Scatter plot

In a scatter plot, markers are presented in a two-dimensional coordinate system. It is useful for getting an overview of how your data is distributed across two dimensions, and to spot correlations.

By default, each row in your loaded data table is represented by a marker. The marker's position in the coordinate system is defined by the row's values in the columns selected on the X-axis and the Y-axis. For example, the marker's position for the last row in the data table below is the coordinate 186 on the X-axis, and 84 on the Y-axis.

You can adjust the scales and scale labels, as well as other axis settings, from the visualization properties for each axis, and you can add features such as gridlines, zoom sliders or error bars, and so on.

More variables can be set to correspond to the color, size or shape of the markers, thereby adding more dimensions to the plot.

Correlation in a scatter plot

The relationship between two variables is called their correlation. If the markers are close to making a straight line in the scatter plot, the two variables have a high correlation. If the markers are equally distributed in the scatter plot, the correlation is low, or zero. However, even though a correlation might seem present, this might not always be the case. Both variables could be related to some third variable, which explains their variation, or, the distribution could be due to pure coincidence.

In the scatter plot example below, sales is plotted against cost for a number of different products (colored by product). It shows a low positive correlation.

Each product can be shown separately using trellising:

Aggregated markers

The scatter plot can also be used together with aggregation (for example, Sum or Average) by using the setting Marker by. In this case, the values for a certain category are bundled together to show a single marker for each category.

Labels

Labels can be used in visualizations to identify and describe markers and the data associated with them. In the scatter plot below, labels show which category each of the marked markers belongs to:

In a scatter plot, you can interact with the labels and move them by dragging them to a new position. If you move a label, it will stay in the new position until you reset the label positions from the right-click menu in the visualization (installed client only).

All visualizations can be configured to show data limited by one or more markings in other visualizations only (details visualizations). Scatter plots can also be limited by one or more filterings. Another alternative is to configure a scatter plot without any filtering at all. See Adding data limitations for a visualization for more information.

You can show data from multiple data tables in the same visualization if a proper data table matching is available. For more information, see Multiple data tables in one visualization and Column matches.