Thanks for the illustrations and response @CarlyRae!
In future work creating plots of poop, you might consider the color palette named in an xkcd survey (yes, you bet there's a baby poop green color...)
Just to confirm what you were saying about overlapping confidence intervals:
- With the first plot (beta_Ratsamples) if group centroids overlap based on 95% confidence intervals: when they (mostly) fail to overlap the group centroids are likely different.
- In the case of the second plot (jacc_dispnest) you're plotting the convex hull, not the ellipse, correct? I'm guessing by code that you're using the base R
plot()call tied in with vegan, which generates the hulls by default. Is there any reason why you prefer to do it this way? I'm guessing you could also have amended the plot to include both the hull and ellipse data with:
plot(jacc_dispnest, hull = T, ellipse = T)
What I'm curious about is whether you like using the same overlapping characteristics as a visual check for both the dispersions and group centroids.
Cheers