"We must confess that your proposal seems less like science and more like science fiction," declares an executive in Contact, the movie about scientists who detect alien transmissions in outer space. Grant reviewers may confess the same of application abstracts that are filled with wonderful ideas but lack practical, nuts-and-bolts details. A good abstract is like a postcard-sized reprint of a famous work of art: It captures and illustrates the entire research picture without leaving the reader puzzled or confused.
In their efforts to spruce up and dress the body of the research plan, many grant applicants--postdocs and faculty alike--often fail to include essential pieces of the abstract, such as research data and methods. Because the abstract is the first glimpse a reader gets of an application's worth, such oversights can raise unnecessary questions, and may even create the impression that the research plan itself may be incomplete. The key to designing a winning grant application is to start off with a well-rounded, concise summary of your whole application: To accomplish that in a few hundred words, however, takes skill.
What's in an Abstract?
Ellen Barrett, a professor of physiology and biophysics at the University of Miami School of Medicine suggests four key components of a well-rounded abstract: The abstract should introduce the reader to the problems you are addressing, the overall hypotheses you are testing, the main techniques you will be using, and your overall experimental plan. Barrett's advice echoes the Public Health Service grant application instructions, which state that a grant applicant should address the following questions:
What do you intend to do? Why is the work important? What has already been done? How are you going to do the work?
What do you intend to do?
Why is the work important?
What has already been done?
How are you going to do the work?
The abstract should provide succinct answers to all such questions.
"Most good research is hypothesis-driven," writes Barrett in an online grant-writing guide  she put together for researchers at her university. In an abstract, those hypotheses should describe "your overview of the mechanisms underlying the process you are studying, not just your prediction" about how experiments will turn out, she says. Barrett warns against writing abstracts with the assumption that your hypotheses are true--a costly error that "has doomed many applications," she says.
With many abstracts, "the problem is that applicants don't summarize the full proposal," says consultant Bob Lucas, a former university research administrator, who is now director of the Institute for Scholarly Productivity based in California. What many applicants do is simply cut- and-paste the first two paragraphs of the introduction into the space set aside for the abstract, he discloses. And while those two paragraphs may be beautifully constructed, they don't typically explain the whole project. Applicants wind up "saying what they are going to do, but not how they're going to do it," reveals Lucas. Physicist Scott Bergeson, for example, was 0 for 3 before attending one of Lucas's workshops. "I made the typical mistakes," explains Bergeson, an assistant professor at Brigham Young University in Salt Lake City. "I had great ideas but my proposals were really vague, too general," he says. As a new member of faculty, Bergeson tried hard but failed to secure three major grants that totaled more than $1 million. But then last March, he went to one of Lucas's workshops, began applying the professor's writing practices, and instantly turned his game around.
Dog Walker or Cocktail Talker?
Lucas suggests beginning with a four-page description--a "concept paper summary"--of what you want to accomplish. By adding more specific detail to this document, you end up with a draft of your research plan. Conversely, by "boiling it down," you wind up creating a concise research abstract that fits with and reflects the entire research plan. This practice keeps you focused and "helps remind you what the point of your research application is," Lucas says. It helps you become disciplined, too: Word limits on abstracts forces you to delete, rephrase, and chop up information that is not essential to the abstract, says Lucas. "You might be making an interesting point," he states, "but it might not be relevant," so out it goes. Bergeson used such writing techniques for a grant application due the following May, just 6 weeks after the workshop. It worked--his application was funded.
Lucas lets his eager scholars in on a couple of insightful anecdotes: Writing, he says, isn't an ordeal if you realize it's like "Walking the Dog." You don't think about taking the dog out for a stroll, you just do it. With writing, simply set aside time to "walk the dog": Sit down and write every day and soon writing will be as natural as handling radioactivity. "An idea with a plan is a grant application," Lucas continues. "An idea without a plan is simply cocktail talk."
"Make the specific aims and the ultimate goal very clear" to reinforce ideas for reviewers, says Suzanne Fisher, director of the Division of Receipt and Referral at the National Institutes of Health's (NIH's)
Fisher directs the team of referral officials and administrators who process and review applications at the federal funding agency. By virtue of having to handle tens of thousands of submissions every year, she is sensitive to mistakes that scientists continue to make--especially when it comes to writing in plain English. "Descriptions," she says (the abstract is now called the description at NIH) "should avoid excessive use of jargon and abbreviations," because "unless you are an insider, you have no idea what the application is about." Science gobbledygook--" We will study the MLC2 Ser-18-Ala Nyquist B-process at pCas 7.5-5.5 +/-MLCK"--will not enamor reviewers, even if they do understand what you're talking about.
Keywords Perhaps Not Key
Fisher tries to dispel the commonly held belief that shrewd keywords win over the hearts and minds of reviewers and officials: "The referral office uses more than just the title or description to make assignments," she enlightens, "so there is no point in trying to direct assignments by judicious word choices." Starting off your abstract with the word "Aging" for example does not mean you should expect your application to be automatically routed through to the National Institute on Aging, she clarifies. Fisher makes another point that "if an award is made," the description "will be public information" deposited in the federal CRISP  (Computer Retrieval of Information on Scientific Projects) awards database. "So it should be clear, concise, accurate, and not contain proprietary information."
Rate Your Abstract
Perhaps the most important reason to write a succinct and complete abstract is because not all reviewers on a panel will be formally assigned to read your proposal: Generally, the primary and secondary reviewers report their analysis and give it a rating or predetermined classification. The remaining reviewers must also rate your application, and unless they have previously scrutinized it, they may pass judgement only on what they read in the abstract.
It's understandable that to make sense of alien messages, some pretty advanced deciphering technology might be required. Research abstracts on the other hand shouldn't need that level of decoding. "All I'm asking is for you to just have the tiniest bit of vision," replies Contact star Jodie Foster to the skeptical executive. "Just sit back for one minute and look at the big picture," she pleads. If you can bring those--vision and the big picture--together in your abstract, your next grant application could be out of this world!
For many young scientists, the research plan itself can appear to be an alien landscape! Next week we begin a series of head-first plunges into the nitty-gritty of your actual research plan : How to structure it, what reviewers are looking for and what irritates them the most. Stay tuned...