### Introducing Scatter Plots

Posted on by

Hello readers! We hope you are doing well, and thank you for your continued support of SimplyAnalytics.  We are excited to announce that scatterplots are officially live! Scatter plots are a great way to visualize the relationship between two different data variables, and we know you will enjoy them as much as we do.

Let’s take an in-depth look at this new feature.

What is a scatter plot?

A scatter plot is a graphical representation where the values of two data variables are plotted along the x and y axis. Each dot represents both the x and y values for a single location, such as a ZIP Code or county.

Why use a scatter plot?

Scatter plots enable users to identify correlations between two different variables. Let’s take a look at an example below using SimplyAnalytics where we’ll use the % of Adults (25+) with a college degree and Median Household Income to see if there’s a correlation between the variables for Counties in the USA.

Getting Started

First, click on New View > Create under the Scatter plot option:

The Edit View page displays your data variables and locations in the project.

Here you can choose which data variables to display along which axis. Of course, this can be edited directly on the scatter plot as well, but for now, select Done to generate the scatter plot.

Voila! Your first scatter plot is created. Now, what exactly is this showing?

The top of the view explains what each point represents – in this example, Counties in the USA. The legend towards the right also displays helpful information.

TIP: You can click on any point to display the name and underlying data.

Looking at this scatter plot, there is a strong positive correlation between median household income and the % of adults who have a college degree within CDs in the USA.

The legend has a section heading titled Correlation that contains an “r” value. What does the r-value mean? In short, that’s displaying Pearson’s R – this is a correlation coefficient that’s used in linear regression. The “r” value will always be on a scale from -1 to +1, and you can use these values to understand the relationship between the variables.

A generalization of the scales and how to think of them is:

Positive Direction – The points looks like they are going uphill

1 – perfect correlation

0.75 to 1 – very strong correlation

0.5 to 0.75 – moderate correlation

0.25 – 0.5 –  weak correlation

Less than 0.25 – none/no correlation

Negative Direction – The points looks like they are going downhill

1 – perfect correlation

-0.75 to -1 – very strong correlation

-0.5 to -0.75 – moderate correlation

-0.25 to -0.5 – weak correlation

Less than -0.25 – none/no correlation

The scatter plot above has an r value of 0.697. This means there is a moderate, positive correlation. Does a negative direction/value mean anything bad or wrong? Nope! It just means as the x axis increases, the y axis decreases – nothing negative or incorrect.

Renaming

You can rename your scatter plot by either clicking on the heading at the top of the graphic, or by selecting Edit on the legend and renaming there.

Changing Variables

You are welcome to select either axis in the legend, and change the variable(s) you want to analyze.

Changing Point Size and Color

Feel free to edit these options within the Edit Legend page to change the appearance of your graphic.

Toggle Off/On the Line of Best Fit

Use this button to toggle between whether or not the line of best fit is present. What is the line of best fit? In short, it is a straight line that best represents the data on a scatter plot.