mine the natural history of a disorder over time, to determine whether the problem is transient or chronic, and how various risk factors relate to outcomes. Earlier studies of representative school-age children are those by Rutter and Yule (1975) in the important Isle of Wight and London studies and, later, studies by Rodgers (1983) in Great Britain and Northern Ireland and by Silva et al. (1985)in Dunedin, New Zealand.
More recently, Shaywitz et al. (1990, 1992) have reported on the results of a sample survey of Connecticut schoolchildren followed longitudinally from kindergarten through high school, and Catts et al. (1997) have reported on reading difficulties in a representative sample of children in Iowa. Together, these studies provide the strongest basis for estimating the prevalence of reading difficulties in childhood. It is also of interest, of course, to compare estimates of reading problems from studies like the National Assessment of Educational Progress (NAEP) and the Prospects Study to those from prospective sample surveys.
In identifying, studying, and treating reading problems, two main kinds of reading difficulties have traditionally been distinguished. Reading disability, also called "dyslexia" and "specific reading retardation," was at first considered to be a qualitatively and etiologically distinct condition that an individual either had or did not have. The condition was viewed as having a biological and perhaps genetic basis, as being invariant over time, and as affecting a small group of children, primarily boys.
A key criterion for identifying dyslexia was the existence of a substantial discrepancy between the child's aptitude (operationalized as IQ) and his or her achievement, reflecting the assumption that the reading problems of a bright and otherwise capable youngster are different in nature from those of a child who is generally less able to cope with schoolwork. In this traditional conceptual model, poor readers who do not meet the criteria for a reading disability are characterized instead as having garden-variety reading problems (or "general reading backwardness"), arising from such causes as poor