posted
A co-worker asked me about the Flesch stats that come up when you do a spelling/grammar check in Word. I had to look a few things up to properly explain them and stumbled across this part of an article by Rudolf Flesch, inventor of the Flesch scale.
While I like to look at the readability stats for my stories, I never really understood how they were generated or what I really needed to do to change them (not that they need to be changed).
posted
Whether you should change them depends on audience. If you're writing for children or young adults, then you probably do need to up your score. Otherwise, you really don't need to bother with it.
Granted, this is why some things score negative six And if you're scoring in the same range as things that you find incomprehensible (because of language, not content), you might want to consider adjusting your usages a bit.
posted
After reading this article I decided to check the stats on some of my own writing...
The following list is just what I could find on short notice on my hard-drive, I've written far more, printed it off into hard-copy, and deleted the digital copy... There is also a lot of stuff that hasn't gotten to the digital copy stage yet (I prefer to hand-write first... something about the flow of the arm movements helps my creativity...)
No Man is an Island (Super short essay I wrote in 12th grade for 11th grade English, don't ask why I took 11 English in 12th grade, long story there.):
I'm not sure how much this means, though. I would far rather hear from wise readers who can tell me whether or not I'm relaying the message I'm attempting to relay and whether or not the story is interesting and compelling.
"In Memory" (WOTF published finalist -- 3100 word Science Fiction) Passive Sentences: 2% Flesch Reading Ease: 73.1 Flesch-Kincaid Grade Level: 5.6
"Betrayer of Trees" (WOTF winner -- 7400 word Fantasy) Passive Sentences: 3% Flesch Reading Ease: 76.6 Flesch-Kincaid Grade Level: 5.6
"The Man Who Moved the Moon" (Phobos winner -- 6900 word Science Fiction) Passive Sentences: 3% Flesch Reading Ease: 71.6 Flesch-Kincaid Grade Level: 5.8
Stats for the stories I currently have submitted somewhere:
"Taint of Treason" (800 word Fantasy) Passive Sentences: 4% Flesch Reading Ease: 78.0 Flesch-Kincaid Grade Level: 5.2
"Protector of the Line" (7400 word Fantasy) Passive Sentences: 4% Flesch Reading Ease: 74.5 Flesch-Kincaid Grade Level: 6.0
"Tabloid Reporter to the Stars" (7800 word Science Fiction) Passive Sentences: 4% Flesch Reading Ease: 65.0 Flesch-Kincaid Grade Level: 7.8
"Salt of Judas" (5500 word Fantasy) Passive Sentences: 2% Flesch Reading Ease: 79.9 Flesch-Kincaid Grade Level: 4.3
"Spectacles of Death" (2400 word Fantasy) Passive Sentences: 1% Flesch Reading Ease: 69.7 Flesch-Kincaid Grade Level: 6.9
"By the Hands of Juan Peron" (5000 word Science Fiction) Passive Sentences: 4% Flesch Reading Ease: 66.4 Flesch-Kincaid Grade Level: 6.7
Hmmm. It looks like "Tabloid" is probably written at too high a grade level (among its other flaws.) And people writing for WOTF might want to aim for a grade level of about 5.6.
quote:No Man is an Island (Super short essay I wrote in 12th grade...Why "No Man is an Island" is still on my hard-drive, I can assure you I do not know...
Egad, that makes me feel old. We didn't HAVE hard drives when *I* was in 12th grade. We were grooving with 8-track tapes.
By the way, I've always heard it called the "Fog" index - when you are obscuring clear writing with verbosity.
"Wings" and "Gruff Bluff" both have higher Flesch scores than "Something in the Air", but "Wings" somehow scored a higher grade level while "Gruff Bluff" scored a lower grade level. Perhaps the high percentage of passive sentences in "Wings" affected the grade level...
In the past, whenever I've scored "too high" or "too low" a grade level for my liking, I've always just adjusted my vocabulary. Now that I know a little bit more about how the formulas work, I might try adjusting sentence length and structure as well.
I also think it's interesting that the stories I've written in the last year or so are all within 5 points on the Flesch Reading Ease Scale.
The most disturbing thing for me is the high grade level on my Nonfiction as compared to my Fiction... Am I picking up bad habits from all the legaleze I read?
My Fiction's lower grade level could probably be easily explained as probably the most influential person on that is and was OSC, and as we all know, he is quite big on the "plain English" thing.
I seem to average around 4/5% on my passive sentences... EJS is published and he seems to average around 2/3%... should I start working on reducing my passive sentences?
For the most part I seem to average around a 4th/5th grade reading level and a 70-80 on my Flesch Reading Ease on my Fiction. That seems to be OK.
quote:Is that really 14% passive on "Wings", or did you mistype?
Yes, it really is 14%. However I forgot to mention that Wings is a flash story with a word count of 585. I just manually checked, and the 14% is actually only 5 sentences out of about 35.
I guess that's one of the things you have to be careful of when reading the stats.
quote:should I start working on reducing my passive sentences?
What is the range of acceptability on passive sentences? I don't know. I heard a few different things. Something in the 5% range, anything less than 10%...it just depends who you ask. While active sentences can give you a lot of bang for your literary buck, there are times when a passive sentence just works better, gives you a better flow.
For example, when I first wrote "Something in the Air", I wrote it without any passive sentences, just to see if I could. Some of the sentences I had just didn't work when I wrote them as active. On subsequent re-writes, I put a few actives back into passive and it reads better.
If you want to try working on passive sentences (recognizing them and re-writing them) I believe there is an exercise or two in the Writing Class.
posted
Honestly, in fiction writing I believe passive voice sentences need to be near 0%....but even as I say that, I will also say this much: percentages are meaningless in this case. When it comes to passive voice you have to look at every case separately and ask yourself why it is written that way. If there isn't a good reason, then it needs to be active. It is the case by case that matters, not the percentage. The only reason I say it should be near 0% is that there are very few acceptable cases for passive voice in fiction.
Mostly, it's ok to have a passive sentence when the perpetrator of the action does not matter AND the object of the action is what the paragraph/passage is about. My classic example:
Lilith was arrested for murder.
Actively, this would be "The police arrested Lilith for murder." but maybe you don't care. Unless this is a police procedural murder mystery (in which case the police are quite important) then Lilith may be what this is about and you want to keep the focus on her.
Even so, these sentences should be rare. If you are using passive voice ten percent of the time it may even be that you picked the wrong protagonist...because they are not causing enough of the action.
I think the stats on the last one are a load of bunk. For sure it is not a 12th-grade piece level work! I should try to find my soft copy of my masters thesis. That should crash windows.
I'm also willing to bet that Word confuses passive voice with past perfect tense. That's probably why the non-fiction article scored so high on that stat.
[This message has been edited by Spaceman (edited July 07, 2005).]
Well it looks like I am pretty consistent. I should probably watch for passive sentences a bit more closely. I seem to pretty much have 3% across the board, from 1k to 40k stories. Hmm. This was interesting.
posted
Fun! I dug up a lot of stuff that I've written over the years.
Here are scholarly works:
intro to my PhD dissertation (published by Routledge as part of the series "Outstanding Dissertations in Linguistics") 33 % passives 23.6 Flesch reading ease 12.0 Flesch-Kincaid grade level
intro section of a chapter published in a book of scholarly articles: 35% passives 11.7 Flesch reading ease 12.0 Flesch-Kincaid grade level
intro of my latest submitted research paper 25% passives 21.5 Flesch reading ease 12.0 Flesch-Kincaid grade level
Here is some fiction:
Peter’s Snowy Day (story I wrote for an ESL class I was teaching for college-educated Russian senior citizens) 3% 81.3 4.3
posted
Nice kk... So maybe my higher-than-average-in-fiction grade level and lower flesch reading ease is more normal than I thought.
Yours is quite extreme. 12th grade in your non-fiction and 5th in your fiction. 20 in your non-fiction and 80 in your fiction.
In direct comparison, my one piece of non-fiction that I tested was 11th grade and had a flesch score of 55. My fiction averages 4th/5th grade and 70-80 flesch.
So my non-fiction still has a long ways to go in its quest of unreadability.
posted
I must be a passive voice Nazi: 0% on everything I checked. My Flesch scores are consistantly 78-81%, and Grade Level of about 4.5 - 5.0, though the most recent chapter of a WIP spiked up to 7.2.
That raises an interesting question: should a novel be consistant reading level throughout, or is fluctuation to be expected?
[This message has been edited by TheoPhileo (edited July 09, 2005).]
posted
Yeah, the grade levels you get in Word max out at 12. The link supplied in the original post gives the following grade equivalences for the readability statistic:
90 to 100 5th grade 80 to 90 6th grade 70 to 80 7th grade 60 to 70 8th and 9th grade 50 to 60 10th to 12th grade (high school) 30 to 50 college 0 to 30 college graduate
It also gives some averages for different types of writing. Here is just some of the examples it gives:
Consumer ads in magazines 82 Seventeen 67 Reader's Digest 65 Time 52 Newsweek 50 Harvard Business Review 43 New York Times 39 New York Review of Books 35 Harvard Law Review 32 Standard auto insurance policy 10 Internal Revenue Code minus 6
After reading how it is calculated, I can see it might be tricky running articles and reports through it. If you use in-text citations (as is standard in linguistics), this will artificially lower your readability score. Removing citations boosts my scholarly article readability scores significantly.
posted
That's a good point. Bulleted text will also increase the difficulty resulting in a lower score because it reads the entire list as a single sentence (unless you put a period at the end of every bullet).
In general, essays and non-fiction will tend to score lower in terms of readability. In journalism, the goal is to write at an eighth grade level. Compared to alot of the fiction stats we're all listing, that is two or three grade levels higher.
Printer-friendly view of this topic
And if you're scoring in the same range as things that you find incomprehensible (because of language, not content), you might want to consider adjusting your usages a bit.
