Checking the coactivation of school and uni closures

school-uni-coactivation.py

import pandas as pd 

# From https://www.notion.so/977d5e5be0434bf996704ec361ad621d?v=fe54f89ca9e04ac799af42b39e1efc4b
path = "COVID 19 Containment measures data.csv"
df = pd.read_csv(path)

withoutUS = df[~df["Country"].str.contains("US:")]
withoutUS = withoutUS[~withoutUS["Country"].str.contains("United States")]
numCountries = withoutUS.Country.unique().shape[0]

#df[df.Country.isnull()].shape[0]

withoutUSorNulls = withoutUS[~withoutUS.Keywords.isnull()]
countries = withoutUSorNulls.groupby("Country")

sameRow = lambda df : df.Keywords.str.contains("school closure") \
                        & df.Keywords.str.contains("university closure")

simultaneousCountries = countries.apply(sameRow)
numCoactivations = simultaneousCountries.sum()

withSchoolRows = withoutUSorNulls[withoutUSorNulls.Keywords.str.contains("school closure")]
numCountriesWithSchoolClosureInOurData = withSchoolRows.Country.unique() \
                                          .shape[0]

print( numCoactivations / numCountries, "of all countries in our sample" )
print( numCoactivations / numCountriesWithSchoolClosureInOurData, "of countries with school closure noted in our sample" )