BREAKOUT SESSION #3: ADAPTIVE LEARNING: EARLY IMPLEMENTATION MODELS AND SUCCESSES
Rebecca Orr, Ph.D.
Professor of Biology, Collin College
Background Information• Professor of Biology at Collin College in Plano, Texas
• Two year public college, three campuses serving approximately 27,000 credit students
• Total enrollment in Biology 1406 is ≈2000/year
• Most of my students’ goals:• Nursing/Dental Hygiene/Surgery Technology/etc.
• Transfer to a university
No two students are identical — they learn and forget at different rates, come from different educational
backgrounds, and have different intellectual capabilities, attention spans, and modes of learning. As a result, designing a real-time recommendation engine that is sensitive to the characteristics of each student is an
immensely complex task.
http://www.knewton.com/assets-v2/downloads/knewton-adaptive-learning-intro.pdf
My Implementation of Adaptive Follow-Up: Summer 2013
• No Adaptive Follow-Up was offered during first two exam periods.• Exam 1 & 2 scores were averaged and utilized to compare
student ability of Spring and Summer student cohorts.
• Adaptive Follow-Up was offered for second two exam periods and was made “optional.”
• Two sets of items were selected for Follow-Up.• Follow-Up was due two days after parent assignment.• Test-out option was allowed and set at 95%.• Value was set as “extra credit” and added to homework
component of the lecture grade.
Adaptive Follow-up in Mastering• “Parent” Assignment created
• Item selection, difficulty, and estimated time for completion is instructor directed
• Adaptive Follow-Up is enabled for selected “parent” assignments• Time spent in follow-up and number of problem sets issued is
instructor directed
• Knewton engine creates Adaptive Follow-Up sets for the student using IRT based engine
• Items selected are individual to the student, based on:• Demonstrated proficiencies and lack of proficiencies in current and
previous assignments• Content prerequisites necessary for future, scheduled assignments
Student Participation in Adaptive Follow-Up: Summer 2013
• Almost 92% of students either tested out of or actively worked on the first Adaptive Follow-Up assignment.
• Average participation rates observed over the course of Adaptive Follow-Up offerings were:• 16.7% tested out of Adaptive Follow-Up by scoring 95% or
better on the parent assignment. • 58.1% chose to actively work on the Adaptive Follow-Up sets
after completing the parent assignment.• 25.2% did not participate in Adaptive Follow-up.
Addition of Adaptive Follow-up Increased Exam Averages: Summer 2013 (error bars- standard error)
Exam 1&2 Exam 3 Exam 405
10152025303540455055606570758085
Spring 2013: No Adaptive Follow-up for Exam 3 & 4, n=121
Summer 2013: With Adaptive Follow-up for Exam 3 & 4, n=37
Student Feedback on Adaptive Follow-Up
• “I really like how it [Adaptive Follow-Up] takes me back to the basics so I know where I need to study to build my strengths.”
• “I originally thought that the Adaptive Follow-Up assignments were going to be a waste of time, but it actually is more of a benefit.”
• “It was helpful, plus I felt confident when taking the test.”
Interesting Effect of Adaptive Follow-Up
• Students reported putting more effort into the Mastering parent homework:
“…Adaptive Follow-Up questions served as motivation to learn the material better…I really think it's just the idea of "testing out" of something that makes me feel smarter and encourages me to get a better grade on the [Mastering parent] homework.
Conclusions
• Adaptive Follow-Up provides individualized recommendations to increase student proficiency in content.
• Adaptive Follow-Up is based on demonstrated understanding of topics as well as the content graphed as prerequisite for success of future assignments.
• Adaptive Follow-Up may increase student success.
• Effect of Adaptive Follow-up on student success is being examined this semester implementing required completion of Adaptive Follow-ups and using a larger sample size.