Table 3.
Exposure information from the validation study (Van Roosbroeck et al. 2008a) available for measurement error correction.
| Data | PM2.5 absorbancea (10–5/m) | NO2 (μg/m3) | PM2.5 (μg/m3) |
|---|---|---|---|
| n | 172 | 209 | 174 |
| Measured personal exposure (mean ± SD) | 1.71 ± 0.70 | 26.9 ± 11.3 | 16.8 ± 11.2 |
| Measured ambient exposure (mean ± SD) | 1.61 ± 0.63 | 32.0 ± 8.4 | 18.2 ± 10.0 |
| Ratio of personal and ambient SDs | 1.11 | 1.35 | 1.12 |
| Correlation of personal and ambient exposures | 0.78 | 0.04 | 0.45 |
| Validation model R2 | 0.62 | 0.22 | 0.21 |
| p-Value for test of heteroscedasticity | 0.74 | 0.14 | 0.77 |
| σ2 | 0.044 | 0.326 | 1.004 |
| Deattenuation factorb | 0.87 | 0.05 | 0.50 |
| aPM2.5 absorbance was measured in the validation study and is used to adjust models for black smoke. bThe deattenuation factor is calculated by multiplying the ratio of the personal and ambient exposure standard deviations by the correlation between the personal and ambient measures. | |||