by Xan Gregg xan@jmp, Director, Research and Development, SAS
With JMP, you can assign the marker of your choice to data points in plots to highlight your data. And you aren’t limited by the 32 built-in symbols – you can also select any character (including letters) as a marker for a scatterplot. The easiest way to set a letter as the symbol for a row is with the "Other..." item in the Marker submenu. Here's an example with Big Class. After selecting all males, right-click in the row-state area to get the option shown in Figure 1.
Figure 1 Assigning markers to rows
You then see a dialog in which you can type a letter, shown in Figure 2.
Figure 2 Defining the custom marker
Setting the male rows to "M" and the female rows to "F" yields the data table shown in Figure 3.
Figure 3 Custom markers assigned to rows
Figure 4 shows the scatterplot.
Figure 4 Scatterplot with male and female markers
Any character in the current Marker Font (set in Preferences) is allowed, so there is a wide variety of Unicode characters available. You can change the font preference to a symbol font for even more specialty options. While special characters can be tricky to type from the keyboard, you can usually find them in a Unicode table and paste them into the dialog.
You can also set the symbols through scripting. Here's a scripting example that uses the Mars/Venus symbols to designate male and female, showing two different techniques for specifying the character as shown in Figure 5.
dt = Open("$SAMPLE_DATA/Big Class.jmp"); r = dt << Select Where(:sex =="M"); r << Markers("\!U2642");// hexadecimal r = dt << Select Where(:sex =="F"); r << Markers("♀");// actual character
Figure 5 JSL script for defining symbol markers
Figure 6 shows the results of running the preceding JSL script.
Figure 6 Custom markers assigned to rows
Figure 7 shows the scatterplot that this script creates.
Figure 7 Scatterplot with Mars and Venus symbol markers
By customizing the markers, you can enhance the visualization of your data, allowing you to make sense of the data in a glance.