Overview
When sharing outcomes data on your Virtual Career Center, protecting student privacy — especially around small sample sizes — is a top priority. uConnect includes built-in settings that let you control how outcomes data is reported in AI Search, so you can share meaningful insights without exposing identifiable or overly granular information.
This article covers how to manage these settings and what they do (and don't) affect.
Why This Matters
Small sample sizes can create privacy risk. For example, reporting "2 out of 3 graduates from the Art History program are employed" could make it easy to identify individual students in a small cohort. Reporting that same data as a percentage reduces that risk while still giving students and stakeholders useful, directional insight.
Where to Manage This Setting
- Go to Analyze > Student Outcomes in your admin dashboard.
- Click into the Data Exclusion tab.
- Locate the Graduate Counts section.
Report Percentages Instead of Raw Counts
Under Graduate Counts, you'll find this setting:
☑️ Report percentages instead of raw graduate counts in AI Search
When this box is checked:
- AI Search will report outcomes as percentages (e.g., "25% of graduates") rather than raw headcounts (e.g., "10 out of 40 graduates").
- This reduces the risk of small sample sizes being reverse-engineered to identify individual students.
What this setting does not affect:
- Salary and other non-headcount figures — these are reported independently and are unaffected by this setting.
- Student Outcomes charts on your site — this setting is specific to AI Search responses and does not change how data is visualized on your public-facing Outcomes page.
Additional Data Governance Control: Exclusion Rules
For more granular control, you can also use Exclusion Rules in the same Data Exclusion tab:
- Click + Add Exclusion Rule.
- Define the rule criteria for rows you want excluded.
Each rule you create will exclude matching rows from both the paid ODV module (Outcomes Data Visualization) front end and AI Search — giving you an additional layer of control beyond the percentage-reporting setting, such as excluding programs with very small graduating classes altogether.
Sharing Outcomes Data with Confidence
If your institutional research team or data governance colleagues have questions about how student privacy is protected, you can point to the following built-in safeguards:
- ✅ Raw headcounts can be suppressed in favor of percentages in AI Search, reducing the risk of small cohorts being identifiable
- ✅ Specific rows or cohorts can be excluded entirely from both AI Search and the front end, giving you control over what's shared at all
- ✅ Salary data reporting is kept separate and is unaffected by these privacy controls, so you can manage each type of data independently
These controls are designed to put governance decisions in your institution's hands — you decide the level of granularity that's appropriate for your student population, based on your own data governance policies and IRB/FERPA considerations.