How Accurate are MS Word's Readability stats?

[presidebt]

I have MS Office 2003 and I have Word’s readibility stats eneabled for whenever I spell check. First, I’d like to know if Flesch-Kincaid Grade Level is a) accurate and b) does it go higher than 12.0? (that’s the highest score I’ve gotten)

Also, how does the Flesch Reading Ease function work? Is it accurate? I just scored 39.2% on something I wrote. Is that good? Bad? Meaningless?

Thanks!

[Agape]

@presidebt

I have MS Office 2003 and I have Word’s readibility stats eneabled for whenever I spell check. First, I’d like to know if Flesch-Kincaid Grade Level is a) accurate and b) does it go higher than 12.0? (that’s the highest score I’ve gotten)

Also, how does the Flesch Reading Ease function work? Is it accurate? I just scored 39.2% on something I wrote. Is that good? Bad? Meaningless?

Thanks!

To chip away at your many questions, I found this:

One of the formulas that is similar to the SMOG formula is the Flesch-Kincaid formula. This index computes readability based on the average number of syllables per word and the average number of words per sentence. The score in this case indicates a grade-school level. For example, a score of 8.0 means that an eighth grader would understand the document. Standard writing approximately equates to the seventh- to eighth-grade level. However, for Patient Education for U documents, the recommended reading level is around the sixth grade.

Since it uses grade-school levels to determine ease of reading, I’m thinking that 12 is the highest it can go.

Source: http://www.med.utah.edu/pated/authors/readability.html

[SmackFu]

By “accurate”, do you mean whether Word is computing it accurately, or whether the stat itself accurately reflects real people’s reading levels?

[Agape]

Chipping away at them further,

The Flesch Reading Ease Score identifies readability based on the average number of syllables per word (ASL) and the average number of words per sentence (AWS). The formula is -

206.835 - (1.015 X ASL) - (84.6 X ASW)

The formula results in a reproducible and predictable score in the range from zero to 100, with higher scores signifying greater reading ease. The average Reading Ease score for comic strips is around 90, with the Australian Financial Review in the 40s, non-specialist legal journals around 30, standard insurance policies around 15 and some of the wilder deconstructionist journals around 10.

The lower your Flesch Reading Ease Score, the more difficult it is to read your work. According to the site, your work is about as interesting as the Australian Financial Review

Ah but don’t worry; my latest paper scored 22.4. It’s insipid.

[Shalmanese]

Huh, how does that produce a number between 0 and 100?

Theoretically, you can have an infinite number of syllables per word/line so the lowest score is negative infinity.

[Futile_Gesture]

Flesch-Kincaid may be ok for school as a measurement of writing development, and a very rough guide to the writing’s complexity, but I refuse to accept it’s ‘accurate’. Readability depends on lots of aspects of the writing, quite apart from sentence length and number of syllables. It’s quite possible to write an unreadable tract using short sentences and little words.

I don’t think I’d use this, in the same way I don’t use Word’s grammar checker. I won’t be told how to write by a computer.

[RealityChuck]

@presidebt

I have MS Office 2003 and I have Word’s readibility stats eneabled for whenever I spell check. First, I’d like to know if Flesch-Kincaid Grade Level is a) accurate and b) does it go higher than 12.0? (that’s the highest score I’ve gotten)

I’m more familiar with the Fog Index, which, as someone mentioned, is similar. The point is to get a low readability index. The lower the grade level, the easier it is to read. I’ve seen Fog indexes in the 20s – the writing was deadly.

The indexes are accurate as far as they go. Word calculates them correctly (it’s merely doing the math). And usually a lower Grade Level index is better than a higher one. IIRC, something like the New York Times gets around 8-10 grade level; if you’re going above that, then you’re needlessly complex. You shouldn’t write your prose specifically with these scores in mind, but if you consistently get a high Fleish-Kinkade grade level, you are doing something very wrong.

[presidebt]

@Agape

The lower your Flesch Reading Ease Score, the more difficult it is to read your work. According to the site, your work is about as interesting as the Australian Financial Review

Ah but don’t worry; my latest paper scored 22.4. It’s insipid.

Ha! I guess I better start dumbing down my Women’s History Bios.

Anyway, thanks for these answers Dopers. And SmackFu, I meant “whether Word is computing it accurately.” Sorry for the ambiguous use of language.

[presidebt]

@RealityChuck

but if you consistently get a high Fleish-Kinkade grade level, you are doing something very wrong.

But wouldn’t this depend on the audience? For instance, I continue to get straight A’s on research papers for school. Almost all of my papers indicate 12.0 and low scores in readibility (not quite 22.4!), generally 30s to 40s. On the other hand, most of my creative writing seem to fall bewteen the 6-9 grade level, with anywhere from 50-80% readibility.