Percentile Rank of a Column in a Pandas DataFrame

The percentile rank shows what percentage of values in a dataset are less than or equal to a given value. In pandas, we can calculate percentile ranks using the rank() method or scipy's percentileofscore() function.

What is Percentile Rank?

If a student scores in the 80th percentile, it means their score is greater than or equal to 80% of all other scores in the dataset.

Using rank() Method

The most common approach is using pandas' rank() method with pct=True parameter ?

import pandas as pd

# Create sample DataFrame
data = {'Name': ['Ram', 'Shyam', 'Shrey', 'Mohan', 'Navya'],
        'Score': [55, 92, 68, 70, 88]}
df = pd.DataFrame(data)

# Calculate percentile rank
df['Percentile_Rank'] = df['Score'].rank(pct=True)

print(df)

    Name  Score  Percentile_Rank
0    Ram     55              0.2
1  Shyam     92              1.0
2  Shrey     68              0.4
3  Mohan     70              0.6
4  Navya     88              0.8

Converting to Percentage

To express percentile ranks as percentages, multiply by 100 ?

import pandas as pd

data = {'Name': ['Ram', 'Shyam', 'Shrey', 'Mohan', 'Navya'],
        'Score': [55, 92, 68, 70, 88]}
df = pd.DataFrame(data)

# Calculate percentile rank as percentage
df['Percentile_Rank'] = (df['Score'].rank(pct=True) * 100).round(1)

print(df)

    Name  Score  Percentile_Rank
0    Ram     55             20.0
1  Shyam     92            100.0
2  Shrey     68             40.0
3  Mohan     70             60.0
4  Navya     88             80.0

Using scipy.stats.percentileofscore

For more control over the calculation method, use scipy's percentileofscore() ?

import pandas as pd
from scipy.stats import percentileofscore

data = {'Name': ['Ram', 'Shyam', 'Shrey', 'Mohan', 'Navya'],
        'Balance': [5500, 9200, 6800, 7000, 8800]}
df = pd.DataFrame(data)

# Calculate percentile rank using scipy
df['Percentile_Rank'] = df['Balance'].apply(
    lambda x: percentileofscore(df['Balance'], x)
)

print(df)

    Name  Balance  Percentile_Rank
0    Ram     5500             20.0
1  Shyam     9200            100.0
2  Shrey     6800             40.0
3  Mohan     7000             60.0
4  Navya     8800             80.0

Comparison of Methods

Conclusion

Use rank(pct=True) for simple percentile rank calculations in pandas. For more advanced options or percentage output, use scipy's percentileofscore() function.