Adaptive Analytics: Improving the Odds
Concurrent Session 2
We present cooperative research about how adaptive analytics combined with two variable domains improves the odds of student success in college algebra. We will show that demographic predictors such as grade point average can be used as mediators for metrics that respond well to instruction.
Adaptive platforms facilitate the design of courses and programs that personalize the student learning experience by customizing content while continually assessing learning outcomes. Requisite learning time for students varies, impacted by variables such as students who face challenges with mathematics readiness or students expressing ambivalence for the subject matter. However, the structure of an adaptive course equips students with unique, real-time, adjusted learning paths underpinned by continuous assessment that can accelerate their learning or extend their learning space beyond traditional boundaries such as semesters, depending on their achievement levels.
This presentation will describe a longitudinal, cross-institutional, adaptive learning evaluation between the research unit of the platform provider (Realizeit), the Research Initiative for Teaching Effectiveness at the University of Central Florida, and Colorado Technical University. Our 5-year continuing collaboration has taken place with the understanding that the three organizations must stay the course to achieve valid and meaningful information.
After a contextualization of adaptive learning at our respective campuses and how we have accomplished collaborative research, we will discuss the methods and results from our ongoing analyses of the impact of educational analytics arguing that individual student prediction is not a tractable approach. We contend that integrating student information system data with real-time class data provided by Realizeit effectively improves the odds of success for like student cohorts. We will present data about:
1. Assessing the granularity of adaptive courses,
2. Prototype analysis of student learning path behaviors,
3. Student success based on the institutional contexts,
4. The possibility of real-time adaptive predictive analytics,
5. Simulating student behavior based on learning analytics, and
6. Implications of adaptive learning for helping underserved student populations.
Preliminary analyses indicates that adaptive learning has the potential to help alter students’ learning pathways to maximize their chances of success. Used effectively, this innovation can positively impact teaching and learning in higher education.
The work in this presentation relates the concept of intersectionality that emerged in the late 20th century addressing educational and financial inequity citing the impossibility of decoupling concepts such and poverty and racism and that their interaction is more impactful that either concept considered separately. This line of thinking is rooted in several other disciplines--for instance, in Douglas Engelbart’s theory of integrated domains, interaction effect in statistical models, the emergent property of complex systems, entanglement in quantum physics, and in learning the intersection of cognitive, behaviors and affective characteristics of students. Understanding these principles and the basis of adaptive learning analytics can help us understand that one-to-one student prediction of risk or success must be based on the intersection of many components and that clearly the most effective approach is one that levels the playing field by increasing the odds of success. Nothing in higher education appears to operate independently because higher education is a complex system. Consider Taleb’s perspective: “The main idea behind complex systems is that the ensemble behaves in ways not predicted by its components. The interactions matter more than the nature of the units. Studying individual ants will almost never give us a clear indication of how the ant colony operates. For that, one needs to understand an ant colony as an ant colony, no less, no more, not a collection of ants. This is called an “emergent” property of the whole, but the parts and whole are different because what matters are the interactions between such parts. “