Mr. Alvarez has a goal that says "improve reading fluency." Last year, he measured it with a weekly quiz he wrote himself. This year, a colleague asks which passage level he used, how long the student read, and what the scores were compared with. He realizes his quiz cannot answer any of that.
Curriculum-based measurement, or CBM, exists to answer those questions. This guide explains what it is, what to look for in a tool, and how a probe is given.
What CBM is
The Texas Education Agency's computational fluency system describes CBMs as timed tests that represent an important indicator of general proficiency in a domain. For math, that indicator is often computational fluency. For reading, it is often how many words a student reads correctly in a minute. Understood describes CBM as a common method that tracks skills such as reading, spelling, or math, with a first test that sets a baseline and later tests that track progress toward a goal.
Two features set CBM apart from a classroom quiz. The probes are standardized, so the same directions and timing apply each week. And there are many alternate forms, so you can give one again and again.
General outcome versus single skill
NCII's goal-setting guide separates two kinds of measure.
| Single-skill measure | General outcome measure (GOM) | |
|---|---|---|
| What it tests | A discrete skill, such as two-digit addition | A broad indicator of overall competence |
| Example | A mastery test for a unit | Passage reading fluency |
| Good for | Short-term objectives, especially for students working on discrete skills | Tracking growth across the year |
| Limit | Less useful for maintenance and growth across a broad set of skills | Does not test one narrow skill |
NCII says the goal-setting strategies in its guide, such as using benchmarks or national rates of improvement, are for GOMs and should not be used with single-skill measures.
What makes a tool fit for an IEP goal
NCII says a tool used for an IEP goal should be a reliable and valid measure, sensitive to change over time, and designed for frequent and ongoing use. It should have enough alternate forms of equal and controlled difficulty, specify minimum acceptable growth, provide benchmarks for minimum acceptable end-of-year performance, and report reliability and validity for performance level and growth for students with intensive needs.
The MTSS Center's chart of key features says much the same: tools should be brief, reliable, valid, sensitive to growth, allow repeated measurement, and have typical growth rates and end-of-year benchmarks. Both point to NCII's Academic Progress Monitoring Tools Chart, where you can check how a tool rates. NCII notes it does not endorse any tool on the chart.
Which skills, at which grades
NCII's Exhibit 1 lists sample target skills with recommended grades.
| Domain | Target skill | Recommended grades |
|---|---|---|
| Reading | Letter naming fluency; letter sound fluency; phoneme segmentation fluency | Kindergarten |
| Reading | Nonsense word fluency | Late kindergarten to grade 1 |
| Reading | Word identification fluency | Grade 1 |
| Reading | Passage reading fluency (oral reading fluency) | Late grade 1 to grade 4 |
| Reading | Maze | Grade 4 and up |
| Mathematics | Oral counting, number identification, quantity discrimination, missing number | Kindergarten to grade 1 |
| Mathematics | Mathematical computation | Grades 1 to 8 |
| Mathematics | Number concepts and applications | Grades 2 to 8 |
| Written language | Total words written; words spelled correctly; correct word sequence; correct letter sequence | Grades 1 to 6 |
The details of each are in the next guides: oral reading fluency progress monitoring, math fact fluency progress data, and writing progress monitoring.
How a probe is given
Texas's math system shows what standardized means in practice. Students get a fixed time, and the timing is set by grade: for example two 1-minute tests in kindergarten and first grade, 2 minutes in grades 2 and 3, and 4 minutes in grades 4 to 6. The document says it is important to follow the timing precisely and consistently to allow valid interpretation. Administrators read a script, start a timer at "Start," avoid answering questions about the items, and say "Stop" at the end of the time. The number of items correct in the time is plotted on a graph.
The same logic applies to any CBM. Use the publisher's script and timing, the same way each week, every time. A change in directions, time, or scorer changes what the score means.
Scoring and graphing
Score with the answer key. Record the score and the date, then plot it. Always add a goal line, as covered in aimline and trendline explained. If your tool gives benchmarks or growth rates, use them in setting the goal, as in rate of improvement explained.
A ten-minute weekly routine
A routine keeps probes from being skipped. Pick the same day and time each week. Have the next form, a timer, and the score sheet ready before the student sits down. Read the script. Time it. Score it right away with the key and write the score and date on the sheet. Plot it before the next student comes in, so the graph is never a week behind. Then put the used form away, so you do not give it twice. The whole thing takes about ten minutes per student, and most of that is the setup you did once.
A caution on homemade measures
NCII's guide says many teacher-developed measures, such as spelling tests or other common formative assessments, have not been validated for individual progress monitoring. It also says measures built into an intervention usually show progress inside that intervention, not on a broader measure of the domain. Using only those, it warns, can lead a team to wrong conclusions about progress toward grade-level expectations. For that reason NCII recommends curriculum-independent tools.
The MTSS Center's article makes the same point: avoid unvalidated "homegrown" measures when a validated tool is available.
Common probe mistakes
- Changing the form type. Keep the same kind of probe all year.
- Changing the time limit. One minute and two minutes are not interchangeable.
- Coaching during the probe. Texas says to avoid feedback because the results will be used to evaluate progress.
- Giving a probe to a tired or distracted student without noting it. Write the condition on the sheet.
- Using a screener as a monitoring tool. Check that your tool has enough forms.
Tool choice and monitoring frequency
The sources do not say which tool your district should buy or how often you must monitor. Those depend on your district and on the student's goal. Terms are in the glossary.