jacob-long
plot_summs: expressing significance levels
Would it be feasible to express significance levels in plots generate with plot_summs()?
- For a single model plot, one could envision different plot symbols conditional on the significance level. For example, all non-significant coefficients could be plotted with a hallow square symbol, all significant coefficients with filled symbols (* = upside triangle, ** = downside triangle, *** up-down triangles)
- For multiple models, one could continue this coding, with each model having a different color. Alternatively one could consider plotting estimates with different opaqueness, i.e. all non-significant estimates more transparently. Thanks a lot for considering this feature request
The "point.shape" argument to plot_summs() currently permits only true/false. The symbol is either given in advance (by a fixed symbol order) or the same for all models. Would it be feasible to pass a vector of PCH symbols to manually choose the symbol for each model? That would be very helpful, many thanks!
The latest commit in the dev branch adds the ability to choose points by passing a vector of shapes to point.shape. The original request is quite a bit more complicated and I'm not sure if/when I'll get to it — I've generally been moving towards removing features that focus on statistical significance if they have any maintenance burden.
