. Author manuscript; available in PMC: 2022 May 6.

Published in final edited form as: Tob Regul Sci. 2020 Jul;6(4):266–278. doi: 10.18001/trs.6.4.4

Table 4.

Statistical Power of the Unadjusted Structural Equation Model (SEM) and Simple Linear Regression (SLR) Models for Random Samples of Different Sample Sizes

	Power^a
	CPD		TNE
	N^b = 50	N = 100	N = 50	N = 100
SEM
Latent MA exposure	0.890	0.989	0.996	0.999
SLR with each biomarker
log(3-HPMA)	0.780	0.965	0.972	0.999
log(CEMA)	0.812	0.985	0.993	0.999
log(HMPMA)	0.805	0.980	0.979	0.998
log(2-HPMA)	0.269	0.503	0.495	0.678
log(SPMA)	0.548	0.865	0.878	0.963
SLR with the sum of 5 biomarkers
log(Total)	0.847	0.987	0.985	0.999

Note.

^a:

Power is the probability of correctly accepting that two variables are related.

^b:

N: the sample size within each Monte-Carlo simulation.