Toward Practical Procedures for Predicting and Promoting Success in Virtual School Students

Abstract

Two lines of research have emerged to study the causes and prevention of student failure in virtual (K-12) schools: studies of learner characteristics and studies of learning environment characteristics. To develop a useful model for predicting failure and promoting success in virtual school environments, a study was designed to measure the relation between a combination of student and environmental factors derived from previous research and successful course completion during one semester at a large (N = 4,100) virtual school. Study findings yielded a model that can discriminate between successful and unsuccessful online school students and is especially effective at identifying those likely to succeed.Early in the development of distance learning programs, it became apparent that dropout and failure rates tended to be significantly higher for distance students than for those in traditional, face-to-face classrooms (Bernard and Amundsen 1989;Cyrs 1997;Dille and Mezack 1991). That trend has persisted throughout the development of newer distance delivery technologies ). Two very different lines of research have emerged to study the causes of the problem and how to address it: studies of Correspondence should be sent to M. D. Roblyer, University of Tennessee at Chattanooga, College of Health, Education, and Professional Studies, 615 McCallie Ave., Chattanooga, TN 37403. SUCCESS IN VIRTUAL SCHOOL STUDENTS 91learner characteristics and studies of the characteristics of learning environments. However, neither of these lines of research has yielded a consistently helpful model for addressing the problem.This study was undertaken to explore whether a combination of learner characteristics and learning environment variables derived from past research could predict success in one kind of distance learning population (virtual school students) and how organizations that offer distance courses might use findings from such a model to facilitate online learning success for future students. STUDYING THE ROLE OF LEARNER CHARACTERISTICSDuring the pre-Internet days, research on distance learning success developed around the influence of learner characteristics. The assumption was that, in light of the highly abstract quality of these learning environments, successful learners were those who were simply better equipped to deal with the complexity of learning in this new format.As distance learning expanded to include more diverse populations and younger learners at earlier levels of education, research on learner characteristics began to assume even greater significance. Hartley and Bendixen (2001) felt that if the primary reason for success or failure in distance learning lies in learner characteristics, this finding has the potential to widen an already troublesome Digital Divide. STUDYING THE ROLE OF NONSTUDENT VARIABLESNot everyone agrees, however, that learner characteristics are the overarching contributors to success. P. Smith and Dillon (1999) felt that the characteristics of distance learning delivery systems and course de...

Prediction of Student Dropout in E-Learning Program Through the Use of Machine Learning Method

“…Roblyer et al used a Likert scale that contained 60 measurable variables for conducting a survey of 2,162 E-learning students. Binary logistic regression analysis of the sample data revealed a relatively strong correlation between demographic attributions (such as age) and learning achievement (such as GPA) and student dropout [20]. Nichols et al made use of ordinal regression for analyzing the data obtained from 187questionnaires returned by distance education students; the results showed that among four sets of attributions (namely computer usage attitudes, learning motivation, perception of satisfaction and previous academic performance), only previous academic performance exhibited a strong correlation with dropout [21].…”

Measuring e-Learning Readiness Concept: Scale Development and Validation Using Structural Equation Modeling

“…Also, this study shows that the predictive validity of the ELR tool is clearly superior to those tools as [39] with 83%, [34] with 67% and [12] with 79.3 %.…”

“…The first group called "Success" includes students who have received a grade of D to A +, and the group "Failed" those withdrew from the class with a grade of "AR" or "AX" (AR: withdrawn without academic penalty, AX: withdrawn with academic penalty) or completed it with a grade of E (failure). The student ranking criterion in both groups was almost similar to that of [12]. The independent variables are the five dimensions of ELR, and we created a dichotomous variable (Group) which takes the value "0" if the student has failed and "1" in the case, this variable represents the dependent variable model.…”

“…Moreover, according to these authors, it is important to investigate the motivation of online learners, including its antecedents and outcomes. In addition, several studies have shown that cognitive factors such as motivation contribute to the online learner success [12].…”

Walk, Don't Run — to Online Learning

“…Forcing students to take courses online may set up less intrinsically motivated students for failure. Moreover, before students register for online courses, have them take a type of personality test that can predict their level of success based on their answers to questions probing their learning preferences and styles (Roblyer 1999;Roblyer, Davis, Mills, Marshall, & Pape, 2008). Then, students can be better counseled into online or face-to-face courses based on their academic strengths and weaknesses.…”