Report model fit in some way

[Aariq]

R1 suggests adding to results:

• raw data added to plots in some way
• Table of output from anova()

In the results, I was surprised to see only a few p-values and not much else about the models in terms of how they were assessed for goodness of fit. While I am not at all suggesting that p-values are the solution, I would have liked a little more sense of how well the models fit. One way to do that might be to show the raw data plotted with the model fits so the reader can see, for example, what kind of functions were drawn. Another might be to present something in table format (the comparable version of a table of AICs or of the results of frequentist tests).

R3:

Line 260 -- you mention the CI of the model estimated coefficients, but these aren't provided anywhere that I could find. I think a table based on the summary of the two fitted models discussed here (and elsewhere too) should be added. And report the variance of the random effect.

[Aariq]

maybe add some kind of table of model performance (r2, RMSE) to model validation supplemental.

[Aariq]

One possibility that addresses a part of this is to have a table for each vital rate like so:

[Aariq]

Possibly another table showing all intercepts ± 84% CIs like so:

(bold indicates non-overlapping CIs)