Density plot question
A very basic question from a new JASP user...I produced the attached histogram and density plot of a measure consisting of the mean score of a 3-item scale with anchors ranging from 1 to 4. As you can see, the density plot is very "wavy" and I wonder about its usefulness for such a scale. This is my first time reporting densities (I've used SPSS to report the histogram with the normal curve) so advice around the intereptation and usefulness of the plot here would be appreciated. Also, why are there no values on the y-axis? Thanks!
Comments
Hi MJS,
For a scale this discrete, a nonparametric density plot (using "density" in R) is not useful. There are no values on the y-axis because the units are not relevant -- the histogram has been rescaled so that the area under the curve is 1. In other words, don't use the general density estimator here. I would just stick to the histogram. Hmm I see that we might want to add "histogram" as a separate option in JASP. Clearly the plotting facilities are still very simple.
Cheers,
E.J.
Thanks. Yes, a regular histrogram would be great. I can make JASP do it if I change the variable to ordinal, but that rounds all my data and doesn't seem to be the right approach. By the way, once I change it to ordinal, it permanently changes the data and there is no way to change it back to continuous. Seems like a big inconvenience.
Yes that is a big inconvenience. This may be a "feature" of the latest release. I'll make a note on GitHub.