Diet quality and the income gradient: a nonparametric analysis

Utilizing NHANES data, our study seeks to further understand the relationship between income and diet quality by examining how the correlation between these two variables changes nonparametrically across the income spectrum using local polynomial regressions. Employing similar methods, we also attempt to explain the association between income and substitution between healthy and unhealthy food categories, and how this relationship changes with increasing levels of income. Preliminary results suggest that diet quality improvements associated with small income increases are much stronger and more positive for those at the low and high ends of the income distribution than for those at medium income levels. Analysis of the substitution between unhealthy and healthy food categories suggests that the relatively large improvements in diet quality among low-income individuals are likely due to substitution away from low-quality food categories as income increases, while the positive association observed at high income levels are the result of increases in consumption of high-quality foods.