the mother had her first live birth of a child who had ever lived with her. Hence any
births which were reported as having been given away in adoption or previous still births
are not counted. The youngest group of teenaged mothers, who form the reference
category in regressions, were mostly aged 17 or 18 at the time of their first birth (n = 836
and 1180 respectively). Sample numbers at ages up to 15 were 146, and at age 16, 383.
The oldest age group shown in table 1, those aged 31 or over are mostly in their early
thirties. This group is split further in the regressions into those aged 31-33 (n = 1894),
946 aged 34-36 and 409 aged 37 or more (of whom 93 were 40 or over).
Table 2 gives a few indications of the demographic characteristics of the cohort families
by mother’s age at first birth, excluding only 27 families where the informant was not the
cohort child’s natural mother, and 9 families where it was not possible to infer her age at
first birth. At the time of the survey, not necessarily the situation at the first birth, 14% of
the mothers were living without a partner. There is a near ten-fold difference in the
proportion of youngest and oldest entrants to motherhood who were ‘lone mothers’ at the
time of the survey - 39% where the first child had been born before the mother was 19
compared to 4.3 percent where she had been 31 or more. This is partly because early
childbearing is more likely to be unpartnered, and partly because of the greater fragility
of early partnerships. Where there is a two parent family, the chances of the couple being
married rises with age at motherhood, as do the (high) chances of the mother’s partner
being the natural father of the cohort child. Reconstituted families are most common
among mothers who started early. 22% of cohort children with a mother whose first
child was born before age 19 had half or step siblings in the family compared with around