The externalities of preschool attendees in middle school classes

Using a nationally representative survey dataset on middle school students in China, this paper examines whether and how exposure to more classmates with a preschool experience affects students' cognitive and non-cognitive outcomes. We make use of the random class assignment rule, required by China's education law, to address the potential selection bias from class sorting. Our results show that exposure to more classmates with a longer preschool duration significantly improves students' cognitive and non-cognitive outcomes, whereas exposure to more classmates with a shorter preschool duration does not have such significantly positive effects, offering evidence of the dosage effects of attending a preschool program from the perspective of externalities. We further demonstrate three possible mechanisms through which the effects operate and quantify the explanatory power of each mechanism in driving the effects.