This study examines whether attending
preschool enhances the cognitive abilities, health and socialization of junior
high school students in China. Using propensity score matching to control for a
rich data set of student, family and school characteristics, I find that
attending preschool enhances cognition among seventh graders but those gains
fade among ninth graders. The greatest benefits from preschool accrue to both
seventh and ninth graders from economically disadvantaged families. Results for
non-cognitive categories are mixed, and no evidence suggests superior health
outcomes. Evidence also shows cognitive benefits among adolescents – especially
those economically disadvantaged – who entered preschool earlier.
JEL classification: I21; I28; J13
junior high school outcomes; propensity score matching; China
The effects of preschool education have
attracted considerable attention from economists, especially in the United
States, because preschools provide childcare and build children’s human
capital, which predicts their future outcomes (Heckman 2006). Numerous
studies examine how early interventions,such as Head Start and Perry Preschool,
affect social and educational outcomes of students from disadvantaged families
(Currie and Thomas 1995; Garces, Thomas, and Currie 2002;
Heckman et al. 2010). These studies generally associate early
intervention with higher test scores, earnings and less likelihood of criminality
in later life (Duncan and Magnuson2013).
Despite these findings and their significance
for students from disadvantaged families, the case for universal preschool
education has not been researched extensively (Elango et al. 2015).
This article offers new evidence from China’s adolescents. Preschool is not
compulsory in China, and there is substantial variation in attendance among
children between ages three and six.
This analysis employs newly collected data
from the China Education Panel Survey (CEPS), an ongoing nationwide survey. The
CEPS contains rich data about junior high school students, families, teachers
and schools. Moreover, it administers internationally standardized cognitive
tests and surveys self-reported health and social behaviours.
I use propensity score matching (PSM) to
estimate the causal effects of attending preschool on later outcomes. PSM
provides a good means to estimate the treatment impact underlying the
assumption of selection on observables, and it is used in recent literature to
evaluate the effects of early childhood education (Goodman and Sianesi 2005;
Hawkinson et al. 2013; Apps, Mendolia, and Walker 2013).
This study relates closely to that by Apps, Mendolia, and Walker (2013),
who investigate the relation between nursery attendance and outcomes among
adolescents. Using matching methods, they find that preschool childcare
outcomes at ages 11, 14, and 16, and it has a
lasting effect on intentions towards further education and economic activity at
ages 19 and 20. They also find positive effects on health behaviours. Despite
its potential importance, little research investigates the ‘dosage effects’ of
attending preschool (Loeb et al. 2007; Domitrovich et al. 2013).
In this analysis, I replace the binary
variable of preschool attendance by a series of age of entrance dummies, using the OLS
method to gauge the causal effects.
My results show that
attending preschool is associated with an increase of 0.163 in standard deviations
of cognitive scores for the seventh graders, but these gains fade among ninth
graders. The greatest benefits of preschool attendance appear among seventh and
ninth graders from economically disadvantaged families. Results for
non-cognitive outcomes are mixed, with some positive effects on communication
skills, but no evidence suggests improvement on self-reported health. Evidence
also shows that the age at which students enter preschool matters: adolescents
who entered preschool at younger ages exhibit cognitive benefits later,
especially if they belong to economically disadvantaged families.
This study advances
the literature in several respects. It is the first to evaluate the effects of attending
preschool using a representative sample of Chinese students. Two previous
studies focus on impoverished Chinese rural areas: Rao et al. (2012) examine
preschool attendance and academic achievement of first graders. Wong et al.
(2013) evaluate how a 1-year voucher/conditonal cash transfers (CCT)
intervention affects preschool attendance and academic readiness in a poor
county in Henan Province. These studies document positive impacts of preschool
attendance on student outcomes, but small sample sizes limit their results.
Also, they examine
only short-term outcomes and offer no confident basis on which to frame
educational policy for China as a whole.
Second, this study
analyses how the age at which children enter preschool influences later
Third, it examines
heterogeneous effects among families of different socio-economic status. The study’s
collective findings have policy implications for education design, especially
during the current debate about universal preschool education in China.
This study proceeds as
follows. Section II introduces China’s childcare and education system. Section
III describes data and defines variables. Section IV delineates identification
strategies. Section V presents baseline estimates of relationships between
attending preschool and child outcomes. Then it examines the heterogeneous
effects of attending preschool on students from differing economic strata and
the effect of the age at which children enter preschool. Section VI concludes
and suggests further research. The Appendix provides a confirmatory robustness