The Long-Term Impact of High-Quality Early Childhood Education

Would you believe me if I told you that having gone to a high-quality kindergarten can have long-term effects on behavioral dimensions such as effort, initiative and participation?

In the 1985-6 school year, children entering kindergarten at 79 Tennessee public schools participated in a curious experiment. Once registered in the school system, each of these children was randomly assigned to one of two class types: large classes (with 20 to 25 children) or small classes (with 13 to 17 children). Over the next four years, as the children progressed to third grade, they continued in the same type of class they’d been assigned to previously. During this same period, the teachers who taught these children were randomly assigned to their respective classes. This experiment has been widely cited in the educational literature to document the positive impact of small class size and high-quality teachers on student achievement, as measured by standardized tests. This experiment is known as Project STAR (Student/Teacher Achievement Ratio).

Using the original experimental design, research published in 2011 in the Quarterly Journal of Economics explores the long-term effects of the early years of education. The researchers successfully managed to merge records from about 95% of the more than 11,000 children who participated in STAR with their tax records from the years 2005-7, that is, when the original subjects were between 25 and 27 years old. With this information and a rigorous methodology, they managed to rather convincingly document new evidence on the importance of high-quality learning experiences during early childhood.

These are the main findings of the study:

• Children who received a higher quality education in early childhood (measured by being in a class with fewer children) are more likely to be enrolled in college at age 20 than those who had a lower quality education (the difference between the two groups is 1.8 percentage points, with the average college enrollment at age 20 for the sample at 26.4%). But in addition to college enrollment, the children from small classes also fare better in other areas: homeownership, savings and social mobility, to name a few.
• Other observable characteristics of high quality education also appear to have significant long-term impacts. For example, those children whose kindergarten teacher had at least 10 years of experience earn 6.9% more at age 27 than those children whose kindergarten teachers were less experienced.
• By measuring the average performance of the classmates of each of the children in the STAR sample at the end of kindergarten, the authors construct an aggregate measure of the quality of the educational experience that combines teacher effects, group effects and other characteristics of the class that influence the children’s performance. Those children who were randomly assigned to a class one standard deviation better in this measure of quality earn 3% more at age 27. It is also more likely that these children will go to college and that they will attend better universities.
• The effects of assignment to a high quality kindergarten class on a child’s performance on standardized tests quickly fade in time. By the time children from the STAR project were in eighth grade, no significant differences were detected in the test scores of those who were assigned to higher quality classes versus those who weren’t. However, despite the fact that differences in academic performance do not persist over time, as explained above, long-term impacts on labor market performance in adulthood are detected in those who have had an experience of high quality early childhood education. This leads us to explore the possibility that a quality kindergarten experience has helped these children to develop better non-cognitive skills that later prove useful for achieving success in work and life. As mentioned at the beginning of this post, the empirical evidence from this study seems to be consistent with this hypothesis. Available data suggests that STAR children exhibit better non-cognitive skills (effort, initiative and participation) several years later. However, the authors recognize that this is an issue that requires further and better analysis.

This is an important time in Latin America and the Caribbean in terms of the degree of public policy interest in issues related to early childhood. It’s wonderful to see the efforts undertaken by countries in the region to expand the coverage of child care services for this segment of the population. The long-term impact of the STAR Project reminds us that increasing access to child development services is a necessary yet insufficient condition for investment in order for human capital to yield the expected results. It is essential to guarantee the quality of these services, but this is no easy task.