How do you de-identify student rosters and essays locally under FERPA?
Load your .csv, .tsv, or .xlsx class roster (or drop student .pdf and .docx submissions), verify the flagged PII columns, pick an Alias format, and download a clean ZIP bundle containing your private lookup key CSV. Every byte of student data stays inside your local browser tab.
Teachers, instructional coaches, and department chairs regularly share pre-test and post-test growth tables with PLC colleagues, grant evaluators, or spreadsheet scripts. Stripping direct identifiers on your own laptop supports U.S. Department of Education FERPA student privacy guide safeguards while keeping a local lookup key for follow-up student support.
If your scores live across separate Canvas and Kahoot exports, run multi-file LMS gradebook spreadsheet joins first so every student row is consolidated before masking PII. The table below compares all three alias formats and two workflow modes.
| Mode or alias option | Sample generated identifier | Re-identification key columns | Alphabetical leak guard | Primary classroom use case |
|---|---|---|---|---|
Sequential code |
Student 001, Essay_002 |
Anonymous Code + masked PII headers |
Sort by alias (001..N) |
Department PLC tables and numbered lab packets |
Neutral nature pseudonyms |
River Harbor 001 |
Anonymous Code + masked PII headers |
Lexicographic alias sort | Readable case-study discussions and IEP exemplars |
Salted SHA-256 token |
ANON-7F3A91C2 |
Anonymous Code + masked PII headers |
Pseudo-random hex sort | Grant evaluations and external research datasets |
Spreadsheet roster mode |
Participant ID column 1 |
Collapses multiple PII cols into 1 ID |
Enabled by default | .csv, .tsv, and .xlsx score spreadsheets |
Blind-grading files mode |
Submission_001.pdf |
Anonymous Filename, Original Filename |
Shuffled via SHA-256 token |
Double-blind essay, capstone, and portfolio zips |
Comparing sequential codes, nature pseudonyms, and salted SHA-256 tokens
Different review workflows call for different participant labels. This student roster anonymizer offers three deterministic Alias format options that apply consistently across both tabular spreadsheets and batch submission archives.
Numbered participant prefixes and neutral nature aliases
Selecting Sequential code (Student 001, Student 002…) unlocks the Code prefix input so you can label rows Student 001, Examinee 001, or BioLab 001 with 3-digit zero-padding. In blind-grading file mode, spaces in your prefix convert to underscores so filenames like Essay_001.pdf open cleanly on Windows, macOS, and Linux.
Choosing Neutral nature pseudonyms (River Harbor 001…) pairs 12 gender-neutral given names (River, Cedar, Rowan, Quinn, Sage, Avery, and 6 others) with 10 landscape surnames (Harbor, Summit, Meadow, Brook, Vale, Ridge, and 4 others). Those 120 combinations plus a 3-digit suffix make PLC conversations far easier to follow than raw serial numbers.
Salted SHA-256 tokens for non-sequential roster masking
When sharing data outside your immediate teaching team, Salted SHA-256 token (ANON-7F3A91C2…) prepends a timestamped session salt (ferpa-<ms>) to each lowercase identity string before computing a 256 bits digest (FIPS 180) from the SHA-2 cryptographic hash family overview. The formatter extracts the leading 4 bytes (32 bits) as 8 uppercase hexadecimal characters prefixed with ANON-.
Because hash outputs distribute uniformly across 0 through F, sorting rows by ANON- token scrambles original roster positions thoroughly. Instructors who scan multiple-choice bubble exams with KR-20 stats often apply salted SHA-256 tokens before posting anonymized score distributions for class review.
Masking spreadsheet PII columns and hiding alphabetical surname order
Deleting only a Student Name column in Excel frequently leaves SIS ID, SIS Login, or Parent Email columns exposed further to the right. Worse, keeping rows in their original order leaks identity whenever Student 001 corresponds to Adams and Student 028 corresponds to Zimmerman.
Auto-detecting name, SIS ID, parent email, and birth-date headers
As soon as you drop a .csv, .tsv, or .xlsx sheet, detectPiiColumnIndices scans every column header and pre-checks fields matching names, logins, emails, phone numbers, street addresses, parent or guardian contacts, and birth dates. Assessment columns containing score, grade, points, percent, quiz, exam, lab, total, section, or period stay unchecked so quantitative metrics remain intact.
You can toggle individual column checkboxes before exporting, while at least 1 identifier column stays active to anchor the lookup table. All checked PII columns collapse into a single leading Participant ID column inside the De-Identified Data worksheet, while daily paper roll calls can stay on separate printable weekly and monthly class attendance grids in your classroom binder.
Sorting by alias to break alphabetical surname order
Leaving Sort anonymized rows by alias (hides alphabetical surname order) checked re-orders the exported rows lexicographically by Participant ID after aliases are assigned. When paired with Salted SHA-256 token or Neutral nature pseudonyms, row 1 of the anonymized sheet no longer reflects row 1 of your alphabetical SIS roll.
Under Export package, the default ZIP bundle writes three separate files compressed at deflate level 6: <stem>-anonymized.xlsx, <stem>-anonymized.csv, and PRIVATE-reidentification-key.csv. This operational tool assists with classroom privacy workflows and does not constitute formal district legal counsel; verify institutional data-sharing policies before releasing records externally.
Batch-renaming student essay and lab files for blind grading
Downloading bulk assignment submissions from Canvas, Moodle, or Google Classroom typically names every file lastname_firstname_essay.pdf. Seeing familiar student names before opening a rubric introduces measurable grading halo effects on subjective writing and capstone projects.
Packaging anonymized PDF and DOCX submissions with a private key
Switching to Blind-grading files (.pdf / .docx) lets you drop dozens of .pdf, .docx, .txt, .md, .png, .jpg, or .zip files at once. Inside edu.worker.ts, hashing runs asynchronously through the browser’s built-in MDN Web Crypto SubtleCrypto.digest() reference engine without transmitting a single document over the internet.
Clicking Download blind-grading ZIP + key produces blind-grading-submissions.zip containing an anonymized-submissions/ folder alongside PRIVATE-reidentification-key.csv. After scoring every coded paper, open the key CSV or merge scores back onto your master roster in a few clicks.
Checking internal document headers before committee review
Filename anonymization changes only the outer archive entry name while preserving exact binary document contents, digital signatures, and formatting. Before running department-wide blind calibration, remind learners to omit their names from MLA page-one headers and Microsoft Word Author properties.
Elementary interventionists compiling anonymized RTI progress portfolios frequently pair de-identified benchmark tables with vector fraction bars ready for small-group packets during grade-level team meetings. Keeping demographic identifiers out of shared folders lets specialists focus strictly on skill growth.
Tip: Store
PRIVATE-reidentification-key.csvin an encrypted local folder before emailing<stem>-anonymized.xlsxto colleagues, and choose Salted SHA-256 token whenever original roster rows were sorted A-to-Z by last name.