LINGUIST List 8.228

Tue Feb 18 1997

Sum: Readability

Editor for this issue: Ann Dizdar <>


  1.>, readability responses

Message 1: readability responses

Date: Wed, 12 Feb 1997 09:44:27 GMT
From:> <>
Subject: readability responses

From: <>

Thanks to the following who responded to my question (LINGUIST List
Vol-8-141) about "text readability formulae which do not employ
sentence (or word) length":

- ---------------------------------------------------------------
"Arendse Bernth" <>

Also interested in answers to this question!

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

Who has created: "a general formality index based primarily on
frequencies of sets of common wordforms, but without explicit
reference to word length or sentence length.". Potential problem:
"index tends to treat conversation as highly informal.".

- ---------------------------------------------------------------
From: (Kurt Godden CS/50)

Knows none, but would welcome info on this topic.

- ---------------------------------------------------------------
From: David Lee <>

Confirmed my suspicions and has not encountered any that are
essentially mathematical formulae. Hovever, he mentioned, "Other
possible ways of looking at 'text readability'" such as
 (i) type-token ratio (which might measure "one tiny aspect of it")
(ii) ratio of 'content words' to 'function words' (giving only
"relative 'density' of the lexis").
(iii) ratio of 'core vocabulary' to the total no. of words. (Something
he's been working on.)

Apparently 'core vocabulary' is not clearly defined. "However, a
rough- and-ready list of words can be drawn up and used once you've
settled on your preferred definition of 'coreness'."

Our colleauge used "the list of defining vocabulary included at the
back of the Longman Dictionary of Contemporary English. This is a
controlled-vocabulary word list (about 2,000 words) based on various
criteria such as frequency, unambiguousness, usefulness for
definitions, etc. This list (word+part-of-speech) was operationalised
and used to rate (very roughly) how difficult the texts of the British
National Corpus are in terms of lexis.".

- ---------------------------------------------------------------
From: Koutsomitopoulou Elenh <>

Mentioned searching through psycholinguistic theories of text
interpretation, considering a cognitive approach for a "broader view
of a text (e.g. Kintch & van Dijk work in the '80s).", and
textlinguistics. DeBeaugrande's (1983?) 'text communication'article
was also mentioned.

- ---------------------------------------------------------------
From: Dick Hudson <>
 home page =
 unpublished papers available by ftp =

Had no information, but was interested in what I receive. He is
working in this area. And he reminded me about summarising back to the

Also mentioned was the survey article `Readability' in the
Enclyclopedia of Language and Linguistics (ed. Asher). "It gives quite
a lot of formulae which go beyond sentence-length."

- ---------------------------------------------------------------
From: "James L. Fidelholtz" <>

Did not know of any formulae, but said that they are working "on a
project to develop a readability formula for Spanish (specifically, we
are working with physics texts)," and wants to know if I hear of
anything else.

- ---------------------------------------------------------------
From: Stuart Luppescu <>

Just did a dissertation that relied on the lexile readability formula
(which "is weighted rather heavily on the log of the mean sentence
length"). He reckons that "sentence length is a rather good predictor
of passage difficulty.", and expressed interest in any readability
formula that do not use it.

Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue