Reliability of DIBELS
Some important concerns are:
- DIBELS cannot be scored consistently. The tester must time responses ( three seconds on a stopwatch), mark a score sheet, and listen to the student all at the same time.
- DIBELS does not test the reading quality. No test evaluated what the reader comprehends.
- DIBELS misrepresents students' abilities. Children who already comprehend print are undervalued, and those who race through eat test with no comprehension are overrated.
- The focus on improving performance on DIBELS is likely to contribute little or nothing to reading development and could actually interfere. It pushes students to do everything fast.
Administrative considerations meant to ensure reliability of DIBELS results:
- Brief, repeatable measures can be aggregated to increase reliability.
- When more reliable information is needed, the average of 3 to 5 probes on different days can be used.
- For each measure, note how many probes would be necessary to reach .90 reliability.
- Even the least reliable DIBELS measure, Initial Sound Fluency, yields a reliability for .90 when administered 4 or 5 times for an approximate total of 5 minutes of assessment. Therefore the reliability of 5 minutes of Initial Sound Fluency would be .90
- This differs from the Woodcock-Johnson, which must reach accepted reliability in one assessment because it is not brief and repeatable.