## Writing concisely about indices of extinction risk

Edit: An updated version of the paper mentioned in this post is available at bioRxiv: http://dx.doi.org/10.1101/000760 (and now it is published: http://dx.doi.org/10.1111/cobi.12308)

A post on the joys of editing science.

For three years I have worked on a paper in which we develop indices of ecological communities that relate to extinction risk. One of these indices depends on the geometric mean abundance of species (see box), an increasingly used biodiversity metric. For example, bird monitoring in Europe and North America reports the geometric mean, and the Living Planet Index is also essentially a geometric mean. I even use it in a recent paper to help guide fire management.

What is a geometric mean? How is it an index?

I guess readers will know how to calculate the arithmetic mean of n numbers – add them and divided by n. In contrast, multiply the numbers together and raise the product to the power 1/n to calculate the geometric mean

The geometric mean weights smaller numbers more than the arithmetic mean. For example, the arithmetic mean of 1 and 9 is 5 (10/2), while the geometric mean is 3 (the square root of 9).

The geometric mean abundance of species summarizes the state of a suite of species. Rather than simply the average number of individuals per species (the arithmetic mean), the geometric mean decreases with unevenness in abundance. Combining abundance data from multiple species, the geometric mean becomes an index of the biodiversity in a community. It has various useful heuristic properties (see references in our recent fire paper), but our new paper derives indices from models of extinction. One of these indices depends on the geometric mean abundance.

Well, finally I submitted the paper to Conservation Biology. In the paper, co-authored with Alana Moore, John Morgan, Jochen Krauss, and Chris Clements, we develop indices of extinction risk in communities by using very simple population models. If you have read some of my previous posts on indices, you might know my concern with ecologists sometimes developing indices in ways that defy logic. Our new paper argues indices will tend to be more logical when based on models.

Now I won’t say much more about the paper – a topic for another post (though read the abstract below, and try the entire article if you want to read the paper; also see my INTECOL talk). Here I discuss my less than concise communication of our paper. Though a fan of concise communication, I don’t claim expertise.

In my graduate seminar subject, we spend a hour or two critiquing pieces of writing and asking how to write them more concisely. The first sentence of my paper’s abstract provided a good example:

Monitoring programs often combine abundance data from different species to generate biodiversity indices to inform aggregate trends in the state of the species.

The sentence starts quite well, but the second half stinks! It just goes on and on and on and on and … too many prepositions, like ants on a honey pot:

Monitoring programs often combine abundance data from different species to* generate biodiversity indices to inform aggregate trends in the state of the species.

And what are aggregate trends anyway?

However, I must admit, I submitted that sentence to Conservation Biology just the other day. Conservation Biology has a 6000 word limit for regular papers (from abstract through to the literature cited). My paper had about 6600 words; I hoped a 10% margin was close enough. Sorry! The paper came back with its tail between its legs – I needed to remove 600 words.

Apart from anything, this experience proves Conservation Biology doesn’t preferentially favour its editors! So, last Friday I worked through the paper, studying every sentence. Could sentences be removed, or reduced and merged? Could I re-write them with fewer words? Was that paragraph really necessary?

I was ruthless, but remained stuck about 20 words over the limit. I couldn’t cut it more without losing content. But the awful abstract remained intact, with some flab to trim. And what better tool to help me than the Writer’s Diet! If you don’t know Writer’s Diet, then pay a visit. It identifies problems in five areas: overuse of complex nouns; conjugations of the verb “be”; prepositions; adjectives/adverbs; and words such as it, this, that and there.

Complex nouns (e.g., often nouns ending in -ion and formed from other words, e.g., extinction), especially necessary technical terms, can be difficult to eliminate in scientific writing. Also, some adjectives are critical.

So, I set Writer’s Diet to ignore the following four words in my abstract:

abundance, biodiversity, extinction, and geometric.

I ran the test – my abstract bordered on flabby:

The score for my first abstract. Bordering on flabby

Writer’s Diet highlights these cases for different reasons. You can often replace “be” verbs with active verbs. (Note, I first wrote that sentence as ‘The verb “be” can often be replaced by more active verbs’!). Too many prepositions impair writing. And why use adverbs and adjectives instead of more relevant verbs and nouns? For example, use “sprinting”, not “fast running”.

Conveniently, Writer’s Diet highlights the problematic words:

Some kindness please for my poor first abstract!

Including the words that Writer’s Diet ignored, my abstract contained 241 words.

So, I set to work, editing for brevity and trying to write more clearly. With some effort, I reduced the abstract to 194 words:

Biodiversity indices often combine data from different species to inform monitoring programs. Heuristic properties can suggest preferred indices, but we lack objective ways to discriminate between indices with similar heuristic properties. Biodiversity indices can be evaluated by determining how well they reflect management objectives that a monitoring program aims to support. For example, the Convention on Biological Diversity (CBD) requires reporting about extinction rates, so simple indices that reflect extinction risk would be valuable. Here we develop three biodiversity indices based on simple models of population viability that relate extinction risk to abundance. The first index is based on the geometric mean abundance of species. A second uses a more general power mean. A third integrates both the geometric mean abundance and trend. These indices require the same data as previous indices, but they also relate directly to extinction risk. Field data for butterflies and woodland plants, and experimental studies of protozoan communities show that the indices correlate with local extinction rates. Applying the index based on the geometric mean to global data on changes in avian abundance suggests that the average extinction probability of birds has increased approximately 1% from 1970 to 2009.

Here’s the Writer’s Diet result:

Fit & trim!

I’m not 100% happy with the result; I still see room for more improvement. But with fewer words and a better first sentence, I’d met my goal. And I even enjoyed the challenge! What do you think – a better abstract?

The lessons I learnt:

1. Leaner writing  is possible even if you have already spent years working on a paper. In fact, in that case the room for trimming might be even greater.

By the way, I ran this post through Writer’s Diet. For the win…

There is only a little of this and that! ;-)

* I think “to” is regarded as a preposition when used in an infinitive.