The PAR Framework Proof of Concept: Initial Findings from a Multi-Institutional Analysis of Federated Postsecondary Data

Volume, Issue - Date: 
Volume 16, Issue 3 - June 2012
Author(s): 
Phil Ice
Author(s): 
Sebastián Díaz
Author(s): 
Karen Swan
Author(s): 
Melissa Burgess
Author(s): 
Mike Sharkey
Author(s): 
Jonathan Sherrill
Author(s): 
Dan Huston
Author(s): 
Hae Okimoto
Oganization: 
Phil Ice, American Public University System
Oganization: 
Sebastián Díaz, West Virginia University
Oganization: 
Karen Swan, University of Illinois Springfield
Oganization: 
Melissa Burgess, American Public University System
Oganization: 
Mike Sharkey, The Apollo Group (University of Phoenix)
Oganization: 
Jonathan Sherrill, Colorado Community College System
Oganization: 
Dan Huston, Rio Salado Community College
Oganization: 
Hae Okimoto, University of Hawaii System
Full article - Free: Click on the file to download.: 
Keywords: 
retention, progression, completion, online learning, postsecondary, predictive analytics, data repositories
Abstract: 

Despite high enrollment numbers, postsecondary completion rates have generally remained unchanged for the past 30 years and half of these students do not attain a degree within six years of initial enrollment. Although online learning has provided students with a convenient alternative to face-to-face instruction, there remain significant questions regarding online learning program quality, particularly when considering patterns of student retention and progression. By aggregating student and course data into one dataset, six postsecondary institutions worked together toward determining factors that contribute to retention, progression, and completion of online learners with specific purposes: (1) to reach consensus on a common set of variables among the six institutions that inform student retention, progression and completion; (2) to explore advantages and/or disadvantages of particular statistical and methodological approaches to assessing factors related to retention, progression and completion. In the relatively short timeframe of the study, 33 convenience variables informing retention, progression, and completion were identified and defined by the six participating institutions. This initiative, named the Predictive Analytics Reporting Framework (PAR) and the initial statistical analyses utilized are described in this paper.