This is an interesting data point. And like most things GPA related it can get messy really quickly. There's only one place in the current system where any type of major GPA is visible, and that's the degree audit. To more specifically the audit has an upper division major GPA which generally excludes all lower division classes. Except that the actual logic is based on classes used for upper division requirements, not the classes themselves, which means in rare cases when a lower division class is used for an upper division requirement, it does pull that in. So far as I know there's no other widely accepted calculation for major GPA, at least in the policy or in ISIS. As we dig in I think it's likely we'll find many subtlety different calculations related to major GPA. But it's absolutely worth having a conversation about.

I'm not aware of something called "time to candidacy", but the official record of advancement to candidacy is in ISIS. It's entered on AH STDNACEV. There is an academic events table in the data warehouse which pulls the information.

Blank major names are an error or an artifact of something else that's weird on the record. Except for pre-ISIS records, where blank majors are probably more common than completed rows. Modern records (Fall 1991 and later) generally should always have a primary major code, and so a Primary Major Name.

I'm also in favor of something more specific than ID. PID is just how it's labeled in ISIS.

If we are considering splitting StudentID from InstructorID, keep in mind those values will be the same for people who have both types of records as it's the exact same field in ISIS. I'm fine with labeling them differently for ease of use, presuming we add both as related terms to PID in the IGC.

Staff in my area currently use the Data Warehouse to produce a number of reports related to quality assurance and enforcement of policy which require that we have visibility of student name. The reports identify individual records which need to be reviewed or updated manually in ISIS. It's technically possible that we could complete our processing without knowing the name, but matching the name from the report to the record that you pull up in ISIS based on the PID is an important validation. If we make a mistake and pull the wrong record and process an incorrect update, there's very little in the way of an audit trail which would help us make the correction. It's measure twice, cut once type of situation. I have 20 or 30 queries that fall into this category and result in hundreds of updates like this.