Presentation Pair: Learning Data Supports Student Success
Presentation Pair: Learning Data Supports Student Success
Monday, January 29 | 3:00PM–3:45PM CT | Galerie 3, Second Floor
Session Type:
Breakout Session
Delivery Format:
Presentation Pair
Riding the Data Wave: Using Predictive Analytics and LMS Data to Support Student Success In fall 2017, a few CSU campuses will launch pilots of Blackboard Predict (Bb Learn) and X-Ray Learning Analytics (Moodle) to forecast student success and struggles in designated high-failure-rate courses. The CSU System has embarked on a targeted graduation initiative and this cutting-edge way of leveraging LMS data with timely interventions offers a potentially powerful mechanism of students' identification at the point and time of failure, before it is too late, and offering students strategies for course correction.
Outcomes: Identify the conditions under which it is best to invest in predictive analytics at your institution * Define strategies for a successful implementation, including course and faculty selection considerations * Describe high-impact practices that are likely to translate predictive analytics into student success
From Data to Predictions: What Are We Missing in Learning Data? Predictive learning analytics, which can fuel adaptive learning technologies, has been identified as being of great importance for higher education in the past several years by the annual NMC Horizon Report. However, developing robust predictive models based on one or more digital learning environments (DLEs) has proven to be nontrivial, as DLEs do not capture contextual and learner information. In this session, we will report our findings from our e-text research to date and discuss the importance of including other DLE and contextual data in predictive models to advance our understanding of student engagement and achievement more broadly.
Outcomes: Learn how predictive models work in a higher ed learning context * Explore the research findings from IU e-text research project as they relate to predictive learning analytics * Discuss the types of additional data sources to include in predictive models
Presenters
Serdar Abaci
Educational Research and Evaluation Specialist, Indiana University
JP Bayard
Director Systemwide Learning Technologies and Program Services, California State University, Office of the Chancellor
Kathy Fernandes
Academic Technology Officer, California State University, Chico