Leveraging Data to Create Personalized Learning Environments
David B. Knight, Engineering Education, Higher Education, ICAS, HCD, HCI, Virginia Tech
Troy Abel, Design, ICAT, HCD, HCI, Virginia Tech
Cory Brozina, Engineering Education, Virginia Tech
Chris Frisina, Computer Science, HCD, Virginia Tech
Eric M. Stauffer, Instructional Design and Technology, HCD, Virginia Tech
Abstract: A convergence of pressures has led researchers to seek innovative ways to measure and
track student learning outcomes and empirically identify the conditions that lead to their
development. Learning analytics is an emerging field of inquiry that uses existing student traces
to aggregate and illuminate student data through visualizations and dashboards in an attempt to
improve learning outcomes. While there are currently efforts both in vendor and academic arenas
to try and understand the long-term learning and decision-making effects of such dashboards, there
appears to be a missed opportunity in the development of these dashboards in vivo using humancentered usability practices to develop these new tools for learning. Practices that select relevant
data traces and develop dashboards with learners instead of for learners may lead to stronger
student self-efficacy, build on existing social learning theory, and benefit from perspectives found
within human-centered design practices. Our interdisciplinary team of faculty and graduate
students from engineering education, computer science, human computer interaction, human
centered design, the learning sciences, and visual communications are following a mixed-methods,
human-centered approach to developing a dashboard breaks new ground in learning analytics by
involving the end users throughout the design and development process. Qualitative data will be
generated and analyzed through participatory design action research methodologies exploring user
feelings and attitudes towards success and how that information could inform the selection of data
sources. Quantitative research will identify patterns across currently disparate institutional
databases, including incoming student surveys, longitudinal surveys within the first year
engineering courses, and existing data sets found within the Scholar learning management system,
This study identifies what learners and instructors desire in learning dashboards; illuminating
patterns across currently disparate institutional databases helps learners (and instructors) regulate
learning processes and make empirically supported adjustments to current practices.
General Formatting Guidance
Proposal Formatting
The poster proposal should address the information indicated on the Poster Session page on the conference website:
http://www.cider.vt.edu/conference/cfp1. In general, the poster proposal should adhere to the following guidelines:





1 inch margin left, right, top and bottom
No headers or footers
Fonts should be 10-point and Times New Roman
Single spaced text should be used throughout
Maximum length of a proposal is 1 page
Proposal Contents
A poster proposal should include the author(s) name; department, institution; abstract (not to exceed 300 words);
literature review; description of pedagogical practice (if practice focused); methodology, data analysis and results (if
research focused); and, a discussion. In addition, poster proposals will be evaluated based on the following
questions: Does the proposal demonstrate an appropriate knowledge of the literature? Is the description of the
practice clear? (Practice) Are the research methods appropriate? (Research) Is the proposal well written?