Date of Award
Doctor of Philosophy
Richard L. Allington
Anne McGill-Franzen, Vincent Anfara, Amy Broemmel
In recent years, increased attention has been paid to accelerating the development of struggling young adolescents’ reading skills (Franzak, 2006). It has been widely acknowledged that these students require intensive instruction in reading in order to meet changing societal demands (Allington, 2002; Afflerbach, 2004; Alvermann, 2001; Biancarosa & Snow, 2004). Score reporting from the National Assessment of Educational Progress (NAEP) may demonstrate a dichotomy amongst our young adolescent readers, but the scores do not tell us about the specific needs of individual students. In other words, these levels essentially create two groups: those who can read and those who cannot. Further, instructional decisions are being made based on the limited proficiency scores of state mandated standardized assessments. This method of reporting scores creates a notion of homogeneity amongst the reading skills of young adolescents.
The purpose of this multivariate correlational study was to determine the patterns of reading abilities amongst struggling young adolescent readers in an attempt to demonstrate the heterogeneous nature of these students and the variability of reading skills they bring to middle school classrooms (grades 6-8), in an effort to influence both policy and instruction at this level. Data were collected during the 2005-2006 academic year. Each student participant (n=94) was administered five assessments that measured alphabetics (phonemic awareness and phonics), fluency, vocabulary, and comprehension, which were representative of both the highly and less constrained skills (Paris, 2005) presented as essential components of reading instruction by the National Reading Panel (NRP, 2000). Independent samples t-tests were used to compare the assessment means of several subgroups, students who qualified for special education, free and reduced price lunch, and English Language Learner services, and those who did not qualify for these services. Results indicated all of these students scored below grade level on the assessments administered. However, all of the students represented varying abilities and needs that required further analysis. Factor analysis was then utilized to determine which reading skills assessed were most directly related to student performance on TCAP. Three factors emerged, meaning, decoding, and rate and accuracy. Finally, cluster analysis presented four distinct clusters of struggling young adolescents, which represented heterogeneous abilities in various reading skills.
Results indicated one-size-fits-all approaches to policy and instruction relating to struggling young adolescent readers do not meet the heterogeneous needs of this population of students. Rather, in-depth assessment and diagnoses are necessary to determine the most appropriate instructional tools for individual students. Further, by suggesting the use of state mandated standardized assessment scores be the sole indicator of student placement in remedial reading courses, policy fails to address the multifaceted process of reading and the differing trajectories of young adolescent reading development.
Dennis, Danielle V., "Putting the Horse Before the Cart: Utilizing What Assessment Data Reveal About Struggling Young Adolescent Readers to Inform Policy and Instruction. " PhD diss., University of Tennessee, 2007.