Add new rules for de-identified participant data

[jacobmpeters]

Rules are available here:

Action Items:

• [x] Test in Stage and deliver to Box
• [x] Review the stage report with Madhuri

[jacobmpeters]

@mnataraj92

Hi Madhuri,

A while ago you guys wrote up some new Recruitment QAQC rules in the Rules to Add tab of qc_rules_recruitment.xlsx on box.

I tested these rules in stage on 11/21 and again today 12/28.

• qc_report_recruitment_stg_NEW_RULES_ONLY_2023-11-21.xlsx
• qc_report_recruitment_stg_NEW_RULES_ONLY_2023-12-28.xlsx

The new rules flagged a lot of issues that are likely not real. We never got around to polishing these rules and implementing them in production, primarily because I was preoccupied with memory issues that I have been addressing this month.

Anyway, I want to make sure that we polish these up and get them into production.

Would you like to meet to go over the results of the stage report?

[mnataraj92]

Thanks @jacobmpeters! I'll reach out to schedule some time to go over these!

[jacobmpeters]

@jacobmpeters and @mnataraj92 met to discuss the results of qc_report_recruitment_stg_NEW_RULES_ONLY_2023-12-28.xlsx.

Action items:

Jake:

• [ ] "Pre-consent Opt-out" rule --> Check if "crossvalidDate" this is the wrong qctype, if so, Kelsey might need to code a custom QC metric
• [ ] "Not Active No Age" rule --> Check why this is flagging participants that do not have values for the state variable
• [ ] "Suffix" --> This one has "string", "integer", "provided" subfields in Stage. Test in prod.

Madhuri:

• [ ] "Not Active No Verification" rule --> Check w/ Nicole which values should be included in ValidValues

Once these items are done, I can run this again in stage.