world: looking for a heart of gameness. In DIGRA
Conf.
Kauffman, L., & Rousseeuw, P. J. (1990). Finding groups in data. An Introduction to Cluster
Analysis. New York: John Willey & Sons.
Knutas, A., Ikonen, J., & Porras, J. (2013). Communication Patterns in Collaborative Software
Engineering Courses: A Case for Computer-supported Collaboration. In Proceedings of
the 13th Koli Calling International Conference on Computing Education Research (pp.
169–177). ACM.
Koster, R. (2005). Theory of fun for game design. Paraglyph Press.
Maggiorini, D., Nigro, A., Ripamonti, L. A., & Trubian, M. (2012). The Perfect Looting System:
Looking for a Phoenix? In IEEE Conference on Computational Intelligence and Games
(pp. 371–378).
Maggiorini, D., Previti, S., Ripamonti, L. A., & Trubian, M. (2013). Resources Optimization in
(Video) Games: A Novel Approach to Teach Applied Mathematics? In Learning and
Intelligent Optimization (pp. 189–195). Springer.
Michael, D. R., & Chen, S. L. (2005). Serious games: Games that educate, train, and inform.
Muska & Lipman/Premier-Trade.
Miller, G. A. (1956). The magical number seven, plus or minus two: some limits on our capacity
for processing information. Psychological Review, 63(2), 81.
Mitchell, A., & Savill-Smith, C. (2004). The use of computer and video games for learning: A
review of the literature.
Moccozet, L., Tardy, C., Opprecht, W., & Léonard, M. (2013). Gamification-based assessment
of group work. In Interactive Collaborative Learning (ICL), 2013 International
Conference on (pp. 171–179).
Monterrat, B., Desmarais, M., Lavoué, É., & George, S. (2015). A Player Model for Adaptive
Gamification in Learning Environments. In C. Conati, N. Heffernan, A. Mitrovic, & M.
F. Verdejo (Eds.), Artificial Intelligence in Education (pp. 297–306). Springer
International Publishing.
Monterrat, B., Lavoué, É., & George, S. (2015). Toward an Adaptive Gamification System for
Learning Environments. In S. Zvacek, M. T. Restivo, J. Uhomoibhi, & M. Helfert (Eds.),
Computer Supported Education (pp. 115–129). Springer International Publishing.
Okamoto, T. (2004). Collaborative technology and new e-pedagogy. In IEEE International
Conference on Advanced Learning Technologies, 2004. Proceedings (pp. 1046 – 1047).
Otte, E., & Rousseau, R. (2002). Social network analysis: a powerful strategy, also for the
information sciences. Journal of Information Science, 28(6), 441–453.
Prensky, M. (2006). Don’t Bother Me, Mom, I’m Learning!: How Computer and Video Games
are Preparing Your Kids for 21st Century Success and how You Can Help! Paragon
House New York.
Resta, P., & Laferrière, T. (2007). Technology in support of collaborative learning. Educational
Psychology Review, 19(1), 65–83.
Rigby, S., & Ryan, R. (2011). Glued to games: How video games draw us in and hold us
spellbound. ABC-CLIO.
Ripamonti, L. A., & Maggiorini, D. (2011). Learning in virtual worlds: a new path for supporting
cognitive impaired children. In Foundations of Augmented Cognition. Directing the
Future of Adaptive Systems (pp. 462–471). Springer.
Ripamonti, L. A., & Peraboni, C. A. (2010). Managing the design–manufacturing interface in
virtual enterprises through multi user virtual environments: a perspective approach.
International Journal of Computer Integrated Manufacturing, 23(8-9), 758–776.
Ritterfeld, U., & Weber, R. (2006). Video games for entertainment and education. Playing Video
Games. Motives, Responses, and Consequences. Mahwah, NJ: Lawrence Erlbaum
Associates, 399–413.
Rollings, A., & Adams, E. (2003). Andrew Rollings and Ernest Adams on game design. New
Riders.
Romero, M., & Usart, M. (2013). Serious Games Integration in an Entrepreneurship Massive
Online Open Course (MOOC). In Serious Games Development and Applications (pp.
212–225). Springer.
Romero, M., Usart, M., Ott, M., & Earp, J. (2012). Learning through playing for or against each
other? Promoting collaborative learning in digital game based learning. LEARNING, 5,
15–2012.
Rubens, N., Vilenius, M., & Okamoto, T. (2009). Automatic Group Formation for Informal
Collaborative Learning. In Proceedings of the 2009 IEEE/WIC/ACM International Joint
Conference on Web Intelligence and Intelligent Agent Technology - Volume 03 (pp. 231–
234). Washington, DC, USA: IEEE Computer Society.
Salen, K., & Zimmerman, E. (2004). Rules of play: Game design fundamentals. MIT press.
Salen, K., & Zimmerman, E. (2005). Game design and meaningful play. Handbook of Computer
Game Studies, 59–79.
Schell, J. (2011, November). The Pleasure Revolution: Why Games Will Lead the Way.
Scott, J. (2012). Social network analysis. SAGE Publications Limited.
Serçe, F. C., Swigger, K., Alpaslan, F. N., Brazile, R., Dafoulas, G., & Lopez, V. (2011). Online
collaboration: Collaborative behavior patterns and factors affecting globally distributed
team performance. Computers in Human Behavior, 27(1), 490–503.
Sheth, S. K., Bell, J. S., & Kaiser, G. E. (2012). Increasing Student Engagement in Software
Engineering with Gamification.
Singer, L., & Schneider, K. (2011). Supporting the cooperation of end-user programmers through
social development environments. In Proceedings of the 2nd international workshop on
Web 2.0 for software engineering (pp. 13–18).
Singer, L., & Schneider, K. (2012). It was a bit of a race: Gamification of version control. In
2012 2nd International Workshop on Games and Software Engineering (GAS) (pp. 5–8).
Squire, K., & Jenkins, H. (2003). Harnessing the power of games in education. Insight, 3(1), 5–
33.
Susi, T., Johannesson, M., & Backlund, P. (2007). Serious games: An overview.
Tan, C. T. (2013). Towards a MOOC game. In Proceedings of The 9th Australasian Conference
on Interactive Entertainment: Matters of Life and Death (p. 30).
van Tienen Marijn, F. V. F., & Rothkrantz, G. D. D. L. (2015). A Peer matching tool using social
media. In International Conference on e-Learning (Vol. 14, p. 201).
Vivian, R., Falkner, K., & Falkner, N. (2013). Analysing Computer Science Students’ Teamwork
Role Adoption in an Online Self-organised Teamwork Activity. In Proceedings of the
13th Koli Calling International Conference on Computing Education Research (pp. 105–
114). ACM.
Wagstaff, K., Cardie, C., Rogers, S., & Schrödl, S. (2001). Constrained k-means clustering with
background knowledge. In ICML (Vol. 1, pp. 577–584).
Wong, L.-H., & Looi, C.-K. (2011). What seams do we remove in mobile-assisted seamless
learning? A critical review of the literature. Computers & Education, 57(4), 2364–2381.
Yamawaki, S. (2013). Gamified mobile learning with OCW—Effective learning though
competition and collaboration. In OCWC Global Conference 2013.
Zyda, M. (2005). From visual simulation to virtual reality to games. Computer, 38(9), 25–32.
APPENDIX A. COMMUNICATION VISUALIZED AS SOCIAL NETWORK GRAPH