Bibliographic record
Abstract
IntroductionAt the Manitoba Centre for Health Policy (MCHP), we have been performing data linkage for over 25 years. Over time, the Manitoba Population Research Data Repository (MPRDR) has expanded to over 80 datasets. Data linkage methods are key to bringing all this data together for population-based research. Objectives and ApproachThe presentation will include a detailed description of the individual steps involved in the data linkage process and provide information about the methods developed and knowledge gained over time at MCHP. We will present different scenarios linking health, education, social and justice data and the choices that are made prior to and during data linkage. The data linkage process and linkage methods, including data validation techniques, will be illustrated with examples from our work. ResultsThe presentation will describe the different types of data we have in the MPRDR and illustrate how the data are processed in a de-identified manner so that privacy and confidentiality are maintained. The presentation will provide details on the data linkage methods used, dependent on the type of data sources being linked. This involves identifying and describing a 5-step data linkage process, including: pre-processing (gaining knowledge about the data and cleaning/standardization techniques); searching for and selecting the appropriate linkage variables; applying different linkage techniques (e.g.: deterministic, probabilistic, “fuzzy matching” and manual review) to the data, “rules” for deciding when data linkage should occur, and reporting and Interpreting linkage outcome metrics and quality. Conclusion/ImplicationsOur ability to link different data sources provides the capacity to study questions and complex issues related to health, social, education and justice from a population perspective. The techniques and methods described in this presentation should be applicable to other organizations linking administrative data.
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 | 0.025 | 0.049 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.017 |
| Science and technology studies | 0.006 | 0.002 |
| Scholarly communication | 0.005 | 0.002 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.023 | 0.003 |
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".