PROMOTING POTENTIAL 303United States, the National Institute of Child Health and Human Devel-
opment’s Study of Early Child Care and Youth Development. In the
United Kingdom, the best example is the National Child Development
Study. It includes all the individuals— around sixteen thousand— born in
one week in 1958 through a number of follow- ups into adulthood.
Th e aim of these studies is not merely to replicate the already
well- known correlations between environmental factors and indices of
development, which they do. Th ey are also able, with such large samples,
to perform more detailed statistical analyses that might be more indicative
of causes, and, therefore, of targets for intervention.
However, that aim has again been mitigated by the need, with such
large samples, to use only broad categories of environmental factors (with
the exception of some easily identifi able specifi cs like smoking in preg-
nancy). For example, in relation to reading and math attainments at
eleven years, the National Child Development Study found school per for-
mance to be statistically associated with the following variables: SES, de-
gree of parental initiative in contacts with school, housing tenure (owned
or rented), geographic region, amenities in the home (e.g., bathroom, toi-
let), father’s education, mother’s education, family size, and crowding.
With such a large sample, the eff ects on attainment of each of these cou ld
be statistically “distinguished” from one another. However, although such
associations are indicative in some impor tant re spects, the true nature of
the factors, and how they work, is diffi cult to establish. For example, one of
the biggest eff ects on gain in reading attainment at age eleven seemed to be
having, or not having, sole use of amenities like an indoor toilet. In other
words, these factors leave much room for speculation.
In such studies, the implicit input → output model is suggested by the
assumption that children in the same home automatically experience the
same “environment.” Yet children in the same home turn out to be almost
as diff er ent from one another, cognitively, as those from diff er ent homes.
In a well- known paper, Robert Plomin and Denise Daniels suggested this
was due to children in the same family actually experiencing diff er ent
environments.^23
But identifying such “nonshared” environments has been diffi cult. For
example, a meta- analy sis of forty- three papers that addressed associations
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:58:05 UTC