Standard approaches for computing population exposures due to specific sour
ces of air pollutants are relatively complex. In many cases, more simple an
d approximate methods would be useful. This paper develops an approach, bas
ed on the concept of exposure efficiency, that may be used for estimating t
he impact of a source (or source class) on the integrated population exposu
re. The approach is illustrated by an example, which uses the concept of ex
posure efficiency to examine the impact of perchloroethylene emissions from
dry cleaners in the United States. The paper explores the geographic varia
bility of exposure efficiency by evaluating it for each of 100 randomly sel
ected dry cleaners. For perchloroethylene, which has a long atmospheric res
idence time, the site-to-site variability in exposure efficiency is found t
o be relatively small. This suggests that simple exposure assessments, base
d on generic distributional characterizations of exposure efficiency, may b
e used in risk assessments without introducing appreciable uncertainty. For
many compounds, like perchloroethylene, the uncertainty inherent in the es
timation of cancer potency or source emissions would dominate these small e
rrors.