Reading level design: Conceptual and methodological issues in reading research.

Abstract

Researchers have recently begun to use the reading level design, in which readingdisabled children are compared with younger normal children at the same level of reading achievement, in an attempt to overcome some of the interpretational problems in the field. The potential of this design for testing hypotheses is explored by using examples from current research, and a more general design is introduced which extends the utility of the design from studying reading disability to investigating normal or superior reading acquisition. Limitations and practical problems associated with this approach are discussed.

The /O/ in OVER is different from the /O/ in OTTER: Phonological effects in children with and without dyslexia.

“…The 40 remaining children were without reading problems and attended a regular primary school. These children were matched to the children with dyslexia on either chronological age (age-match group, 20 children), or on word-reading level (reading-match group, 20 children; see Backman, Mamen, &Ferguson, 1984, andRack et al, 1992).…”

Is a cerebellar deficit the underlying cause of reading disabilities?

“…As Backman, Mamen, and Ferguson (1984) have argued, in many studies on dyslexia, such as some of those reviewed here (e.g., Fawcett & Nicolson, 1995cFawcett et al, 2001;Moe-Nilssen et al, 2003;Nicolson & Fawcett, 1990;Raberger & Wimmer, 2003;Wimmer et al, 1999;White et al, 2006), normal readers have been used as a control group and matched to a dyslexia group on chronological age and sometimes also on overall level of intellectual functioning. Differences found between the groups on the measures in question have been then assumed to reflect the underlying causes that explain difficulties in reading (Backman et al, 1984). Nonetheless, the positive results obtained from these studies (i.e., overall poorer performances of children with dyslexia as compared to the control group) are difficult to interpret and may not really reflect causes of reading difficulties (Bryant & Goswami, 1986), due to the other possible between-group differences that have not been controlled (Backman et al, 1984).…”

“…Differences found between the groups on the measures in question have been then assumed to reflect the underlying causes that explain difficulties in reading (Backman et al, 1984). Nonetheless, the positive results obtained from these studies (i.e., overall poorer performances of children with dyslexia as compared to the control group) are difficult to interpret and may not really reflect causes of reading difficulties (Bryant & Goswami, 1986), due to the other possible between-group differences that have not been controlled (Backman et al, 1984). One major difference, as Bryant and Goswami (1986) have pointed out, is the fact that the groups compared in these studies differ on their reading level, such as in MoeNilssen et al"s (2003) study described earlier or some of the other studies reviewed.…”

“…For example, the choice of reading test used to match the groups was not evident. Reading is generally considered to be a complex phenomenon and performance on even a simple task of accurate word reading can reflect complex cognitive central processes such as the activation of spelling rules for decoding or word-specific information as well as possibly more peripheral processes such as attention (Backman et al, 1984;Jackson & Butterfield, 1989). The processes underlying reading can become even more complex when tasks such as text level reading or reading comprehension are used for matching the groups (Jackson & Butterfield, 1989).…”

A Comparison of Analogy-and Rule-Based Decoding Strategies in Normal and Dyslexic Children

“…The use of an RA control group has gained increasing acceptance in the literature on reading disabilities. The advantage of such a design is that it controls for differences in experience with print and in relative performance levels that inevitably occur when dyslexic children are compared to CA controls on any reading-related measure (Backman, Mamen, & Ferguson, 1984). In the present study, this design made it possible to determine whether dyslexic children were qualitatively distinct in their decoding strategies from young normal readers or whether their reading behavior was essentially normal, given their level of reading experience and knowledge.…”