Image 20150225 1774 19gp36s.jpg?ixlib=rb 1.1

The words that make a successful research grant application

What words appear most frequently in successful research grant applications? And how have they changed over the years? Sean MacEntee/Flickr, CC BY-SA

The words that make a successful research grant application

Summer is ending, and for researchers in Australia that means one thing: grant writing season is hotting up. It’s an intense period of creative thought and brutal focus. The decisions we make in February determine the sort of research Australia pursues for the next three to five years. It also affects where your tax dollars are spent.

And since the Australian Research Council’s Discovery Projects scheme funds only 20% of the applications that are submitted to it, the decisions we make also affect the lives and livelihoods of researchers across the country for the next three to five years.

As such, the decision of what to put in a grant application is one we spend a lot of time thinking about.

So, I thought we would have a look at the successful grants from the last six years to see what they said in their project summary. This not only reflects the nature of the research that was proposed but also the way the applicants talked about that research.

To do this, I extracted the project summaries of successful applications for Discovery Projects between 2010 and 2016 from the Australian Research Council (ARC) and parsed them to find the frequency of each word. I then removed some basic stop words and graphed the results below.

New and improved

One message is clear: if you want your research to be funded, it has to be “new”. That is, after all, the point of a Discovery Project, so no big surprises there.

Likewise, these are grants funded by Australian taxpayers and written and reviewed by Australian researchers, so mentioning that the research will have something to do with “Australia” or things “Australian” seems like a good idea.

Author provided

I also wondered about how the research subjects might change due to shifts in the bureaucratic sands of the Australian funding organisations themselves. So I thought it might be interesting to see if there were any trends in the use of different words between the years 2010 to 2015. So, I split up the data by year and chose some words that I thought might have changed in their frequency of use between 2010 and 2015.

As a neuroscientist I am well aware of the increasing assertion made by the ARC that medical research should be funded by the National Health and Medical Research Council (NHMRC). As can be seen below, there was a slight drop in the mention of the words “health” and “disease”.

However, taking into account the fact that the number of funded Discovery Projects also dropped between these years (from 925 in 2010 to 665 in 2015 –- largely because of the introduction in 2012 of the Discovery Early Career Researcher Award) that reflects a continuing commitment to these fields by the Australian Research Council.

Author provided

In context

It is interesting to note these shifts in selected words, but what about the corpus of project summaries in general? Considering the changing number of funded projects, the most informative way of interpreting these data is not in simply comparing each word between the years 2010 and 2015, but in making that comparison in the context of all the other words used.

To that end, I subtracted the frequency of use of each word in the year 2010 from the frequency of use in the year 2015 and graphed the results. As would be expected, most words were used less in 2015 because there were less projects. However, some have fallen further than others.

Author provided

The words are listed alphabetically from left to right. Starting from the left, I thought at first that we might be becoming less parochial as “Australia” and “Australian” have both seen a large decrease in the frequency of their use. Then I read across and saw that “global”, “international” and “world” have seen dramatic decreases as well.

Perhaps the most striking aspect of the graph is the words which have seen large increases from 2010 to 2015. Of particular prominence is “aims”. However, after a bit of digging, I am confident this is simply a stylistic change. When I read back over the project summaries from 2015, the great majority began with a variation on the phrase “This project aims to…”.

However, this does not explain the rise of our great expectations. Although around 300 less Discovery Projects were funded in 2015 than 2010, it seems were are expecting around 300 more things to come out of this funding round.

Grant writing is a strange business. So much work is spent in the preparation. So much depends of the outcome. And yet, if we are to get through this process with some semblance of sanity, perhaps we should consider our expectations. And keep them from being too great.

Note: The code I used to find the data and generate the above graphs is available here