Investigating the Validity and Fluency Threshold of the Decoding Proficiency Assessment Tool
2006 to 2007

CCT will conduct two small studies to examine student performance on Scholastic's Decoding and Proficiency Assessment (DPA). The first study will determine the degree to which there is a threshold for accuracy and speed for students at three different grade levels and one adult sample. Scholastic will provide CCT with source code necessary to administer the DPA prototype and resultant data collected at local school sites. Data will be analyzed within grade-level sub-samples. Basic descriptive statistics for each item will be generated. Response time files will be analyzed to determine average response time, deviations from the average, and the relationship to accuracy; response patterns in terms of incorrect responses and distracters; and different strands of the scope and sequence. The response time analyses will provide preliminary findings that can feed into a determination of appropriate cutoff points for fluent responses on the DPA. The second study will examine the validity of the DPA in relation to other existing, similar, and established measures for students at two identified grade levels. A primary selection criterion set out by Scholastic is to identify students who have shown evidence of poor decoding skills. CCT will administer three measures to each student, including the DPA prototype. Results from the three tests will be analyzed and summarized with a focus on the inter-correlations among the three measures, computed within grade level.


Ellen Mandinach (PI)
Harouna Ba
Merav Dechaume