« Infographing the cost of iPad | Main | An enemy of the message »


Steve Kern

To make a profile chart in JMP, you will need a data table with three columns. First a continuous numeric column containing the data, second a categorical column for the primary categorization, and for the third a categorical column to identify the observations for each series. In this case, the data column would be "Proportion", and categorical columns would be "Year" and "Category". The example data table should have 20 rows. From here, make a Oneway plot using Analyze>>Fit Y by X. Put the data column, "Proportion" in the "Y, Response" box. Put the "Year" column in the "X, Factor" box and hit "OK". After the plot is generated, click the little red triangle next to the graph title to bring up the platform options menu. Click on "Matching column..." which will bring up a small column selection dialog. Here you can select the "Category" column, and the graph will now display colored lines connecting the two observations for each category. This procedure can be extended to other examples with more X categories or continuous X variables, and grouping with "Matching columns..." can allow you to perform analyses like fitting regressions to each individual series rather than the entire data set.


Hold Ctrl while you right click to apply marker size changes to everything in the window. And you can define every aspect of the axis marking system by right clicking on the axis and choosing "Axis settings." You should be able to set the min, max, increment, etc. You can also just click and drag!

Xan Gregg

From the creator of Graph Builder, many thanks for the kind works and suggestions for refinement. Graph Builder, and JMP in general, is focused on fast interactive exploration, but we're always trying to make improvements in visual quality as well. For the legends, it's possible to clean them up a bit by right-clicking on the legend title and choosing Legend Settings.

Graph Builder does allow line elements even with categorical X values. Despite the rule you refer to, I've come to see that the ability of the line slopes to show variation patterns often outweighs any misleading suggestion of continuity between the categories.

With such lines you can make Profile Charts if the data is in the right form: a single response variable as Y with one categorical variable as X and another categorical variable as Overlay. If you have multiple variables, you can use Tables:Stack to put them in the form Graph Builder wants. Or just use the Parallel Coordinates plot in JMP, especially if the responses have different scales.

The continuous axis is not flexible enough to show only "odd" values like 1999 and 2009, but you can make year categorical ("nominal" in JMP) and get a two level categorical axis with those years.

Pavlov's data analyst

The biggest shortcoming of donut charts is in fact that they make me salivate.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)

Marketing analytics and data visualization expert. Author and Speaker. Currently at Vimeo and NYU. See my full bio.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Good Books

Keep in Touch

follow me on Twitter