Data relationships

Another task that is common in data exploration is understanding the relationship between data features. The display types below can be used to plot two or more variables against one another to observe trends and patterns between them.

Scatter plot

A scatter plot uses dots to represent values for two different numeric variables. The position of each dot on the horizontal and vertical axis indicates values for an individual data point. Scatter plots are used to observe relationships between variables.

A scatter plot's primary use is to observe and show relationships between two numeric variables. The dots in a scatter plot not only report the values of individual data points, but also patterns when the data are taken as a whole.

Identification of correlational relationships is common with scatter plots. In these cases you may want to know, for a particular horizontal value, what a good prediction would be for the vertical value. Relationships between variables can be described in many ways: positive or negative, strong or weak, linear or nonlinear.

A scatter plot can also be useful for identifying other patterns in data. You can divide data points into groups based on how closely sets of points cluster together. Scatter plots can also show if there are any unexpected gaps in the data and if there are any outlier points.

Figure 27: Scatter plot

To create a scatter plot, the curve on which you apply the display needs to be set as an XY curve. Refer to the XY curve conversion and XY curve graph type sections for more information.

A scatter plot is mainly useful when you are dealing with cross results. Please refer to the cross results section for more information.

By default all the points are connected, it is therefore recommend that you disable the display of the line. You can access the options through the graph key. Double-click the graph key to open the window of the curve properties.

Figure 28: Curve properties of scatter plot

The line and the symbols settings allow you to customize the display.

You can uncheck the visibility of the line to disable the display on the graph. For a scatter plot you want to display only the symbols.

Check the visibility of the symbols, you can then pick one color from a palette and choose in between symbol shapes and symbol size. It is recommended to leave the density to 100% in order to see all the points for this kind of chart.

Bubble plot

A bubble plot is an extension of the scatter plot used to look at relationships between three numeric variables. Each dot in a bubble chart corresponds with a single data point, and the variables’ values for each point are indicated by horizontal position, vertical position, and dot size or dot color.

Bubble Charts are typically used to compare and show the relationships between categorized circles, by the use of positioning and proportions. The overall picture of bubble plot can be used to analyze for patterns or correlations like a scatter plot.

Figure 29: Bubble plot

To create a bubble plot, the curve on which you apply the display needs to be set as an XYZ curve or a M1D curve. Refer to the XYZ curve conversion, M1D curves conversion, and XYZ/M1D curves graph type sections for more information.

BA bubble plot is mainly useful when you are dealing with cross results. Please refer to the cross results section for more information.

Once the bubble plot is created, you can access the options through the graph key. Double-click the graph key to open the window of the curve properties.

Figure 30: Curve properties of bubble plot

The bubble settings allow you to customize the display.

The style allows you to select the appearance of the bubble. If you have chosen a bubble plot size type you can pick one color from a palette, else if you have chosen a bubble plot color type you can pick a colorscale to define the variation of the color gradient. The size ratio allows you to tune the maximal dimension of the bubbles that are drawn on the graph.

Note

For the bubble plot style, you can change between size style and color style through the display option in the curve properties window.

Connected scatter plot

When a scatter plot is used to look at correlational relationship between two variables, it is common to add lines to the plot showing a trend. This is the case for plotting a component's characteristics for example.

In that situation the size of your data needs to be small and the sequential order of the X data item is usually monotonic.

To create a connected scatter plot, the curve on which you apply the display needs to be set as an XY curve. Refer to the XY curve conversion and XY curve graph type sections for more information.

Figure 31: Connected scatter plot

By default all the points are connected, it is also recommend to enable the display of the symbol. You can access the options through the graph key. Double-click the graph key to open the window of the curve properties.

Figure 32: Curve properties of connected scatter plot

The line and the symbols settings allow you to customize the display.

You can pick one color from a palette for the line, and choose in between line styles and line thicknesses.

Check the visibility of the symbols, you can then pick one color from a palette and choose between symbol shapes and symbol size. It is recommended to leave the density to 100% in order to see all the points for this kind of chart.

Note

While a "connected bubble plot" does not really make sense, you can plot a component performance map to look at the correlational relationship between three variables. Please refer to the 2D colormap and 3D surface sections for more information.