. Author manuscript; available in PMC: 2017 Feb 1.

Published in final edited form as: J Occup Environ Hyg. 2017 Jan;14(1):31–39. doi: 10.1080/15459624.2016.1211284

Table 2.

Statistical analyses to assess agreement between the two sampling methods

Exposure measurements		N	Median of concentration ratios (Passive/Active)	Concordance correlation coefficient (CCC)			Linear regression
Exposure measurements		N	Median of concentration ratios (Passive/Active)	CCC-total	CCC- Precision	CCC- Bias	Slope (β)	p- value for β=1
All data	Personal	49	1.24	0.680	0.742	0.916	0.895	0.380
	Area	17	1.14	0.940	0.970	0.969	0.839	0.010
	Combined	66	1.19	0.742	0.788	0.941	0.877	0.157
Data without outliers (Cook’s distance method)	Personal	45	1.20	0.835	0.858	0.973	0.894	0.200
	Area	17	1.14	0.940	0.970	0.969	0.839	0.010
	Combined	62	1.16	0.866	0.888	0.975	0.875	0.037

N: Number of sample pairs

CCC-total = CCC-Precision multiplied by CCC-Bias