http://www.acf.hhs.gov/sites/default/files/opre/grade5.pdf
page 47
Conclusions:
Consistent with the impact analyses presented in Chapter III, the nonexperimental analyses
show that Early Head Start enrollment was not significantly associated with key child and family
outcomes when children were in fifth grade. The analyses revealed a few significant associations
within racial/ethnic groups, however, with several favorable associations among white and African American children but unfavorable associations among Hispanic children. Among white and African American children, these associations reflected those found in the impact analyses. Among Hispanic children, a negative association between Early Head Start and ECLS-K Mathematics scores did not reach statistical significance in the impact models, although the direction was the same. Participation in a formal early care and education program at ages 3 and 4 was not significantly associated with cognitive outcomes in fifth grade, but it was associated with more externalizing behaviors, internalizing behaviors, and attention problems. Similar associations were found when
children were entering kindergarten; however, the associations were small and did not rise to the
level of clinical importance
(
Kisker
et al.,
unpublished manuscript
).
Among African American
children, formal program participation at ages 3 and 4 was associated with more family involveme
nt
in school when children were in fifth grade.
Attending a higher
-
poverty school in fifth grade was associated with poorer academic outcomes
and a lower cumulative success index. These associations were found within all of the racial/ethnic
subgroups. Attending a school with a higher concentration of minority students was associated with
lower PPVT scores and more externalizing behavior problems. Generally, minority students fared
better when in schools with a moderate concentration of minorities compared to those in schools
with high or low concentrations of
same
-
minority
students. For example, Hispanic students
attending mixed
-
ethnicity
schools performed better on language and literacy outcomes (PPVT and
ECLS
-
K Reading) than those attending schools with a low or high concentration of Hispanics.
African American students attending schools with a
low
concentration of African Americans tended
to have
more
externalizing and attention problems compared to those attending mixed-
race schools.
For the most part,
the individual early care and school
experiences
of children
prior to fifth
grade
were not strongly associated with children’s academic and social
-
emotional outcomes
for the
sample overall
relative to elementary school characteristics
at fifth grade
. Scho
ol characteristics at
fifth grade were individually associated with both social emotional and academic outcomes.
However
cumulatively, having
a higher number of potentially supportive program and education
al
Early Head Start Children in Grade 5
experiences (Early Head Start, formal early care and education at ages 3 and 4, and attendance at a lower-poverty school) was associated with better academic outcomes in fifth grade.
