Readability statistics - what limitations?

[Steve Jong/Lightbridge <Steve_Jong/Lightbridge*LIGHTBRIDGE -at- NOTES -dot- CREDTECH -dot- COM>]

Aussie Peter Dann <PJDann -at- VCRPMAP -dot- TELECOM -dot- COM -dot- AU> asked:

>> Can anyone point me to a book or other source which explains how these
>> statistics are derived, and which discusses their usefulness and
>> limitations?
>> Does anyone incorporate these statistics in their own documentation
>> standards, or in document specifications (for example, by setting a maximum
>> Flesch grade level for a particular piece of work)?

I would recommend going first to the source:

 Flesch, Rudolf. _The Art of Readable Writing, 25th Anniversary Edition,
Revised and Enlarged_.
 New York: Harper & Row, Publishers, 1974.

Rudolf Flesch describes his Flesch Index and how it was created.  Then I would
go to the single most important
critique of readability indexes for technical documentation:

 Redish, Janice C., and Selzer, Jack.  "The Place of Readability Formulas in
Technical Communication."
 _Technical Communication_, Vol 32, No. 4 (Fourth Quarter 1985), 46-52.

In my opinion, I would support the use of readability indexes as part of a
suite of quality metrics,
because I believe they are reasonably grounded and correlate positively to
readability,
but I would not advocate using them on their own to judge technical documents.
However, to answer your other question, yes,
some managers do believe in using them as a singular metric.
Watch out for them 8^)
          -- Steve
================================================================
Steven Jong, Documentation Specialist  ("Typo?  What tpyo?")
Lightbridge, Inc, 281 Winter St., Waltham, MA 02154 USA
<jong -at- lightbridge -dot- com>, 617.672.4902 [voice], 617.890.2681 [FAX]