Universal preschool and cognitive skills – the role of school starting age as a moderating factor.

• Previous empirical evidence on the impact of universal preschool on cognitive skills is ambiguous. • Preschool and primary school enrollment age may be correlated, which can lead to a negative impact of earlier preschool enrollment. • We estimate the impact of earlier preschool enrollment on 6th, 8th, and 10th grade mathematics and reading test scores, while controlling for any indirect impact through primary school enrollment age. • Preschool enrollment has a significant positive impact on test scores, especially among students with less advantaged backgrounds, as long as it does not lead to earlier school enrollment. • These results explain some of the variation seen in previous evidence, support the beneficial impact of preschool, and present an important policy consideration for planned preschool expansion. Previous empirical evidence is mixed regarding the impact of universal preschool on cognitive skills. We show that preschool enrollment can impact test scores positively if it does not lead to earlier school enrollment. We examine rich student data and use different enrollment cutoff dates in Hungary to separate the beneficial direct effect of earlier preschool enrollment from a negative indirect effect that may occur through earlier school enrollment. We find significant direct impacts: 6th-grade reading (math) test scores increase by 9.0 (6.3) percent of a standard deviation for children who enroll in preschool a year earlier. This impact persists through 10th grade and is larger among disadvantaged children. The findings support the importance of universal preschool for improving cognitive skills and equity. They highlight a key consideration for policy evaluation and design and help reconcile ambiguities in the previous empirical evidence. [ABSTRACT FROM AUTHOR]