Toggle menu
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

EdSights Risk Rating levels indicate how at risk a student is of not persisting to the following term compared to other students at our institution. There are 3 levels of risk: High, Medium, and Low.

A high-risk student is less likely to persist compared to a low-risk student.

Risk Levels are calculated exclusively based on students' interactions with the AI. The only input that contributes to a student's risk level is how they respond to the bot's check-ins.

A Not Applicable value represents students with insufficient interactions to establish a risk rating.

Metadata
Data Custodian
Institutional Assessment and Research
Data Source
Source System:CMS - SA
Source Table Name:N/A
Source Field Name:N/A
Census Process:N/A
Logical Transformation / Calculation
N/A
Dataset Folder and Field Name
Dataset Folder:Edsights Risk
Field Name:EdSights Risk Rating


Business Rules

EdSights is an AI chatbot (Monte) that engages our undergraduate students through supportive, personalized text message conversations with campaigns every ~2 weeks.

Risk Levels come only from students' responses to our proactive framework check-ins[1], and only students with at least one response are rated. Risk points also expire over time, with the timing depending on the question and category, since many answers only reflect a student's situation at that moment. So a continuing student who hasn't answered a check-in this term can show as Risk: Not Applicable even if they had a risk score last semester. Their risk level returns once they respond to a new check-in.

Custom campaigns can be used to engage students (either from EdSights custom campaign library[2] or may be designed custom by the campus) but these custom campaigns do not generate risk ratings.

Examples

  • High Risk
  • Medium Risk
  • Low Risk

See Also

Definition Source

Risk Levels indicate how at risk a student is of not persisting to the following term compared to

other students at your institution. There are 3 levels of risk: High, Medium, and Low. A high-risk

student is less likely to persist compared to a low-risk student.

Risk Levels are calculated exclusively based on students' interactions with the AI. The only input

that contributes to a student's risk level is how they respond to the bot's check-ins. The goal of

this approach is to capture the student voice, which is a critical, yet often missing data point in

most predictive analytics.[3][4][1]

References

Cookies help us deliver our services. By using our services, you agree to our use of cookies.