COVID-19 data are messy: analytic methods for rigorous impact analyses with imperfect data
Bibliographic record
Abstract
BACKGROUND: The COVID-19 pandemic has led to an avalanche of scientific studies, drawing on many different types of data. However, studies addressing the effectiveness of government actions against COVID-19, especially non-pharmaceutical interventions, often exhibit data problems that threaten the validity of their results. This review is thus intended to help epidemiologists and other researchers identify a set of data issues that, in our view, must be addressed in order for their work to be credible. We further intend to help journal editors and peer reviewers when evaluating studies, to apprise policy-makers, journalists, and other research consumers about the strengths and weaknesses of published studies, and to inform the wider debate about the scientific quality of COVID-19 research. RESULTS: To this end, we describe common challenges in the collection, reporting, and use of epidemiologic, policy, and other data, including completeness and representativeness of outcomes data; their comparability over time and among jurisdictions; the adequacy of policy variables and data on intermediate outcomes such as mobility and mask use; and a mismatch between level of intervention and outcome variables. We urge researchers to think critically about potential problems with the COVID-19 data sources over the specific time periods and particular locations they have chosen to analyze, and to choose not only appropriate study designs but also to conduct appropriate checks and sensitivity analyses to investigate the impact(s) of potential threats on study findings. CONCLUSIONS: In an effort to encourage high quality research, we provide recommendations on how to address the issues we identify. Our first recommendation is for researchers to choose an appropriate design (and the data it requires). This review describes considerations and issues in order to identify the strongest analytical designs and demonstrates how interrupted time-series and comparative longitudinal studies can be particularly useful. Furthermore, we recommend that researchers conduct checks or sensitivity analyses of the results to data source and design choices, which we illustrate. Regardless of the approaches taken, researchers should be explicit about the kind of data problems or other biases that the design choice and sensitivity analyses are addressing.
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 | 0.627 | 0.881 |
| Meta-epidemiology (narrow) | 0.004 | 0.006 |
| Meta-epidemiology (broad) | 0.008 | 0.008 |
| Bibliometrics | 0.018 | 0.023 |
| Science and technology studies | 0.005 | 0.024 |
| Scholarly communication | 0.020 | 0.023 |
| Open science | 0.012 | 0.019 |
| Research integrity | 0.009 | 0.025 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".