answered by verbal nutritionist Dr. Helen Sword
Q. How did you come up with the scoring values for the WritersDiet Test?
A. I chose examples of the liveliest and stodgiest academic writing I could find and extrapolated the initial values from there. Since then, I have tweaked, tested, streamlined, and elaborated the WritersDiet Test algorithms based on more than 1,000 writing samples -- a process of informed evaluation based on extensive reading, rhetorical analysis, intuition, and, yes, a dollop of subjectivity.
Q. Can effective writing really be reduced to a numerical formula?
A. No, of course not. The WritersDiet Test does not attempt to measure for vividness of expression, clarity of thought, fluidity of style, or any of the other elements that matter most in engaging writing. The test does, however, provide a user-friendly method for identifying some of the sentence-level grammatical features that most frequently weigh down academic prose.
Q. I ran the WritersDiet Test on a passage by my favorite author, and it came out "flabby." Doesn't this mean your test is flawed?
A. I developed the test to help stodgy academic writers write more clearly and energetically. Many fabulous pieces of prose will receive scores of "flabby" or even "heart attack" on the test, because stylish writers have the confidence and skill to play around with language in ways that the WritersDiet Test is not designed, and therefore should not be expected, to evaluate.
Q. How does the WritersDiet Test differ from other quantitative measures of readability, such as the Fog Index or the Flesch Index?
A. The Gunning-Fog Index, developed in 1952 by Robert Gunning, uses an algorithm based on word and sentence length to calculate the number of years of formal education required for a person to understand a given text. The Flesch–Kincaid Grade Level Formula makes a similar calculation keyed to United States grade levels. The WritersDiet Test, by contrast, has nothing to do with grade levels or average word length. Complex scholarly essays can score very well on the WritersDiet Test, whereas a children’s story filled with weak verbs and fluffy adjectives could score badly.
Q. Is it better to test shorter writing samples or longer ones?
A. On average, longer samples generate more favorable WritersDiet Test scores. I recommend that you test relatively short samples (100-500 words) if you are trying to figure out how to strengthen and tone a specific paragraph. Longer samples (500-1,000 words) give more of a "big picture" view and help you spot recurring patterns in your writing.
Q. Why can't I run the test on a full-length document?
The WritersDiet Test was originally calibrated for passages of 300 words and works best on writing samples of around that length. With samples of more than 1000 words, the scores flatten out to a point where they eventually become meaningless. We are currently developing a downloadable Windows application for PC and Mac that will allow you to run the test on a full-length Word document and spot the flabby zones. Watch this space!