> Can someone who knows more about readability scores than I do help me
> out?  I am working with a number of pre-postsecondary students who are
> struggling with their biology text.  When I ran the readability check
> provided by MSWord, I got these scores:  Flesch Reading Ease:  49.4 ;
> Flesch Kincaid Grade level 10.8.
> I was surprised, since my assessment of the text was that it was much
> more difficult than an eleventh grade reading level.  I would guess
> there are other factors that these scales do not measure? Is there a
> good reference that I could use to fill in my own knowledge gaps as to
> how these and other scales compare?
Although I'm still trying to figure out the difference between a
"secondary" student and a "pre-postsecondary" student, the issue of
Flesch-Kincaid scores came up fairly recently in this listserv.  The
following is a partial copy of a message I posted April 24th of this
year -- I hope it is of value to you.

- --- -- --- -- ---

If I understand Flesch's work (and note that I'm a math teacher, so I'm
not particularly well versed in this topic), he was attempting to come
up with a simple, mathematical formula that could give an indication of
the readability of a text.  His result (modified by Kincaid):  [(11.8 *
average_syllables_per_word) +  (0.39 * average_words_per_sentence) -
15.59] was based on taking many different existing texts and studying
their sentence and syllabic length.  Flesch-Kincaid scores are just an
example of what we mathematicans call curve fitting.  If you plot the
grade a teacher is using a text in as the X-axis, the average number of
syllables per word as the Y-axis, and the average number of words per
sentence as the Z-axis, you get a three dimensional point that
represents that book.  Repeat this for hundreds of books and you have
points all over the place, but there is a trend line which corresponds
to the formula above.

To summarize, knowing a Flesch-Kincaid score is 6.9 is about as useful
as knowing a student's GPA is 3.2.  Distilling a book (or a student) to
a single number leaves much to be desired, but still has value for such
things as initial screenings.

Some have attempted to come up with better scoring devices.  I found
this site ( via a brief Yahoo
search, but I have no idea whether ATOS is better or worse than

Prof. Eric Kaljumagi
Mt. San Antonio College