Duplicate IDs Data Issue Documentation

Issue Summary

Duplicates occur in several instances, both with students and with staff; for example, when students are applying to multiple campuses, when students transfer from one campus to another, and when former students become employees (and vice versa). If an individual is not identified as already having an existing EMPLID in PeopleSoft and a new EMPLID is created, this creates confusion and poor experiences for the student or employee. Students’ credits are not all tracked toward graduation in a single space, financial aid awarded and payments made may credit to the wrong account, emails may arrive at unmonitored addresses, and pay history and tax-filing forms may have incomplete or duplicate data. Ultimately, all of these outcomes diminish the experiences of our students and staff and create work for multiple university departments.

Background

Duplicate IDs have been an existing issue since the implementation of PeopleSoft within the UMS; over the years, more than 11,000 duplicate IDs have been reported. Over the last decade, 6,000 duplicates have been cleaned up and deleted. Currently, there are 6,843 records needing attention. It is the goal of this workgroup to continue the clean-up and deletion process, as well as reduce the number of duplicates being created.

November 2020 Proposal Forum

Resources

Duplicate IDs Proposal

Proposal Forum Presentations Slides presented by Linda Reid and Janet Boucouvalas

Proposal Summary

To address duplicate IDs (EMPLIDS) within the UMS, two projects are proposed:

• Revise and enhance duplicate resolution processes to clean up the thousands of already-existing duplicate IDs in the system (PeopleSoft).

• Create processes & documentation, as well as designate responsible parties, to help decrease the frequency of duplicate IDs being created on an ongoing basis. 

Background

Duplicate IDs have been an existing issue since the implementation of PeopleSoft within the UMS; over the years, more than 11,000 duplicate IDs have been reported. Over the last decade, 6,000 duplicates have been cleaned up and deleted. It is the goal of this workgroup to continue the clean-up and deletion process, as well as reduce the number of duplicates being created.  

Janet shared current system process flowcharts: centralized processes, then campus processes. These are only two of many maps for this issue which also includes HR. 

Solution & Next Steps 

• Establish ownership for creation, resolution, decision making and communication related to Duplicate IDs. This would include assigning responsibilities to Duplicate ID Coordinators at each campus to work with all functional areas to resolve duplicates identified. 

• Create documentation on creation of records, resolution of Duplicate IDs, and update training information based on annual review of documentation.  
• Establish consistent System-wide communication between parties involved in Duplicate ID resolution.
• Increase collaboration between HR and CS to minimize Duplicate ID creation within the two separate databases.
• Consistency in capturing names on forms used to create new EMPLIDs (i.e., registration forms and applications).

Comments & Questions

• Is there a consideration for census dates and any prioritization to ensure that duplicates are cleaned up by census days? For example, if a student becomes part of an initial cohort with a duplicate ID and then disappears down the road, that would affect our retention and graduation rates, etc. 
• Can IR and HR be brought in to collaborate? These groups might be able to help spot check and ensure that the correct duplicates are being removed and that appropriate fixes are in place to account for any inflation in cohorts or to prevent inappropriate deletions.
• Is there any definition for Duplicate IDs--at times we want to eliminate an ID and at other times keep them. What is the difference between eliminating and merging, etc.?
• Is it ever the case that we intentionally have duplicates within the system? For example for students who study at multiple campuses?
  • None are intentional. Duplicates arise from nicknames or former names, for example.
  • UMS-wide---one EMPLID per one person.
  • Definition documentation would be great to have---what is merging?, what is eliminating duplicates?, etc. Establishing documentation is a primary next step.
• It is overwhelming that there are 6000 duplicates across the UMS. It hasn’t occurred often in individuals’ experiences.
• In the documentation, it shouldn’t be using technical terms or data elements---it's one ID per person and we fix the data to achieve that goal.
• Would an EMPLID ever change over time for any reason?
  • No, it doesn’t ever change by design. Sometimes a future ID becomes the keeper though---in that case, it is a mistaken change in ID over time. Sometimes an ID can be requested to change if there are objections to the ID itself.
• Why can’t we use SSN?
  • Students do not always provide SSNs anymore; we cannot require that they provide it.
  • SSNs could have typos. 
  • SSNs could get reassigned over time
• When we converted our data into PeopleSoft, we had 3000 duplicates and reduced that down to almost zero.
• Also of note, this is not unique to the UMS---other institutions who use PeopleSoft have similar issues.
• Do Early College students get an EMPLID?
  • Yes. These students come through the SPC and are usually correctly matched; this process requires hand matching on names, locations where the student has lived, etc.
• Can we do better with preventing duplicates in the first place? For example, POIs do not have SSNs, but now require DOB. Working on improving in this way.
• I wonder if there is a group of students who represent roughly 80% of the problem.
• Importance of training on Search/Match. This tool, when used carefully, could prevent a lot of duplicates.
• Batch processes loading, e.g., SAT scores–do these affect duplicates?
  • Yes, there are lots of these. Lots of hand matching goes on in this area.

Proposed Solutions

• Establish ownership for creation, resolution, decision making and communication related to Duplicate IDs. This would include assigning responsibilities to Duplicate ID Coordinators at each campus to work with all functional areas to resolve duplicates identified. 
• Create documentation on creation of records, resolution of Duplicate IDs, and update training information based on annual review of documentation.  
• Establish consistent System-wide communication between parties involved in Duplicate ID resolution.
• Increase collaboration between HR and CS to minimize Duplicate ID creation within the two separate databases.
• Consistency in capturing names on forms used to create new EMPLIDs (i.e., registration forms and applications).