This paper discusses the use of partial association measures for carry
ing out path analyses on categorical data. The measures considered are
essentially PRE (proportion of reduction in error of prediction) meas
ures for nominal variables and concordance-discordance indices for ord
inal ones. These measures provide a natural way to evaluate the streng
th of the path linking a non-measurable response variable to one of se
veral categorical explanatory factors. Concerning the decomposition of
raw association into direct and indirect effects, it is shown, howeve
r, that they do not share the properties of conventional path coeffici
ents for measurable variables. Especially purely nominal association m
easures need to be interpreted with care. The scope of the partial mea
sures for path analysis is illustrated through a study of the relation
ships between the educational styles experienced by swiss adolescents
and their self-esteem.