Student Predictive Analytics Services docs

To Whom It May Concern:

Pursuant to the Georgia Open Records Act, I hereby request records related to student predictive analytics services used by the university.

For the purpose of this request, the term "student predictive analytics services" means software or computer code that employ machine learning models to generate predictions about individual students or groups of students for one or more of the following purposes:

a) To generate predictions about which prospective student names/contact information to buy from vendors such as the College Board, ACT, Cappex, etc. for advertising and recruiting.
b) To generate predictions about prospective students' likelihood to apply to, be admitted to, and/or enroll at the university.
c) To generate predictions about prospective students' financial aid needs and/or likely response to different financial aid packages.
d) To generate predictions about enrolled students' likelihood of passing a class, re-enrolling for another term (retention), and/or graduation (these tools are often called early warning systems).
e) To generate predictions about students' behavior and/or safety (e.g. whether a student is at risk of harming themselves or others).
f) To generate predictions about academic integrity (e.g. test proctoring software and generative AI detection).
g) To generate predictions about which majors or careers students are best suited to.

My request applies to student predictive analytics services used by the university during the 2022-23 academic year, including those developed in-house by the university or purchased by the university from a vendor.

For each student predictive analytics service the university used during the 2022-23 academic year, I request:

1) All active contracts, statements of work, and data sharing agreements between the university and the vendor regarding the student predictive analytics service.

2) Records defining the input variables and their weights/relative power for each student predictive analytics service model the university used during the 2022-23 academic year (In other words, records defining the data points fed into each predictive model and their respective impacts on the model's output).

3) Records documenting the accuracy (AUC, true/false positive, specificity, lift, etc.) of each student predictive analytics service the university used during the 2022-23 academic year.

4) Any validation studies, disparate impact studies, or other records documenting how the student predictive analytics services the university used during the 2022-23 academic year performed across different student groups (e.g. Black students vs. Hispanic students, male students vs. female students).

I ask that all fees be waived as I am a journalist and intend to use the requested records to publish articles in the public interest about the operations of a government agency. If you choose to impose fees, I ask that you provide an explanation of the fees, including the hourly wage of the lowest-paid employee capable of fulfilling the request.

If you choose to reject this request or redact portions of responsive documents, I ask that you cite the statutory exemptions and associated case law underlying your decision to withhold each portion from public review.

The requested documents will be made available to the general public, and this request is not being made for commercial purposes.

In the event that there are fees, I would be grateful if you would inform me of the total charges in advance of fulfilling my request. I would prefer the request filled electronically, by e-mail attachment if available or CD-ROM if not.

Thank you in advance for your anticipated cooperation in this matter. I look forward to receiving your response to this request within 3 business days, as the statute requires.

Sincerely,

Todd Feathers