"Improve reading fluency by the end of the year." Nobody can mark that goal as met. There is no skill to count, no number to reach, and no way to say when to check.
This guide covers the parts of a measurable goal, then three ways to choose the number. The first step, a baseline, is in how to set an IEP baseline.
What the rule asks for
Section 300.320(a)(2) requires a statement of measurable annual goals, academic and functional, designed to meet the child's needs from the disability. Section 300.320(a)(3) requires a description of how progress toward those goals will be measured. A goal with no number cannot do the second job.
Three parts of a goal
Utah's technical guide says each annual goal has a condition, a target skill or behavior, and a criterion. Anyone reading it should be able to see what the student will do and how progress will be measured.
| Part | What it says | Example |
|---|---|---|
| Condition | Where, with what material | When given a grade 3 reading passage |
| Target skill | What the student does, in active words | Maya will read aloud |
| Criterion | How well, how often, by when | 77 words correct per minute by the spring benchmark |
NCII's template is similar: when given the grade level and tool, the student will do the observable behavior at a level of proficiency within a time frame. Its sample goal is: "When given a standardized third-grade level reading probe, Michael will read 99 words read correctly in 1 minute by spring benchmarking."
NCII also offers a quick test. Is it quantifiable? Can it be observed? Can data be collected systematically? Can the collected data be graphed?
Choosing the number: three options
NCII says each option works for goals built on general outcome measures, such as oral reading probes, and that the IEP team should decide which to use. It recommends presenting each option to the team.
| Option | How you get the number | Best when | Watch for |
|---|---|---|---|
| Benchmark | Use the end-of-year or mid-year benchmark for your tool | Student is at or near grade level | Too far a stretch for students far below it |
| National rate of improvement (ROI) | ROI times weeks left, plus baseline | The benchmark is out of reach this year | Typical growth may not close a gap |
| Intra-individual | Student's own slope times 1.5, times weeks, plus baseline | Student is far below grade level | Needs six to nine data points first |
Option 1: benchmark
Look up the benchmark for your tool at the end of the year. NCII notes that most published progress monitoring tools supply benchmarks in their data systems. Before you commit to one, calculate the weekly growth the student would need. NCII says to compare that rate with typical rates of improvement from national norms to see if the benchmark is appropriate.
Option 2: national rate of improvement
The formula is: ROI × number of weeks + baseline score = goal. NCII's example is a math probe: 2 digits per week × 10 weeks + a baseline of 30 digits = a goal of 50 digits. Weeks with no instruction, such as breaks, are not counted.
With invented numbers: a baseline of 47, a published ROI of 1.0 word per week for the tool and grade, and 30 weeks of instruction gives 47 + 30 = 77 words correct per minute. Use your own tool's published ROI, not this one. Rate of improvement has its own guide: rate of improvement explained.
Option 3: intra-individual
For a small group of students with intensive needs, NCII says benchmarks or national ROI may give unrealistic goals. The alternative uses the student's own growth. Collect six to nine data points to find the slope, then use: Goal = slope × 1.5 × weeks + baseline score.
NCII's worked example: scores of 2, 3, 5, 5, 5, 6, 7, 4 over eight weeks. The median of the first three is 3, and of the last three is 6. The slope is (6 minus 3) divided by 7, or 0.43. Multiply by 1.5 for 0.645, then by 10 weeks left for 6.45, then add the baseline of 6. The goal is 12. NCII notes the multiplier of 1.5 means 50 percent more growth than the student is making now, and that a team may choose another multiplier.
Checking whether the number is realistic
Take invented numbers. A student's baseline is 47 words correct per minute, the benchmark you are considering is 100, and 30 instructional weeks remain. The growth needed is (100 minus 47) divided by 30, or about 1.8 words per week. If the published rate of improvement for your tool and grade is 1.0 per week, the benchmark asks for nearly twice typical growth.
That does not make the benchmark wrong. NCII says more than typical growth is often needed to close a gap. But it is a number the team should see and discuss, along with what extra instruction would support it. Writing down the weekly growth needed next to the goal also gives you a ready-made aimline later.
Writing the criterion so it can be checked
- Say how many times. NCII lists "three consecutive probes" as an example of a proficiency level. Utah gives "on two formative assessments and one summative assessment per trimester."
- Say when. "By the spring benchmark" or "by the end of the IEP year."
- Count the opportunities. Utah says percent-based measures should not be used when occurrence is open-ended. "80 percent of directions" is measurable. "80 percent of the time" is not, unless the time is defined.
- Use the same tool as the baseline. NCII says to use the tool that you will use for monitoring.
A note on behavior goals
Utah's guide says a behavior goal should focus on teaching a replacement behavior that matches the function of the behavior, and may mention reducing the problem behavior as a secondary result. It also offers a "potato test": a meaningful goal does not describe something a potato could do, such as sitting quietly with hands to self. Data for these goals are covered in behavior frequency and duration data.
Pitfalls
- Picking a round number. A criterion that is simply "80 percent" tells the reader nothing about whether it was derived from data.
- Copying last year's criterion. The baseline changed, so the number should be checked.
- Mixing a single-skill measure with these formulas. NCII says the strategies are for general outcome measures and should not be used with single-skill measures.
- Skipping the team. NCII says all team members, including families, should be involved.
Utah also notes that the criterion should be set from data the team has gathered, with current data stated in the present levels. If those are missing, return to baseline vs present levels.
Terms used here are in the glossary.