PRODUCT SUITE

Identifying potential dropouts at an early stage gives you an edge in retaining the student. Earlier the identification, greater the retention rate. Student attrition rates are not just statistics. They represent students with problems on various grounds, problems that if left unresolved over a period of time create disaffection. Therefore, the challenge is not only early identification of such students but also the determination of the accurate moment at which the system raises the alert to the appropriate stakeholders.

ProRetention™ adopts four simple steps to help student retention:

Identify

ProRetention™ helps you identify factors that could have adverse impact on the student. Even small symptoms of such factors could add up and become triggers for a student to eventually drop out. ProRetention™ helps you identify the probable risk attributes that could directly or indirectly be a factor predisposing a student to be at risk. The risk attributes identified are transformed into structured data, facilitating in defining at-risk rules that would consequently yield conclusions with a better probability of predicting dropouts.

The following are some sample factors that influence attrition:

• Trace on certain student (community) demographic attributes
  • Being single-parent children
  • Speaking English as a second language

• Trace on social attributes
  • Living in high-growth states
  • Living in unstable school districts

• Trace on student academics
  • Low or drop in GPA
  • Low attendance

Define

ProRetention™ helps define runtime triggers in identifying students at risk. On occurrence of the event the system automatically raises an alert to the identified key stakeholders for immediate intervention. The identification of risks is not a onetime activity - the system provides an ongoing facility to your school to collect data and evolve as you change and grow.

ProRetention™ application offers the following capabilities:

• Configurable attributes to participate in the making of "rules"
  • GPA
  • Attendance
  • Demographic information
  • Assignment details

• Configurable rules to raise alerts as it happens"
  • GPA
  • Low or drop in GPA
  • Missing (consecutive) classes
  • Non-completion of assignments

• Configurable alert mechanisms (SMS, email, or web)
• Configurable escalations on the alerts if no intervention is made
• Configurable actions in the event of a rule (alerts, surveys, appointment, or ticket)

Manage

ProRetention™ has a well-defined process that allows key stakeholders to manage alerts that are raised for students at risk. These alerts appear in the inbox of the user. The system also helps the user by suggesting appropriate actions that need to be carried out depending on the type of the alert.

Following are some examples:

• Initiate voice call and/or email communication
• Raise an issue ticket
• Initiate survey

ProRetention™ has an inbuilt mechanism to send reminders to users when they fail to act upon alerts on time. It also has the capability to send escalations to department heads when an alert remains unattended for more than an acceptable timeframe. Speed is of essence while acting on an at-risk student. ProRetention™ will guarantee that no prospective intervention will fall through the cracks.

Refine

ProRetention™ builds intelligence of factors that determine at-risk students. The self-improvement engine uses historical information of student dropouts in determining possible exception rules to be added to the predefined set. This engine constantly learns from patterns that evolve over a period of time. Since there is no standard or global set of reasons for dropping out, this engine becomes extremely important in a dynamic, ever-changing school environment.

What does our engine offer?

• Corrective: Identifying dropouts and evaluating emerging patterns (which are subsequently created as rules for at-risk triggers)
• Preventive: Proactively finding students at risk and raising alerts for appropriate actions

What do you get?

• Early identification of elements influencing dropouts
• Dynamic creation of checkpoints in early identification of dropouts
• Early alerts on occurrences of checkpoint violations
• Identify exceptions and refine rules set to increase probability of predicting dropouts