A data science project
about American diet over the years
Average American Diet in Calories
We followed USDA’s categorization of food groups and visualized each year’s consumption patterns below. Interact with the graph below to see whether Americans are eating healthier by USDA’s standards.
Average American Diet by Food Group
Explore in detail the changing composition of each food group of the American diet.
Advisors: Prof. Jennifer Mankoff, Nikola Banovic
Team: Chong Shen, Junyu Huang, Qiao Yin
My role: Collected Data, Designed and developed data visualization using D3.js
According to the National Center for Health Statistics, the obesity rate among American adults has more than doubled since 1970. The video below described in more detail how American diets are changing and affecting the overall health of the population. Hence we decided to look into what Americans are consuming each year and whether they are eating healthy or not.
Data for food consumption is retrieved from the United States Department of Agriculture Economic Research Service’s Food Availability (Per Capita) Data System. In particular, we used the ERS Loss-Adjusted Food Availability Data Series, which is derived from food availability data by adjusting for food spoilage, plate waste, and other losses to more closely approximate actual intake. We define healthy through the standards of USDA Dietary Guidelines for Americans, which have been published every 5 years since 1990.
The USDA Loss-adjusted availability data set has complete data for all food groups except for of grains for 2011 – 2014. We decided to shorten the time span of our study to 1970 – 2010, since we couldn’t find a replacement for this part of data.
The dataset is coherent as we verified that the subgroup consumptions of each food group add up to the total consumption of each food group. The data is correct as we cross-examined the trends of our data with publications on food trends from USDA’s Economic Research Service. The dataset is public on USDA’s website and anyone with internet access can use it.
For the section of Average American Diet by Food Group, we originally had one graph per food group, where we combined the recommended amount, the total consumption and the consumptions of some outstanding subgroups of the food group. However, our users reported that there were too many things to keep track of on the chart and was hard to digest visually. Hence we split up each food group’s visualization into two parts: one that only describes how the total consumption and the recommended amount have changed over time, the other that shows the trends in the outstanding subgroups.
In addition, some users reflected that they didn’t understand how the main question was answered on the website. Therefore, we introduced labels that evaluate each food group on whether they show improvement in healthy diets or not. This helps us link the visualization and analyses more closely to our question and clarify our narrative.
Lastly, our users displayed confusion when seeing food consumption being compared to “The Guidelines” because they tended to overlook the text section and jumped to the graphs. To emphasize the presence and importance of USDA’s Guidelines, we added pictures of the guidelines and explanation of how the guidelines were incorporated into the graphs in visualization. We specifically pointed out that the shaded area is the range of a recommended amount of daily consumption for each person.
Americans have been eating healthier from 1970 to 2010 by consuming more fruit and vegetables, but they have also been eating more added fat and sugar.
For Americans to meet these USDA’s healthy diet standard, they would need to substantially lower their intake of added fats, refined grains, and added sugars and sweeteners and increase their consumption of fruits, vegetables, whole grains, and lower fat milk and milk products.