Misrepresentations of COVID19 Data (and how to fix them)

# import relevant packages
import pandas as pd
import matplotlib.pyplot as plt

# init the numbers by hardcoding, (it is not publicly available so...)
No_mask = [9.75, 9.2, 9.4, 9.75, 9.9, 9.5, 9.5, 9, 8.5, 8.5, 8.75, 8.5, 9.9, 9.9, 10.1, 9.5, 9.55, 9.6, 10, 8.8, 9, 8.8, 9.2]
mask = [25.25, 19.75, 19.7, 20.5, 19.75,19.75, 20.5, 19.9,20.6,21.5,19.75,19.75, 20.5, 19,19.6,17,16.2, 16.4, 16.5, 16,16.1,15.75,15.9]

# Generate the same dates using pandas
date = pd.date_range(start="2020-07-12",end="2020-08-03")

# Initiate a subplot because we know we want two axis
fig, ax1 = plt.subplots(figsize = (9,5))

# This converter is here to allow for matplotlib to convert the date properly
pd.plotting.register_matplotlib_converters()

# Plotting the Mask Mandates first
color = 'red'
ax1.plot(date, mask, color=color)
ax1.set_xlabel('date')
ax1.set_ylabel('Mask', color=color)
ax1.tick_params(axis='y', labelcolor=color) # This colours the axis too!

ax2 = ax1.twinx()  # Create a new Axes object which uses the same x-axis as ax1

color = 'blue'
ax2.plot(date, No_mask, color=color)
ax2.set_ylabel('No Mask', color=color)  # x-label is already defined above so we just do y-label here
ax2.tick_params(axis='y', labelcolor=color) # This colours the axis too!


ax1.set_ylim(15,25.5) # setting individual ylims
ax2.set_ylim(4,14.5) # for both axis so we get the "misrepresentation"

plt.suptitle('Kansas COVID-19 7-Day Rolling Average of Daily Cases/Per 100K Population', fontsize = 18)
plt.title('\nMask Counties Vs. No-Mask Mandate Counties')
fig.tight_layout()  # makes everything nice and uncropped
plt.grid()
plt.show()

import pandas as pd
import matplotlib.pyplot as plt

# init the numbers by hardcoding, (it is not publicly available so...)
No_mask = [9.75, 9.2, 9.4, 9.75, 9.9, 9.5, 9.5, 9, 8.5, 8.5, 8.75, 8.5, 9.9, 9.9, 10.1, 9.5, 9.55, 9.6, 10, 8.8, 9, 8.8, 9.2]
mask = [25.25, 19.75, 19.7, 20.5, 19.75,19.75, 20.5, 19.9,20.6,21.5,19.75,19.75, 20.5, 19,19.6,17,16.2, 16.4, 16.5, 16,16.1,15.75,15.9]

# Generate the same dates using pandas
date = pd.date_range(start="2020-07-12",end="2020-08-03")

# There is actually no need to create an Axes object, but I want to show what changed
# hence I did it.
fig, ax1 = plt.subplots(figsize = (9,5))

# This converter is here to allow for matplotlib to convert the date properly
pd.plotting.register_matplotlib_converters()

##################################################################
# This is start of the part that changed when you refer to above #
##################################################################
# Plotting the numbers using ax1 object
ax1.plot(date, mask, label = 'With mask', color = 'red')
ax1.plot(date, No_mask, label = 'Without mask', color = 'blue')

# Setting the labels
ax1.set_xlabel('date')
ax1.set_ylabel('Number of Cases')
ax1.tick_params(axis='y')
################################################################
# This is end of the part that changed when you refer to above #
################################################################

plt.suptitle('Kansas COVID-19 7-Day Rolling Average of Daily Cases/Per 100K Population', fontsize = 18)
plt.title('\nMask Counties Vs. No-Mask Mandate Counties')
fig.tight_layout()  # makes everything nice and uncropped
plt.grid()
plt.show()

US_link = 'https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_US.csv'
covid19_US_data = pd.read_csv(US_link)
georgia_df = covid19_US_data.loc[covid19_US_data['Province_State'] == 'Georgia']
georgia_df.head()

# Subsetting our dataset further to include the counties we want using .isin
georgia_df_filter = georgia_df[georgia_df['Admin2'].isin(['Hall', 'Cobb', 'Fulton','Gwinnett','DeKalb'])]

df_t = georgia_df_filter.T
df_t.head(6)

#Changing the column name
df_t.columns = [df_t.loc['Admin2']]

df_t_date = df_t.loc['4/26/20':'5/9/20']

# 26 April to 9 May 2020


df_t_date.head()

# You can specify the colour using the color argument, alpha is the transparency of the graph
ax = df_t_date.plot.bar(figsize=(15, 6), color=['#5553B1', '#23878E', '#988739','#95572C', '#236CA9'], alpha = 0.9)

# This sets the background colour of the graph
ax.set_facecolor('#0D2E4F')

# Redo the legend so it looks nicer
ax.legend(["Cobb", "DeKalb", "Fulton", "Gwinnett", "Hall"], title = 'Georgia Counties')
ax.set_ylabel('Confirmed Cases', fontsize = 20)
ax.set_xlabel('Dates', fontsize = 20)
ax.set_title('Top 5 Counties with the Greatest Number of Confirmed COVID-19 Cases', fontsize = 25)
plt.show()