Literacy Assessment and Data-Based Decision Making
Assessment items ask you to match an instrument to a purpose, then interpret scores without over-claiming. Expect scenarios giving several data points and asking what the specialist should do next.
Four purposes, four kinds of tools
- Screening: brief, class-wide, identifies who is at risk. It does not diagnose.
- Diagnostic: pinpoints the gap - phonics inventory, phonemic awareness tasks, spelling inventory, oral reading analysis.
- Progress monitoring: short, repeated, equivalent forms, sensitive to weeks of growth (oral reading fluency, letter-sound fluency, maze).
- Outcome/evaluation: annual standardized measures used for program decisions.
Reading the numbers
- Standard scores center on 100 with a standard deviation of 15; roughly 85-115 is the average band.
- A percentile rank is the percentage of the norm group scoring at or below that score - it is not percent correct.
- The standard error of measurement means every score is a band; a small change may be noise.
- Informal reading inventory levels are conventionally independent at about 95-100% accuracy, instructional at 90-94%, frustration below 90%.
Worked example
Oral reading fluency probes show 44 words correct per minute in fall, 47 in winter, 50 in spring. The goal slope was 1.5 words per week, so 24 weeks should have produced about 36 words of growth; actual growth is 6. Reasoning: the trend line is flat, so the intervention is not working. The defensible move is to change the intervention or grouping now, not to wait. Accuracy and comprehension should also be checked, because a fast, inaccurate reader is not fluent.
Traps that cost marks
- Changing placement on the basis of one probe.
- Treating grade-equivalent scores as equal-interval growth.
- Equating percentile rank with percent correct.
- Letting a running record's accuracy figure stand alone when the reader is guessing from pictures and context.