One-minute Fluency Measures - Skewed Views

I just finished reading “One-Minute Fluency Measures: Mixed Messages in Assessment and Instruction” by Theresa A. Deeney in the March 2010 issue of “The Reading Teacher”. She has some excellent points, among them:

One-minute fluency measures (such as the DIBELS Oral Reading Fluency or ORF and AIMSweb) only measure rate and accuracy, but they don’t measure what are arguably two other very important aspects of fluency: prosody, or what we’ve traditionally called expression, and comprehension, which has a recursive relationship with fluency in that readers who comprehend what they read usually end up sounding fluent, and readers who sound fluent are more easily able to comprehend their own reading. Since traditional short fluency measures don’t measure prosody and comprehension, our working definition of what fluency actually is has gotten skewed. She argues that we’re confusing fluency assessment with fluency instruction, and because we’re focusing only on rate and accuracy, that’s what we end up teaching as well. In an effort to increase the scores on these quick assessments, we give students more short texts to practice reading quickly and accurately. Students end up with a skewed message about what reading is and how to know when you’re good at it.

Her other point about one-minute fluency measures is that they don’t take into account the need to develop reading endurance in students. She demonstrates this using the readings of eight students reading at least one year below grade level. Each of these students was able to read within the recommended guidelines for accuracy and speed on a one-minute reading. But when asked to read for longer periods (4-5 minutes) most of the students read much slower and much less accurately. Her point is that we’ve helped improve readers’ speed and accuracy on short texts at the expense of these same qualities on longer texts. Some students were able to keep their accuracy fairly high over time, though their speed decreased. These students most likely needed practice reading longer texts and developing stamina. However, some students decreased both in their speed and their accuracy, indicating the problem might lie deeper. The reader has evidently not become automatic in her word recognition and would need help with either sight words or decoding unfamiliar words. To determine this we have to go deeper than typical one-minute assessments allow, and do a miscue analysis on the selection.

One-minute fluency assessments have their place – they allow us to see which students need further work on rate and accuracy. But they don’t tell us how the reader does on longer texts or why the student might be having difficulty. As always in education, we have to find a balance. In this case, we need a balanced view of the definition of fluency and the causes of dysfluency.