ABSTRACT
Wybrow & Hanley (2015) reported that proportions of phonological and surface dyslexics change depending on how control groups are selected. This observation questions the appropriateness of the reading-level match design for establishing causality in cognitive studies of reading. Here, I focus on three features: (1) the lack of an explicit definition of the reading-level concept; (2) the metric problems associated with using this design; and (3) the ambiguity of the delay-deviance contrast in interpreting reading deficits. I also delineate alternative methodological features that could effectively inform developmental designs. Thus, I argue that (a) control variables should be as independent of the target-dependent measure as possible; (b) they should be shaped within the theoretical aims of the study and be explicitly considered in the interpretation of findings; and, (c) conditions of interest should be viewed along with critically associated conditions using approaches that allow predicting the size of the expected deficit..
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 In the perspective presented here, it should be added that the categorizations used in international manuals (such as “dyslexia”) are typically underspecified in cognitive terms and therefore present limitations similar to those outlined above for the concept of “reading level”. An extensive discussion of this issue is beyond the scope of the present paper. At any rate, it can be observed that these categorizations refer to clinical aims and the measures used are taken as indications of the presence of a behavioural problem to be further specified through cognitive analysis.