How To Draw A Scatter Plot In R

This tutorial will explain how to create a scatter plot in R with ggplot2.

It volition explain the syntax for a ggplot scatterplot, and will likewise bear witness you footstep-by-step examples.

If you need something specific, you can click on whatever of the following links …

Tabular array of Contents:

• A Quick Review of Scatterplots
• Syntax
• Examples

But information technology's probably better if you read the whole tutorial. Everything will make more sense that fashion.

A Quick Review of Scatterplots

Let's quickly review what a scatterplot is.

Scatterplots visualize numeric data. Specifically, a scatterplot show the relationship between two numeric variables, where the values of i variable are plotted on the x-axis and the values of the other variable are plotted on the y-axis.

Scatterplots are extremely useful tools for showing the relationship between two numeric variables. For data visualization, reporting, and analytics, you'll use them over and over.

Scatter Plots in R

If you demand to create a scatter plot in R, you take at least two major options, which I'll talk over briefly.

• base R
• ggplot

I strongly adopt the ggplot2 scatterplot, simply allow me chop-chop talk about both.

base R scatterplots

Y'all can create a scatterplot in R using the plot() function.

I'k going to be honest: I strongly dislike the base of operations R scatterplot, and I strongly discourage you from using the plot() function.

Like many tools from base R, the plot() part is difficult to use and hard to modify beyond making uncomplicated modifications. The syntax is clumsy, difficult to remember, and often inflexible.

I haven't used the plot() part to create a scatterplot in R in almost a decade. There'south a amend manner …

ggplot2 scatterplots

If I need to make a scatter plot in R, I ever utilize ggplot2.

If yous're an R user, you lot've probably heard of ggplot2. The ggplot2 package is a toolkit for doing data visualization in R, and it's probably the best toolkit for making charts and graphs in R. In fact, one time you know how to use it, ggplot2 is arguably one of the best data visualization toolkits on the market place, for any programming language.

ggplot2 is powerful, flexible, and the syntax is extremely intuitive, once you know how the system works.

If y'all need to brand a scatterplot in R, I strongly recommend that you use ggplot2.

Having said all of that, let's accept a expect at the syntax for a ggplot scatterplot.

The syntax for a ggplot scatterplot

The clandestine to using ggplot2 properly is agreement how the syntax works.

If y'all're non familiar with how the ggplot2 organization works, y'all might want to read our introduction to ggplot2 tutorial. That tutorial explains most of the basics of the ggplot system.

At a high level, the syntax for a ggplot2 scatterplot looks something like this:

There are a few critical pieces to this syntax that you lot need to know:

• The ggplot() function
• The information = parameter
• The aes() function
• Geometric objects (AKA, "geoms")

Let's take a expect at each of those separately.

The ggplot part

The ggplot() function is simply the role that we use to initiate a ggplot2 plot.

You'll use this every time that you lot want to brand whatever type of data visualization with ggplot2. Nevertheless, the other parameters and functions you utilize along with it volition dictate exactly what visualization gets created.

The data parameter

The data parameter tells ggplot2 the proper noun of the dataframe that you desire to visualize. When yous use ggplot2, you need to use variables that are contained inside a dataframe. The data parameter tells ggplot where to observe those variables.

So for example, if your dataframe is named my_dataframe, you will ready data = my_dataframe.

Recall: ggplot2 operates on dataframes.

The aes function

The aes() function tells ggplot() the "variable mappings." This might audio complex, but it's really straightforward one time y'all understand.

When nosotros visualize data, we are essentially connecting variables in a dataframe to parts of the plot. For example, when we make a scatter plot, we "map" one numeric variable to the 10 axis, and another numeric variable to the y axis. We map these variables to dissimilar axes within the visualization.

The aes() office allows us to specify those mappings; it enables us to specify which variables in a dataframe should connect to which parts of the visualization. If this doesn't make sense, just sit tight. I'll prove y'all an example in a minute.

(For more than detailed explanation of the aes() part, read the department almost the aes() function in our ggplot2 tutorial.)

The point "geom"

Finally, a geometric object is the thing that we depict.

When yous create a bar chart, you draw "bar geoms." When you create a line nautical chart, yous describe "line geoms." And when you create a scatter plot, you are describe "point geoms."

The geom is the thing that you describe.

In ggplot2, nosotros need to explicitly land the type of geometric object that we want to draw (i.e., confined, lines, points, etc).

When create a besprinkle plot, we draw indicate geoms (i.e., points). To specify that we desire to draw points, we call geom_point().

Additional parameters

At that place are likewise a few boosted parameters that you tin can use to control the appearance of the points in your scatterplot.

Specifically, the most of import parameters yous should know are:

• colour
• size
• alpha

Let me apace discuss each of these.

Colour

The color parameter controls the color of the points.

When you provide an statement to this parameter, you tin provide a "named" color similar ruddy, green, bluish, etc. R has a variety of named colors, then explore them and observe some you like.

Keep in mind, that when you provide the colour name, it needs to exist enclosed inside of quotation marks. Then for instance, you'll set colour = 'ruby-red'.

Size

The size parameter enables you to specify the size of the points.

If yous want to play with this parameter, at that place'southward non a perfect mode to choose a good size, then I recommend that y'all use some trial and fault to find one that works.

Yous can too use this parameter to create a bubble nautical chart, merely that's slightly more complicated, so we won't comprehend it here.

Blastoff

The blastoff parameter enables yous to alter the opacity of the points (i.e., how transparent the points are).

This value needs to be between 0 and 1, where:

• 1 is fully opaque
• 0 is fully transparent

By default, this parameter is set to alpha = one.

This parameter is very useful when you have a large number of points, and your scatterplot has an consequence with overplotting. Dealing with overplotting is somewhat of a nuanced consequence, but one way to handle it is past decreasing the alpha value.

I'll show you an case of this in the examples.

Examples: How to make scatterplots with ggplot2

Ok. Now that I've quickly reviewed how the syntax works for a ggplot2 scatterplot, allow's accept a look at some examples of how to create a scatter plots in R with ggplot.

Examples:

• Create a simple scatterplot with ggplot2
• Change the Color of the Points
• Change the Size of the Points
• Add a LOESS Smooth Line
• Add a Linear Regression Line

Run this lawmaking first!

A few quick things before you run the examples.

You'll need to run some code to load ggplot2 and also to create the dataset that we'll exist working with.

Load the tidyverse package

First, you need to make sure that you've loaded the ggplot2 package.

Really, I recommend that yous load the tidyverse package. Remember that the tidyverse package includes ggplot2.

Keep in mind that this also assumes that you've installed the tidyverse package on in RStudio.

library(tidyverse)        

Create a sample dataset

Next, nosotros'll need to create a dataset to plot.

Here, nosotros're going to create a new dataframe called, scatter_data.

set.seed(55) scatter_data <- tibble(x_var = runif(100, min = 0, max = 25)              ,y_var = log2(x_var) + rnorm(100)              )        

We can take a expect at this dataframe with the following code:

scatter_data %>% glimpse()        

OUT:

Rows: 100 Columns: 2 $ x_var                      xiii.6953379, 5.4539920, 0.8740999, xix.7887324, xiv.0060519, one.8556294, three.2… $ y_var                          ii.6122496, two.7738665, -1.2230670, iii.6239948, 3.6479324, i.1145059, 2.244…                              

As you can see, this dataframe has two variables, x_var and y_var. Nosotros'll be able to plot these variables as a scatterplot.

Case 1: Create a simple scatterplot with ggplot2

Now that we have our dataframe, scatter_data, we'll plot it with ggplot2.

Allow's run the lawmaking first, and and then I'll explain.

ggplot(data = scatter_data, aes(ten = x_var, y = y_var)) +   geom_point()        

OUT:

Explanation

Every bit you can see, this lawmaking has created a simple scatter plot. It's pretty straightforward, but let me explain it.

We're initiating the ggplot2 plotting organization by calling the ggplot() function.

Inside of the ggplot2() function, we're telling ggplot that we'll be plotting data in the scatter_data dataframe. Nosotros do this with the syntax information = scatter_data.

Adjacent, inside the ggplot2() function, we're calling the aes() function. Think, the aes() function enables united states of america to specify the "variable mappings." Here, we're telling ggplot2 to put our variable x_var on the x-axis, and put y_var on the y-axis. Syntactically, nosotros're doing that with the lawmaking ten = x_var, which maps x_var to the 10-axis, and y = y_var, which maps y_var to the y-axis.

Finally, on the second line, we're using geom_point() to tell ggplot that we want to draw point geoms (i.e., points).

That's it. That'southward all there is to it. The syntax might look a little arcane to beginners, but one time you empathize how it works, information technology'south pretty easy.

Having said that, there are still a few enhancements nosotros could make to improve the chart. Allow'south talk about a few of those.

EXAMPLE 2: Alter the Colour of the Points

Now, we'll make a simple modification by changing the colour of the scatterplot points.

To modify the colour of the points to a solid color, we need to use the color parameter.

ggplot(data = scatter_data, aes(x = x_var, y = y_var)) +   geom_point(color = 'red')        

OUT:

Explanation

Again, this is very straightforward.

To create this, we only set color = 'red' inside of geom_point(). We practice this within of geom_point() because we're irresolute the color of the points. (At that place are more complex examples were we have multiple geoms, and we need to be able to specify how to alter one geom layer at a fourth dimension.)

Case three: Change the Size of the Points

In this instance, we'll change the size of the points.

We can exercise that with the size parameter.

ggplot(information = scatter_data, aes(10 = x_var, y = y_var)) +   geom_point(colour = 'cherry-red', size = four)        

OUT:

Explanation

Hither, we've increased the size of the points by setting size = four inside of geom_point().

At present, to be clear: I'm not sure that I like this scatterplot with larger points. I actually think that the defaults were but fine.

Having said that, sometimes, yous need to increase or decrease the size of your scatterplot points, and so I wanted to bear witness y'all how information technology's done.

As a side note, decreasing the size of your points tin be a great way to bargain with overplotting. Try it with the diamonds dataframe from ggplot2.

EXAMPLE iv: Add a Smooth Trend Line

Now, we'll add together a smoothen tendency line.

To add a smooth line, nosotros can use the statistical operation stat_smooth().

ggplot(information = scatter_data, aes(x = x_var, y = y_var)) +   geom_point(color = 'red') +   stat_smooth()        

OUT:

Explanation

Hither, we added a smooth line by adding the lawmaking stat_smooth() after the scatterplot code.

Observe that the outset ii lines are exactly the aforementioned as the code for our simple scatterplot (with red points).

And then to add the smooth line, we but use the '+' and so stat_smooth().

This is one of the reasons that ggplot2 is and then great. Frequently, modifications to a elementary plot only require you to tack on a telephone call to an boosted office. So you can build the base of operations version of a plot, and so enhance it by adding new lines of code.

Keep in mind that the default trend line is a LOESS smooth line, which ways that information technology will capture not-linear relationships.

But, you can besides add together a linear trend line. Allow's do that side by side.

Example five: Add a Linear Trend Line

To add a linear trend line, y'all can use stat_smooth() and specify the exact method for creating a trend line using the method parameter.

Specifically, you'll use the code method = 'lm' equally follows:

ggplot(information = scatter_data, aes(ten = x_var, y = y_var)) +   geom_point(color = 'ruby') +   stat_smooth(method = 'lm')        

Caption

The lawmaking for this example is essentially the same as the lawmaking for case iv.

The only difference is that we've added the code method = 'lm' inside of stat_smooth(). This causes stat_smooth() to add a linear regression line to the scatterplot, instead of a LOESS polish line.

Leave your other questions in the comments beneath

Do you lot have more than questions well-nigh how to create a scatterplot in R with ggplot2?

Is there something you need to practice that I didn't cover here?

If and so, go out your question in the comments department virtually the lesser of the page.

Sign Upward to Learn More Data Scientific discipline in R

This tutorial should give y'all a good overview of how to create a scatter plot in R, but if you really want to master data visualization in R, there's a lot more than to learn.

And there's even more than if yous demand to acquire data manipulation and machine learning.

The skilful news is that here at Sharp Sight, we publish free data scientific discipline tutorials every week.

If you lot sign up for our free newsletter, y'all'll get our free data science tutorials delivered right to your inbox.

When you sign up, you'll get complimentary tutorials on:

• ggplot2
• dplyr
• data wrangling
• car learning
• … and more.

Nosotros have tutorials about data science in Python too.

So if you're serious nigh learning information science, just sign up for our free newsletter.

Check your email inbox to confirm your subscription ...

Source: https://www.sharpsightlabs.com/blog/scatter-plot-in-r-ggplot2/

Posted by: greentheopect.blogspot.com

Share this post