Pandas: Calculate Correlation with corr()

Compute pairwise correlation of columns in a DataFrame using the Pandas corr() method. Analyze relationships between numerical variables using Pearson, Spearman, or Kendall correlation coefficients, and manage missing data thresholds.

Show Table of Contents ↓

Introduction to corr() 🔝

The corr() method computes pairwise correlation of numerical columns, excluding missing values by default. It returns a correlation matrix where values range from -1.0 to 1.0, indicating the strength and direction of linear or monotonic relationships.

Create numerical sample DataFrame 🔝

We start with a multi-column numerical DataFrame representing study hours, attendance, and exam scores.

import pandas as pd

df = pd.DataFrame({
    "HOURS": [2, 4, 6, 8, 10],
    "ATTENDANCE": [50, 60, 75, 90, 95],
    "SCORE": [55, 65, 78, 92, 98]
})
print(df)
assert df.shape == (5, 3)

Expected output

   HOURS  ATTENDANCE  SCORE
0      2          50     55
1      4          60     65
2      6          75     78
3      8          90     92
4     10          95     98

Basic Pearson correlation matrix 🔝

Call corr() on the DataFrame to calculate the standard Pearson correlation coefficient matrix.

corr_matrix = df.corr()
print(corr_matrix)
assert corr_matrix.loc["HOURS", "SCORE"] > 0.99

Expected output

               HOURS  ATTENDANCE     SCORE
HOURS       1.000000    0.985955  0.994273
ATTENDANCE  0.985955    1.000000  0.993427
SCORE       0.994273    0.993427  1.000000

Alternative correlation methods (Spearman & Kendall) 🔝

Use the method parameter to compute rank-based non-parametric correlations such as "spearman" or "kendall".

spearman_corr = df.corr(method="spearman")
print(spearman_corr)
assert spearman_corr.loc["HOURS", "SCORE"] == 1.0

Expected output

            HOURS  ATTENDANCE  SCORE
HOURS         1.0         1.0    1.0
ATTENDANCE    1.0         1.0    1.0
SCORE         1.0         1.0    1.0

Exercise: compute advertising sales correlation 🔝

Given a DataFrame with AD_SPEND = [10, 20, 30, 40] and SALES = [15, 28, 42, 55], compute the correlation matrix and verify that the correlation between AD_SPEND and SALES exceeds 0.99.

Exercise solution 🔝

Solution constructing the DataFrame and applying corr().

ad_df = pd.DataFrame({"AD_SPEND": [10, 20, 30, 40], "SALES": [15, 28, 42, 55]})
answer = ad_df.corr()
print(answer)
assert answer.loc["AD_SPEND", "SALES"] > 0.99

Expected output

          AD_SPEND     SALES
AD_SPEND  1.000000  0.999298
SALES     0.999298  1.000000

Practice in Google Colab 🔝

Interactive Colab Notebook

Practice corr() examples interactively directly in your browser:

View on GitHub

Reference: Pandas corr() documentation.

Subhendu Mohapatra

Author & Instructor at plus2net

I write and maintain practical tutorials on Python, PHP, SQL, JavaScript, HTML, jQuery, and web development at plus2net. The tutorials focus on clear explanations, working examples, and code that readers can test and adapt while learning.

About the author Watch on YouTube

Connect:

Subscribe to our YouTube Channel here