In this homework, we’ll adapt data from Chicago-IL communities. The original source is the GeoDa Data and Lab. The first dataset Airbnb.RDS refers to Airbnb rentals, socioeconomic indicators, and crime by community area in Chicago. The Communities.xls dataset contains health and socioeconomic indicators for the 77 community areas of Chicago, 2012-2014.
Import both datasets and use the function head() to see only the first five rows from each one. This might help you to import the Communities data.
Merge the datasets using the function full_join(). What is the dimension of your new dataset? What variables do they have in common (variables with the same column name)?
Important: You can find more information about the variables in the Airbnb and Communities datasets here and here. I added two more variables to the Airbnb data: area and dist. They represent the total community area and the distance (in km) from the community to Chicago downtown, respectively.
You want to work only with the following columns: community, price_pp, num_spots, rev_rating, PerCInc14 , num_theft, FirearmM ,unemployed, harship_in, Pop2014, BirthRate, Over65, dist, and area. Select only those variables and store them in a new data frame. [5 points]
Create the new variable theft_rate dividing the total number of thefts by the population in 2014. [5 points]
First, divide the total population in 2014 by the community’s area to get values for population density (number of people per square mile). Then, create the new variable logdens, taking the natural logarithm of population density. [5 points]
Filter your new dataset to identify Central Chicago. In other words, you want to filter communities within \(3 km\) from Chicago downtown.
What is the average number of Airbnb spots in Central Chicago? What are the average Airbnb prices, per capita income, theft rate, firearm-related deaths, population density, and birth rate in Central Chicago? [15 points]
Compare those numbers you just got from Central Chicago (Airbnb spots and prices, income, crime, and birth rate) with the values from Central Chicago adjacent communities - communities that share a border with Central Chicago. Discuss the results. [15 points]
Finally, compare the values for the same variables in Central Chicago with the average numbers from “Far from downtown” - the communities that have a distance from downtown higher than 19 km. [15 points]
Do you have any explanation for the presence of birth rate in d., e., and f. (i.e did you expect to find a relationship between birth rate and distance from downtown)? [5 points]
One of the regularities within cities predicted by the Alonso-Muth-Mills model is the decline of population density with distance to the Central Business District. Build a scatterplot to show that this is happening in Chicago, i.e., there is a negative relationship between logdens (y-axis) and dist (x-axis) in our dataset. [10 points]
Now, let’s take a look at the relationship between Airbnb prices and the distance to Downtown Chicago. Construct a scatter plot between those variables (y-axis prices and x-axis distance). What do you see? What is the reasoning behind this relationship? [10 points]