Home
Resource
Investigating the relationship between reading achievement, and state-level ecological variables and educational reform
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(827.09 kb)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

Yurecko, Michele. Investigating the relationship between reading achievement, and state-level ecological variables and educational reform. Retrieved from https://doi.org/doi:10.7282/T39G5N1X

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

TitleInvestigating the relationship between reading achievement, and state-level ecological variables and educational reform
NameYurecko, Michele (author); Camilli, Gregory (chair); Kuhn, Melanie (internal member); Penfield, Douglas (internal member); Twing, Jon (outside member); Rutgers University; Graduate School - New Brunswick
Date Created
Other Date2009-01 (degree)
SubjectEducation, Reading (Elementary), Academic achievement
Extentx, 125 p.
DescriptionThis study identifies profiles in fourth grade reading achievement across states as measured by the 2002 National Assessment of Educational Progress (NAEP), and examines the link between these profiles and state policy and ecological variables. A series of multilevel models (MLM) that extend procedures traditionally employed in the analysis of differential item functioning (DIF) were applied in order to evaluate state-level performance at the item level. The variability of states' performances on individual items was estimated while controlling for overall state reading proficiency, and a residual variance statistic, item difficulty variation (IDV), was estimated for each item. A subset of items with relatively large IDVs was included in a second tier of analyses, and aggregated "parcel scores" were estimated. These parcel scores represent empirical clusters of items for which state membership influences student performance beyond what would be expected given state-level reading proficiency estimates. Two parcel scores were constructed in this analysis. Parcel 1 represented a cluster of items associated with long, fictional reading passages. Parcel 2 represented a cluster of items associated with short, non-fictional reading passages. Parcel scores can be interpreted as value-added scores, suggesting that states with high Parcel 1 scores performed better on items associated with long, fictional passages than would have been predicted by overall state reading proficiency. Similarly, states with low Parcel 1 scores performed worse than expected. The interpretation of Parcel 2 follows in kind. As rates of non-native speakers and poverty increased across states, scores on Parcel 1 decreased and scores on Parcel 2 increased. These quantitative results, coupled with a qualitative case study of Maryland, New York and Texas, suggest the following major theme: larger, more populous states, with higher levels of poverty and non-native speakers of English exhibit a distinctive pattern in parcel score performance, scoring lower than predicted on items associated with long, fictional reading passages, and higher than predicted on items associated with short, non-fictional passages.
NotePh.D.
NoteIncludes bibliographical references (p. 120-123)
Noteby Michele Yurecko
Genretheses, ETD doctoral
Persistent URLhttps://doi.org/doi:10.7282/T39G5N1X
Languageeng
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024