| Despite significant regulations regarding the reading proficiency of students in the United States, more than half of students graduating high school are continually reading below proficiency (National Center for Education Statistics, 2010). In 2000 the National Reading Panel (NRP) identified phonemic awareness, phonics, fluency, vocabulary and comprehension as the five main areas of reading development. As a result of the NRP's findings and the low rates of literacy, No Child Left Behind (2002) mandated that all students are proficient readers by 2014 and that proficiency is measured on a yearly basis by state mandated high-stakes assessments, beginning in Grade 3. This study presents findings of statistical analysis examining the relationship between early literacy benchmark data in kindergarten and Grade 1 and statewide high-stakes achievement tests taken two to seven years later. One hundred thirty kindergarten students, ages five to six in 2002, were followed through 2010 and were included in this study sample. Results indicated that the DIBELS demonstrates moderate predictability when estimating future PSSA performance, based upon kindergarten and Grade 1 winter benchmark testing. Oral Reading Fluency at the winter Grade 1 benchmark moderately predicted PSSA Reading domain performance in Grades 3 and 5. Letter Naming Fluency at the winter kindergarten benchmark moderately predicted PSSA Reading domain performance in Grade 7. Diagnostic accuracy of the DIBELS, using ROC analysis, was more acceptable than the diagnostic accuracy of the DIBELS using the recommended cut scores when screening for the performance on the PSSA Reading domain. These findings imply that the assessment of early literacy skills, as early as the winter of kindergarten, predicts future reading performance up to seven years later. Implications of these findings and recommendations for future research were discussed. |
