To visually evaluate the distribution, you need to build **histograms**. If the distributions are far from **normal**, we should notice it right away. 

Picture time! Let's build distributions for two groups on one graph.

# Import the libraries 
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Read .csv files 
df_control = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/c3b98ad3-420d-403f-908d-6ab8facc3e28/ab_control.csv', delimiter=';')
df_test = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/c3b98ad3-420d-403f-908d-6ab8facc3e28/ab_test.csv', delimiter=';')

# Plotting hists of Impression columns 
sns.histplot(df_control['Impression'], color='#1e2635', label='Group A')
sns.histplot(df_test['Impression'], color='#ff8a00', label='Group B')

# Add the legend to the graph
plt.legend(title='Groups')
plt.xlabel('Impression')
plt.ylabel('Frequency')
plt.title('Distribution of Impressions')

# Show the graph
plt.show()

In this code, we use the `sns.histplot` function from the `seaborn` library. We pass it to the desired column `df_control['Impression']` to compare with `df_test['Impression']`.

Are these distributions **normal**? Hard to tell... 

Let's look at **box plots**:


# Import libraries
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Read .csv files 
df_control = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/c3b98ad3-420d-403f-908d-6ab8facc3e28/ab_control.csv', delimiter=';')
df_test = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/c3b98ad3-420d-403f-908d-6ab8facc3e28/ab_test.csv', delimiter=';')

# Add to the dataframes columns-labels, which mean belonging to either the control or the test group
df_control['group'] = 'Contol group'
df_test['group'] = 'Test group'

# Concat control and test dataframes 
df_combined = pd.concat([df_control, df_test])
sns.boxplot(data=df_combined, x='group', y='Impression', palette=['#1e2635', '#ff8a00'],
            medianprops={'color': 'red'})

# Sign the axes
plt.xlabel('')
plt.ylabel('Impression')
plt.title('Comparison of Impressions')

# Show the results
plt.show()

Even after **boxplots**, it is not clear whether the distributions are **normal**.

In order to display two **boxplots** on the same chart, we combine the data frames using the `pd.contact` function. 

Next, we use the `sns.boxplot` function, passing the combined data frame `df_combined` to it. On the x-axis are the values of the column `'Impression'`, and on the y-axis are the **Сontrol** and **Test group**. With the help of the `matplotlib` library, we sign the plot and axes.

Even after **boxplots**, it is not clear whether the distributions are **normal**. But in normality, we need to be sure. 

How to do it? <u>Statistical tests come to the rescue</u>, which we will discuss in the next chapter.


Welcome to the A/B testing course. In this section, we will tell you what controlled experiments are, where they are used, and why they are used. We will look at a specific example of a controlled experiment and talk about A/A testing. The Python programming language will help us with this. Well, let's get started!

Before conducting a controlled experiment, we need to determine the characteristics of the sample. The type of tests that we will conduct a little later will depend on these characteristics. You will learn to build distribution histograms, box plots, and check the normality of the distribution. Let's get started right now!

The last thing to do before conducting a controlled experiment is to check the sampling variances. Some types of graphs and Levene's test will help us with this. Let's talk about it in more detail in this section!

So we got to A/B testing. Soon you will make your first parametric tests, or as they are also called - T-tests. We'll use all the knowledge we've learned in the previous sections: metric definition, normality testing, variance testing, and plotting. All this will help us get statistically significant results. Well, to work!

It turns out that sometimes the samples do not have a normal distribution. This means that we cannot perform parametric testing. What to do in such a situation? Non-parametric testing. We will talk about him in the last section.

The Art of A/B Testing

Histograms and Box Plots