An Emergent Digital Teaching and Learning Model
Concurrent Session 2
Presenters discuss results from our 4-year, collaborative, adaptive learning research including student outcomes and models that proved most effective for our online, adaptive, college algebra courses. We present an emergent teaching and learning model that frames digital and adaptive learning with predictive analytics.
An Emergent Learning Model
This session will address three innovations in higher education: digital learning, adaptive learning, and predictive analytics and describe how they intersect to form an emergent teaching and learning model. The concept of emergence is key to complexity theory where final results are more than the sum of their individual elements. In the model, the two-way intersection of digital and adaptive learning formulate a supply chain theory of knowledge acquisition where the adaptive process corresponds almost exactly to the elements of supply chain management. Learning analytics and digital learning coalesce into a predictive analytics approach where students’ learning behavior is indexed in the online, adaptive learning environment. Adaptive learning and predictive analytics intersect to produce real-time learning information. The comprehensive intersection of all elements converge into an environment where students experience the advantages and challenges of online learning supported by the adaptive and predictive processes. This is an emergent instructional environment that exceeds the sum of its individual elements.
A Emergent Collaborative Research Model
The learning model grew out of a research partnership, between a large public university, a for-profit, largely online institution, and their common adaptive learning platform provider. While the two universities vary, they have found collaborative research to provide valuable outcomes and actionable results to help inform instruction with the goal of improving student success.
The public university has over 69,000 students, with many direct transfers from partner state colleges. The average student age is 24 with 22% over the age of 25. The university began investigating predictive analytics in online and adaptive learning to help improve student success while allowing faculty control and flexibility over course content. A team of Personalized Adaptive Learning instructional designers (ID) provides support and technical and pedagogical guidance to faculty as they implement the course design process. Faculty who wish to use adaptive learning participate in a required faculty development program and are then assigned an ID to help with the adaptive platform.
The for-profit university provides industry-relevant programs to a diverse student population of approximately 25,000 students. The university began offering online courses in 2000 and now offers over 50 online or blended programs. The older student population has an average age of 36 and is 60% female. The university’s open enrollment results in students who enter with varying levels of expertise; therefore, the university began investigating adaptive learning in 2012. This approach provides students with individualized learning paths that adjust to their varied knowledge and preferences to improve their online instructional experience while providing a range of analytic data. They are introduced to adaptive learning during orientation and are provided help guides and additional training in using the technology. Also, faculty must successfully complete a separate asynchronous training prior to teaching a course with adaptive technology.
The platform provider creates a student customized learning and predictive environment, providing their clients with a comprehensive array of supportive data generating real-time prediction models. Institutions can import existing courses into the platform, creating customized student learning paths, or they can build adaptive courses from scratch. The platform is adaptable in that it does not impose a pedagogical approach on the course, so that instruction can be customized to suit the needs of each pedagogical approach, course or institution. The platform supports approaches ranging from competency-based learning to self-directed models, as well as various models of learning in corporate settings.
The principle underlying all these approaches is the separation of curriculum from content. Traditionally, learning is content-driven, with structure typically dictating the same linear pathway through the material for all students. The curriculum drives the direction of learning and uses content to help students acquire knowledge. The platform defines the curriculum using a hierarchical model and a structure known as the Curriculum Prerequisite Network--a directed acyclic graph where the nodes represent the concepts to be learned, and the boundaries represent the prerequisite relationships that exist between them.
The two universities and the platform provider complement each other’s approaches to online and adaptive learning for a cooperative research and implementation perspective. The public university is research intensive, subscribing to a data, information, insight, action paradigm. The private university has refined the online and adaptive learning scaling process. The platform provider develops comprehensive predictive data bases for both universities and is expert at the most current graphical and animated data analysis techniques. The cooperative partnership results in a research process of which the individual institution would not be capable.
The Emergent Study
This presentation will present data describing student outcomes in college algebra at the two universities, emphasizing online and adaptive learning using the context of predictive analytics with data provided by the platform provider. The researchers will share some of the outcomes from the 4-year collaborative relationship, including student knowledge acquisition and behaviors in each university’s online, adaptive college algebra course. In addition, the presenters will describe the predictive learning analytics models (using adaptive learning indices) that were found most effective at both universities. One of the models was post hoc and identified actionable predictive variables that respond well to instruction, allowing for early intervention. This was particularly encouraging because these actionable variables could raise the chances of students who might normall have a 3:1 odds of failure to better then even--making them comparable to the class as a whole's success odds. The second real-time predictive model tidentified the “point of no return” where the opportunity costs of remediation or tutoring for students at risk would no longer be effective. The real-time model demonstrated that the predictive power of the adaptive learning indices changed throughout the course. To our knowledge, this is the first time this phenomenon has been identified. In addition, we discuss the impact that course design has on these findings and the utility of this approach for the future.
Online college algebra that is adaptive and capitalizes on predictive learning analytics creates a fluid and emergent educational environment that responds to the needs of many student cohorts. Features like initial student knowledge baselines, continuous assessment and feedback, iteratively designed learning paths, mastery certification and instructional format preference make a more flexible and responsive educational landscape. Adam (2006), in her work with temporal culture provides insights into what can happen with consistent outcomes and variable learning time.
The emergent learning environment transforms a student’s:
- Time frame: The time boundaries of a course or program of study
- Timing: When learning will take place
- Tempo: The pace of learning
- Duration: How long learning will take place
- Sequence: In what order learning will take pace
The faculty member's role in this environment allows them to better facilitate learning as they can specify learning objectives for students through course design, and analytics data provided by the adaptive system can help them iteratively improve these objectives and address students’ weaknesses at key points in the learning timeline. Therefore, instructors can more accurately suggest effective interaction and intervention with students in areas that require support or additional instruction. Faculty members have a real-time view of student progress that is not readily available in other methods of teaching. For instance, adaptive systems can reliably identify skills or concepts with which the class on average is excelling or having difficulty allowing faculty to quickly address any deficiencies. In addition, instructors can track individual student progression through course content. This provides faculty the opportunity to adapt their instructional procedures, activities, or homework assignments to personalize instruction in the online college algebra environment by capitalizing on what predicts and responds to instruction.
The virtual session chatroom will be monitored actively by co-presenters. Also, we hope to solicit active conversation around adaptive learning research via an active Twitter backchannel through establishment of a session hashtag (in addition to the conference hashtag) which will allow researchers to connect. The presenters will solicit input and feedback from the participants throughout the session (and afterward) via online polling, question-and-answer, and through the backchannel conversations. Session materials and evaluation resources will be made available to participants.