Several ecologic studies have stressed the association between adolesc
ent fertility and socioeconomic indicators, yet the true magnitude of
the association and the intermediate factors involved remain unclear.
The objectives were to describe in an urban context the ecologic assoc
iation between the adolescent (15-19 year olds) fertility rates distri
bution and trends and to compare the effect of using different aggrega
tes. The results show that the adolescent fertility rates distribution
and trends between 1980 and 1992 in 10 Barcelona districts parallel t
he socioeconomic distribution as measured by a socioeconomic index (av
erage family income index). The average family income index is negativ
ely associated with fertility rates, with Pearson coefficients of -0.7
8 and 0.67 at the district and neighbourhood levels respectively. With
in Spain,, using local (Barcelona's districts or neighbourhoods) or re
gional (province or communities) aggregates yields a different degree
of ecologic association, with local analysis giving a higher ratio bet
ween the highest and lowest rates (1:34.1 at the neighbourhood level,
1:3.2 at the community level) and a higher correlation with income (-0
.70 at the neighbourhood level, -0.82 at the district level), which is
lower and statistically non-significant at the province or community
level. This study supports the use of the smaller available units of a
nalysis in ecologic studies. The influence of ecologic bias and the po
tential role of socioeconomic factors leading to pregnancy and childbe
aring among adolescents are discussed.