# Pandas Data Correlations

A correlation between two variables is a statistical measure that describes how those variables are related to each other.

It is frequently applied in data analysis to understand how variables are related to one another.

## Find Relationship

Pandas correlations are one of the most powerful features of the Pandas module, and one of its most impressive features is the corr() method.

This method computes the correlation between each column in your data set by utilizing the corr() function.

For example, we will be working with a file known as ‘language_data.csv’ which is a CSV file.

#### Example:

import pandas as pdsmrx_df = pds.read_csv('empWrong_data.csv')print(mrx_df.corr())
Reminder: It is important to note that the corr() method does not consider “not numeric” columns while computing Pandas correlations.

### Example Explanation

Firstly, the Pandas library is imported using the abbreviation ‘pds‘. Then, a CSV file named ‘empWrong_data.csv‘ is read using the read_csv() function of Pandas library and stored in a Pandas DataFrame object named ‘mrx_df‘.

Next, the corr() function of Pandas DataFrame is applied on the ‘mrx_df’ object to calculate the pairwise Pearson correlation coefficients between all pairs of variables in the DataFrame. The correlation coefficients are computed using the default parameter values for the corr() function. The output of this function is a correlation matrix that displays the correlation coefficients between each pair of variables in the DataFrame.

Finally, the correlation matrix is printed using the print() function. The output of the code will be a table of correlation coefficients for all pairs of variables in the ‘mrx_df’ DataFrame.

There is a range of numbers between -1 and 1.

The value 1 indicates a 1 to 1 relationship (a perfect correlation), and for this data set, each time the value in the first column increased, the other value also increased.

• A relationship between 0.9 and 1 is also considered to be favorable one, because if you increase one value, the other will probably increase as well.
• A relationship of -0.9 is just as good as one of 0.9, but if one value increases, the other will likely decrease.
• A relationship of 0.2 means that the one value going up does not necessarily mean the other will follow.

The following table shows how to interpret the correlation coefficients:

Correlation CoefficientStrength of Relationship
-1.0 to -0.7Strong negative
-0.7 to -0.3Moderate negative
-0.3 to -0.1Weak negative
-0.1 to 0.1No correlation
0.1 to 0.3Weak positive
0.3 to 0.7Moderate positive
0.7 to 1.0Strong positive

## Pandas Correlations – Specify One

I believe it is reasonable to conclude that if you are comparing Pandas correlations with any other correlations, you need to have a correlation of at least 0.6 (or -0.6) to consider it a good correlation according to Pandas correlations.

### Perfect Correlation

Each column in Pandas correlations always has a perfect correlation with itself, so it is obvious that SALARY(\$) and SALARY(\$) produce the number 1.000000.

### Good Correlation

In Pandas, a correlation coefficient of 0.9 represents a very good positive relationship between two variables.

In other words, there is a high degree of correlation between the two variables. 0.9 suggests that as one variable increases, the other variable will also increase in a highly predictable manner as a result of the increase in the first variable.

According to Pandas, a correlation coefficient of 0.009 indicates that there is a very weak or negligible relationship between two variables.

In other words, it means that there is very little or no correlation between the two variables that are being analyzed.

The value of 0.009 suggests that there may be a small increase in the other variable as one variable increases, but the relationship may not be statistically significant or meaningful in a practical sense.

+1
0
+1
0
+1
0
+1
0
+1
0
+1
0
+1
0