Catalog Home Page

Prediction and prevention of reading failure in first grade children

White, Margaret Reynolds (1989) Prediction and prevention of reading failure in first grade children. PhD thesis, Murdoch University.

PDF - Whole Thesis
Available Upon Request


The present thesis is concerned with prediction and prevention of reading failure, in particular, prediction and prevention prior to the commencement of formal reading instruction. This necessarily involves the identification of reliable predictors of reading failure in the prereading child, so that effective early intervention can alleviate later difficulties for children who are at risk of reading failure. A review of the literature and techniques of preschool screening revealed many theoretical and methodological problems. Inadequate longitudinal evaluation of predictive validity, inadequate assessment of predictive utility, and unacceptably high false positive or false negative rates were cited frequently. Despite these problems, some predictive batteries have demonstrated that they are more effective than observational methods, in early identification of the children who are most at risk.

The first study in the present research was an Australian cross validation of the then-unpublished Satz Predictive Test Battery (Satz, Taylor, Friel and Fletcher, 1976). This battery was selected for its close attention to underlying theory and its methodological strengths. Satz hypothesises that children who have difficulty acquiring early reading skills have a neurodevelopmental lag, primarily affecting the left cerebral hemisphere, which deferentially delays their acquisition of these skills. In testing his theory, Satz paid particular attention to methodological weaknesses observed in other studies. He therefore tested the entire white, male, kindergarten population of Alachua County, Florida, and conducted follow-up studies over six years, in order to assess the predictive validity and utility of the test battery at each age level. The Australian cross validation involved assessment of reading risk for 407 five- year-old children, from a predominantly lower-middle socioeconomic group in Western Australia. The predictive efficiency of the tests was assessed by measuring the children's reading competence at the end of first, second and third grades, and relating their actual scores to their predicted performance. Data indicated that the accuracy of the Australian predictions was consistent with those reported in the original experimental studies at the University of Florida. Cross validation data confirmed the finding of the Florida studies that predictive accuracy was high in the extreme groups, and lower in the less extreme groups. The Australian data also confirmed that tests were more effective than teachers in successfully identifying children with the highest risk of failure. Factor analysis of predictive and criterion variables employed in both the Florida and Australian studies, suggested a sensory-motor-integrative factor as the principal factor discriminating between successful and unsuccessful beginning readers.

Two interventions which aim to enhance sensory-motor-integrative functioning were therefore selected for evaluation. The first intervention, Sensory Integrative Therapy (SIT) (Ayres, 1972a) was administered to the experimental group of children identified as being at risk of failure . The results of this intervention, assessed at the end of first grade, revealed significant advantage in reading competence for the treatment group of children receiving SIT. Their advantage over the control group was maintained for a further two years, without further intervention. By contrast, the second intervention - A Motor Integration Programme (MIP) (Lefroy, 1975) - did not result in any consistent advantage for the treatment group of five year old children. The third stage of the present studies sought to identify alternative predictors which could be expected to contribute to more effective prediction. A predictive model with a different theoretical base was therefore investigated. The Savage model focuses on learning level and speed as critical factors in early reading success (Savage, 1979), and it is reasonable to expect that these factors could have predictive value. A further group of school-entry children was therefore tested, using the Savage Test Battery. The new parameters contributed by the Savage tests did not, however, improve successful identification rates significantly.

It was concluded that the Satz Predictive Test Battery is useful in identifying children who are at risk of reading failure, and intervention with a programme of Sensory Integrative Therapy appears beneficial to later reading levels of some of these identified children. The problem of false positive and false negative predictions still persists, and further refining of the test battery is necessary, if children are not to be labelled incorrectly and given interventions which are unnecessary. However, data from both the American and Australian studies indicate that predictive tests have value for identifying the children with the highest risk of failure. Teachers' predictions are consistently successful in predicting good readers, but are less successful than the tests in predicting children who are at risk of failure. Within the limitations of this small study, intervention with a programme of Sensory Integrative Therapy appears beneficial for five-year-old children who are identified as being at risk by the Satz Predictive Test Battery. Replication with larger numbers, and in other settings, would allow present findings to be generalised over other populations of children at risk of reading failure.

Item Type: Thesis (PhD)
Murdoch Affiliation: School of Social Sciences
Notes: Note to the author: If you would like to make your thesis openly available on Murdoch University Library's Research Repository, please contact: Thank you.
Supervisor(s): Savage, Doug
Item Control Page Item Control Page