Achieving Data Quality in a Statistical Agency: A Methodological Perspective DATA DETECTIVES: UNCOVERING SYSTEMATIC ERRORS IN ADMINISTRATIVE DATABASES
Bibliographic record
Abstract
Secondary users of health information often assume that administrative data provides a relatively sound basis when making important planning and policy decisions. If errors are evenly and/or randomly distributed this may have little impact. This assumption is betrayed when information sources contain systematic errors, or when systematic errors are introduced in the creation of master files. The most common systematic errors involve underreporting of activity for a specific population, inaccurate re-coding of spatial information, or differences in data entry protocols. The Central East Health Information Partnership (CEHIP) provides information support for development of public health programs and health system planning through a partnership in Ontario’s most populous health planning region. CEHIP has identified a number of systematic errors in administrative databases and has documented many of these in reports distributed to partner organizations. Failures to register births and incorrect assignment of geographic codes in vital statistics files have been studied. Misclassification of cause of death has also been explored, particularly with respect to delays in determining cause of death and the effect this has on official data sets. Differences in data entry protocols for reportable disease data have been researched, raising questions about the consistency of data submitted by different tracking agencies. This paper will describe how some of these errors were identified, and note processes that give rise to such losses in data integrity. The conclusion will address some of the impacts these problems have for health planners, program managers and policy makers.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.726 | 0.854 |
| Meta-epidemiology (narrow) | 0.002 | 0.004 |
| Meta-epidemiology (broad) | 0.006 | 0.004 |
| Bibliometrics | 0.019 | 0.023 |
| Science and technology studies | 0.008 | 0.046 |
| Scholarly communication | 0.031 | 0.027 |
| Open science | 0.009 | 0.019 |
| Research integrity | 0.009 | 0.014 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".