API_ACS_data_request

A Jupyter Notebook to pull the data for American Community Survey

Description of the project

This project allows user to pull the data on the male population grouped by race and age in American cities from American Community Survey (ACS) by the US Census Bureau. Users can input the year and the city of the ACS they desire and get the data accordingly.

1. Create a virtual environment

• (a) Install the virtualenv package

pip install virtualenv

• (b) Create a directory and set up the virtual environment

mkdir Environments
cd ./Environments
virtualenv project_acs 

• (c) Activate the virtual environment

source project_acs/Scripts/activate # For Mac/Linux users, use: source project_acs/bin/activate

You should be able to see the prompt with (project_acs), and you can check that the Python interpreter is in this virtual environment by typing:

which python

Reference for Winodws vs. Unix-based OS: https://stackoverflow.com/questions/8921188/issue-with-virtualenv-cannot-activate

2. Install all the dependencies packages

Packages we need for this project:

• pandas
• requests
• jupyter (if you are using jupyter notebook)

pip install pandas
pip install requests
pip install jupyter

We can check that now we have the packages we need in this virtual environment:

pip list

Now, we can export the packages and versions needed for this project to requirements.txt

pip freeze --local > requirements.txt
cat requirements.txt

To install all the packages (and the corresponding verisons) required for this project in a new virtual environment, move the requirements.txt to the working directory and type the following in shell:

pip install -r requirements.txt

To exit the virtual environment:

deactivate

3. Select the virtual environment as the Kernel on Jupyter Notebook

Install the virtual environment as a kernel onto Jupyter Notebook

python -m ipykernel install --name=project_acs

Then, select "Kernel" > "Change Kernel" > name of your virtual environment

4. Get an API Key

Go to: https://api.census.gov/data/key_signup.html;

Input the name of the university (University of Chicago) and your email address in the required fields; the key will be sent to the email you entered; save the key as a .txt file.

5. Run the Jupyter Notebook

Notebook for using API to Pull the Data from ACS in US Census Bureau

Follow along the markdowns in the notebook acs_data_pull.ipynb to query the data for the male population in Chicago city by age and race by inputting the year desired (the latest is 2019). The notebook also demonstrates how to query the data for other American cities.

Notebook for cross-validating the results

The notebook data_cross_validation_CDC_wonder.ipynb contains the dataset downloaded from the CDS Wonder Database for cross-validating the results from the API data pull, follow along the markdowns to query the dataset for reference point check to ensure that the data results from the API data pull is sensible.