|
|
2012 » Papers » Volume 2 » Assessing Student Retention and Progression: A Multi-Modal Approach 1. ASSESSING STUDENT RETENTION AND PROGRESSION: A MULTI-MODAL APPROACH Authors: Ice Phil Volume 2 | DOI: 10.12753/2066-026X-12-119 | Pages: 170-176 | Download PDF | Abstract
Student retention and progression are important measures of success for postsecondary education. They are key factors by which online programs, in particular, are currently scrutinized. This presentation reviews a successfully implemented, multi-modal approach that leverages data mining and quantitative analysis, supported by text analytics.
American Public University System, began assessing student retention with an exploratory model utilizing regression analysis with 32 variables from the student information system. Quickly, the initiative expanded include the use of data mining across all campus systems touched by students and the integration of criterion nodes into neural network models. Despite an 87% degree of accuracy in predicting retention within a 125 hour window across 187 variables, the issue of causality remained opaque.
Incorporation of text analytics to student input provided a means of ontological ordering of qualitative data that could then be converged back onto relevant data points across high probability nodes of disenrollment. The merging of these techniques has provided APUS with both a means of creating actionable business intelligence to assist in retaining students, as well as a causal understanding of systemic issues.
As previously noted, the merging of data mining, neural network analysis, conventional regression analysis and text analytics has provided a robust framework for intervention at both the short term and long term horizons. Through actionable intelligence, provided by the explanatory data derived from text analytics and semantic analysis, the APUS data team has been able to provide insight to the instructional design team, faculty members and administrative stakeholders. This has translated into a richer basis for continuous quality improvement of course materials, pedagogical strategies and student services. The impact on retention and student satisfaction has been considerable with 31% and 19% increases respectively since implementation.
Participants will be introduced to the data collection, federation and modeling techniques utilized at APUS. This will include exploration of methodology and required technical infrastructure. The presentation will be in case study format with numerous examples and resource links. Participants will be encourage to raise questions at any point and to consider how similar techniques might be used at their institution. Significant coverage will be given to exploration of perceived problems associated with both technical infrastructure and stakeholder buy-in. | Keywords
Online Learning, Text Analytics, Retention, Community of Inquiry |
|
|
|