evaluating adaptive learning: implementation & impact
TRANSCRIPT
Cristi Ford, Ph.D.
Katrice A. Hawthorne, Ph.D.
WICHE Cooperative for Educational Technologies
November 11 - 13, 2015
Evaluating Adaptive Learning:
Implementation and Impact
◼Primarily online, but serving 50% military including on bases throughout world
◼87,000 Undergrad & Grad Enrollment▪47% minority students ▪20% Pell eligible▪ 74% work full-time▪ 54% working parents▪ 31 - Median age
University of Maryland University College
◼2010, Carnegie grant with CMU for Open Learning Initiative (cognitive tutors)
◼2014, launched three vendor tools for piloting (Cogbooks, Realizeit, and EdReady)
◼2014, Next Gen Courseware subgrantee to Stanford OLI and Lumen Learning
History with Adaptive Learning
4
• Every innovation project should start with research and evaluation!
• Determine the types of questions you want to answer about your innovation project▪ Why will the new innovation make a
difference on our campus?▪ What questions are we trying to answer as a
university?
Create a Research Focus
Evaluation Logic Model
6
UMUC Innovation Cycle for Adaptive
7
8
What’s Next?
Pilot: Courses Deployed
Results - Course Completion
Adaptive N = 138
Traditional N = 147
Results - Course Completion
Adaptive N = 430
Traditional N = 425
Key Findings
Study Time and Student Achievement
◼FINC 330: Significant correlations between ▪ # of activities completed & midterm exam
scores▪ Study time & final grade▪ # of activities completed & final grade
◼CMIS 141: Significant correlations between▪ Study time & score gain▪ # of activities completed & score gain
◼MATH 009: Significant correlations between▪ Study time & score gain▪ Study time & post-assessment scores▪ Study time & midterm exam scores, final exam
scores, and final grades
Score Gain
N = 156
Student Perceptions of Adaptive
Summer 2015
N = 39
Response rate = 28%
Student Comments
“I know so much more than I did before and this is an accomplishment for me in math. I have always stayed away from math until this class. This class helped me to move through math in a way I did not see me doing just weeks ago.”
“I really appreciated the A.I. It assesses my knowledge level as I complete questions and adjusts accordingly allowing me to progress at my own pace.”
“It presented the lecture in bite size, easy to digest chunks of information and provided quizzes to assess comprehension with immediate feedback on answers.”
What We Have Learned
◼Learning is not linear
◼Combining learning science and data science are essential
◼Effective, flexible faculty and rich feedback are integral
◼Not enough content in some areas
◼Mapping and course organization are important
Questions & Contact
Cristi Ford, Assistant Vice Provost for Learning Innovation InitiativesCenter for Innovation in Learning and Student SuccessUniversity of Maryland University [email protected]
Katrice A. Hawthorne, Associate Director, Evaluation and ResearchCenter for Innovation in Learning and Student SuccessUniversity of Maryland University [email protected]