The impact of image resolution on power, bias, and confounding
Bibliographic record
Abstract
Studies of the impact of environmental pollutants on health outcomes can be compromised by mismeasured exposures or unmeasured confounding with other environmental exposures. Both problems can be exacerbated by measuring exposure from data sources with low spatial resolution. Artificial light at night, for example, is often estimated from low-resolution satellite images, which may result in substantial measurement error and increased correlation with air or noise pollution. METHODS: Light at night exposure was considered in simulated epidemiologic studies in Vancouver, British Columbia. First, we assessed statistical power and bias for hypothetical studies that replaced true light exposure with estimates from sources with low resolution. Next, health status was simulated based on pollutants other than light exposure, and we assessed the frequency with which studies might incorrectly attribute negative health impacts to light exposure as a result of unmeasured confounding by the other environmental exposures. RESULTS: When light was simulated to be the causal agent, studies relying on low-resolution data suffered from lower statistical power and biased estimates. Additionally, correlations between light and other pollutants increased as the spatial resolution of the light exposure map decreased, so studies estimating light exposure from images with lower spatial resolution were more prone to confounding. CONCLUSIONS: Studies estimating exposure to pollutants from data with lower spatial resolution are prone to increased bias, increased confounding, and reduced power. Studies examining effects of light at night should avoid using exposure estimates based on low-resolution maps, and should consider potential confounding with other environmental pollutants.
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 | 0.307 | 0.562 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.003 | 0.006 |
| Science and technology studies | 0.002 | 0.010 |
| Scholarly communication | 0.006 | 0.004 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 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 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".