Table 3.

Stepwise multiple regression models, adjusted r²,^a of air concentrations and pesticide use (lb/mi²).

Variables added sequentiallyb	Chlorpyrifos	Chlorpyrifos oxon	Diazinon	Malathion
Model series 1c
Use within 3 miles: day i	0.07	0.13*	0.12*	0.08
Use within 3 miles: day i – 1	0.08	0.21#	0.17*	0.14
Use within 3 miles: day i – 2	0.09	0.24#	0.23**d	0.14
Use within 3 miles: day i – 3	Removede	Removede	Removede	0.15d
Model series 2f
Use within 1.5 miles: day i	0.05	0.16#	0.01	0.05
Use within 1.6–3 miles: day i	0.07	0.19#	0.05	0.07
Use within 1.5 miles: day i – 1	0.09	0.24#	0.06	Removede
Use within 1.6–3 miles: day i – 1	Removede	0.25#	0.1	0.13
Use within 1.5 miles: day i – 2	0.13*	0.31#	0.12	Removede
Use within 1.6–3 miles: day i – 2	Removede	Removede	0.13	0.16
Use within 1.5 miles: day i – 3	0.18*	0.39#	0.14	Removede
Use within 1.6–3 miles: day i – 3	Removede	Removede	Removede	Removede
Use within 1.5 miles: day i – 4	0.21d	0.41#d	NA	NA

NA, not applicable (i.e., analysis was not conducted).

^aAdjusted for the number of variables in the model; statistical significance is a test of the overall significance of the multivariate model.

^bEach row within a model series represents a regression model with the named variable (the added variable) and the variables named in the rows above it included. Day i is the sample collection day. Day i – 1, i – 2, i – 3, and i – 4 represent each of the 4 days before the sample collection day.

^cFor model series 1, pesticide use in PLSS sections within a 3-mi radius of monitor was entered as one variable.

^dPesticide use variables in this model selected for subsequent analysis.

^eThe adjusted r² decreased compared with previous model or the use parameter coefficients were negative; the variable was eliminated from subsequent models.

^fFor model series 2, use in PLSS sections within a 1.5- and a 1.5- to 3-mi radius of monitor were entered as two separate variables.

^*p < 0.01;

^**p < 0.001;

^#p < 0.0001.