Analytics to Improve Persistence, Motivation, and Learning

• In Event: 77.028 - Can Learning Analytics Improve Learning?

Abstract

Why do some people give up in the face of adversity while others keep working toward their goals? This presentation addresses the question of how analytics can be used to provide answers to this question and to support the improvement of academic persistence, defined as the focused pursuit of both short and long-term academic goals while avoiding seductive distractions.
A main goal of the presentation is to sort through the welter of ideas about persistence and related constructs and give a coherent organization to that knowledge so that it can be used as a foundation for building analytics that collect and generate useful, usable and valid information. The paper highlights contemporary perspectives on academic persistence and critically analyzes the pragmatic value of these different perspectives for driving analytic approaches that can lead to insights and improvements in persistence. Ultimately the aim is to provide a preliminary answer to the question of what sort of analytic approaches are best suited to ameliorating the persistence problem.
To improve motivation to persist, measures, analytics, and interventions can be organized to focus on the three factors that operationalize motivation at the task level: starting, persisting, and investing adequate mental effort (Pintrich and Schunk, 2002). Using this three-part framework helps bring some coherence to the daunting array of both formal research and informal constructs surrounding persistence and motivation, and also provides a target for analytic possibilities. We present examples of validated measures and research-tested interventions for persistence motivation for students who need them most. Online instructional technologies have opened many new opportunities to capture information about students’ persistence motivation and learning as they work, and to provide personalized support to individual students. This can happen in both online and traditional classroom contexts and we will describe examples from both.
As an example of how persistence-related data may be collected and used to guide online instruction, researchers in an online program designed a system to monitor beliefs, emotions, and behaviors associated with motivation, based on the work of Bandura (1977), Clark & Saxberg (2012), Dweck (2006), Eccles & Wigfield (2002) and others. In this system, motivation data are combined with performance data to generate personalized guidance for students and teachers. The data are also used by instructors to separate low performing students into those with value, confidence or mood issues. Different feedback and guidance is given to students with different motivational problems or profiles.
Technology provides many affordances for persistence analytics and data mining, but the new data and analytic approaches will need to be validated to show that they are actually providing accurate and reliable info that can be used to improve academic persistence. In addition, learning analytics may discover new insights about student differences and environmental factors that impact persistence and add to the interventions currently available. The growth of online instruction across the world also makes it possible to conduct rapid experimentation on new analytic strategies and the measures and interventions linked to them.

Authors

• David Michael Niemi, Kaplan

• Richard E. Clark, University of Southern California