Intellischool uses anonymised data to provide insights and comparisons to users.
In order to provide comparitive insights, as well as train predictive models and use machine learning, a massive amount of data is required. No single school can provide enough data to create statistically sound predictive models, or a suitable environment to train an artificial intelligence on recognising patterns in data.
To circumvent this issue, Intellischool uses anonymised data from many schools to facilitate these practices.
The de-identifcation process
When creating anonymised datasets, Intellischool goes through a process that:
- Removes any identifiable information from the top-level school record;
- Strips personally identifiable information from the student/staff/parent records associated with each school, including name, date of birth, e-mail, phone numbers, addresses, and any unique identifiers (such as student IDs - including CASES21 identifiers);
- Searches for and redacts any identifiable information from free text fields, such as comments and task names, including names of persons, school names, postal codes, addresses, phone numbers and e-mail addresses; and
- Deletes any temporary data created while anonymising the data set, removing any remaining paths to identifying data.
Types of de-identified data that Intellischool uses
Student and school data that are used in anonymised analysis include:
- Academic results
- Aboriginal / Torres Strait Islander status
- Geographic location (to a Local Government Area / region within a state)
- Language background at home
- Socioeconomic status (SES)
- Standardised test results
- Wellbeing indicators
- Year levels
Privacy and data security are critical at Intellischool. For more information on how we anonymise datasets, or a technical description of the process, please reach out to your Intellischool representative.