Scatter Plots

A scatter plot explores the relation between two variables. As they are usually used when data forms natural pairs, scatter plots illustrate the relationship between two values. If X is changed, what happens to Y?

Highlight the dataset to view and click the icon for Scatter Plot View from the Views menu. Alternatively, right click on the dataset, select View > Scatter Plot.

Fields of the following data types can be included:

Integers
Doubles
Strings

Scatter plot view

The data of the defined fields are plotted, showing the selected configurations. Look for relationships between the fields, evident in distinct patterns formed. Sometimes these illustrate an association between two fields, but no causal relationship.

Depending on the number of data points to be represented, data is represented in different ways. Read more about the data point threshold.

A triangle represents an actual data point in a cell.
- Hover over the triangle to see information on the absolute x and y coordinates of the point.
- If a large number of points (i.e. triangles) are plotted close together and in the same cell, they will be grouped and represented as a circle. Hovering over this circle will show information on how many points there are in this grouping and what their average x and y values are.
A circle represents more than one data triangle in the same cell grouped together.
- The greater the circle size, the greater the number of data points represented.
- The circle centre is plotted on the average of all the represented data values, and is therefore not necessarily within the centre of the cell.
- Hover over the circle to see information on the count of how many points are grouped and the average x and y values of all those grouped points.
- Zooming in with the mouse on the circle might show individual points if the zoomed area doesn't have too many points close together.

Data point threshold

There is a point threshold of 10 000 points per plot.

To determine if the values in a cell will be represented as a triangle or as a circle:

Multiply the count of data points in a cell, by the number of cells configured for the x axis, and by the number of cells configured for the y axis.
If this value is greater that the threshold value of 10 000, then a circle will be used to indicate the data values.
If this value is less that the threshold value of 10 000, then a triangle will be used to indicate the data values.
In one plot, it is possible to have data represented by both triangles and circles.

Bad quality data

Bad quality data is not represented in the scatter plot.

Zooming

Click and drag with the left mouse button on the view to zoom in and magnify the axes.
Click the [Zoom out] button to return to the original view.

Configuration

General

Select the fields to be used for the scatter plots, to determine if a relationship exists. Configure the fields and charts, by selecting the relevant information required.

X axis and Y axis

The plot is divided into a number of cells in order to best represent all the possible data. These cells from a grid on the plot, and data values are plotted within the cells.

Select the number of cells to be plotted on the scatter plot, and either use the field min/max values, or customize the required minimum and maximum values.

A default number of cells is plotted on each axis, if no number of cells is defined during configuration.