Improving Your Data Quality | Completeness

To begin improving your Completeness score, remember that this data quality dimension is comprised of three metrics, listed below. Each metric can be evaluated and addressed to improve your overall Completeness score.

1. Median records per user per year for “core screens”
2. Percent of "core fields" populated in records for "core screens"
3. Percent of dated records for "core screens"

You can review the attached Data Quality packet or the article, Data Quality: The Basics, for a more thorough explanation of each metric.

To begin improving these metrics, reach out to your Success Consultant who can provide a “Data Quality/Completeness” report. This report provides a breakdown for each user account and can help you identify the following issues:

• Users who have a low record count for core screens
• Records that are missing data in core fields
• Records that are un-dated

After reviewing and analyzing the report results, your Success Consultant can help you determine the best action items to address your particular issues. A few potential outcomes may be:

1. Follow up with specific individuals, asking them to add more detail to records that were flagged as “low quality”.
2. Send a general faculty reminder to encourage continuous data entry, especially for core screens.
3. Review the screens where you load data on behalf of faculty via Web Services or CSV data imports (Yearly Data, Scheduled Teaching, etc.). Depending on the time of the year, confirm that you have loaded the most up-to-date data you have available.
4. Identify new opportunities to load data on behalf of faculty.
5. Review the core fields listed in your data quality packet, and consider making some or all of these fields required. When editing an existing record or creating a new record, faculty will be required to enter data into the required field before they are allowed to save the record.

Improving your completeness score will give you confidence that there are enough records in your system to feed your desired outputs and that the records that will be included in reports has enough information to be meaningful.