Proposal view
| Proposal Type: | Individual Paper |
|---|---|
| Domain: | Learning and Instructional Technology |
| SIG: | Learning and Instruction with Computers |
| Scheduling category: | Self regulation |
| Equipment |
Computer and data projector / beamer |
| Paper Details |
|---|
| Paper type | Theoretical |
|---|---|
| Title | The Learning Kit Project: Lessons Learned and Implications for Future Uses of Technology in Researching Self-Regulated Learning |
| Abstract | In many studies of self-regulated learning (SRL), learners are asked to describe the extent to which they use one or another cognitive process – e.g., memorizing concepts in an ordered list using a serial mnemonic – when they are presented with a particular kind of learning task. Then, learners experience instruction where the various cognitive processes they described have some chance to be used as they address various tasks embedded in instruction. Their achievement is correlated with their self-reports. Larger correlations are interpreted to indicate that learners practiced SRL. This approach to research omits critical information, namely: (a) Did learners actually engage in the cognitive processes they reported? (b) Did learners modulate cognitive processes in relation to different tasks and varying conditions of instruction? Without these data, there is no direct evidence that learners engaged in SRL. Moreover, given some empirical findings that learners are inaccurate reporters about their cognition, inferring how SRL unfolded during instruction is clearly hazardous. The Learning Kit Project sought to develop instruments and methods for analyzing data that involved direct evidence of SRL. Our instrument was software, called gStudy, that collects highly detailed, time-stamped traces of cognitive processes. Our tool for analyzing trace data, called LogAnalyzer, provides methods for examining trace data in ways that identify not only whether particular cognitive processes were used but also patterns that characterize modulation of cognition as a function of prior conditions. This presentation gives a more thorough account of the logic that undergirded the Learning Kit Project. It also demonstrates cognitive tools supplied by gStudy that learners could use to study, and explains how trace data of cognitive processes provide a more sturdy foundation for building and testing hypotheses about SRL. Projections are cast for the future of using technologies in researching SRL. |
| Summary | The Learning Kit Project grew from a premise that learning from instruction entails an interaction between two intelligent agents – instructors and learners (Winne, 1989). Instructors build instructional designs that blend their experience with, we hope, findings from empirical research about features of instructional designs that promote learning. Learners cognitively mediate features of those instructional designs (Winne, 2001). They selectively apply their knowledge and skills to choose whether to use features of instructional design, how to use them, and how to modulate what they do to optimize their goals (Winne, 1995). That is, learners engage in self-regulated learning (SRL). The Learning Kit Project, a collaboration among researchers in Canada, Finland, New Zealand, and the U.S., sought to build instruments for researching SRL and planned research that could contribute to advancing understandings about what SRL is and how it manifests in learning. A major product of the Learning Kit Project was a software application called gStudy (Winne, Hadwin, Nesbit, Kumar, & Beaudoin, 2005) and several supporting systems for organizing data collection and analyzing data. The suite of gStudy tools realized suggestions Winne (1992) made about software technologies that help pull up research on learning by its bootstraps. gStudy is a shell in which instructors and researchers install information in structured forms called learning kits. A kit can be about any topic. Information can be presented using text, diagrams, photos, charts, tables, audio and video clips—any information format found in libraries and on the Internet. gStudy provides a variety of cognitive tools learners can use to engage with and learn information by indexing, annotating, analyzing, classifying, organizing, evaluating, cross-referencing and searching it. Tools for studying information were designed, as much as possible, to satisfy two criteria. First, they afford opportunities to apply findings that research demonstrates can positively influence solo and collaborative learning and problem solving. Second, as learners use tools, they generate data that traces features of their cognitive and motivational engagements that represent processes of learning. For example, one of gStudy’s tools was note tool. To make a note that annotates information in a learning kit, a learner selects information (a string of text, a region in a diagram, a frame in a video clip or any of the whole “information objects” that can be created in gStudy) and pops up a contextual menu by using a keyboard combination. gStudy notes are structured packages of information. An instructional designer can create a template (schema) that learners fill in to structure their annotations about the selected information. For example, debate note template might present text fields that instantiate a 6-slot schema: issue, position A, evidence for A, position B, evidence for B, my position. Such schemas provide learners with standards for metacognitively monitoring comprehension and for elaborating information in ways that may enhance its retrievability (see Bruning, Schraw, Norby, & Ronning, 2004). When a note is created, the information the learner selected to annotate and the information the learner enters into the note’s slots are automatically linked. Links allow learners to navigate from a selection of information to a note, and vice versa. Learners (and instructional designers) can create new note templates at will by using a template editor. As learners create notes, gStudy logs a variety of data: What information (the selection) triggered a note-taking event? What template (schema) did the learner choose as the best organization of information about that selection? What information did the learner enter into which slots of the template? At what time points was the selection made, the template chosen, and the note completed? Data like these provide raw materials for tracing fine-grained processes that constitute learning events; by identifying patterns among these data and observing changes in those patterns, gStudy provides rich accounts of whether and how learners engage in SRL (see Winne, 2001). Contextualizing these kinds of data with other data gathered outside gStudy, such as indicators of motivation and indexes of achievement, it is possible to paint a much fuller portrait of what SRL is, how learners enact SRL, and what effects arise when learners self-regulate. gStudy provides a variety of tools beyond structure notes. These include tools to label (tag) information, index information by keywords (or other methods), and to link any of these “information objects” to one another (a la hyperlinks). As learners navigate across complexly linked architectures of information, gStudy records their paths. gStudy also provide tools for learners to build and manipulate concept maps, and to participate with partners in chats. As well, gStudy offers learners a sophisticated tool for searching information in their learning kit whenever desired. The spectrum of trace data that can be collected spans rather a large variety of ways that learners can approach and self-regulate learning. The Learning Kit Project led to many findings, a few of which are described in the papers comprising this symposium. In my presentation, two other kinds of findings will be examined: (a) issues in developing software technologies that provide critical instrumentation for researching complex activities such as SRL; and, (b) challenges that arise in characterizing learning processes when a researcher is able to gather extraordinarily detailed, sequential data that proximally reflect processes that are key mechanisms in theoretical accounts of learning and SRL. The presentation ends with “lessons learned” and an overview of a successor to the gStudy software that, by the time of the conference, will be completed. |
| Keywords | Computer supported Learning Environments Self regulation |
| Appendices | |
| Authors | ||||||
|---|---|---|---|---|---|---|
| Name | Surname | Institution | Country | EARLI Number | Presenting | |
| Philip | Winne | Simon Fraser University | Canada | winne@ubc.ca | * | |
