Vault: Merging relational learning and mobile learning with the philosophy of Parkour

Teknik och samhälle

Datavetenskap och medieteknik

Thesis

15 HP, basic level

Vault: Merging relational learning and mobile learning

with the philosophy of Parkour

Vault: Kombinerar relational learning och mobile learning med filosofin hos parkour

Kristoffer Linderman

Examen: Kandidatexamen 180 hp Huvudområde: Datavetenskap Program: Applikationsutveckling Datum for slutseminarium: 2018-05-28

Handledare: José Font Examinator: Nancy Russo

Abstract

Denna uppsats presenterar forskning som belyser de förändringar inom utbildningsystemen som följt i och med utvecklingen av internet, sociala medier, online kommunikation och utbyte av kunskap online. Under det senaste årtiondet har sättet människor lär sig på förändrats från det traditionella klassrummet, som enbart använder sig av tryckt material, till det moderna

klassrummet som nyttjar digitalt material [1]. Electronic learning (e-learning) innebär av att använda sig av digitala material istället för det traditionella tryckt materialet [2]. Mobile learning tar detta koncept ett steg längre genom att kombinera individuellt lärande med varsomhelst och närsomhelst lärande [2]. Detta möjliggörs till stor del av den teknologiska utvecklingen inom mobila enheter [3]. Relational learning är ett sätt att lära där individen lär från andra genom ett gemensamt utbyte av idéer och kunskap [4].

Parkour, eller konsten av rörelse, har av O’Grady blivit presenterat som en kollaborativ läromodell [5].Utövare av parkour kallas traceurs och när traceurs utövar parkour blir parkour ett verktyg för aktivt lärande. Med detta lärande blir förståelse och överkommandet av hinder en läromodell. Detta är en läromodell där vetandet och upprepandet är nyckeln till traceurs strävan efter lärandets berusning. Med lek förvandlar parkour sin omgivning till en miljö för lärande. Med hjälp av parkour tar traceurs över ägandet av sitt egna lärande och hittar

möjligheter att kunna uttrycka sig själva [6]. En naturlig del av parkour är vikten som läggs på att vara medveten om sina egna förmågor och mål med sitt lärande. Genom att dela med sig av resultatet av detta lärande skapas basen av den kollaborativa läromiljön hos parkour. I denna uppsats presenteras en mobil Android applikation: Vault. Vault kombinerar mobile learning med relational learning och använder sig av fördelarna av den kollaborativa

läromodellen som existerar inom parkour. Tre populära mobila applikationer ämnade för lärande har analyserats. Denna analys i samband resterande resultat har varit med och format designen och utveckling av den presenterade applikationen Vault.

Vault är även testad för att mäta dess potentiella fördelar av att användas som en allmän utbildningsapplikation, med ett fokus på relational learning. Detta test är beskrivet och

resultaten är presenterade. Avslutningsvis återfinns en diskussion angående resultaten från både testet av applikationen samt analysen och det teoretiska materialet som presenterats i

uppsatsen. Denna diskussion följs av en sammanfattning som även innehåller förslag på framtida forskningsområden som kan utforskas vidare.

Sökord: Relational learning, mobile learning (m-learning), electronic learning (e-learning), parkour, mobilapplikation

Abstract

This thesis presents research that addresses the educational change that arises in the era of internet, social media, online communities, and knowledge sharing on the web. During the last decade, the way people learn has seen a big shift from the traditional classroom that purely uses printed material to the contemporary classroom that utilizes digital technologies for the teaching material [1]. Electronic learning is teaching using electronic resources instead of the traditional printed material [2]. Mobile learning takes this concept one step further by combining individualized learning with anytime and anywhere learning [2], enabled by the technological advances of mobile devices [3]. Relational learning is a way of learning in which the individuals involved learn from each other through the bilateral exchange of experiences and ideas [4].

Parkour, or the art of movement, has been presented as a collaborative learning model by O’Grady [5]. As parkour practitioner, also known as traceurs, play parkour they also provide a platform for active learning where knowing and overcoming obstacles composes an educational model. In this model, knowing and repeating is the key to the pursuit of learning. With play, parkour appropriates the spaces in which it takes place into an environment of learning. By playing, traceurs take ownership of their own learning process, finding the flow path that lets them express themselves [6]. Inherent to the practice of parkour is the importance of self-awareness of one’s skills and learning goals, as well as recording and sharing the learning outcomes. This, in turn, becomes the basic construct of a collaborative learning environment. In this thesis, an Android mobile application, called Vault is presented. Vault combines mobile learning and relational learning, while at the same time reaping the reward of the community-based learning model existing in parkour. The thesis also provides an analysis of popular mobile learning apps. This analysis aides in shaping the design and development of the presented application, Vault.

Vault is also tested in order to gauge the potential benefits of using an application designed to be a general-purpose educational application with a focus on relational learning. This test is detailed, and the results are presented. The findings from these results, and the results from the aforementioned analysis and the theory presented in this thesis, are discussed and future lines of research are presented.

Keywords: Relational learning, mobile learning (m-learning), electronic learning (e-learning), parkour, mobile application

Table of Contents

1 Introduction ... 1

1.1 Background and Previous Research ... 1

1.2 Parkour as an instructional model ... 2

1.3 Research Questions ... 3

1.4 Purpose of research ... 3

2 Method and Motivation ... 4

2.1 Method description ... 4

2.1.1 Research Question 1 ... 4

2.1.2 Research Question 2 ... 4

2.1.3 Research Question 3 ... 4

2.2 Method discussion ... 5

2.2.1 Design and Creation ... 5

2.2.2 Questionnaire ... 5

2.2.3 Application Analysis ... 6

2.3 Motivation ... 6

3 Solution ... 8

3.1 Relational learning features ... 8

3.1.1 Authoring ... 8

3.1.2 Content visibility ... 8

3.1.3 Interactivity ... 9

3.1.4 Ability to give and respond to feedback ... 9

3.2 Current Applications ... 9 3.2.1 Selection Process ... 9 3.2.2 Selected Applications ... 10 3.2.3 Analysis Process ... 10 3.2.4 Question motivations ... 11 3.3 Application Development ... 12 3.3.1 Front-End ... 12 3.3.2 Back-End ... 13

3.4 Testing and Questionnaire ... 14

3.4.1 Question motivations ... 15

4 Results ... 17

4.1 Analysis of selected applications ... 17

4.1.2 Coursera ... 18 4.1.3 Duolingo ... 18 4.2 App development ... 18 4.2.1 Content Creation ... 19 4.2.2 Content consumption ... 21 4.2.3 Backend ... 23 4.3 Questionnaire ... 24

5 Analysis and Discussion... 27

5.1 How do the current mobile learning applications implement relational learning? ... 27

5.2 How can a mobile application implement mobile learning with a special focus on relational learning? ... 28

5.3 What are the potential benefits of relational learning from the perspective of mobile learning? ... 29

6 Conclusions and future research ... 31

6.1 Future research ... 32

1

1

Introduction

1.1 Background and Previous Research

During the last decade, the way people learn has seen a big shift from the traditional classroom that purely uses printed material to the contemporary classroom that utilizes digital technologies for the teaching material[1]. Electronic learning (e-learning) is teaching using electronic resources instead of the traditional printed material [2]. Using electronic materials instead of printed materials is convenient for both the content provider and the learner. The content provider, the individual that provides the

educational content or facilitates it, does not have to deal with printing out material or printing enough of them. Meanwhile, the learner does not have to remember to bring the material as they can access it digitally. The modern educational technologies attempt to provide a better way for the delivery of learning from the content provider to the learner. Mobile learning is a new term that is still much in its infancy. Motiwalla defined mobile learning as “[it] combines individualized learning with anytime and anywhere learning”[2]. Other researchers have defined the term mobile learning as learning enabled by mobile devices [3]. For this thesis, a combined definition will be used. So, in this thesis, mobile learning is defined as a formal and informal learning that is performed on a smart phone, and available anywhere at any time. Mobile learning takes the concept of electronic learning and tries to make it more flexible and comfortable by the use of personal electronic devices. This removes the previous geographical limitations as well as the limitation on time; now a student can learn anywhere and anytime. Mobile learning offers benefits that are inherent to mobile devices such as connectivity, mobility, geolocation and social networking [7].

Relational learning is a way of learning in which the individuals involved learn from each other through sharing experiences and ideas [4]. In this way teachers, students, mentors and professors can all learn from each other, disregarding the hierarchy that exists in the traditional educational relationship. This opens the space for a more collaborative learning experience as the involved parties can create their desired learning environment[4].

Smart phones allow for the learners to interact with each other, regardless of their location[8]. By interacting, the learners can collaborate on projects, and discuss content with both fellow students but also the content provider. This can benefit all of the actors involved in the educational process. The content provider can get insight into how the content is received and interpreted, and the learners can contribute with new meaning and understanding, aiding the learning process of other learners.

The use of mobile computing devices in project-based courses has been shown to be beneficial to the sense of connectivity with students, instructors and their clients [7]. This is created by allowing a continuous connection to the internet to update on progress, share photos and use instant communication. This, in addition to relational learning, makes for a new type of learning environment. An environment in which the exchange of knowledge has a possibility to be mutual and thus mutually beneficial.

2 This thesis will be looking at how to combine the benefits of mobile learning and relational learning. Also analyzing how the current popular learning applications utilize relational learning. The apps that will be analyzed will be Coursera: Online courses, Duolingo: Learn Languages Free and Udemy Online Courses. These are all very popular apps in the

category of education on the Google Play Store, boasting over 1,000,000 downloads each, with Duolingo: Learn Languages Free having over 100,000,000 downloads [9][10][11]. Although the number of downloads is not in direct correlation to their current level of popularity, it is, if nothing else, a sign of having an application that has been deemed interesting enough to install for a lot of users. These apps will be analyzed from the perspective of how they implement relational learning into their application and how the users can experience and utilize these implementations.

In addition, this thesis will develop a learning application that aims to implement the concept of both mobile learning and at the same time reap the reward of relational learning. This application will be developed based on findings from the analysis as well as the theory of relational learning, parkour and mobile learning. The application will be developed on the Android platform and will thus be limited to smart phones running the Android operating system. The application’s demographic will be all users of mobile Android devices that want to become learners and/or content providers. The application is called ‘Vault’ and will hereby be referred to as such. After the development phase, Vault will be tested on several teachers in an elementary school in Höör, Sweden. The test participants will test and evaluate Vault in order to see the potential benefits it can have. The teachers will use Vault in both their professional life and their personal life in order to see if both aspects can benefit from the use of the application. The testing will conclude in a questionnaire [12]. This questionnaire will collect the responses and impressions of the users.

1.2 Parkour as an instructional model

The philosophy of parkour is based on the transformation of obstacles into opportunities [13]. The traceur, the parkour practitioner, is aware of his or her own constraints as well as of the spatial constraints present within the environment: the city observed as a playground. These constraints in intersection with the leisure practice of parkour, shape the traceur’s learning path to the attainment of freedom and autonomy. This path sometimes builds upon the traceur’s acceptance of such constraints, but also on the rejection, utilization, or transformation of these constraints.

Belonging to a collaborative learning community of active practitioners is also essential for the practice of parkour [5]. Members in a parkour group promote and enjoy a fluent collaborative learning environment by providing instant cross-feedback in a significantly horizontal social construct that opposes to traditional hierarchical instructional models. As a playful learning activity, parkour offers traceurs a community of teachers and learners. This community dares the traceurs to look for and walk through the challenging path to their self-achievement. Simultaneously, the community emphasizes rewarding play and fun over completion and competition [14].

3 Looking at parkour from the lens of education reveals that several of these features are present in the relational learning paradigm [4]. In this model, all participants involved in education learn from each other inside a community of relational learners. This model bases on a constant dialog among all community members about learning and

deconstructing the traditional and hierarchical relationship between teacher and student. This opens the scope for a more collaborative learning experience as the involved parties can create their desired learning environment.

1.3 Research Questions

RQ1 - How do the current mobile learning applications implement relational learning? RQ2 - How can a mobile application implement mobile learning with a special focus on relational learning?

RQ3 - What are the potential benefits of relational learning from the perspective of mobile learning?

1.4 Purpose of research

The concepts behind this research came from the philosophy and educational model found within parkour. Within parkour there are no set teachers, no set rules of how to

communicate the knowledge to others or who is allowed to provide this knowledge. Upon starting the research this concept was found to overlap with the theory presented in relational learning.

The purpose of this thesis is to further the research of relational learning. More

specifically, further the application of relational learning on a mobile learning platform and being inspired by the learning community prevalent within the parkour movement.

Furthermore, the thesis aims to obtain an understanding of how some of the currently popular mobile learning applications implement relational learning, although not explicitly.

4

2

Method and Motivation

As this thesis aims to answer several research questions, there will be the need to apply several different methods to achieve a satisfactory and clear answer. These methods will be applied to different parts of the project in order to get the necessary data.

2.1 Method description

2.1.1 Research Question 1

The thesis also aims to evaluate how some of the most popular mobile learning

applications currently employ a relational learning model. The chosen applications will be installed and used from the perspective of a user that is interested in learning on a smart phone. The applications will be evaluated on to what extent they implement relational learning features. These features will be defined based on the theory on relational learning. A form will be devised with boxes for different features or attributes the application can have that would be considered an aspect of relational learning. Each application will be analyzed using the same form in order to be able to compare them in a more measurable way.

2.1.2 Research Question 2

The primary selected method for this project will be the development of a new artifact. This will be accomplished according to the design and creation method [15]. Design and creation is the process of developing new IT artifacts. These artifacts can either be Constructs, Models, Methods or Instantiations.

• Constructs is an artifact that creates new vocabulary to a certain field of

computer science. These artifacts are used in order to describe the fields problems and specify their solutions.

• Models are a combination of constructs that together can be used to describe a relationship between different constructs.

• Methods are descriptive instructions needed in order to complete a specific task. These instructions, or algorithms, rely on the supportive vocabulary and models, provided by constructs and models.

• Instantiations are a working, realized, system that demonstrates other IT artifacts. To be able to solve the questions presented in this thesis, an instantiation-artifact will be developed in the shape of a mobile application named Vault. This artifact will be used to answer research question three, regarding how to implement a mobile learning application with the main focus of employing a relational learning model.

2.1.3 Research Question 3

Vault, as part of the design and creation method, will be tested on elementary school teachers to see how the they might be able to implement the application in to their work and private lives. The selection of teachers was to be able to get information on their use of the application in both an educational environment and a personal environment. In

5 order to evaluate how beneficial Vault is in both the professional and private lives of the elementary school teachers, the teachers will subsequently receive a questionnaire concerning their experiences with Vault. In this questionnaire, they will answer questions regarding if the application did or did not improve their educational work and/or if the see potential for the application in an elementary school setting. The questionnaire will

contain a strict set of questions regarding the experience with Vault. In addition to the strict questions, the questionnaire will contain some open questions in which the teachers are able to share their own thoughts and impressions more unrestricted. This is to ensure that the teachers can give any and all feedback they deem relevant along with the

information given by the strict questions.

2.2 Method discussion

The choice of method, or methods, to use in a project of this scale are both crucial for the actual development process of Vault and the thesis. The choice is also fundamental in creating the possibility to gather data and being able to answer the research questions of this thesis.

2.2.1 Design and Creation

As this project aims to answer how an application can be developed in order to make use of both mobile learning and relational learning, actually developing an application was deemed necessary. This application is intended to implement relational learning into a mobile learning application. In addition, this application is for testing and determining the potential benefits and potential drawbacks of the concept.

As with most methods, there are pros and cons for the design and creation method. The method has a high level of relevance as it is used in solving a real problem or using an artifact to prove or test an idea. In this instance, the developed artifact was important as it laid the foundation for both the testing and the realization of the theory behind

relational learning. There can, however, be situations where the scientific contribution of an artifact can be hard to distinguish [15]. To try to remedy this potential problem, the artifact will, in addition to its own scientific contribution, be used as the platform to conduct the test. For the test, the developed artifact will be the tool used in order to gauge the potential benefits of merging relational learning and mobile learning.

2.2.2 Questionnaire

When doing a survey the goal is to be able to generalize to a broader demographic [15]. To accomplish this, it is preferred to survey a large amount of people in order to try to get enough data to be able to generalize. For this testing the number of participants will be below ten. This is not ideal for generalizing to all elementary teachers, nor for

generalizing to people in their social-, economic- and age-group. As a result, the

participants of the test will fill out a questionnaire as this study does not seek to gather information from a broad sample [12]. The results acquired from the questionnaire will be seen as an indication of potential benefits and drawbacks of utilizing relational learning in a mobile learning platform. The thoughts and critique gained from the questionnaire will

6 be held in relation to both the number of participants and to other research findings that have been discovered throughout the span of this project.

2.2.3 Application Analysis

The analysis of the three chosen educational applications will aid in the assessment of the current level of implementation of relational learning in mobile learning platforms. The aim is to be able to detect some features they have in common and see how the

developers included the social aspect and relational learning aspect into their applications. The applications in question are not advertised as being applications based on the

relational learning model. However, the applications are some of the most installed applications within their category of Education on the Google Play Store (as of March 2018). This will give an insight into how some of the leading applications view and prioritize the aspect of relational learning as well as the social aspect of learning.

2.3 Motivation

A meta-analysis on the effect of integrating mobile devices with teaching and learning by Y.Sung et.al showed overall moderately good results [16]. The analysis also concludes that the effects of using mobile devices in education is better than using desktop computer or not using mobile devices. However, the article also points out some areas that needed for future development:

• Providing instant feedback and easy interaction between devices and users. • Providing the functionality of authoring tools that leads to flexible teaching and

learning flows.

• Finding the key to integrating mobile devices with instructional strategies, matching mobile devices’ advantages to specific pedagogical challenges.

This thesis proposes the relational learning approach as a general-purpose instructional model, an instructional model that already much of the philosophy found within parkour – parkour being the initial inspiration for this thesis. By utilizing these instructional

paradigms and implementing them into an mobile application it is possible to attempt to address the points that Y. Sung et.al claim need future development [16].

As showed by Y. Sung et.al mobile learning has been shown to give better results than when using a desktop computer [16]. Mobile learning offers several aspects that this thesis proposes to be suitable to utilize in conjunction with relational learning. These features include ubiquity, ease of contribution and socialness [7]. Ubiquity is suitable as it speaks much to the nature of relational learning, in which the concept of opening up for a bilateral exchange of knowledge proposes that knowledge is possible to be found everywhere, being ubiquitous. The ease of contribution found within mobile learning makes it possible to provide knowledge for others to partake in anywhere and everywhere. The social aspect that can be utilized within mobile learning is also one of the

cornerstones in both relational learning as well as within the philosophy of parkour.

Therefore, this thesis chooses to merge these, in order to evaluate the potential benefit of this merge.

7 Elementary school teachers were chosen for this testing due to two factors. One factor was that the test would show potential benefits of individuals in their private life. The other factor was a complementary one, to see if Vault could be utilized within the educational system. Using Vault in the educational system was not a research question that was strived to be answered but merely to act as anecdotal evidence of potential future implementations. The teachers are people that spend their working days doing just that: teaching. As Vault is an educational application made for learning and teaching, it was deemed suitable to have teachers testing it. This could provide a deeper insight into how well the application preformed when it came to creating, and consuming educational content.

8

3

Solution

In order to answer the three research questions presented, three different parts had to be constructed:

• The analysis of how current popular mobile educational applications use relational learning in their application.

• The application, Vault, that would implement the theory of using relational learning in a mobile learning application.

• The questionnaire to gather the required information from the participants.

3.1 Relational learning features

In order to combine the theory of relational learning with mobile learning, relational learning needs to be translated into some core features. These are features that could be implemented into an application, and thus evaluated amongst other applications. These features should be based on, and tied to, the theory of relational learning.

In the relational learning paradigm all participants learn from each other inside a

community, as described by Chorba et al. [4]. This paradigm is based on a constant non-hierarchical dialog, amongst all of the community members. To achieve this, some basic features are extrapolated: authoring, content visibility, interactivity and the ability to give and respond to feedback.

3.1.1 Authoring

In order to have educational content there needs to be a way to contribute with new knowledge or ideas in the shape of authoring. This feature involves users being able to create their own educational content for others to consume. As this is the basis for all of the other parts of relational learning, authoring will be viewed as the quintessential feature of relational learning in a mobile learning context.

3.1.2 Content visibility

For other learners to be able to partake in the educational content gathered from

authoring, there must exist content visibility. Content visibility is the feature of having the educational content visible and available for the other learners. Content visibility can be one of three levels: full content visibility, partial content visibility and no content visibility. Full content visibility means that all users on the specific platform can view and consume all of the content provided by the other users. Partial content visibility means that that the visibility is limited to a selection of learners. This selection of learners can be defined by several different aspects such as learners who have paid for the content, learners that are part of a specific institution etc. No content visibility is educational content that is not visible to any learners.

9

3.1.3 Interactivity

The bilateral exchange of knowledge within relational learning requires interactivity between learners and content providers. This exchange also exists between learners, as they can profit from other learners’ thoughts on educational content. The interactivity can take many shapes, such as comments, forums or message boards. As long as all users are able to openly interact and communicate with each other there is interactivity.

3.1.4 Ability to give and respond to feedback

As relational learning exists in a non-hierarchical setting, the learners are supposed to be able to give feedback to the content provider authoring the educational content. This feedback enables the bilateral exchange of knowledge between the learner(s) and the content provider. Through this, both the content provider and the learners can gain new knowledge and/or see things from a new perspective.

3.2 Current Applications

3.2.1 Selection Process

A report out of Pew Research Center done in 2015 analyzed the available applications for Android devices. The report analyzed over 1 million applications and found that 8.06% of the applications were in the category of educational application[17]. Those numbers are data based on the available applications of 2014. The current number of available applications are far greater, as shown below.

Figure 1 Number of available applications in the Google Play Store from December 2009 to December 2017

10 Looking at the data provided by Statista in figure 1 the number of available applications has almost tripled from about 1.3 million in July of 2014 to 3.5 million December of 2017. Based on the huge number of educational applications available, the process of selecting the applications to analyze was done based on popularity and installs. By looking at the top charts for free educational applications in the Google Play Store there are a lot of applications that are very niched [19]. These include applications for learning laws and regulation for driving and applications containing student discounts. As this thesis aims to analyze applications for learning in general the aforementioned type of applications was not taken in consideration. Instead, three applications were selected based on their area of education and their number of installs.

3.2.2 Selected Applications

Duolingo: Learn Languages Free is a language learning application that also has an online version for use in web-browsers. It is one of the most popular educational applications in the Google Play Store with over 100 million installs[9]. Duolingo offers courses in over twenty languages, ranging from Italian to Swahili. The majority of courses are created for English speakers, although they do offer several courses for non-English speakers. The application is mostly free but does offer unlockable content within the application. This content is unlocked using ‘Lingots’, an in-app currency that is acquired by simply using the application and completing lessons. In addition, the application does have Duolingo Plus that is bought using real money. However, using Duolingo Plus does not add additional content to the application. Instead, it simply removes ads and enables offline use of the application by downloading the lessons.

Coursera: Online courses is a more general learning application with subject areas from math to music, to medicine. The application has more than 5 million installs from the Google Play Store[10]. The courses offered on Coursera are primarily created by other institutions and companies. There are several universities that have provided courses one can enroll in. In contrast to Duolingo, who provides most material for free, Coursera has less free material. Coursera offers courses that will earn the user course and specialization certificates, these courses can also be accessed through a web-browser. The user will have to pay to access these certificates, but some courses have a free option in which the user is not given a certificate upon completing the course. In addition to this several courses have no certificate but still cost money to enroll, and some courses are completely free. Udemy Online Courses is an application that is also aimed towards more general learning. Udemy offers over 2,000 topics ranging from web development to yoga. The application has over one million installs on the Google Play Store [11]. Udemy also has an online platform for taking part of the courses in a web-browser. The available courses are created by experts within their field. Some of the courses are completely free while other cost money to enroll.

3.2.3 Analysis Process

In order to answer RQ1 the analysis should be consistent and reliable. To keep the analysis of the application as consistent and reliable as possible a form was created. The

11 form contains only polar questions regarding the presence of features that are based on the aforementioned list of features deemed to be relational learning features. The basic configuration will be one individual acting as the content provider and others as the learners. This is then built upon with the learners being able to provide additional information and feedback that might be deemed relevant and beneficial to the original content provided.

The form contains six polar questions. Each yes answer grants one point. The sum total of the points is then the final score the application gets. These scores are merely to be seen as giving a way to numerically compare the different applications i.e. a score of five is better than a score of two. The form in its entirety can be found in Appendix 1. The features that are deemed to be important to the relational model, are described in more detail in the section Front-End.

3.2.4 Question motivations

The questions making up the questionnaire are constructed in order to evaluate how well the selected applications implement the main features of the relational learning paradigm: authoring, content visibility, interactivity and ability to provide and respond to feedback [4]. Below is the motivation for all of the six questions that make up the form used for the analysis. Each question is motivated by its ties to the theory presented above. Does the application enable every user to create learning material?

One of the key aspects in both relational learning and parkour is the concept of learning from each other and creating and environment in which everyone can provide and partake in new information [4][5]. This makes being able to create content an important feature in terms of relational learning. It is imperative to not limit who can provide content as this would limit the relational learning aspect. This question directly addresses the aspect of authoring as described above in section 3.1.1.

Does the application enable every user to view ALL learning material?

This question is to directly address the aspect of content visibility and to analyze whether the application in questions offers full content visibility or not. As previously mentioned content visibility has three levels: full, partial and no content visibility. This question only answers if the application in question has full content visibility or not. If the answer is ‘No’ then additional comment should be made in order to determine what level of content visibility the application provides. As mentioned by Waern, Balan and Nevelsteen parkour belonging to a collaborative learning community is essential for the parkour philosophy [14]. This would also include allowing all of the participants to have access to all of the learning material.

Does the application enable every user to ask questions regarding the material? Does the application enable every user to leave feedback to the content provider? The ability to provide feedback is part of the relational learning paradigm, in which both parties involved in the learning are open to learn. Asking questions regarding the material in order to get additional information regarding the educational content can minimize

12 potential misunderstandings regarding the material. These questions are to cover the aspect of giving feedback to the content provider as well as the interactivity between the learner and the content provider.

Does the application enable every user to interact with the content provider?

Does the application enable every user to interact with other users learning the same thing?

Both of these two questions are addressing the same important aspects: bilateral

communication. These questions are based in the relational learning features interactivity. The bilateral communication can also found in the horizontal structure prevalent within the parkour community [5]. By allowing everyone to openly communicate their own thoughts, feedback and knowledge with each other it allows for a mutual beneficial situation, in that all parties can learn something.

3.3 Application Development

To be able to answer RQ2, a mobile application named Vault is developed. Vault’s development is based on the results from both the previously written theory on the subject and the analysis of the currently available educational applications. Vault is to implement features that make it embrace the concepts of relational learning to a mobile learning platform. To be able to implement any relational learning features, Vault is required to be able to communicate with other devices. This is necessary in order to share knowledge and ideas between users regardless of the geographical location of the users. This development project is modeled from the Client-Server model. The client-server model describes how a client can obtain resources or data from a server [20]. The server can tailor to the needs of several clients. In the client-server model the client would act as the front-end [21] and the server would act as the back-end [22].

Communication between the back-end and the front-end is done using an API. An API (Application Programming Interface) is an intermediary that allows two different applications to transfer data [23]. For this project the API will manage the

communication between the database and Vault. The database will in turn hold all of the content that the users have provided in order to make it available to other users.

3.3.1 Front-End

The front-end part of the development process means the development of the interface with which the user interacts [21]. As this development process follows the client-server model, the entire client will be seen as front-end and the entire server will be seen as back-end. In this project Vault acts as the client.

Vault is developed for the Google Android operating system. The choice of Android operating system was made based on previous experience with both the operating system and the main programing language Java. In order to develop Vault, the software Android Studio will be used. Android Studio is the official Integrated Development Environment (IDE) for Android [24].

13 With the aim of successfully creating a mobile learning application with a focus on the relational learning model there are some features that will be needed. These are features that are derived from the list of relational learning features found in section 3.1.

The user should be able to become a content-provider with ease. This means that the requirement and process of getting the application and becoming a content provider is quickly and easily completed by the user. By enabling the users to become content providers, the application achieves the first relational learning feature: authoring.

In addition, users of the application must be able to view content provided by other users. Vault will contain no limitation to the content visibility outside the requirement of the user being registered in the application. This feature achieves the relational learning feature: content visibility.

Comments will also be a part of the application. By enabling the ability to comment on the content that a user has provided Vault will facilitate the bilateral part of the

knowledge exchange. As other users partake in the discussion the original content-provider can in turn learn something new, either pedagogical or regarding to the subject at hand. This feature achieves both the relational learning feature interactivity as well as the ability to give and respond to feedback.

With these features implemented the application is created using the aspects of relational learning as a foundation. Using these relational learning features as the basis of the design and development is done in order to ensure that the application can be seen as an

application implementing relational learning.

3.3.2 Back-End

The back-end of the project will be the server. The server will handle the management of the data created by the users using Vault. As the users create content it will be uploaded to the server and stored there so other users can retrieve it. The server will contain several different parts: first, it will require an API so that Vault can communicate with the server, and secondly, there will be the actual server software. This software will handle the retrieval and modification of the data stored in the database. Lastly, there needs to be a database to store all the information from all of the user’s content.

The API and server are tightly joined as they work in conjunction with each other. Both API and server are written in the programming language Go, also known as Golang. The programming language Go is an open source project developed by employees at Google [25]. The choice of programming language for the back-end was made due to several factors. Go is a modern language that is created for usage in, amongst other applications, servers and API. Go is a language that compiles to machine code and does so very quickly and efficiently [26]. The server has a connection to a MySQL database will all of the data. The motivation behind choosing MySQL lies in it being the most popular open source database [27]. This leads to having a knowledgeable and sizable community to provide explanations and assistance in the development process.

14

3.4 Testing and Questionnaire

To answer RQ3, regarding the potential benefits of the developed application, a test is conducted with six elementary school teachers. This test spans over the course of nine days. At the start of the test, the participants are given phones with Vault installed. Prior to the start of the test, the participants were explained the nature of the test. The explanation went through the concepts of relational learning and how these concepts can be found in the parkour philosophy as well. Some examples of how Vault could be used was given and an open discussion was held. The discussion addressed additional ideas from the participants of potential uses and discussion regarding how, or if, they were suitable for Vault.

Accompanied with the application are two instructional videos. These videos offer a detailed guide on how to use the application. The guide walks through the process of creating educational content containing only one picture and some text. After this explanation an additional video continues to explain how to create more complex content containing videos, links to websites in addition to plain text. The videos are made

available to the participants, so they can re-watch them whenever they want and how many times they want. The use of these instructional videos is intended to eliminate any learning curve in using Vault. The learning curve of the application is minimized in order to maximize the time the participants can spend actually using the application as intended. As this test focuses on the concepts of relational learning and the parkour philosophy the emphasis is made on getting the participants using the application as quickly and effortlessly as possible.

The participants are instructed to use Vault as much as they want. The use of Vault is not limited to work hours or work-related content. The participants are encouraged to explore the potential of Vault and the different usages it might provide. For participants that are not familiar with the Android operating system, a quick introduction to the basics of the operating system will take place. After the introduction to the operating system a walk-through of Vault shows how to both view and create content provided by others on the platform. The participants are then free to take the devices and use Vault freely for the duration of the testing phase. After the testing phase is complete, all devices are returned, and a questionnaire is given to the participants.

Providing the participants with a phone is to remove the obstacle of the participants not owning an Android smartphone themselves. This can however lead to some unwanted drawbacks. Part of the benefit of having a mobile learning application is the ability to easily and quickly use the application and learn something new or create content. This low intrusion is disturbed when the participants cannot use their regular smartphone. The act of having to either get or just carry around the testing device might in and of itself make Vault less appealing to the participants to use and test. Due to this possible issue, the questionnaire contains questions of how frequent the application was used. In addition, the question asks what could have been done in order to make the participants use Vault more. This in order to try to be able to gauge whether the addition of a new device was a

15 causing factor in Vault not being used more. The question also opens up for additional explanations in regard to what could be done to promote added use of Vault.

The motivation behind offering a walk-through of Vault lies in the purpose of the test. The test is not primarily to see how easy or intuitive Vault is to use. The test aims to evaluate the potential of the relational learning implemented into a mobile learning

platform. Therefore, the main objective is to make it easy to understand how Vault works and how to use it. This in order to encourage the participants to really explore the

benefits of the theory and concepts of relational learning.

All of the participants are given a questionnaire when the test is over. This questionnaire is created using Google Forms and contains 21 questions regarding their experience with Vault. The aim of the questions is to get data to be able to answer RQ3. The questions address primarily the potential benefits of Vault, in both the private life as well as the professional life. This is done with multiple-choice and open questions where the participants can give feedback that is difficult to quantify. The questionnaire in its completion can be found in Appendix 2.

The results from the questionnaire are collected and processed. As the closed questions have numerical value assigned to them an average is derived and shown. It is more difficult to calculate an average when it some to the open questions. This is due to the nature of the open questions: they are there so that the participant can give information that the numerical values cannot. The open questions are individually assessed, and the complete answers are provided as an appendix in the results. The open questions are analyzed by looking for things they have in common. This in an attempt to find a more general conclusion of the potential benefit. The analysis is done for every open question that is relevant to the project. This means that questions regarding the user experienced are not the focus of the analysis; these questions will, instead, be used to aid the future development of Vault.

3.4.1 Question motivations

Below is the motivation for the questions that make up the test questionnaire. The questionnaire can be found in Appendix 2. All of the questionnaires are anonymous and no personal data, other than what is stated in the questionnaire, is collected. Motivation below may not include all the questions found in the questionnaire as some questions were asked purely to benefit future development of the application.

Firstly, the gender and age of the participant is asked. This is to get a better

understanding of what part of the population they would represent, if a generalization would take place. As previously mentioned due to the small number of participants the results will not be generalized for the population at a whole. The question on gender and age are followed by asking if the participants usually create online content. This content could be video-content, content for blogs, forums etc. This is done in order to get a better sense of the participants online activity, especially in shape of creation.

Next are questions regarding overall satisfaction of the application and the ease of the content-creation. These two questions are answered on a linear scale from 1-5, one being

16 not very satisfied/easy and five being very satisfied/easy. These questions are aimed at answering both general satisfaction of the application and also see if the process of

creating educational content was intuitive. As Vault only has educational content provided by users it is beneficial if the process of adding content is easy. Additionally, a question is asked on how easy it was to connect the application to other sources like YouTube, blogs, websites etc. This is asked for the same reason as the one on creating content. Afterwards, two questions regarding community and external sources are asked. The participants are asked if they felt part of a community while using the application. As previously mentioned, being part of a community is a big aspect within parkour [5].It is therefore interesting to see if the socialness attributed to mobile learning [2] can create a sense of community.

Questions regarding the educational benefits are asked. The questions ask if the user could see themselves using Vault for learning and teaching respectively. As this is the goal of the application these are open questions, so the participant can answer the most suitable answer. This is done in order to let the participant fully explain why, or why not, this is the case. Thereafter, questions regarding how relevant and beneficial the

application can be in both their personal and professional lives are asked. This is done to get insight into the potential benefits of the application and in what aspects they

manifest. These questions are answered using a linear scale of 1-5, one being least and five being most beneficial/relevant.

After having some questions regarding the user experience, used for future development of Vault, there are some questions for additional feedback from the participants. These questions include any additional feedback on user experience and any overall feedback for the application. This is followed by two open questions the potential benefits of the application in both private and professional settings. These questions are open in case the participants have any additional thoughts or comments on the application.

17

4

Results

The results of this thesis consist of three different parts: firstly, the results from the analysis of the popular mobile-learning applications; secondly, the results from the development of a mobile-learning application that focuses on relational learning; and finally, the results from the application test.

4.1 Analysis of selected applications

The three selected mobile-learning applications were analyzed following the structured form as described previously, the form can be found in its entirety in appendix 1. The filled-out forms can be found in appendix 3.

Table 1 Table for the results from the form

Question/App Udemy Coursera Duolingo

Can all users create

content? NO NO NO

Can all users view all

content? NO NO YES

Can all users interact

with content provider? YES YES YES

Can users interact

with each other? YES YES YES

Can users ask questions about content?

YES YES YES

Can users give

feedback to content YES YES YES

Score: 4 4 5

As seen in Table 1 the results show that the mobile-learning applications analyzed all achieved a high score on the form. This means that they all have a lot of features that would constitute as relational learning features. There are, however, some comments that should accompany these results. As the questions provided in the forms are of a general nature some complementary comments were provided to each question. The comments are needed to further explain the extent of the availability of the feature.

4.1.1 Udemy

Udemy features courses that are created by users, or in some cases by companies. These courses are, however, subject to the review of the company Udemy. Therefore, some content will be rejected or required to be altered in order to be published. This caveat would in turn make it so that not all users are able to provide content for others to partake, thus resulting in Udemy not getting a point in this section. Udemy has a lot of content that is only available to users that have paid for it. This limitation makes it so that it cannot be considered to make all of its content available for all of the users. Udemy uses a Q&A (questions and answers) section for the communication between users. In this section, the user can post a question and get it answered by either other

18 users or by the content provider. It is using this section that the user can both give

feedback to the content provider and interact with the other users that are in that course.

4.1.2 Coursera

Coursera’s content is provided by individuals or a group of individuals. The content

provider has to be affiliated with one of Coursera’s 150 partner institutions. If the content provider is an instructor or professor from an affiliated partner institution they are still required to send in an application and await approval. This makes creating content on Coursera difficult and definitely not something that all of the users can do. The same is true for the consumption of the content on Coursera. Most courses offered are

subscription-based courses. This means that the user must pay a fee for every month that s/he is enrolled in the course.

Coursera uses a forum for all of its communication, both between users and with the content provider(s). Yet, these forums do not follow a specified format. Some courses have forums dedicated to giving feedback to the content providers while other courses merely have forums for the discussion of a week’s lecture/assignment. This makes it so, in theory, users can communicate feedback with the content provider, assuming that the content provider has created a forum for feedback and/or reads through the posts in the other forums.

4.1.3 Duolingo

Duolingo’s content is mainly provided by Duolingo themselves. Although a lot of the content is provided by so called Course Contributors. These are regular individuals that apply to volunteer as contributors to a specific course. The user is required to send in an application in order to be considered. This makes creating content difficult for regular users and makes most users merely consume the content as learners and not share their own knowledge.

Duolingo is the only application, out of the applications analyzed, that offers all of its content for free. The application does have some small parts, referred to as Bonus Skills, these are sections that the user can buy using Lingots. Lingots are an in-app currency that the user gets by completing more sections of each language course. This, in turn, makes all of the content on the platform free as a user will earn Lingots from using the application.

Duolingo has a comment function built into each section of a language course. This is used to enable discussion amongst the users. In the comments, the users can ask

questions if they do not understand a certain word or why it is used. Other users can then share their own knowledge in order to answer the questions or clear up uncertainties. The feedback to the content provider is done by reporting a problem with a specific exercise.

4.2 App development

The development of Vault uses the features addressed in the analysis of the selected applications, which in turn are based on the relational learning features found in 3.1.

19 These features included enabling all of the users to be able to become content providers and create content based on their interests and knowledge, this achieves the feature of authoring. All users of Vault are able to view all of the content in Vault, this is to achieve the feature of content visibility. In addition to allowing all users to be able to view

content, it is important to provide all of the users with the ability to comment on the content provided by other users, this feature achieves the interactivity and the ability to give and respond to feedback. The comments are needed in order to enable the bilateral exchange of information and opinion alike.

The design of the application is using some guidelines provided by Google in their Material Design. The application has no main menu as all of the content of the application is accessed from the main screen (as seen in figure 4).

4.2.1 Content Creation

As the analysis of the selected applications showed, enabling all users to be able to provide content is not something that is to be expected from every mobile learning application. Vault allows for all users to become content providers. The only requirement is to create an account on the platform. This is to be able to tie the post to a content provider. Being a mobile learning application that focuses on relational learning, having content for the users to consume is paramount. Therefore, providing a simple to use interface for the content creation was developed.

Each educational content is called a post. A post is meant to contain one coherent piece of educational content that other users can consume and discuss. Each post, in turn, contains post items. Post items are components within the posts. These post items can be either a picture-, video-, link- or text-item depending of the content. The post items can contain plain text, or text accompanied with either pictures, video or a link to an outside reference. Having these smaller components simplifies the process of creating content and having a sensible structure at the same time.

In the example displayed in figure 2, a post is created with the title “Real life uses of arithmetic”. This example is to show how a content provider creates a post and the steps required in this process. The example given is a post containing a link to an outside source with real life problems in which arithmetic is required. The goal of the post is to give the learner an understanding of the benefits of learning arithmetic, and how important they can be to understand in real life, as well as in academic life.

20 As seen in Figure 2 the user has entered both the title and the description of the post. The description acts as an abstract, giving the learner a quick understanding of what the whole post is going to contain. After adding both the title and the description the user proceeds to create the first post item. This item will be a link post item as it is intended to contain a link to an outside website.

As seen in Figure 3 the user has completed the post item. Each post item contains a header to give more context and information of its content. After the heading of the post item the user has added the full address to the desired website. Under the heading “Add your Text” the user can add additional information regarding the, in this case, link

associated with the post item. This text can give context to the post item or expand upon what is presented in the post item i.e. additional explanation of a picture, explaining why a video is good etc.

Additionally, the user can attach a GPS location (as seen in the lower part of Figure 3) to the post item if it is relevant for the learner to see or know of a certain location. The addition of GPS location is not a feature that has been deemed to be a relational learning feature, as it does not help to achieve either of the four major features of authoring, content visibility, interactivity or ability to give and respond to feedback. Instead, GPS location is only there for those users that feel like it would be relevant to their educational content. The GPS location can be chosen by either searching for a location by name or using the current position of the device. By searching for a location, the user can attach a location that they are not presently at and thus allows for more versatile use by being able to, for example, add the location of the Colosseum in Rome if the post item talks about

21 the Colosseum. In the current state of the application other users cannot search for posts based on location, although this would be a valuable feature for future implementations. In Figure 2 the newly created post item is visible at the bottom of the screen. If the users choose to create additional post items to the post they will appear below the first in a list. These post items can be rearranged in order to create a better structure and to maintain a red line in the content. In addition to the header of the post item, these is a small icon showing the user what type of post item it is. This is to help the user keep track of what each post item is. If the user wishes to view or modify the content of a post item they can click the edit pencil on the right-hand side. This will open up the view in Figure 3 again where the user can change or simply re-read the content of the post item.

4.2.2 Content consumption

After the content provider is satisfied with the post they simply finish by clicking the double checkmark (top right corner in Figure 2). This creates the post and uploads to the server. From here, all of the other users of Vault will see the post in their Feed. This feature achieves full content visibility as all of the users can view all of the content on the platform. The feed is the main page of the application, in which all the posts from other users will appear (see Figure 4). This main page is the page where users land after logging in and from this page users can access all the features in the application. Therefore, no menu is needed in order to navigate to different parts of the application. In the feed the title, image and the short description of the post is displayed. An image will only be shown if the post contained a picture post item. The name of the user that created the post will be shown, in combination to their profile picture. There is also a button for adding the post to the user’s favorites. The favorites are saved under the tab Favorites that can be found in the top of Figure 4. Favorites are a way for the user to save posts that are the most relevant in order to find them easier at a later date.

22 Upon clicking on a post, the post will be displayed in its entirety (see Figure 5). The post will be displayed in the same order as the content provider had arranged the post items when creating the post. Each post item will be separated as seen in Figure 5 in order to clearly delineate between two post items. The view of the link post item is shown I the bottom part of Figure 5. The view of picture, video and text post item can be found in Appendix 4, in addition to several other figures showing the different views of Vault. The view of the link post item shows the heading, text and finally the linked site. The link is shown as a small summary of the website, with a header and text, as seen in Figure 6. The header and text are extracted from the website, so the content of this is not in the control of the content provider.

If the user clicks on the link they will be taken to an internal web-browser. This internal browser displays the website in its entirety, as seen in Figure 7. The internal web-browser offers a fully functional website with basic controls such as back, forward and refresh. In addition, the user can opt to open the website in their native browser of choice by clicking the icon in the top right corner. This will open up their selected default

browser on the smart phone. Figure 4 Main feed of Vault, with newly

23 At the bottom of the post there is a comment section, as seen in Figure 6. It is by

utilizing the comment that Vault achieves both the feature of interactivity as well as the ability to give and provide feedback to the content provider. Any and all users can write a comment to share feedback or information, including the content provider. This enables the bilateral exchange of information between content provider and learner, as well as interactivity between learners.

4.2.3 Backend

The backend is an API written in Golang that has a database containing all the data regarding the posts, comments and users. The API has eight different endpoints. An endpoint is a unique address that goes to the specific function declared in the

backend[28]. Each endpoint refers the backend to a different function, each function is a set of code that executes upon request. The API gives Vault the ability to make API requests either get or post information. A request is a call to an endpoint of the API. There are four different types of API request that Vault can make to the API: GET, PUT, POST and DELETE. The GET type is used when Vault wants to get information stored in the database. The PUT type is used when Vault wants to put information to an existing item, i.e. update the information in the database. The POST type is used to post create new content in the database. The DELETE type is for deleting content in the database.

By using the same endpoint but a different request-type the backend will call on different functions and thus perform different actions. When each new post is added to the

Figure 6 View of the comment section

Figure 7 View of the link opened in the internal web-browser

24 database they are given a unique ID to identify them with. The same happens when creating a new user.

Table 2 Showing and explaining the different endpoints of the API

Endpoint

Type

Description

/posts/X/Y

GET Gets the posts in the order they were added to the database. Starting from position X in the database and getting the following Y posts.

/posts

/posts

PUT Updates a specific post in the database. The ID of the post it sent in the attached post, so no ID is required in the endpoint.

/posts/X

/user

/user/X

/comment

/comment/X

of X from the database.

The GET request for getting posts only operates by the IDs of the posts that are published. The GET request is used every time the application gets new posts from the server. This happens both on updating the feed to get new posts as well as after uploading a post to update the feed with the newly created post. As the backend is designed there is no way to get posts with a certain topic or to get posts for a specific location. Adding the capability of getting posts based on these two factors should be considered for future implementations. By adding the feature to search for posts as well as adding backend support for getting topic or location specific posts would enable users to search and find relevant content easier.

4.3 Questionnaire

The questionnaire was sent out to the test participants at the end of the test. As the questionnaire was created using Google Forms, all of the responses were collected

digitally. The results were displayed in both a summary and in an individual version. All of the six filled questionnaires can be found in Appendix 5. As the participants of the test were native Swedish speakers they filled out the questionnaire in Swedish. Therefore, all answers in Appendix 5 are in Swedish. The answers used here will consequently be translated into English. As there was no requirement of answering all of the questions in the questionnaire, some questions do not have six answers.

25 The questionnaire was completed by six individuals. All six individuals are females and with an average age of 45.34 years. The ages range from 26 to 52 years of age. When asked how satisfied the participants were with the application, they scored it on average of 3.67 out of a maximum of 5. This score is similar to the score when asked how easy it was to create content within the application. This question resulted in an average of 3.5 out of a maximum of 5. 66.7% of the participants said that they did feel like an active part in a community when using the application. The remainder said that they did not have enough time with the application to get this feeling of community.

When the participants were asked about the educational benefits of the application the responses showed a general feeling of not having enough time with the application. When asked if the participant could use the application to learn five out of the six responses said that they felt like they needed more time with the application in order to properly

evaluate this aspect of the application. However, all of the responses said that they saw the potential of Vault being used for learning.

When asked if the participants could see themselves using the application for teaching, two participants said that they could not see themselves using the application. Either ‘no’ as in to use it with the students or no due the not feeling secure confident with the forum yet. Four out of the six responses said that they might be able to use the application for teaching, but this would also require more time with the application to properly evaluate. One response said that they could see themselves using Vault for getting tips and ideas from colleagues regarding relevant and fun activities for the school environment.

When asked about the best feature of the application there were four responses. Two of these responses said that the best feature was to upload ideas for other teachers or tips for lessons. The remaining two said that the best feature was the ease of creating the educational content for the application.

When asked about the relevance and benefits of the application in the participants professional life five participants responded. The average score was 3 of 5. The lowest score was two and the highest score was four. This score is significantly higher than the score of the relevance and benefits of the application in the participants private lives. For this question all six participants responded with an average of 1.84 out of 5.