Mobile Learning is Coming of Age - What we have and what we still miss

Mobile learning has left the status of a new born child. It is time now to implement some structure into the complex and various activities by a framework presented in this paper. The framework classifies mobile learning in the five categories free, formalised, digital, physical, and informal context. Examples for each category are given. Overviewing the existing projects, the framework allows some analysis about missed out potentials of mobile learning. It helps to avoid the repeated reinvention of the wheel. 1 How to Classify Mobile Learning Mobile learning is a rather new term which received ongoing attention during the new millenium when mobile technology started its strong impact on society. Until now there were only few attempts from literature reviews to give mobile learning some meta-structure which allowed systematic analysis. Some early literature reviews are limited to a collection and description of diverse bundles of activities and projects [Tri03, LNB03], but without an underlying consistent classification. It is not obvious by which criteria a classification of mobile learning could and should be structured in a meaningful way. A classification by deployed technology leads to a dead-end street, because technology is the least stable component in mobile learning. Technology can often be replaced or varied without changing the didactical concept. Roschelle [Ros03, pg.262] tried a useful classification by types of application, i.e. classroom response systems, participatory simulations, and collaborative data gathering. Even this classification was neither generic nor complete, but it covered a good deal of mobile learning activities at that time and still does today. A very promising classification has been suggested by Naismith et al. [NLVS05]. They classified mobile learning projects by ”six broad theory-based categories of activity”[NLVS05, pg.1], i.e. behaviourist, constructivist, situated, collaborative, informal/lifelong, and support of learning and teaching. Those categories are mainly based on pedagogic paradigms. But there are three weaknesses in this classification. First, the pedagogy within a mobile learning project is not as stable as one would assume. Even small changes in the design might shift the project into another category without having changed anything significant. Second, the categories are not sufficiently distinct. A mobile learning project can for example be collaborative, situated, and informal at the same time, which makes it impossible to place a project clearly in one specific category. Without a clear placement, projects can hardly be bundled and compared to analyse common patterns, specific requirements, similar added values, or potentials. It will be hard then to formulate specific guidelines and recommendations. Third, the categories are not explicitly linked with each other. They only allow at best a static placement, but no direction. There is no suggested ripening path for projects to develop by and by from initial phases towards a stage, which exploits the full potential of mobile learning. This paper suggests a persistent, differentiated, but nevertheless intergradient classification, which is supposed to overcome those weaknesses (see chapter 3). Looking for a sufficient meta criterion, the author suggests context to be adequate. Embedding learning in context is the specific value of mobile learning. A classification by context helps to apply technology more tightly focused and to evaluate it. Nyiri reveals succinctly the fundamental role of context in education saying ”Knowledge is information in context”[Nyi02, pg.4]. Sharples et al. define context in direct relation to mobile learning when they state: ”Context is constructed by learners through interaction: To explore the complexity of mobile learning it is necessary to understand the contexts in which it occurs. Context should be seen not as a shell that surrounds the learner at a given time and location, but as a dynamic entity, constructed by the interactions between learners and their environment. For example, visitors to an art gallery continually create contexts for learning from their paths through the paintings, their goals and interests, and the available resources including curators and other visitors.”[STV05, pg.5] Naismith et al. hold a more pragmatic view of context, writing: ”Mobile devices give us a unique opportunity to have learners embedded in a realistic context at the same time as having access to supporting tools.”[NLVS05, pg.15]. On the basis of the most popular conferences (MLearn 2003-2005, WMTE2004-2005), the Journal of Computer Assisted Learning, recent large research projects (mLearning, Mobilearn) and prior literature reviews [Ros03, NLVS05, Tri03, LNB03] the author detected and bundled about 120 projects by five categories. 51 systems have been chosen to be presented in this paper.1 The five categories are: free, formalised, digital, physical, and social context. An earlier version of the classification has already been used to structure thoughts in a prior publication [SF04]. There it had no value in itself, was not reflected and neither scientifically motivated. It was not introduced as being an instrument for analysis. Chapter 3 catches up with these shortfalls, argues the intergredient nature, and presents a revised version of the classification. But before, in the following chapter each category will be explained in detail and the referring mobile learning projects will be disclosed. Thus it will indicate the persistence and differentiation of each category. In chapter 4 the review from this paper will systematically be used to analyse the vacancies and undiscovered potentials of mobile learning and extrapolate the next steps. 1The chosen systems are either unique, well-known, well-described, or specifically adequate to outline the type of category. Less known systems with little innovation or with a lack of sufficient description have been sorted out. 2 Categories of Context in Mobile Learning