Before generation
Use this as an early screening step before publishing a small aggregate grade table. It identifies cells below the chosen threshold and warns when one masked value can be recovered from the total.
EDUCATION DECISION WORKBENCH
Grade Distribution Privacy Checker
Apply primary small-cell suppression and detect the simplest complementary-disclosure risk.
Runs in this browser tabEditable downloadUp to 500 structured rows and 30,000 characters per field. Use anonymous or fictional labels.
Review output
Your learning artifact appears here
Load the worked example to inspect realistic inputs, or enter anonymous, policy-approved information. Results remain in this browser tab until you copy or download them.
Transparent method
How the grade distribution privacy checker works
Primary suppression masks count < threshold. A complementary-risk warning appears when a visible total accompanies exactly one suppressed cell. More complex linked-table, differencing, and characteristic risks are outside this checker. “No simple risk detected” is not privacy clearance. Review other tables, totals, percentages, prior releases, and local knowledge that could enable re-identification.
- 01Before generation
Obtain the approved disclosure policy and aggregate counts.
- 02During review
Review primary and complementary risks.
- 03Before use
Send the complete release package through the authorized privacy review.
How to interpret the result
Grade Distribution Privacy Checker
Worked example
Counts of 14, 11, 8, 3, and 2 are tested with n<5. The last two cells receive primary suppression. Because more than one cell is suppressed, the simplest single-cell subtraction risk is reduced, but a full disclosure review is still required.
Formula or methodology
Primary suppression masks count < threshold. A complementary-risk warning appears when a visible total accompanies exactly one suppressed cell. More complex linked-table, differencing, and characteristic risks are outside this checker.
How to interpret the result
“No simple risk detected” is not privacy clearance. Review other tables, totals, percentages, prior releases, and local knowledge that could enable re-identification.
Common mistakes
Do not use an arbitrary threshold, publish percentages with tiny denominators, forget complementary suppression, or assume removing names de-identifies a table.
Learning boundary
Mechanical suppression does not prove compliance
FERPA and other requirements depend on context. An authorized privacy professional must approve publication.
Review checklist
Human review before educational use
- Only aggregate counts are entered.
- The threshold is policy-approved.
- Totals and linked tables are reviewed.
- An authorized disclosure review approves release.
Practical questions
Grade Distribution Privacy Checker questions
Why can a total reveal a suppressed cell?
If all other cells and the total are visible, subtraction can recover the hidden count.
Is n<5 always the right threshold?
No. Use the threshold and methods required by the responsible organization.
Does this replace a FERPA review?
No. It screens one table with limited disclosed rules.