In an attempt to investigate the relationship between exposure to carbon black and respiratory
morbidity, a study of the complete carbon black manufacturing industry in Western Europe was
commissioned. As part of this study, a large number of personal inhalable
n 8015 dust
exposure measurements was taken during three phases of data collection between 1987 and
1995. Repeated measurements on the same worker were taken in the last two phases, which
enabled the estimation of the within- and between-worker components of variance.
Simultaneously, the ®xed eects of phase and factory were estimated using mixed-eects
analysis of variance.