Return to Course Home Page

TryPy 01 - Exploring St. Louis Blood Toxicity Data

Part 0. Setup Steps

  • Create a repo on GitHub named eds217-trypy-01
  • Clone to create a version-controlled project
  • Create some subfolder infrastructure (nbs, data, figs)
  • Create and save a new quarto in RStudio called jupyter notebook (.ipynb file) named stl-lead-yourinitials.ipynb in the nbs folder (for example, mine would be stl-lead-kc.ipynb).
  • Make sure to associate the notebook with the eds217_2023 environment.

Part 1 - Get the data


"""
Create a new variable containing 
the link to the .csv file on 
the EDS_221 github repository.
"""
url = 'https://raw.githubusercontent.com/'\
'allisonhorst/EDS_221_programming-essentials/'\
'main/activities/stl_blood_lead.csv'

""" 
pandas can read a csv file into a 
dataframe directly from a url:
"""
stl_lead = pd.read_csv(url)

Read more about the data here.

Part 2 - Explore the data

In your .ipynb file:

  • Create a code cell that imports the numpy and pandas packages and run the cell to import the packages.
  • Use the code above to read the url for stl_blood_lead.csv into a pandas DataFrame called stl_lead
  • Do some basic exploration of the dataset using the DataFrame info and describe commands.
  • In a new code chunk, from stl_lead create a new column called prop_white that contains the percent of each census tract identifying as white (variable white in the dataset divided by variable totalPop, times 100).

Hint: df['new_col'] = df['col_a'] / df['col_b'] will create a new column new_col in the dataframe df that contains the value of col_a / col_b

Part 3 - Create a scatterplot

  • Import matplotlib (import matplotlib.pyplot as plt)
  • Create a scatterplot graph of the percentage of children in each census tract with elevated blood lead levels (pctElevated) versus the percent of each census tract identifying as white.

Part 4 - Create a histogram

  • Create a histogram of only the pctElevated column in the data frame
  • Customize the fill, color, and size aesthetics - test some stuff! Feel free to make it awful.