Learn about the Food Access Research Atlas dataset by checking out this interactive overview. For one perspective about how food insecurity relates to "food deserts," read this blog post.
Download the the dataset. The full dataset has 72,531 rows — one for each of the Census tracts (neighborhoods) in the U.S. This dataset is modified from original Food Access Research Atlas to include only variables related to the demographics of each tract, those related to the income of people living in that tract, and those related to the people living in each tract who have low access to food (defined as living more than 1 mile from a supermarket for urban areas or more than 10 miles for rural areas). The file also has 51 additional tabs — one for each state and the District of Columbia. Since these state datasets are smaller, they might be easier for your statistical software to handle.
Review the "Variable Lookup" tab. This is the data dictionary and contains the the field (variable name in the data set), long name (an expansion of the variable name in words), and description. There are 83 variables in this dataset. Some of them are easy to understand from their name; for example, State is simply the state from which the data was collected***.*** The long description of the variables will help you understand what the data represents***.*** For example, lapop1share is the share of that urban tract population that are beyond 1 mile from supermarket, and TractSNAP is the total count of housing units in the tract that get SNAP benefits.
Work with your team to decide on your focus for the project. It may help to review the judging rubric. The minimum requirement for the challenge is to analyze the data for one state. But you may choose to analyze data from multiple states or to analyze the complete dataset. For some inspiration you may want to read Roger Peng's blog post on Constructing a Data Analysis.
Get to know your data with some exploratory data analyses. If you need help getting started, you might watch this video on how to use the Common Online Data Analysis Platform (CODAP) to do data exploration. R for Data Science by Hadley Wickham and Garrett Grolemund is another great resource. Even if you will not be using R to perform your analysis, the chapter on Exploratory Data Analysis will help you think about what you want to accomplish.
Have fun! (And don't forget to check in with your team's adviser if you have any questions. Or email us at [email protected] with questions about the contest.)