INTRODUCTION
It is widely acknowledged that assessing learning with ICT is a challenging task (Johnson et al. 1994, McDougall 2001, Harrison et al. 2002, Cox et al. 2003). Throughout the development of the use of ICT for learning extensive work has been undertaken on formative evaluation and assessment of software and of innovative projects, but no similar range of effective and rehable ways of assessing real learning gains attributable to or associated with the use of ICT has so far been developed. Nevertheless it is critically important to develop effective techniques for doing this in the light of the major developments and growing investment in ICT resources for learning, including of course real-life learning.
INDIVIDUAL DIFFERENCES AMONG LEARNERS
The matter of individual differences among learners has long been an issue in education. Current research is showing that such differences can be dramatically evident and can have a major impact on assessment in settings where ICT is used. Differences in impact on leaming are being observed in case studies of students involved in exactly the same ICT activity. These are exemplified by findings from a research project investigating the effects of the use of ICT in students’ writing, recently completed by John Vincent at the University of Melboume. Although this study was undertaken with school students, it raises an issue of major importance for leaking beyond school settings as well.
Preliminary analysis of Maine’s data suggests that the teachers do perceive important differences in the leaking experiences provided by the two environments. The assessment issues raised by Vincent apply in the setting in which Mundane is working, but at the time of writing it is too early to establish the relevance of these differences for assessment of programming for robotics and the extent to which they might relate to individual differences in the learners’ cognitive styles.
ASSESSING LEARNING IN MULTIMEDIA CONTEXTS
Vincent described students, almost completely incapable of expression in words, who produced complex and sophisticated narratives when allowed to work in multimedia environments. Judged solely in verbal terms these students appeared to be severely limited in their ability to express their ideas and understandings. However the multimedia artefacts they produced dramatically belie this assessment.
Vincent notes that almost all of the official assessment tools that teachers use are verbally based, so the performance of a leamer is judged by skills with words. He used the Writing Assessment instrument of the Victorian Curriculum and Assessment Authority’s Achievement Improvement Monitor (VCAA 2003) to assess his students’ pieces written with pen and paper and with word processor. However he found that this instrument was completely unsatisfactory for assessing the multimedia products of his students.
ASSESSING LEARNING IN COLLABORATIVE GROUPS
The assessment of group leaming activities, like the matter of individual differences among leaders, is not an issue peculiar to real-life learning. However it is particularly important in real-life leaking contexts as leaming through discussions and group activity are very widely used strategies in real-life leaming settings.
The approach is proving powerful in examining some of the more complex aspects of student learning in classroom and smaller group situations. High-ICT research techniques such as this, investigating the complex communications within groups of learners, will provide insights into group learning processes and – while I am not suggesting the use of three-camera strategies for assessment ~ could provide more sophisticated techniques for assessment of the work of groups and communities of learners.
CONCLUSION
The paper explores some aspects of the complexity of assessing real-life learning with ICT, focusing on three particular issues. The first is the problem of differences in cognitive styles of learners. Some aspects of this were illustrated with a description of research on school students’ writing, and the extension of this work into a current study of adult learners’ preferences for different software environments for programming for robotics.
