Reading and Writing: An Interdisciplinary Journal 12: 31–40, 2000.
© 2000 Kluwer Academic Publishers. Printed in the Netherlands.
Investigating the causes of reading comprehension failure:
The comprehension-age match design
, JANE OAKHILL
& PETER BRYANT
Experimental Psychology, Sussex University, UK;
Experimental Psychology, Oxford
Abstract. The reading-level (or reading-age) match design has become a widely-used tool for
investigating the possible direction of the relation between particular skills and word reading
ability: Cause or consequence. This paper outlines an analogous method for identifying can-
didate causes of reading comprehension failure, the ‘comprehension-age match design’ and
discusses the strengths and limitations of this design.
Keywords: Ability-match designs, Causal relations, Reading comprehension, Word reading
Children’s failure to understand written text can arise at different stages in
the processing of language. For example, children with impoverished vocab-
ularies and/or those who experience difﬁculties in word reading can fail to
understand sentences and longer pieces of text (e.g. Liberman & Shankweiler
1991; Perfetti 1985). In addition, there exists a group of children who demon-
strate reading comprehension problems in the presence of age-appropriate
word reading and vocabulary skills (e.g. Oakhill 1994; Stothard & Hulme
1996). Oakhill and colleagues have shown that such children experience
problems with a range of higher-level reading-related skills, such as infer-
ence making, comprehension monitoring, and anomaly resolution (de Sousa
& Oakhill 1996; Oakhill 1982, 1984; Yuill, Oakhill & Parkin 1989).
However, the ﬁnding that skilled comprehenders make, for example, more
inferences than less skilled comprehenders is essentially correlational. It does
not tell us anything about the direction of the relation between these two
skills. It may be that the skilled comprehenders’ greater experience of reading
and understanding stories has strengthened their inferential skill, relative to
that of the less-skilled comprehenders. Alternatively, it may be that the less-
skilled comprehenders’ poorer inferential skill has (at least in part) caused
their poorer comprehension.