Skip to content

Final Project for Data 512. Focus on COVID-19 and its impacts on domestic violence in Atlanta, Georgia.

License

Notifications You must be signed in to change notification settings

savageGrant/data-512-a7

Repository files navigation

data-512-a7

Final Project for Data 512. Focuses on the impact of COVID-19 on domestic violence in Atlanta, Georgia.

Author: Grant Savage

Project Structure

data-512-a7
│   README.md
│   LICENSE
│   .gitignore
|   A4 Common Analysis.pdf
|   A5 Extension Plan.pdf
|   A6 Presentation.pptx
|   A7 Notebook.ipynb
|   A7 Report.pdf
|   domestic_crime_data.csv
|
└───images
│   │   Yearly_Reported_DV_Cases.png
|   |   pre_post_vaccine_domestic_violence_cases.png
|   |   monthly_domestic_violence_cases.png
|   |   pre_post_pandemic_weekly_domestic_violence_cases.png

Goal of the project

To explore the impact of COVID-19 on domestic violence in Atlanta, Georgia.

I focus on how the rate of reported domestic violence cases was influenced by the pandemic, and the potential impact the vaccine had on reported cases.

Known Limitations

The APD published a weekly PDF which contains the count of domestic violence cases for the week as well as a year-to-date total and the previous year’s year-to-date total. The year-to-date total does not always align with the sum of the weekly count of domestic violence cases for the year, and the previous year’s year-to-date (YTD) totals can be very different than the year-to-date totals achieved by opening up the pdf for the week of that year. For example, if I compare week 40 of 2020’s pdfs (CS-2020-40.pdf) and week 40 of 2019’s pdfs (CS-2019-40.pdf) I will see that the YTD cases reported for 2019 in CS-2020-40.pdf is 324 while the YTD cases reported for 2019 in CS-2019-40.pdf is 329. To remain consistent, I use the reported case counts from the pdfs that were released at the time the cases were first reported, so I’m using CS-2019-40 as the true count of domestic violence cases in week 40 of 2019 to populate my domestic_crime_data.csv.

Data Sources and Description

Column Description
Week The week number from 1 to 52
weekly_21 The weekly reported domestic violence cases in 2021
weekly_20 The weekly reported domestic violence cases in 2020
weekly_19 The weekly reported domestic violence cases in 2019
weekly_18 The weekly reported domestic violence cases in 2018
ytd_21 The year to date reported domestic violence cases in 2021
ytd_20 The year to date reported domestic violence cases in 2020
ytd_19 The year to date reported domestic violence cases in 2019
ytd_18 The year to date reported domestic violence cases in 2018
  • The dataframe "mcc_cobb" is used to plot COVID-19 cases was created using the John Hopkins University COVID-19 dataset and has the following format:
Column Description
Date Python Timestamp Datetime used as index
confirmed_cases The cumulative total confirmed cases of COVID-19
ma The 7 day moving average of cumulative total confirmed cases of COVID-19
ma_delta The diff of ma. This represents smoothed daily cases
at_risk_pop The population of the county minus the confirmed cases
infection_rate Assumes people are infected for 14 days. This is the 14 day rolling sum of ma_delta divided by the at risk population.
infection rate delta The diff of infection_rate
infected The total number of infected population, assuming a 14 day infection.
  • The dataframe "ts" is used to plot domestic violence cases was created using the dataset I created from the APD reports and has the following format:
Column Description
Date The date the data was captured
Year The year the data was captured
pandemic_status indicator of whether the data was captured prepandemic or during the pandemic
weekly_cases The weekly reported domestic violence cases

License and Terms of Use

About

Final Project for Data 512. Focus on COVID-19 and its impacts on domestic violence in Atlanta, Georgia.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published