nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Understanding box plots

fishguy

Hi,

I started working with ordinal data and constructed some comparative blox plots (quantile) of the data. Please help me understand the meaning of the various boxes in me example chart. For example, I see only box with a median displayed and there are tick marks I'm not used to seeing such as the short and long ticks (in addition to the ticks that indicate extreme values).

Thanks!

box plot.png

Accepted answers

P_Bartell

Try using the point and click interactive Help native to JMP. One way to initiate that workflow is to change your cursor to the ? by clicking in the main menu. Then hover over the report sector you want to learn more about and then right click. This action will take you to the JMP Help documentation related to your action.

All comments

P_Bartell

fishguy

Thanks for the suggestion. That clarified the various tick marks. Now I just need to think about the plots with no boxes and those with no median. My guess is those with no box must mean the extreme values are also the quantiles? Those with no median must mean that the median and one of the quantiles must be the same?

Thanks again!

P_Bartell

You are correct. Since you have ordinal data and with a relatively small number individual specific values along the ordinal range, it's entirely plausible to have overlapping horizontal marks along the whisker/box range on the y axis.

And with respect to my original reply...I'll give kudos to my former colleague @gail_massari who taught me something many years ago, before I joined the JMP division of SAS, now retired, "Right click is your friend!"

fishguy

Thanks! and yes, I'm starting to right click more and more!

Mark