The type of graph you show in your example is fairly easy to do in Graph Builder once you know how to add all of the elements you want. I have never had to generate a graph like this since we started using JMP here, but I used an existing data set from one of my projects to see just how such a graph could be set up and it was pretty straightforward. (Well, straightforward in as much as I have learned much more about how Graph Builder works beyond basic functionality over the past few months.)
In the example above I plotted read rate and exclusion rate statistics for multiples replicates of a study. As in your example above, I have boxplots for the read rates—the metric that our client is most interested in—and a line graph of mean exclusion rates—a metric that the survey team quickly realized could bias performance. Creating this graph required the following steps:
That procedure was the basic setup for the graph producing a box plot for read rates and a line graph with points of the mean exclusion rates. I cleaned the graph up by balancing the y-axes so that the gridlines were in sync and there was minimum interference between the two responses. I also contrasted the read rate and exclusion rate graphs by making the boxplots and outliers blue for the read rates and, then the points and line for the mean exclusion rates red.
Experiment with Graph Builder. When you have multiple Ys you can right click and send one over to the right.
The type of graph you show in your example is fairly easy to do in Graph Builder once you know how to add all of the elements you want. I have never had to generate a graph like this since we started using JMP here, but I used an existing data set from one of my projects to see just how such a graph could be set up and it was pretty straightforward. (Well, straightforward in as much as I have learned much more about how Graph Builder works beyond basic functionality over the past few months.)
In the example above I plotted read rate and exclusion rate statistics for multiples replicates of a study. As in your example above, I have boxplots for the read rates—the metric that our client is most interested in—and a line graph of mean exclusion rates—a metric that the survey team quickly realized could bias performance. Creating this graph required the following steps:
That procedure was the basic setup for the graph producing a box plot for read rates and a line graph with points of the mean exclusion rates. I cleaned the graph up by balancing the y-axes so that the gridlines were in sync and there was minimum interference between the two responses. I also contrasted the read rate and exclusion rate graphs by making the boxplots and outliers blue for the read rates and, then the points and line for the mean exclusion rates red.