Skip to main content
. Author manuscript; available in PMC: 2015 Apr 9.
Published in final edited form as: J Expo Sci Environ Epidemiol. 2012 Jan 18;22(2):135–147. doi: 10.1038/jes.2011.45

Table 3.

10-fold cross-validation R2 values from a variety of models based on the eight study communities with complete land-use data.

Model Predictorsa Pollutant
NO2 NO NOx
1 Distances 0.24 0.19 0.24
2 Volume 0.20 0.22 0.23
3 Elevation 0.32 0.31 0.34
4 Population density 0.04 0.05 0.05
5 Land-use categories 0.08 0.03 0.05
6 Distance, elevation, population 0.51 0.45 0.53
7 Distance, elevation, population, land use 0.50 0.46 0.53
8 Distance, volume, elevation, population 0.63 0.55 0.63
9 CALINE-modeled TRP 0.65 0.59 0.67
10 CALINE-modeled TRP, distance, elevation, population 0.66 0.61 0.70
11 CALINE-modeled TRP, distance, volume, elevation, population 0.68 0.61 0.71
12 CALINE-modeled TRP, distance, volume, elevation, population, land use 0.67 0.61 0.71
a

Each set of predictors includes all of the variables listed in Table 2 under the respective heading. Squares of the distance and volume variables were also considered. Stepwise selection was used to obtain the best subset of predictors for each model.