Athena Estimated Grades
Leaving Certificate Predictive Analytics
We have built a model to help estimate what a student would most likely have gotten in the Leaving Certificate in a normal school year, based on their previous results. We have built this model based on 6,000 previous Leaving Certificate students, and all of their previous results.
For these students, the model accurately predicts 71% of grades correctly and predicts 93% within one grade. These estimated results can be used as an unbiased baseline for schools, based on well researched and tested statistical methods. It will incorporate patterns specific to subjects and levels and will recognise outliers in students' previous results - ensuring no student is penalised for one bad result. We will also use the trend of Leaving Certificate results specific to your school, if this differs from the normal trend, where these results are provided.
What output will the school receive?
How do i sign up?
For the purpose of building our academic tracking software, the Athena Tracker, we have done extensive testing on different statistical models to see what best predicts the Junior and Leaving Certificate results. For Leaving Certificate, we have tested this model on over 6,000 students' results from last year.
How accurate are these predictions?
When tested on a dataset of over 6,000 of last year’s Leaving Certificate results, our statistical model accurately predicted 71% of grades, predicted 93% within one grade and predicted 100% within two grades of the actual grade.
What inputs will be used?
All available in-house exam results and mocks are used in predicting these results.
Summary Report - Spreadsheet showing predicted grades for each student