Abstract
This paper introduces a Dutch readability formula and readability tool,
LiNT (Leesbaarheidsinstrument voor Nederlandse Teksten). We f irst
relate readability to the broader notion of comprehensibility, and discuss
classic problems of readability research. The f irst and main focus of
the paper is the LiNT-formula. We discuss how we prepared 120 texts,
collected cloze comprehension data for these texts among 2700 middle
school students, extracted text features and derived the LiNT-formula.
This formula relates cloze comprehension to four text features, all of
which have received support in earlier text comprehension research: word
frequency, word concreteness, clause length and dependency length. We
next discuss the interpretation of LiNT-scores, and the text complexity
levels defined by it. We then examine ten well-known Dutch text genres
in terms of LiNT-scores and LiNT-feature scores; and we report how text
revisions affect both LiNT-scores and text comprehension levels. Our
second topic is a short introduction to the LiNT-tool built around the
formula. The tool aims at providing text complexity metrics to com-
munication practitioners. Finally, we discuss whether our formula and
tool succeed in avoiding the classic readability research problems, and
how they may help to assess readability, diagnose readability problems
and choose between text versions.
LiNT (Leesbaarheidsinstrument voor Nederlandse Teksten). We f irst
relate readability to the broader notion of comprehensibility, and discuss
classic problems of readability research. The f irst and main focus of
the paper is the LiNT-formula. We discuss how we prepared 120 texts,
collected cloze comprehension data for these texts among 2700 middle
school students, extracted text features and derived the LiNT-formula.
This formula relates cloze comprehension to four text features, all of
which have received support in earlier text comprehension research: word
frequency, word concreteness, clause length and dependency length. We
next discuss the interpretation of LiNT-scores, and the text complexity
levels defined by it. We then examine ten well-known Dutch text genres
in terms of LiNT-scores and LiNT-feature scores; and we report how text
revisions affect both LiNT-scores and text comprehension levels. Our
second topic is a short introduction to the LiNT-tool built around the
formula. The tool aims at providing text complexity metrics to com-
munication practitioners. Finally, we discuss whether our formula and
tool succeed in avoiding the classic readability research problems, and
how they may help to assess readability, diagnose readability problems
and choose between text versions.
Original language | English |
---|---|
Pages (from-to) | 2-39 |
Number of pages | 38 |
Journal | Tijdschrift voor Taalbeheersing |
Volume | 45 |
Issue number | 1 |
DOIs | |
Publication status | Published - Dec 2023 |