Reuters

#30DayChartChallenge Day 12

Reuters

I am back from my break and I’m looking forward to squeeze in another #30DaysChartChallenge. Today’s theme is Reuters, so I browsed excellent graphics section and found this article which featured this chart.

I was excited to attempt to mimic this chart because I was very keen on learning how to work with geographical data and visualize maps. Lets-Plot proved to be an excellent library for this type of visualisation.

Reuters Original

My attempt

import pandas as pd
import geopandas as gpd
from lets_plot import *
from lets_plot.geo_data import *

LetsPlot.setup_html()

states = geocode_states().countries('US-48').inc_res().get_boundaries()
reuters = pd.read_csv('./reuters_abortion_access.csv')

df = states.merge(reuters, on='state')

# Adding abortion access data based from https://www.reuters.com/graphics/USA-ABORTION/DISTANCE/jnvwxorwkpw/



p = (ggplot()
    + geom_map(aes(fill='access'), data=df, color='white')
    + scale_fill_manual(
        values = ['#e3dfdf','#c96448','#321f6a','#b695bf'],
        labels = ['Protective','Some restrictions/protections','Most restrictive','Restrictive or very restrictive']
    )
    + ggsize(width=700,height=600)
    + labs(
        title = 'Abortion access when Florida, Arizona limits take effect',
        subtitle = 'U.S. states and the District of Columbia by the restrictiveness of their abortion policies',
        caption = '#30DayChartChallenge #Day12 Reuters\nData: Reuters Graphics\nMade by: www.ddanieltan.com'
    ) 
    + theme_minimal()
    + theme(
        legend_position='top',
        legend_text = element_text(size=12),
        legend_title = element_blank(),
        plot_caption=element_text(size=12, color='grey'),
        plot_title=element_text(size=20, face='bold'),
        axis=element_blank(),
        panel_grid = element_blank(),
    )
)

p

The geodata is provided by © OpenStreetMap contributors and is made available here under the Open Database License (ODbL).

ggsave(p, 'day12.png')

I am really happy with how my attempt came out, even though it’s not a perfect replica. Going through this process, I am more aware of subtleties such as the choice of projection, the availability of text labels and usefulness of annotations that go into making a good map. Looking forward for more practise in the future!

TIL

1. One of the benefits to using Lets-Plot as a plotting library is their great support for maps! You can pull the geopandas files directly from their library for the country, states and even county boundaries for most countries.

2. When joining pandas and geopandas, using df.merge() is preferred. https://geopandas.org/en/stable/docs/user_guide/data_structures.html