This solution uses an automated statistical model to identify students at risk of attrition. This will allow students or administrators to take necessary interventions and steps to ensure student retention and success. Student attrition is a major factor affecting university income, representing a loss of nearly [10% of potential revenues.] (http://seatssoftware.com/student-retention.html) Previous work has shown that early identification of at-risk students followed by student-initiated intervention can [greatly increase retention.] (http://www.itap.purdue.edu/learning/tools/signals/) But identifying at-risk students is a challenging endeavor, particularly at large institutions and in Massive Online Open Courses (MOOCs), where it is difficult to use social interaction as a cue for student risk. Thus, the use of automated statistical models offers a distinct advantage in some educational contexts.
A Statistical Model for Success
The use of this technology is automated and does not require specialized training in statistics, making it very easy to implement. The tool requires the input of student-specific data from past years (e.g. demographics, credit hours taken, prior education, grades in a given course so far, etc.) and a measure of success of each of these students that is of interest to the university (e.g. retention, GPA.). Based on this data, it will predict which students in a new set of data (e.g. first year students) are likely to require intervention.
- Identify at-risk students early in their curriculum to intervene and improve student retention
- Help e-Learning students and MOOC students who do not receive the more personal advising available at a traditional campus
- Becomes more accurate the longer it is used
- Fully automated; does not require specialized training in statistics
- Adapts to the specific qualities of your institution and your students