Skip to content

To all the reviewers we’ve loved before

This post is by Lizzie (I might forget to say that again, when I forget you can see it in the little blue text under the title, or you might just notice it as out of form).

For the end of the year I am saluting the favorite review I received in 2020.

This comes from a paper that included a hierarchical model where we partially pooled by plant species. We had a low and variable replicate number per species, but a pretty good sample size across all species, and we wanted to estimate effects of experimental treatments (things like ‘warm temperature’ and ‘cool temperature’) across species.

We were working on invasive species (species native to somewhere else, in this case Europe, that have been introduced and grow quite well in that somewhere new — in this case North America). There’s a lot of interest in whether evolution post-introduction happens, more specifically evolution that helps the plants do so well somewhere new. We were looking for it in their germination response, by growing seeds we collected in North America or Europe (the seeds’ ‘origin’) in different conditions.

We didn’t find much of an effect of origin. We found effects of our treatments and a few other things, but we didn’t find an origin effect. Maybe because there isn’t a big origin effect, maybe because of the species we picked, or maybe because of lots of things.

In the main text we showed our model estimates, and showed raw data plots in the supplement. I generally think you should try to show raw data in the paper when possible, but this is the first time I got this response from doing it. The reviewer writes:

From [your main text model estimates figure] the main takeaway point that I could garner is that increasing temperatures impact growth and germination speed across species… I find Figure [in the supplement showing the raw data] to be interesting, because the raw data [often for the species with really low sample sizes] suggests to me that you may actually have some origin differences for some species in certain environmental contexts, which may not have come through so clearly in your global, many- leveled models [then details on what specific treatments and for which species this reviewer has discovered some trends].

It’s great to have robust models, but I think it’s worth taking a look at your data and ask whether some more interesting or nuanced stories might come out.

This is the first time I recall when I felt a reviewer was actually taking me by the hand and leading me back to the center of the garden and saying, ‘might you please consider this alternative path? Instead of going left at the fork, perhaps if you go right … you could get the origin effect we all so want to see.’

And to all the reviewers who’ve shared their thoughts
Who now are someone’s else’s reviewer
For helping me to grow
I owe a lot I know….


  1. jd says:

    It is nice when a reviewer actually helps you out!

    I’ve only been at this game for around 4 years now, but overall, I have been disappointed in peer review. I’ve found that the reviewers frequently make suggestions or criticisms that are fundamentally incorrect at the most basic level. I then must write a rebuttal. This then makes me wonder if they are capable of finding a real error or poorly done analysis if there were one. It would be nice to know that they could…that would be nicer than having one’s work appear one day on a blog for all the wrong reasons;)

Leave a Reply