The Outcomes Imperative for Adaptive Learning

Wally Boston university digitalWhen I first joined APUS, conferences were an easy way to get up to speed on many issues. Similar to other industries, there are many different higher ed conferences whose agendas reflect member needs. There are events for college presidents, financial officers, enrollment management and student services staff, academic advisors, accreditation leaders, chief academic officers, faculty, etc. Over time, I’ve reduced my conference schedule as I feel more comfortable with the relevant issues in a given area.

One conference I consistently attend is the Higher Education Leadership Conference at the University of Pennsylvania Graduate School of Education (Penn GSE). The event is unique in that only graduates of Penn’s Executive Doctorate in Higher Education program are invited. Alumni organize the agenda to address some of the current issues in higher education. Because graduates are administrators at colleges and universities (including more than 50 college and university presidents), the dialogue between speakers, panelists, and participants is as stimulating as the presentations.

Last week’s conference, Re-Imagining the Future of Higher Education, maintained the standard of excellence set in previous years. One topic related to the use of technology in improving learning, particularly involving data collection from large datasets, data analysis, and improved learning resulting from sophisticated software providing detailed analysis and guidance to individual students beyond the capacity of their instructors.

One featured speaker on the topic was Candace Thille, Director of Learning Technology and Engineering at, LLC., and a Penn GSE graduate, who has spent her career working with technology, faculty, and students at Carnegie Mellon, Stanford, and now at Amazon. For many years, she has argued that institutions need to be familiar with the algorithms underlying adaptive learning technologies and if vendors will not disclose them, institutions should either develop their own or find  another provider. I fully agree with Candace’s rationale. Clearly, Stanford, Carnegie Mellon, and Amazon have the staff and financial resources to develop learning algorithms internally. However, smaller colleges and universities are going to be dependent on outside providers, and negotiating an open review of how an algorithm works may not be possible for all.

Over time, it’s likely that only a few adaptive learning software packages will prevail. Hopefully, software vendors not controlled by very large universities or companies will choose to share how their algorithms work. We’ve learned enough about how people learn to know that not everyone learns the same way. Beyond the seven learning styles (visual, aural, verbal, physical, logical, social, and solitary) with which many educators are familiar, modern technologies are enabling researchers to determine there may be more. In fact, one recent book by David Schwartz, Jessica Tsang, and Kristen Blair, The ABCs of How We Learn, identifies 26 unique learning styles. As datasets of learners’ activities increase and algorithms improve their abilities to discern different styles, this higher number will likely increase.

Sophisticated software increases the potential to tease out the most effective way to help each person learn. The weakness of today’s educational system is that we often teach to the average, excluding learners on the upper and lower edges. A learner who conforms survives, while non-conformers don’t. As colleges, universities, and corporations develop and refine stronger adaptive learning algorithms, I hope they avoid the bias toward conformity.

Many years ago, I was a board member at a private K-12 school. Annually, we would talk about the tests used to measure the success at various grades of our students. Kindergarten was the only grade where standardized tests were not used to evaluate applicants. Instead, there would be a scheduled session during a tour of the school where the prospective applicants were evaluated for their ability to play with other children and their ability to follow instructions. During one part of the evaluation, a teacher gave each child a box of blocks and asked them to arrange them in a fashion so that one of the blocks was as tall as them (the child). By watching or cooperating, the children usually figured out how to stack the blocks high enough to be deemed satisfactory. One year, during a kindergarten applicant visit, a young girl looked over the blocks, picked one up, threw it up with an underhanded throw and it hit the ceiling. The two teachers observing couldn’t agree as to whether or not the child would fit in based on her different interpretation of the instructions. If it were my choice, I’d admit the child demonstrating independent thinking at four years old over the child who conformed to the norm.

As we embrace adaptive learning software, we have to make sure that we choose learning algorithms that work to the learners’ strengths instead of forcing them to adapt to a norm. In the end, we lose if we are all coached to think alike.

Subjects of Interest


Higher Education

Independent Schools


Student Persistence