What We Know about Early Childhood Interventions
by Janet Currie
Early intervention programs are once again in the spotlight as
policymakers and others debate whether government should fund a
universal preschool program, whether Head Start should be expanded,
and whether government should pursue policies to improve early interventions.
In a recent Joint Center for Poverty Research working paper, "Early
Childhood Intervention Programs: What Do We Know?" Janet
Currie reviews the evaluations of several early childhood intervention
programs. The programs she examines are predominantly center-based
programs that emphasize school readiness. She finds that well-designed,
well-funded early interventions can have large and significant effects
on school readiness and subsequent child outcomes. She also, however,
finds a paucity of high-quality research on early intervention programs;
only four of the many programs evaluated used random assignment.
Effects of Early Interventions
Currie focuses on four model programs that were evaluated using
randomized trials-the Perry Preschool Project, the Carolina Abecedarian
Project, the Early Training Project, and the Milwaukee Project.
In contrast to these small, model programs, she also looks at results
from Head Start, a large, publicly funded program.
The Early Training Project showed dramatic reductions in use of
special education among youth at age 12, although there were no
statistically significant differences between treatments and controls
in achievement test scores, grade retention, or high school graduation.
Although not significant, the outcomes were nevertheless in the
right direction. This intervention was the least intensive of the
The Abecedarian Project found that, at age 15, those children who
had received a preschool intervention scored higher on achievement
tests and had lower incidence of special education and grade retention.
At age 21, the children who received the preschool intervention
were twice as likely to still be in school or to have attended a
four-year college. The effects of interventions for school-aged
children were either small or not statistically significant.
The Perry Preschool Project found positive effects on achievement
tests, grades, high school graduation rates, and earnings, as well
as lowering crime and welfare use, as of age 27. Only the Milwaukee
Project found any long-term effect on IQ. That project found higher
IQs by grade 8, but the control children enjoyed no other advantages.
Head Start has never been evaluated using randomized trials. Moreover,
few studies have attempted to follow children past elementary grades.
The studies that do exist, however, find that in the short-term,
Head Start is contributing to school readiness by improving verbal
skills and health. In short, the results are favorable, but much
less conclusive than evidence from the above model programs.
How They Work: Getting Inside the Black Box of Interventions
Three questions arise when considering how the interventions effect
the outcomes that they do. The first is one of targeting. Where
should policymakers "draw the line" on instituting intervention
programs? Do disadvantaged children benefit more from the program?
Is there an optimal age for intervention? A larger question is what
constitutes quality? If a program is showing better results, what
factors contribute to those results: small class sizes, curriculum,
A tentative answer to the first question-do programs have greater
benefits for more disadvantaged children-is yes, although the answer
likely depends on which aspect of disadvantage is being targeted.
The investigators in the Abecedarian project found program effects
to be twice as large for the most disadvantaged children. In general,
the findings suggest that, where budgets are limited, it is appropriate
to target early intervention to the most disadvantaged children.
What constitutes quality is a more difficult question to pin down.
Typically people speak of quality as constituting a combination
of structure (classroom size, e.g.) and classroom process (e.g.,
teacher interactions, developmentally appropriate material). There
is little proof that the curriculum itself matters. What appears
to matter more is how the teacher relates to and interacts with
Finally, the optimal age of intervention has often been pinpointed
in popular discussion as between birth and 3 years. Brain studies,
however, show that older children are likely to be able to benefit
from intervention as well. Hence, it may be more important to worry
about the quality of the intervention than about the exact timing
Currie concludes that the better studies show larger and more
significant effects of early intervention programs. However, there
are still too few quality evaluations and disappointingly little
consistent evidence of the long-term effectiveness (or lack of it)
of early intervention. Also, the programs for which there are excellent
evaluations are all model programs, and there is a considerable
gap between such interventions and the large-scale, publicly funded
interventions, such as Head Start. Nevertheless, the generally positive
findings support expanded funding for Head Start.
In designing programs, the following guidelines might be kept in
mind. Although it may be useful to intervene before age 3, interventions
for preschool and school-age children can also be effective. Second,
disadvantaged children seem to benefit more from early intervention,
which provides a rationale for targeting such programs to these
children. Finally, the most important aspect of quality is likely
to be the nature of the interaction between teacher and child.
Early interventions with children are designed to provide a secure,
stimulating environment for children to learn. The Perry Preschool
Project involved a half-day of preschool every weekday plus weekly
90-minute home visits over a fairly long time span. Teacher-student
ratios were 1:6, and all teachers had master's degrees and training
in child development. The Carolina Abecedarian, the Early Training
Project, and the Milwaukee Project alter this design in various
ways. For example, the Abecedarian involved full-day intervention
from birth to age 5, followed up with an intervention for school-aged
children. Head Start, of course, is the most famous publicly funded
early intervention program. However, Head Start differs from the
model programs, which typically are better funded and are subject
to more intensive supervision.
Reprinted from the What We Know about Early Childhood Interventions, 2000 Joint Center for Poverty Research, Policy Brief, Vol. 2, No.10