In this paper we report the results of a randomized controlled trial designed to measure the impact of a parental training program on the nutritional status of primary school students in rural Shaanxi Province, in Northwest China. Using hemoglobin (Hb) levels as the outcome variable, we first measure the overall impact of a nutritional training program, then measure the impact separately by gender. We use both descriptive and multivariate analyses.

The results for the descriptive and econometric results were robust and consistent with the literature. Overall, we find no impact on students’ Hb levels when we trained their parents about undernutrition and anemia. In both the descriptive and multivariate results, there was no difference in the change of Hb levels between control and treatment students. Parents in the treatment group did learn more about anemia than parents in the control group, but this increased knowledge did not lead to sharp changes in behavior, in general. We did find, however, that there was a measurable impact of parental training on the Hb levels of female students. In both the descriptive and econometric results we found that the Hb levels of female students rose more than that of male students, and that this difference was statistically significant. We conjecture that the parents of female students may have recognized from the training that they were not providing their daughters with sufficient nutrition. Our data show that parents in the treatment group responded by increasing the daily provision of meat, fish, eggs and beans, relative to parents of girls in the control group.