The trouble with risk assessment is that it is difficult!  It is so often presented as a series of 'flags' that are waved, sometimes manically by institutional specialists who, borrowing from Stevie Smith, may just be drowning, not waving.

​

As we know, institutions are often drowning in data but struggle to know how to use those data to predict which students will accept their places and then enrol, which students will enrol then withdraw early in their study, and then which students are likely to fail units of study that will preclude them continuing towards their qualification and subsequent employment.

​

RADAA provides an solution to these difficult challenges.  It does so in a way that reflects a common-sense approach that allows people struggling with how to build and use risk models to develop simple models initially, to test and then trust, to review and then refine, and then finally to automate and integrate with machine learning.

​

RADAA reflects the (obvious) truth that the factors that are involved in assessing the risks at each stage of the student journey vary.  It also reflects the (obvious) truth that those factors vary depending on the course(s) being studied, how they are being studied, what level of qualification they are at and the background of each students.  Finally, RADAA reflects that risk is a journey and not a point in time model that demands instant reaction when an arbitrary threshold is crossed.

​

Using any data element that Student Pulse stores, including details of comments in phone conversations, responses to micro-surveys, biographic, learning management system usage, grades, marks, assessment attempts, attendance etc etc, RADAA allows you to reflect the relative significance of each 'risk attribute' for any combination of courses/units across any number of time slices.  Using attribute scores and weightings across attribute groups, it calculates risk scores every 24 hours and creates a risk trajectory for every student you wish it to.

​

These risk scores and trajectories can then be used to target and prioritise students to be included in Pulse engagements and automatically to be put highest on the lists of those to be contacted.

​

Academics and support staff can create and share confidential notes about students based on their risk profiles, thus creating a unified case management model that can operate until the risk has reduced and the focus can move to others in greater need.

​

​