The re-identification risk of Canadians from longitudinal demographics
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
BACKGROUND: The public is less willing to allow their personal health information to be disclosed for research purposes if they do not trust researchers and how researchers manage their data. However, the public is more comfortable with their data being used for research if the risk of re-identification is low. There are few studies on the risk of re-identification of Canadians from their basic demographics, and no studies on their risk from their longitudinal data. Our objective was to estimate the risk of re-identification from the basic cross-sectional and longitudinal demographics of Canadians. METHODS: Uniqueness is a common measure of re-identification risk. Demographic data on a 25% random sample of the population of Montreal were analyzed to estimate population uniqueness on postal code, date of birth, and gender as well as their generalizations, for periods ranging from 1 year to 11 years. RESULTS: Almost 98% of the population was unique on full postal code, date of birth and gender: these three variables are effectively a unique identifier for Montrealers. Uniqueness increased for longitudinal data. Considerable generalization was required to reach acceptably low uniqueness levels, especially for longitudinal data. Detailed guidelines and disclosure policies on how to ensure that the re-identification risk is low are provided. CONCLUSIONS: A large percentage of Montreal residents are unique on basic demographics. For non-longitudinal data sets, the three character postal code, gender, and month/year of birth represent sufficiently low re-identification risk. Data custodians need to generalize their demographic information further for longitudinal data sets.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.028 | 0.030 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it