Growth Trajectories in Oral reading Fluency in Semi-Transparent Orthography

Growth Trajectories in Oral reading Fluency in Semi-Transparent Orthography

First Author: Anne Arnesen -- University of Oslo
Additional authors/chairs: 
Johan Braeken; Scott Baker; Wilhelm Meek-Hansen; Terje Ogden; Monica Melby-Lervåg
Keywords: Reading, Fluency, early identification, Assessment, Growth Modeling
Abstract / Summary: 

Introduction: Development of good reading skills are critical for success in school and society, but many students struggle with this. Evidence from English speaking students shows that oral reading fluency is a bridge between decoding and comprehension. Thus, in the US, oral reading fluency (ORF) is among the most used valid measure to screen early readers and to identify struggling readers and monitor their development. However, in more transparent languages there is a lack of studies regarding the validity of ORF.
Method: We examined an adaption of the ORF into a Norwegian semi-transparent orthography in 2228 students in grades 2 to 5 during one school year. We used structural equation modeling to examine longitudinal measurement invariance of ORF, growth in ORF within and across grades, stability of ORF, and the relationship between ORF growth and high stakes national tests of reading proficiency (NTRP).
Results: Results showed that ORF measured the same underlying construct, but some passages stood out regarding the invariance pattern. The ORF growth curves demonstrated a linear growth in grades 2 and 3, and a nonlinear growth in grades 4 and 5. Initial individual differences varied more than growth rates which for all were positive, but largest in grades 3 and 4. High stability in ORF was found. In grade 2, concurrent relations between ORF and NTRP were slightly larger (ORF accounted for up to 56% of the variance) than predictive relations (ORF accounted for up to 48% of the variance). In grades 3 to 5 the concurrent and predictive relations between ORF and NTRP were equal, which ORF accounted for up to 56% of the variance in grade 3, 29% in grade 4 and 30% in grade 5.
Conclusion: The study shows that ORF seems to be a valid and useful assessment tool also in more transparent orthographies. In future studies ORF should be validated against a larger battery of individually administered diagnostic reading tool to examine the ORF's sensitivity and specificity.