Lab 3: Fast Food and Fifth Grade Fitness

Lab 3: Fast Food Access and Fifth Grader Fitness

Background

In 2008, City Councilperson Jan Perry helped pass an ordinance that restricted the number of NEW fast-food restaurants that could open in South Los Angeles, a region of the city that has a high percentage of people of color and a high rate of obesity, and illness associated with poor levels of fitness.

Some economists who studied the situation in the few years following the "fast food ban" found that the law had no effect on the level of obesity and that the number of fast-food restaurants had not diminished, despite the passage of the ordinance.

There were several problems with the "ban".  The biggest problem was that it had so many loopholes that the number of restaurants continued to increase.  It also didn't address other sources of unhealthy food options, like convenience stores.  It also did nothing to help improve access to healthy food options.

One prominent study that declared the "ban" on fast food restaurants a failure had itself a number of flaws. The big problem was that used very simple measures of access to fast food that probably did not accurately measure the level of access to healthy food options.   It also didn't use a very good data set to measure fitness levels.   

Still, the media all thought the study was great and repeatedly cited the study as proof that banning fast-food in a neighborhood was a stupid idea.

You can read about it in the link below:

https://www.npr.org/sections/thesalt/2015/03/20/393943031/why-los-angeles-fast-food-ban-did-nothing-to-check-obesity

What you are going to do in this lab is to take a more robust dataset and evaluate the effect of fast-food access on the level of fitness in neighborhoods (ZIP codes) across Los Angeles using GIS (Geographic Information Systems).

You are going to create a statistical model in which you use four or five variables to predict the level of neighborhood fitness levels.  

You will use a PROXY variable to determine neighborhood fitness level:  This variable is the long-term average of physical fitness test scores earned by 5th graders at local elementary schools within and/or near each ZIP code.

Then you'll run a model in which you examine the EFFECT of things like income, ethnicity, unemployment rate, educational levels, AND the density of local fast-food restaurants on the percent of students who have, over the years, been judged to be in the "healthy fitness zone" by the tests completed in 5th grade.

Student Learning Objectives:

1. The student uses cloud computing software to analyze real-world data to test a hypothesis regarding neighborhood fitness levels.
2. The student recognizes the variety of retail, demographic, income, and fitness patterns in Los Angeles
3. The student successfully uses the ordinary least squares regression tool in ArcGIS to evaluate the relative impact of several variables upon childhood fitness levels in Los Angeles 
4. The student recognizes the utility of model residuals in identifying outlying observations (ZIP codes) 

Begin by watching the video below, then answer some questions