nStudy Traces of Process and Content in Self-Regulated Learning

• In Event: Self-Regulated Learning Analytics: Aligning Data and Their Treatment to the Assumptions of Theory

Abstract

Objectives. I demonstrate nStudy, software that gathers comprehensive, fine-grained, time-stamped data about (a) operations learners apply to information when they study and (b) the information on which they operate. I describe learning analytics that can be generated from these data to support learners’ productive self-regulated learning (SRL).
Theoretical framework. Productive SRL needs process feedback about tactics and strategies learners actually use. Self-reports are too distal and unreliable for this purpose. Data gathered by learning/content management systems, such as resources downloaded or login times, are too coarse. Traces are comprehensive proximal data that (a) support strong inferences about cognitive and motivational features of learning and (b) record precisely which information learners use in learning. nStudy logs traces of practically every observable event describing what and how learning unfolds. Learning analytics and interventions generated using trace data can fuel theory and support SRL.
Methods. Features nStudy offers include: bookmarking, tagging, annotating, chatting with peers, composing essays, concept mapping, searching, organizing artifacts (e.g., content tagged, notes) and more. Most attributes of features are configurable. Learners can use nStudy in the lab or anywhere they connect to the internet. Content can be anything represented in html or pdf format. nStudy supports highly controlled experiments and studying au naturel.
Data. Suppose a learner selects text and tags it #review. Tagging more revealingly enacts highlighting, the most common study tactic. nStudy logs: the time, the text selected, the tag chosen (as provided by a researcher/instructor and/or generated over time by the learner). These data trace: motivation to metacognitively monitor content, the standard used to metacognitively monitor (this is worth reviewing), metacognitive control operationalized by choosing a tag, and planning to review tagged content. Tagging creates a “quote” of the text and stores it in a library. At any time, the learner can search quotes by tag (or by content) and “open” a quote to show it in “highlighted” format within its web page. Learning analytics and prompts to regulate study tactics can be generated using nStudy’s trace data. Examples: goal achievement, “You reviewed only 40% of content you tagged for review.”; opportunity to distribute practice, “It’s been 3 days since you reviewed content you tagged.” Such process feedback and prompts set stages for learners to track and engage in SRL, and for researchers to record all of that process in investigations about SRL.
Results. Studies using nStudy demonstrate traces about motivation differ from self reports, and traces better predict achievement than self reports. We illustrate other analyses of traces that illuminate SRL-as-process and features of content operated on in learning.
Significance. nStudy is free and open source. Distributing nStudy widely can generate truly big data about SRL as learners use it unobtrusively in everyday studying. Learning analytics and prompts to scaffold, delivered through nStudy’s chat feature, can introduce treatments tailored to a learner’s history of learning events. Learners using nStudy partner with researchers to advance learning science as they use learning analytics to improve study skills, SRL and achievement.

Authors

• Philip H. Winne, Simon Fraser University

• John Cale Nesbit, Simon Fraser University

• Zahia Marzouk, Simon Fraser University

• Mladen Rakovic, Simon Fraser University

• Ilana Ram, Simon Fraser University

• Donya Samadi, Simon Fraser University

• Jason Stewart, Simon Fraser University

• Kenny Teng, Simon Fraser University

• Derra Truscott, Simon Fraser University

• Joviat Vytasek, Simon Fraser University

• Sonya Woloshen, Simon Fraser University