Population Estimate Analysis for Finding Incorporated Places with Population < 800

popest.py

import pandas as pd

# Read fails if encoding is left to default
df = pd.read_csv(
    'sub-est2017_all.csv',
    encoding="ISO-8859-1",
    dtype={
        "SUMLEV": str,
        "STATE": str,
        "COUNTY": str,
        "PLACE": str
    })
 
# Subset data to incorporated places
df = df[df['SUMLEV'] == '162']
 
# Concatenate fields to identify unique places
df['unique_place'] = df['STATE'].str.cat(df[['COUNTY', 'PLACE']])
 
# Lazily perform calculation with unique values meeting critera / overall
df[df['POPESTIMATE2017'] < 800]['unique_place'].nunique() / df['unique_place'].nunique()

# Lazily calculate the population of people living in such places
df[df['POPESTIMATE2017'] < 800]['POPESTIMATE2017'].sum() / df['POPESTIMATE2017'].sum()

Data Source: City and Town Population Totals: 2010-2017