READING ACQUISITION: COMPARING THE LONG-TERM IMPACT OF EARLY INTERVENTION PROGRAMS USING LISREL MODELS WITH STRUCTURED MEANS.
Item
-
Title
-
READING ACQUISITION: COMPARING THE LONG-TERM IMPACT OF EARLY INTERVENTION PROGRAMS USING LISREL MODELS WITH STRUCTURED MEANS.
-
Identifier
-
AAI8708300
-
identifier
-
8708300
-
Creator
-
LEON, ANDREW C.
-
Contributor
-
David Rindskopf
-
Date
-
1987
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Education, Tests and Measurements
-
Abstract
-
Two general approaches to compensatory education are used to minimize the number of children who are inadequately prepared for reading instruction. One is used to augment skills deficits. The other works less directly; it is designed to serve those considered at risk on the basis of their socio-economic status (SES). The objectives of this study parallel those approaches. First, the causal relationships among reading readiness and reading achievement abilities were examined in structural equation models of reading acquisition. Second, Head Start and No Preschool samples were compared on SES and reading achievement constructs using structural equation models with structured means.;The data came from the Longitudinal Study of Disadvantaged Children and their First School Experiences, a joint project of the Educational Testing Service and the Head Start Research Office. Five models of reading acquisition were tested. The observed variables were multiple observed measures of each construct, administered from one year prior to preschool through the third grade. Causal relationships among latent constructs of home background, pre-readiness, reading readiness, and reading achievement were hypothesized and tested in the models. The findings support the literature: there are reading readiness skills which facilitate reading achievement. Readiness assessment is shown to be valuable in accounting for individual differences in reading achievement.;One model compared the reading achievement of Head Start and No Preschool samples after controlling for SES differences. Head Start was shown to have no effect on achievement at a given level of SES. This result was compared with that of a more commonly used technique, analysis of covariance. In the latter analysis the groups were not compared using latent variables as in the other analyses, but instead using observed variables. In that analysis, Head Start was shown to have a detrimental effect. This finding appears to contradict that of the other analyses. Reasons for this difference are considered and advantages and disadvantages of the statistical techniques are discussed.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Educational Psychology
