Criteria for Accurate Oral Reading to Indicate Difficulties with Reading Decoding

Research has demonstrated that accurate oral reading data, as measured by percentage of words read correctly, supported intervention decision making. In this study, we examined potential criteria for interpreting oral reading accuracy data by relating them to nonsense word decoding as measured with Nonsense Word Fluency (NWF). Data from over 4,000 first- through third-grade students in one district were analyzed. The study found area under the curve estimates between oral reading accuracy and NWF of .91 (95% CI = .89 to .94) for first grade and .85 (95% CI = .83 to .87) each second and third grades. Receiver operating characteristic analysis resulted in accuracy cut scores of 83% for first grade, 93% for second grade, and 94% for third grade, but different criteria (e.g., 85% for first grade and 90% for second grade) were somewhat equally supported across the three grades. Implications and future directions are discussed.