# 3.7 Lollipop Chart

A **lollipop plot** is a hybrid between a [scatter plot ](/crash-visualization/matplotlib/1.5-pie-chart-plot.md)and a [barplot](/crash-visualization/matplotlib/1.3-column-chart-plot.md). It shows the relationship between a numerical variable and another variable, numerical or categorical. It's better to decrease the order and represent **horizontal lines**.

![](/files/-MBYYScx8m_Gu7zZF4Ny)

First, Let's create a dataset. 

```
# Create a dataset
df = pd.DataFrame({'group':list(map(chr, range(70, 90))), 'values':np.random.uniform(size=20)})

ordered_df = df.sort_values(by='values')  # reorder it by 'values'   
my_range=range(1,len(df.index)+1)         # calcaulate data range
```

### 1. Univariate Lollipop

```
plt.style.use('dark_background')  # set the figure style 
plt.figure(figsize = (14,8))      # set the figure size

#create lollipop bar and lollipop head  
plt.hlines(y=my_range, xmin=0, xmax=ordered_df['values'], color='skyblue')
plt.plot(ordered_df['values'], my_range, "o") 

plt.yticks(my_range, ordered_df['group'])   # reset yticks 
plt.title("Lollipop Plot", loc='left')      # set title and title location
plt.xlabel('Score')
plt.ylabel('Group')
```

![Figure: Univariate Lollipop](/files/-MBYRoFv9CHzdEKYiPuY)

### 2. Highlight Lollipop

Moreover,  we .can add some eye-catching features for storytelling. For example,  highlight a group that specifically interests you with a bigger size and different colors.

```
# Create a color if the group is "A"
A_color=np.where(ordered_df ['group']=='A', 'gold', 'skyblue')
A_size=np.where(ordered_df ['group']=='A', 100, 30)

plt.hlines(y=my_range, xmin=0, xmax=ordered_df['values'], color=my_color, linewidth =2)
plt.scatter(ordered_df['values'], my_range, color=A_color, s=A_size)
 
# Add title and axis names
plt.yticks(my_range, ordered_df['group'])
plt.title("Highlight A Lollipop", loc='left')
plt.xlabel('Score')
plt.ylabel('Group')
```

![Figure: Highlight Lollipop](/files/-MBYT7AhcttjEbzcg7NL)

### 3.  Multivariate Lollipop

We can use multi-variables lollipop if you have **more than one observation** for each group. Instead of displaying the values of both groups one beside each other, show them on the same line, and **represent the difference.**

Now we need to create a new dataset with two variables.

```
# Create a dataframe
var1 =np.random.uniform(size=15)
var2=value1+np.random.uniform(size=15)/ 3
df = pd.DataFrame({'group':list(map(chr, range(70, 80))), 
                'value1':var1 , 'value2':var2 })
 
ordered_df = df.sort_values(by='value1')
my_range=range(1,len(df.index)+1)
```

```
# create white color lollipop bar
plt.hlines(y=my_range, xmin=ordered_df['value1'], xmax=ordered_df['value2'], color='w',alpha = 0.6)

# create lollipop head1 in magnet color and set size =80
plt.scatter(ordered_df['value1'], my_range, color='m', s = 80,label='var 1',marker='^')

# create lollipop head2 in cyan color and set size = 100 
plt.scatter(ordered_df['value2'], my_range, color='c',s = 100, label='var 2')

plt.legend() # show legend 
 
# Add title and axis names
plt.yticks(my_range, ordered_df['group'])
plt.title("Multivariable Lollipop", loc='left')
plt.xlabel('Score')
plt.ylabel('Group')
```

![Figure: Multivariate Lollipop](/files/-MBYXuLaqqrDrW14KOkP)


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://vecco-insight.gitbook.io/crash-visualization/matplotlib/3.7-lollipop-plot.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.