The Interaction of Community Factors and Individual Characteristics on Child Height in China

Feng Zhao, World Bank Group

This objective of this paper is to investigate how individual characteristics (education, wealth and occupation) modify the effects of community factors on child height. Using data from China, we specified a list of community variables including community infrastructure, access-type and efficiency-type health services, food market accessibility, and the strictness of the family planning policy. We applied both the sample-stratifying approach (running regression models on the stratified sample by the individual characteristics) and the interaction-term approach (adding interaction terms to regression models on the unstratified sample). The result of the sample-stratifying approach suggests that community infrastructure and access-type health services are substitutes while food market accessibility and efficiency-type health services are complements of the individual characteristics. The family planning policy is negatively associated with child height only for the less educated, poor and farmer households. The interaction-term approach only confirmed education being able to significantly modify the slope of the selected community factors.