Iterative Design of Mobile Learning Systems for ...

Tech., Inst., Cognition and Learning, Vol. 0, pp. 000–000 Reprints available directly from the publisher Photocopying permitted by license only

© 2008 Old City Publishing, Inc. Published by license under the OCP Science imprint, a member of the Old City Publishing Group

Iterative Design of Mobile Learning Systems for School Projects ANDRÉ MELZER1, LIA HADLEY2, MARIE GLASEMANN3, SILKE GÜNTHER2, THOMAS WINKLER2,* AND MICHAEL HERCZEG2 University of Luxembourg, Campus Walferdange, Research Unit INSIDE, L-7201 Luxembourg, Luxembourg 2 University of Luebeck, Institute for Multimedia and Interactive Systems, D-23538 Luebeck, Germany 3 Aalborg University, Department of Communication, DK-9220 Aalborg, Denmark

1

Different approaches have been proposed on the design and development of digital and interactive technology in mobile learning contexts. In line with recent findings in literature, we suggest that mobile learning systems benefit from a criteria-based iterative design and development process that incorporates evaluation results of mobile learning scenarios based on school curricula. Underlying these scenarios is a holistic, socio-technical system view, which makes allowance for the complex dynamics between teachers, students, researchers, the multiplicity of contextual factors, and the specifications and requirements of the digital devices and applications, as well as their interrelations. Five school project scenarios are described, as well as the evaluation process involved that served as an integral part in creating a dynamic optimization process for the design and development of digital and interactive systems that follow efficiency criteria for mobile learning (i.e., construction, contextualization, communication, and control). In addition, we argue that mobile systems should also allow for users’ or students’ participatory activities in the preparation phase to bridge the gap between indoor and outdoor learning. Keywords: Mobile learning, schools, iterative design, efficiency criteria, digital and interactive devices.

*Corresponding author: [email protected]

1

IADIS-2 (Melzer).indd 1

1/20/2009 2:54:33 PM

2

MELZER ET AL.

1 INTRODUCTION Mobile learning has become a major topic for the learning research community within the last 10 years. The research interest is fuelled by the ever-accelerating progress of technology, the growing pervasiveness of low-cost mobile devices, and their tremendous attractiveness for adults, adolescents, and children. The technological and social changes that accompany the current ubiquitous use of mobile devices are also reflected by the widely accepted notion that learning is continual and happens anywhere at any time (e.g., Gee, 2007). Theoretical approaches acknowledge that learning is by no means limited to formal contexts based on curricula and organized instruction. Rather, additional self-organized emergent learning in informal contexts outside the classroom (e.g., field trips) is also thought to play an important role (for a review on formal and informal learning, see Bransford, Vye, Stevens, Kuhl, Schwartz, Bell, Meltzoff, Barron, Pea, Reeves, Roschelle, & Sabelli, 2006). The prevalent learning situation in schools has been characterized to stand in sharp contrast to, for example, Dillenbourg and Fischer’s (2007) concept of integrated learning. According to Sharples, Corlett, and Westmancott (2002) learning in schools has predominately focused on classroom-bound learning, mediated by a trained teacher presenting their material in a theoretical or an abstract form “out-of-context”, rather than providing coherent learning experiences in outdoor contexts. This abstraction makes it difficult for students to gain immediate access to the inherent core of the learning material. Rather, mental effort is needed to see the relevance or practical implications. Ultimately, the classroom situation may even preclude learners from actively and dynamically constructing knowledge through interaction in the relevant learning context. De-contextualized learning situations provide sub-optimal learning conditions given the importance of successful learning as a constructive process (e.g., Brown, Collins, & Duguid. 1989; Lave & Wenger, 1991; Papert, 1980). To avoid artificial or de-contextualized classroom learning, didactic elements have been devised, such as presenting examples, analogies, or exercises. Alternatively, mobile learning has been suggested for a variety of learning goals providing a rich and meaningful context for immediate or natural interaction with the material (Rogers, Price, Randell, Stanton Fraser, Weal, & Fitzpatrick, 2005). Carefully designed and custom-tailored mobile learning applications have been found valuable to bridge the gap between indoor and outdoor learning (Dillenbourg & Fischer, 2007; Melzer, Hadley, Glasemann, & Herczeg, 2006; Moulin & Piras, 2006; Rogers et al., 2005; Sharples et al., 2002). In particular, mobile learning applications serve as assistive didactic tools that contextualize learning

IADIS-2 (Melzer).indd 2

1/20/2009 2:54:33 PM

ITERATIVE DESIGN AND MOBILE LEARNING

3

contents previously introduced in the classroom. Applying mobile tools for educational purposes is thought to increase students’ opportunities for interaction and sharing ideas (cf. Laru & Järvelä, 2008). This is achieved by exemplifying and augmenting formal classroom learning experiences in outdoor scenarios, thus enabling active and dynamic construction for understanding activities (Dillenbourg & Fischer, 2007; Sharples et al., 2002). Applying mobile learning tools complies with aspects that are shared in several theoretical approaches, for example, activity theory (Vygotsky, 1978), constructivism (Brown et al., 1989), and situated learning (Lave & Wenger, 1991). In the present paper, we report our 5-year-experience with the KiMM (Kids in Media and Motion) initiative; a research and transfer project at the University of Luebeck, Germany. KiMM promotes action-based and body-centred forms of instruction for students and teachers in the classroom (Winkler & Herczeg, 2005). We will confine ourselves to one major aspect of this initiative, which is devoted to mobile and ubiquitous learning in primary and secondary schools respectively1. The KiMM initiative follows an iterative design and development process within a holistic system view to structure the complexity of the factors involved, as well as their interrelations and interactions. The holistic or socio-technical system view acknowledges that successfully bridging indoor and outdoor classroom learning is a complex task, which includes the use of technology in an individual, social, and organizational context (Sharples et al., 2002; Taylor, Sharples, O’Malley, Vavoula, & Waycott, 2006). The socio-technical system view is widely used both in software engineering and organizational psychology (e.g., Cherns, 1987). Within this framework, evaluations of school projects scenarios provide an ever-increasing understanding of how to deploy, implement, and improve mobile learning systems. Iterative design and development ideally results in mobile learning systems that fit seamlessly into the overall situation mentioned above (Sharples et al., 2002; Taylor et al., 2006). In the next part of this paper, we look at key concepts and criteria in mobile learning. Knowledge of these criteria is indispensable for custom-tailored iterative design, because they represent the benchmarks in every step of the development process. In addition, we analyse the role of students’ participation in preparing content with digital and interactive devices in the context of mobile learning. In the third section, a “hands-on” section, five school project scenarios are described, as well as our systematic evaluation process in respect to indoor and outdoor classroom learn-

Other KiMM research activities focus on interactive mixed-reality performances and installations, and web cooperation and pervasive gaming, respectively (see also http://www.kimm.uni-luebeck.de).

1

IADIS-2 (Melzer).indd 3

1/20/2009 2:54:33 PM

4

MELZER ET AL.

ing. The results of each project evaluation process pointed the way to the next iterative step in the design and development of our Moles and Mini Moles mobile learning software system. In the concluding section, we argue that the evaluation results served as an indispensable input for the development process of digital and interactive devices to optimise the KiMM initiative’s mobile learning system.

2 KEY CONCEPTS AND CRITERIA IN MOBILE LEARNING AND THE ROLE OF PARTICIPATION Mobile technologies and devices such as handheld computers, PDAs, or mobile phones, have extended rather than replaced the existing learning tools (Trifonova, 2003). This expedient coexistence benefits from the technology’s advances in computing power. In addition, a plethora of digital and interactive devices is available at dramatically reduced costs compared to systems from just a few years ago. However, effectiveness and availability do not automatically implicate appropriateness or efficiency of technology (Dillenbourg & Fischer, 2007; Nova, Girardin, & Dillenbourg, 2005; Sharples et al., 2002). 2.1 Efficiency Criteria for Mobile Learning With respect to the efficiency of mobile learning, we have already mentioned two specific criteria for the selection and application of useful technology in the introduction, namely construction and contextualization. With respect to construction, mobile technology should encourage students to reflect, explain, and hypothesize about the physical world outside the classroom (Robertson & Good, 2004). In doing so, new experiences relate to existing knowledge. Well-designed mobile learning systems thus entail cognitive processes known to be essential for human memory; namely verifying, elaborating, and deeply processing study material. This provides a basis for retrieving the learned material successfully at a later point in time (e.g., Craik & Lockhart, 1972). Mobile learning systems provide the means for interaction in the relevant learning context with respect to contextualization, thus addressing the aforementioned goal of classroom-bound learning that tries to avoid presenting abstract (i.e., de-contextualized) learning material or information. The context of using mobile systems also has a significant social or even community aspect, which involves the learner, other co-learners, the teachers and their parents (Barnard, Yi, Jacko, & Sears, 2007; Taylor et al., 2006). The social aspect points to a third efficiency criterion of mobile learning systems: supporting learning as a continual communicational process. A shared

IADIS-2 (Melzer).indd 4

1/20/2009 2:54:33 PM

ITERATIVE DESIGN AND MOBILE LEARNING

5

understanding of the world evolves through mutual communication and effortful interactions—a hallmark of collaborative learning (Dillenbourg & Fischer, 2007). With respect to designing mobile learning resources, sophisticated systems provide students with a heightened sense of awareness to the activities of other members of the community context, and/or support multiple conversations between members (Kraut, 2003). This may be achieved, for example, with networked computers that allow for mutual discussions (Dillenbourg & Fischer, 2007; Sharples et al., 2002). 2.2 Participation in Preparing Mobile Learning Content Contextualization, construction, and communication have only been described from a theoretical view so far. This section is devoted to transference of efficiency in designing and implementing practical mobile learning scenarios, which comprise indoor or classroom preparation, the actual outdoors experiences, and the concluding presentation and discussion phase that typically occurs back in the classroom. To this end, Figure 1 shows a simplified representation of two approaches to designing mobile learning tools and deploying mobile learning projects in schools that differ in terms of students’ participation in the preparation phase of mobile learning scenarios.

FIGURE 1 Two learning approaches that differ in students’ participation in creating outdoor learning material. Students are either involved in the construction process that occurs in the classroom (“KiMM”, lower half), or not (“Gap”, upper half).

IADIS-2 (Melzer).indd 5

1/20/2009 2:54:36 PM

6

MELZER ET AL.

Figure 1 illustrates that both approaches place heavy emphasis on active contextualization, communication and construction throughout the outdoor mobile learning experience and the concluding reflective phase. Students use mobile devices (e.g., PDAs, external sensors) for retrieving information and recording their findings, and they make use of the system’s connectivity to other media to present and discuss their experiences and findings with other members of the learning community (i.e., students and teachers both locally and globally). Placing students in a situation in which they need to engage in effortful interactions supports the intrinsic effort of an individual to understand what the other means, which, in turn, leads to the co-construction of shared understanding (cf. Dillenbourg & Fischer, 2007). Hence, students, or learners, have to actively control learning so that they select, use, and modify the technology and devices, which denotes another important efficiency criterion in mobile learning, namely control (e.g., Taylor et al., 2006). Controlling or taking charge of learning by choosing and setting goals and using individual strategies in the context of the learning process is a hallmark of self-regulated learning theory that focuses on learners developing learning skills and using them effectively (Boekarts, Pintrich, & Zeidner, 2000). In both approaches the students are integral participators in the classroom communication. Yet often only the teachers are responsible for the authoring, or construction, process during the technological preparation phase for outdoor leaning scenarios. For instance, the teachers specify the activities and application requirements in terms of the interactions that should occur (Lagos, Alarcón, Nussbaum, & Capponi, 2006). The KiMM approach extends the first approach with respect to the students’ role in creating mobile learning content because it includes their participation in the preparation phase of the mobile learning project onwards, thus bridging the gap between indoor and outdoor learning (Rogers et al., 2005). The gap between classroom learning and outdoor experiences is by no means a novel finding. Rogers and her colleagues (Rogers et al., 2005) have already concluded “this separation of interlinked activities can make it difficult for children to see and understand the connections between what are essentially the same representations and processes being studied, albeit in different contexts” (p. 56). The KiMM approach strives to involve the students from the preparation phase onwards. Thus, they actively construct and control the authoring process of creating the content. This encourages students to reflect and hypothesize about what they have learned (i.e., constructed) so far, and how this information may relate to the physical world they are about to experience. The participatory role of the students from the preparation phase onwards calls for mobile learning systems

IADIS-2 (Melzer).indd 6

1/20/2009 2:54:37 PM

ITERATIVE DESIGN AND MOBILE LEARNING

7

applicable both in the classroom and during field trips, thus covering both indoor and outdoor learning. In addition, the KiMM team invited the students, together with the teachers, to participate in the iterative design process of our mobile learning software system, Moles and Mini Moles (see next section), which complies with the above-mentioned efficiency criteria. The benefits and advantages of students’ early involvement and participation in technologically oriented design processes have already been demonstrated (Guha, Druin, Chipman, Fails, Simms, & Farber, 2005). In the next section, we present the descriptions of five concrete mobile learning projects that corroborate the fact that students’ and teachers’ participation has contributed greatly in helping the KiMM initiative to iteratively design and develop a flexible underlying architecture that covers a broad range of mobile learning scenarios, thus warranting generalization and transfer of the learning system (Melzer et al., 2006).

3 “HANDS-ON”: USING EVALUATION METHODS FOR ITERATIVE MOBILE LEARNING SOFTWARE DEVELOPMENT The socio-technical system view predicts that different media technologies and applications will yield positive learning effects depending on the specific mix ratio of technology and its social and organizational context. Therefore, we designed a flexible evaluation framework (i.e., questionnaires, students’ logs, teachers’ project reports, and interviews; Melzer, Hadley, & Herczeg, 2005a). The evaluation method was adjusted according to the scope of the learning material, the specific age group of the class, the target group (teachers or students), and the range of hardware and software used in the project. 3.1 General Evaluation Method In five KiMM mobile learning projects, a total of 347 students and 26 teachers completed approx. 17,000 questions (either pre-and post-project questionnaires, or only post-project questionnaires). Due to the holistic approach of the KiMM concept the evaluation incorporated various target dimensions (Melzer et al., 2005a). Each dimension used a scale-like measure, contained a varying number of items, and differed in scope according to the specifics of the project in consideration:

s s

Previous exposure to/experiences with media (e.g., “rate your own computer expertise”) Cognition, problem solving (e.g., creativity: “did the device support/trigger new ideas?”)

IADIS-2 (Melzer).indd 7

1/20/2009 2:54:37 PM

8

s s s s s

MELZER ET AL.

Emotion (e.g., motivation, emotion, satisfaction: “satisfied with personal outcome?”) Behaviour (e.g., teamwork: “fair distribution of devices among group members?) Media (e.g., handling, applicability: “device/application well-suited for the task at hand?”) Expectation and apprehension (e.g., concerning the curriculum: “learning goals fully met?”) Judgment of KiMM support (e.g., “sufficient training before using the devices?”)

In addition, students were required to record, in project logs, their observations and reflections throughout the project duration. These protocols listed what tasks they were to do when, their general observations about group dynamics, communication difficulties, technical glitches, end results, and their own feelings about working in a media-supported collaborative work project. The teachers submitted project reports, which were a résumé of the projects, as well as their own experiences and suggestions on how to improve the learning experiences and technology. At the end of the project, the students, teachers, and researchers participated in either a classroom discussion round or a feedback session. The information derived from project logs and reports turned out to be invaluable in clarifying weaknesses and presenting alternative options needed in the iterative design process of the Moles and Mini Moles software system. With respect to the persistence of learning effects, unfortunately, teachers’ reports may not be used as a source of external validation of the long-term success of mobile learning projects. This is due to the fact that teachers test knowledge skills either immediately or shortly after the project ended. Hence, as of yet, we do not have quantitative data from experiments that corroborate long-term learning. Interestingly, at the end of the school term most teachers reported in informal discussions that mobile learning projects persistently changed both the “social and the working climate” in the class. We cannot specify, however, whether the change in students’ behaviour is a direct effect of using mobile learning systems. Alternatively, it may reflect the media-unspecific effect of the intensified group work during the projects that provided a positively connoted social experience that increased group coherence in the class. We have presented detailed analyses of the pedagogical success of various KiMM initiatives’ mobile learning projects elsewhere (Melzer et al., 2006; 2005a; Melzer, Hadley, Winkler, & Herczeg, 2005b). Hence, we will only briefly describe five consecutive school project scenarios, the quintessential information derived

IADIS-2 (Melzer).indd 8

1/20/2009 2:54:37 PM

ITERATIVE DESIGN AND MOBILE LEARNING

9

from the evaluation process that signified another step in the iterative design and development process, and the resulting software application. Each scenario/ design step will be described with respect to its strengths and weaknesses in terms of the efficiency criteria for mobile learning tools. 3.2 Selected Projects and Iterative Design of Mobile Learning Systems The particular subject material of each KiMM mobile learning project in schools was derived and then contextualized from the class curriculum and elaborated in close cooperation with the teachers. This contextualization approach bears close resemblance to the concept of contextual design in software engineering, which has become a well-established approach for consumer-centred systems. Contextual design, which also rests on the socio-technical system view, combines field analyses within the users’ natural environments, requirements analyses, systems conceptualization, and testing (Holtzblatt, 2005). 3.2.1 White Spot History Hunt (WSHH) In this early practical application, members of the KiMM initiative and 8th grade high school students designed a prototype scenario, an interactive history scavenger hunt (Melzer et al., 2005a). WSHH was a pervasive game settled in the time of the Industrial Revolution. The WSHH teams used smart phones and GPS devices to discover a series of white spots on a virtual map of 19th century Luebeck. Navigators and history hunters were located in school classrooms and used an Internet chat program to communicate with their colleagues, the runners, in the city. In the evaluation, the teacher indicated that students’ learning was substantial. However, although both the students and the teacher enjoyed the project, they were overwhelmed with the level of technological complexity. This was mainly due to problems of control and contextualization, for example, students not being able to edit content and the teacher not being able to handle the devices. As a consequence of this, we designed the open source mobile learning exploration system, Moles and Mini Moles. This system serves as a flexible tool for the contextualization of mobile learning projects and combines classroom learning (Moles) and mobile learning (Mini Moles) to further the process of construction (Melzer et al., 2006; 2005a; 2005b). 3.2.2 Mobile Weather Project A 7th grade class created a Mobile Weather Project to determine, analyze, and record the variation of weather elements in their inner city environment. The students researched and used material acquired from three class subjects (geography, mathematics, and physics) to create interactive multimedia questionnaires

IADIS-2 (Melzer).indd 9

1/20/2009 2:54:37 PM

10

MELZER ET AL.

using the JAVA based and XML structured Moles software on laptops (Figure 2). Hence, they actively participated in the construction phase. Using Moles they translated the contents of their prior classroom learning into questionnaires and test kits (or self-tests) for subsequent outdoor learning. During their field day, they measured and recorded weather information at five different work stations distributed throughout the city using the Mini Moles software on PDAs. In this project the students used a variety of software (e.g., Moles, Mini Moles, graphic software, text processing) and hardware (laptops, handheld GPS, PDAs, external sensors to record air pressure, temperature, and wind velocity, digital cameras) (Melzer et al., 2005a; 2005b). The evaluation of the Mobile Weather Project (Melzer et al., 2005b) indicated a substantial increase in students’ self-reported computer skills from pre-project (M  1.76, SD  0.88) to post-project ratings (M  4.28, SD  1.43, six point scale with greater numbers indicating higher level of agreement). The teacher stated, however, that curricular goals were not fully met. In addition to problems with handling the devices, both students and teacher agreed that the mobile learning system did not provide sufficient communication. Hence, a lack of awareness

FIGURE 2 The Moles software GUI with text field for formulating questions (upper arrow), and field for preparing multiple-choice alternatives (lower arrow).

IADIS-2 (Melzer).indd 10

1/20/2009 2:54:40 PM

ITERATIVE DESIGN AND MOBILE LEARNING

11

of the actual status of their colleagues at other work stations resulted. The lack of awareness led to compensatory activities that interfered with actual (curricular) learning goals. Since awareness is a key issue in computer-supported collaborative work (Kraut 2003), and mutual communication is an indispensable prerequisite for establishing shared understanding (Dillenbourg & Fischer, 2007), it is necessary for the software to allow for communication between the geographically separated groups. As a consequence of this, we designed a system version called Mini Moles for PDA that includes a chat function for PDAs with WiFi connections; thus making communication between the groups possible within a limited geographical separation. 3.2.3 Mobile Advent Calendar In this project, 10th grade students created a Mobile Advent Calendar for the first year high school students (5th grade). The 5th graders, new to the school and unfamiliar with the surrounding neighbourhood, went on a digital treasure hunt inside and outside the school to discover the “doors” of their mobile Advent calendar (history, religion, art and music classes). They were equipped with PDAs (Mini Moles), stand-alone GPS, and digital cameras. Each “door” bore a surprise for the 5th graders (e.g., Christmas carols sung in Latin, Spanish, French and English, recited poems by fellow students and teachers of the school). Comparison of pre-project and post-project ratings indicated substantial positive effects on the students’ cognitive level (Melzer et al., 2006). This refers to increases in spatial knowledge and orientation (both in the self-reported data and the teacher’s estimations), and social interaction, respectively. However, the 5th grade students’ ratings indicated that the GUI might be sub-optimal for young students in terms of control. Hence, we developed the Moles for Kids software (Figure 3). Its intuitive GUI (icon-based programming) can be used in mobile learning projects from primary schools grades upwards. 3.2.4 Children, Children! In this museum/school project, a 5th grade class studied the lives of children at the turn of the 18th century (German, history, and music classes). They created aesthetic biographies of five fictitious children (e.g., orphan, pastor’s son, aristocrat’s daughter) in blogs. The students used this material to create a pervasive game (history hunt) with Moles on PCs and then transferred the interactive multimedia questionnaire and test kits to Mini Moles on PDAs. The gamers went to a local castle, now museum, and used the Mini Moles questionnaires and test kits to explore the role of art, education, fashion, hygiene, music, and nature, in the lives of children in the 18th century (Melzer et al., 2006).

IADIS-2 (Melzer).indd 11

1/20/2009 2:54:40 PM

12

MELZER ET AL.

FIGURE 3 GUI for Moles for Kids software. Rather than using the standard folder-and-text-button interface, editing and programming, Moles for Kids is supported by coloured icons and an assistant helper.

Interestingly, post-project ratings indicated that the two teachers (M  5.50; six-point scale with greater numbers indicating higher level of agreement) agreed more strongly to the claim that the mobile learning experience increased the students’ knowledge, than did their students (M  4.20, SD  0.89). The post-project discussions revealed the necessity of developing an automatic or manual localization modus that renders information (i.e., clues and tasks) available to students only if they were physically present at designated locations (i.e., contextualization). This option would create a greater game feeling to scenario-based mobile learning games. The teachers also said they needed assurance (i.e., control) about whether the unsupervised groups actually physically visited the various locations during the outing. Thus, a route-guiding system called Mini Moles Loc was thought to meet the requirements of control and contextualization. 3.2.5 The Fate of Dr. X In this prototype project, 6th grade students (geography and social studies classes) learned that the university hospital is not only a place sick people go to, but also a place where thousands of people work and study. The students played a pervasive detective game using the Mini Moles Loc system (Figure 4), which required navigating their way around the hospital and university campus with PDAs and GPS device. The students went from one hospital location to another, trying to figure out the fate of Dr. X (a fictitious character), who was reported to lie in a coma in the

IADIS-2 (Melzer).indd 12

1/20/2009 2:54:40 PM

ITERATIVE DESIGN AND MOBILE LEARNING

13

FIGURE 4. Students playing “The Fate of Dr. X” (photo middle). Screenshots from the Mini Moles Loc GUI (on a PDA) are displayed on the far left (map navigation), and on the far right (multiple choice questions), respectively.

intensive care ward. Automatic GPS localization or manually entered code words derived from physical indicators (e.g., hospital building numbers or food container id numbers) turned the whole project into a Geogame format (Kiefer, Matyas, & Schlieder, 2006). This enhanced the students’ motivation and learning experiences by combining educational content with location-based information systems and the physicality of their environment. The evaluation results showed that the availability of graphically rich routing information on Mini Moles Loc (i.e., automatic positioning displaying current position and distance between current position and target) did not reduce the time needed for task completion compared to the group equipped with a system version that displayed only the map position of the target in addition to the general map. Rather, the latter group was even slightly faster than the automatic positioning group. We speculate that in the latter condition participants engaged in additional group discussions to compensate for the lack of spatial information. Letting people build their own spatial representations has been suggested to be more efficient than providing them with explicitly displayed spatial information (Nova et al., 2005). The next version of Mini Moles Loc had to extend the limitation to edit the map navigation—a system feature that students reported to miss most. Mini Moles for PDA was therefore extended for use in PDAs and smart phones with built-in GPS. The KiMM team also developed a Mini Moles for Mobile Phone application that allows students to use their personal mobile phones constructively in mobile learning scenarios. Sharples and his colleagues (Sharples et al., 2002) have already stressed the learning advantages of using technology students are familiar with.

IADIS-2 (Melzer).indd 13

1/20/2009 2:54:41 PM

14

MELZER ET AL.

FIGURE 5 Creating a mobile learning game with web-based Web Moles. Web Moles supports the online creation and storage of interactive multimedia questionnaires (via Internet browser on a web server) making them easily accessible through Web Mini Moles (as indicated by the cell phone). Note that Figure 5 depicts a translated screenshot of Web Moles.

The ongoing work with primary and secondary schools will benefit from the next technological iteration called Web Moles and Web Mini Moles, respectively (Figure 5). Web Moles is a web application, which supports educators and learners in jointly creating seamless learning scenarios inside and outside the classroom while using various communication devices, for example mobile phones. Web Mini Moles will run on numerous mobile devices. It simply requires a browser and Internet access. A prototype version of Web Moles and Web Mini Moles is already available (http://moles.mesh.de/).

4 CONCLUSION Educational research has to acknowledge the challenges of advancing technological progress and increasing availability of mobile systems and devices. An expedient response to this challenge ties together technological progress and its growing availability and the conviction that learning is no longer bound to the classroom but takes place anywhere and at any time. It also encourages learning experiences through various mobile learning approaches. In our opinion, the current multiplicity of approaches and methods involved in mobile learning is not a weakness, but analogous to the broader domain of computer-supported collabora-

IADIS-2 (Melzer).indd 14

1/20/2009 2:54:42 PM

ITERATIVE DESIGN AND MOBILE LEARNING

15

tive learning (CSCL), it should be perceived as a sign of maturity that acknowledges the complexity of computer-supported learning (cf. Dillenbourg & Fischer, 2007). The present paper described an approach that addressed the complex situation of mobile learning systems in the context of classroom-oriented curricula. We presented a summary of the 5-year-experience with the optimization process from the KiMM initiative at the University of Luebeck, Germany. This optimization rested on an iterative design and development process; a well-established approach in software engineering. Each iteration cycle increases the realism of the system to be designed (cf. Wickens, & Hollands, 2000), that is, the system will become more appropriate for the particular field of application. This requires the iterative process to include thorough analyses of the requirements (e.g., curricula, goals, and economic factors), participatory design and contextual design, and carrying out and evaluating school projects in which students use systems and devices. Based on a socio-technical system view, five consecutive school projects scenarios were described and analyzed according to efficiency criteria for mobile learning. These project scenarios were designed to support contextualization of curricular material, to enable processes of construction that encourage students to reflect and hypothesize about the physical world, to provide communication functionality needed for establishing a shared understanding, and to provide mobile devices that offer a maximum of control. In addition, the approach of the KiMM initiative extends existing approaches, because it requires students to actively participate in the preparation phase of a mobile learning project, thus bridging the gap between classroom learning and outdoor experiences. Participation warrants that students (a) reflect and hypothesize on material previously learned in the classroom prior their outing, and (b) learn to use mobile technology purposefully; an important yet sometimes overlooked aspect that separates mobile learning devices from mere gadgets. The information derived from each evaluation process represented a step further in the evolution of our mobile learning system. This evolution follows efficiency criteria for mobile learning tools (Sharples et al., 2002; Taylor et al., 2006). Based on lessons learned from the White Spot History Hunt project (3.2.1), we designed our Moles and Mini Moles software system, a flexible basis for the contextualization of mobile learning projects. The Mobile Weather project (3.2.2) and the Fate of Dr. X project (3.2.5) indicated that earlier versions of the system were lacking indispensable functionality in terms of communication (i.e., social awareness, improvement in data accessibility though online access), and contextualization (e.g., displaying geographical information). The Mobile Advent Calendar (3.2.3) and the Children, Children! Projects

IADIS-2 (Melzer).indd 15

1/20/2009 2:54:42 PM

16

MELZER ET AL.

(3.2.4) showed that a “one-size-fits-all” system does not provide each user with the same high level of control (i.e., age-appropriateness in handling, and positioning function, respectively). In this sense, it was our goal to successfully “orchestrate integrated learning environments” (Dillenbourg & Fischer, 2007). However, we do not expect the Moles and Mini Moles mobile learning system ever to reach a “final version”, thus acknowledging the complex and idiosyncratic responsibility for creating mobile learning systems for schools. ACKNOWLEDGEMENT The authors wish to thank the German Bund-Laender Commission for Educational Planning and Research Promotion, the Ministry for Education, Science, Research, and Culture of the Federal State Schleswig-Holstein, Germany, the German Research Foundation (DFG), and the Possehl-Foundation in Luebeck, for their generous financial support. The authors also wish to thank Florian Turck, Thomas Kotewicz and Sebastian Lob for their programming assistance. Simon Werner programmed and evaluated Mini Moles Loc and was involved in the preparation of an earlier version of the present paper. Finally, we want to thank the three anonymous reviewers for their valuable comments. REFERENCES Barnard, L., Yi, S. J., Jacko, S. A., Sears, A. (2007). Capturing the effects of context on human perfomance in mobile capturing systems. Personal and Ubiquitous Computing, 11, 81–96. Boekarts, M., Pintrich, P. R., & Zeidner, M. (Eds.) (2000). Handbook of self-regulation. San Diego, CA: Academic Press. Bransford, J. D., Vye, N. J., Stevens, R., Kuhl, P., Schwartz, D., Bell, P., Meltzoff, A., Barron, B., Pea, R., Reeves, B., Roschelle, J., & Sabelli, N. (2006). Learning theories and education: Toward a decade of synergy. In P. Alexander & P. Winne (Eds.), Handbook of educational psychology (2nd ed.) (pp. 209–244). Mahwah, NJ: Erlbaum. Brown, J. S., Collins, A., Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32–42. Cherns, A. B. (1987). The principles of sociotechnical design revisited. Human Relations, 40, 153–162. Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671–684. Dillenbourg, P. & Fischer, F. (2007). Computer-supported collaborative learning. Zeitschrift für Berufs- und Wirtschaftspädagogik, 21, 111–130. Gee, J. P. (2007). What video games have to teach us about learning and literacy (2nd Ed.). New York: Palgrave MacMillan. Guha, M. L, Druin, A., Chipman, G., Fails, J. A., Simms, S., & Farber, A. (2005). Working with young children as technology design partners. Communications of the ACM, 48(1), 39–42.

IADIS-2 (Melzer).indd 16

1/20/2009 2:54:42 PM

ITERATIVE DESIGN AND MOBILE LEARNING

17

Holtzblatt, K. (2005). Customer-centered design for mobile applications. Personal and Ubiquitous Computing, 9, 227–237. Kiefer, P., Matyas, S., & Schlieder, C. (2006). Learning about cultural heritage by playing Geogames. In R. Harper, M. Rauterberg, and M. Combetto (Eds): Entertainment Computing–ICEC 2006, LNCS 4161 (pp. 217–228). Berlin, Germany: Springer. Kraut, R. E. (2003). Applying social psychological theory to the problems of group work. In J. M. Carroll (Ed.): HCI models, theories, and frameworks. Towards a multidisciplinary science (pp. 325–356). Amsterdam, The Netherlands: Morgan Kaufman Publishers. Lagos, M. E., Alarcón, R., Nussbaum, M., & Capponi, F. (2006). Interaction-based design for mobile collaborative-learning software. IEEE Software, 24(4), 80–89. Laru, J. & Järvelä, S. (2008). Using web 2.0 software and mobile devices for creating shared understanding among virtual learning communities. Proceedings of IEEE WMUTE 2008, Beijing, China (pp. 228–230), IEEE Press. Lave, J. & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, England: Cambridge University Press. Melzer, A., Hadley, L., & Herczeg, M. (2005a). Evaluation of a mixed-reality and high interaction media project in the classroom: Strategies and methods. Proceedings of AACE ED-MEDIA 2005, Montréal, Canada (pp. 3984–3991). Melzer, A. Hadley, L., Winkler, T. & Herczeg, M. (2005b). Developing, implementing, and testing mixed reality and high interaction media applications in schools. Proceedings of IADIS CELDA 2005. Porto, Portugal (pp. 123–130). IADIS Press. Melzer, A. Hadley, L, Glasemann, M., & Herczeg, M. (2006). Using the Moles and Mini Moles software system to bridge the gap between indoor and outdoor learning. IADIS International Journal on WWW/Internet, 4(2), 46–58. Moulin, C. & Piras, A. (2006). Geobook: Enhancing mobile learning with GPS and multimedial book. Proceedings of IADIS Mobile Learning 2006, Dublin, Ireland (pp. 163–169). IADIS Press. Nova, N., Girardin, F., & Dillenbourg, P. (2005). ‘Location is not enough!’: an empirical study of location-awareness in mobile collaboration. Proceedings of the 2005 IEEE WMTE’05 (pp. 21–28), IEEE Press. Papert, S. (1980). Mindstorms: Children, computers, and powerful ideas. New York: Basic Books. Robertson, J. & Good, J. (2004). Children’s narrative development through computer game authoring. Proceedings of IDC 2004: Building a Community. Maryland, USA (pp. 57–64). Rogers, Y., Price, S., Randell, C., Stanton Fraser, D., Weal, M., & Fitzpatrick, G., (2005). Ubi-learning integrates indoor and outdoor experiences. Communications of the ACM, 48(1), 55–59. Sharples, M., Corlett, D., & Westmancott, O. (2002). The design and implementation of a mobile learning resource. Personal and Ubiquitous Computing, 6, 220–234. Taylor, J., Sharples, M., O’Malley, C., Vavoula, G., & Waycott, J. (2006). Towards a task model for mobile learning: A dialectical approach. International Journal of Learning Technology, 2(2/3), 138–158. Trifonova, A. (2003). Mobile learning: Review of the literature. University of Trento – Department of Information and Communication Technology, Technical Report # DIT-03-009. Available under URL, accessed August 11, 2008. Wickens, C. D. & Hollands, J. G. (2000). Engineering psychology and human performance. Upper Saddle River, NJ: Prentice-Hall. Winkler, T. & Herczeg, M. (2005). Pervasive computing in schools—embedding information technology into the ambient complexities of physical group-learning environments. Proceedings of AACE SITE 2005. Norfolk, USA (pp. 2889–2894). Vygotsky, L. (1978). Mind in society: The development of higher psychological processes. Boston: Harvard University Press.

IADIS-2 (Melzer).indd 17

1/20/2009 2:54:42 PM