Since the rise of the obesity epidemic in 1990s, policy makers, news media, scientists, and educators have been puzzling out how to make Americans healthier. One theme that has emerged over and over is status: higher-status Americans seem to be doing just fine, whereas low-status Americans are suffering.

Results from the Sightlines Project¹ suggest similar trends: a significantly higher percentage of survey respondents who graduated from college (29.4%) reported eating the minimum recommended five fruits and vegetables each day compared to those who finished high school but didn’t attend college (19.3%).

The primary hypothesized reason for the discrepancy in healthy eating is about access: Americans who have the resources to buy and cook vegetables tend to own cars, have jobs that allow them time to go grocery shopping, and live in neighborhoods that have easy and safe access to supermarkets — luxuries that are, unfortunately, far from a reality for many Americans.² Previous research shows that the lowest income neighborhoods in America have 30% fewer grocery stores than the highest income neighborhoods,³ and healthy diets are more expensive than less healthy ones. ⁴ The effect of poor diet also shows up in outcome: in America, poverty is directly and strongly linked with obesity and heart disease.⁵

Existing research suggests that, like income and education, higher-status racial groups (particularly Whites) have better access to healthy food than do lower-status racial groups (particularly Blacks), even when controlling for other demographic factors such as income, education, and neighborhood.⁶ When combined with the strong correlation between income, education, and race in America, where Whites tend to be richer and more highly educated than most other groups.⁷ we would expect to see a similar trend in the Sightlines data: non-Whites eating less healthily than Whites.

Surprisingly, the Sightlines data suggest exactly the opposite: despite the strong correlation between education and healthy eating, there is not a correlation between Whiteness and healthy eating. The rate of eating five or more fruits and vegetables each day is comparable for Blacks (23.1%), Hispanics (23.4%), and Whites (24.3%), and significantly higher Asians (29.0%). If healthy eating were primarily driven by privilege markers such as education, we might except the higher healthy eating rate for Asians (the “model minority” who have the higher average income and education than Whites),⁸ but we would expect the rate for Whites to also be higher than the rates for Hispanics and Blacks.

our sample it is possible that the non-White populations surveyed live predominantly in agriculturally-productive parts of the U.S., like California, and therefore have easier access to fresh produce. Further research is necessary to explore these possibilities.

Second: the conclusions we draw from Sightlines data could be limited by sample size and presentation. Our data are not behavioral; they derive from a national survey that asked people to self-report their eating habits. Furthermore, the sample sizes of different demographic groups are not the same, so there are limits to the conclusions we can draw in these data from comparing rates of healthy eating across demographics.

These possibilities raise a broader issue of this type of demographic analysis: ethnicity is not nearly as clean of a variable as is something like education. The concepts of race and ethnicity are categorical variables that confound identity with several continuous dimensions of human variation. Measuring a person’s race is much more complicated than measuring how much money that person makes. It is important to consider that complexity in order to truly understand variation across demographic dimensions.⁹

The Sightlines data are unexpected: contrary to what we would predict based on status, healthy eating is associated with being well-educated, but not with being White. Further research is needed to understand the relationships between healthy eating, status, and race. Such research could be very valuable to applications designed to increase healthy eating among Americans, and to further understand the variation of experience among Americans.

Why might this be? There two primary possibilities:

First: healthy eating might be driven by a hidden third variable. For example, perhaps cultural variation in cuisine explains the unexpected rates of eating fruits and vegetables. Or, perhaps there are regional differences in healthy eating: for example, in our sample it is possible that the non-White populations surveyed live predominantly in agriculturally-productive parts of the U.S., like California, and therefore have easier access to fresh produce. Further research is necessary to explore these possibilities.

Second: the conclusions we draw from Sightlines data could be limited by sample size and presentation. Our data are not behavioral; they derive from a national survey that asked people to self-report their eating habits. Furthermore, the sample sizes of different demographic groups are not the same, so there are limits to the conclusions we can draw in these data from comparing rates of healthy eating across demographics.

These possibilities raise a broader issue of this type of demographic analysis: ethnicity is not nearly as clean of a variable as is something like education. The concepts of race and ethnicity are categorical variables that confound identity with several continuous dimensions of human variation. Measuring a person’s race is much more complicated than measuring how much money that person makes. It is important to consider that complexity in order to truly understand variation across demographic dimensions.⁹

The Sightlines data are unexpected: contrary to what we would predict based on status, healthy eating is associated with being well-educated, but not with being White. Further research is needed to understand the relationships between healthy eating, status, and race. Such research could be very valuable to applications designed to increase healthy eating among Americans, and to further understand the variation of experience among Americans.

1. The Sightlines Project, Stanford Center on Longevity, (February 2016). See http://sightlinesproject.stanford.edu
2. Walker, R. E., Keane, C. R., & Burke, J. G. (2010). Disparities and access to healthy food in the United States: A review of food deserts literature. Health & place, 16(5), 876-884.
3. Weinberg, Z. (1995). No Place to Shop: The Lack of Supermarkets in Low-income Neighborhoods: Analysis of a University of Connecticut Study on Low-income Communitites and Households Receiving Public Assistance in 21 Metropolitan Areas. Public Voice for Food and Health Policy.
4. Drewnowski, A., Darmon, N., & Briend, A. (2004). Replacing fats and sweets with vegetables and fruits-a question of cost. American Journal of Public Health, 94(9), 1555-1559.
5. U.S. Department of Health And Human Services. Obesity and Socioeconomic Status in Children and Adolescents: United States, 2005-2008. National Center for Health Statistics. Dec 2010, available at http://www.cdc.gov/nchs/data/databriefs/db51.pdf.
for Health Statistics. Dec 2010, available at http://www.cdc.gov/nchs/data/databriefs/db51.pdf.
6. Zenk, S. N., Schulz, A. J., Israel, B. A., James, S. A., Bao, S., & Wilson, M. L. (2005). Fruit and vegetable access differs by community racial composition and socioeconomic position in Detroit, Michigan. Ethnicity & disease, 16(1), 275-280.
7. American Community Survey, United States Census, https://www.census.gov/programs-surveys/acs/
8. U.S. Census Bureau 2016 http://www.census.gov/newsroom/press-releases/2016/cb16-56.html
9. Jackson, P. B., & Williams, D. R. (2006). The Intersection of Race, Gender, and SES: Health Paradoxes. In: Schulz, Amy J. (Ed); Mullings, Leith (Ed), Gender, race, class, & health: Intersectional approaches, (2006): 131-162