Fix spelling on one column of one of the dataset based on similar column in other dataset

merge_after_spelling_fix.py

import pandas as pd
import difflib

df1 = pd.read_stata('path to first dataset')
df2 = pd.read_stata('path to second dataset')

def fix_spelling(x):
    try:
        return difflib.get_close_matches(x, df2['common_column'])[0]
    except Exception:
        return x  # Or return "NA", depending on your usecase

# Take the column with smaller number of values in the column to merge data faster,
# otherwise take the column with better spelling
df1['common_column'] = df1['common_column'].apply(lambda x: fix_spelling(x))

merged_df = df1.merge(df2, 'outer', on='common_column')