A plan was written in September. By November the data sheet has two entries in September, six in October, and a note that says "better?" Nobody can tell whether the plan is working, because the sheet was too hard to keep.
This guide is about the sheet. It covers what to put on it, four layouts for the four common measures, and a short fidelity check so you can tell whether the plan was carried out.
Principles that come from the guides
Colorado's guide makes several points about monitoring a plan.
- Frequent beats perfect. Frequent data allow quick, minor adjustments. Without consistent data, a team may spend time on an intervention that is not working. Ideally collect daily, and at a minimum weekly.
- Measure both sides. Best practice is to measure the behavior being reduced and the desired behavior being increased. At minimum, track the target behavior.
- Decide who counts and how. The count should use a clicker, tallies, paper clips moved around a card, or beads, and a named person, because memory is not a data system.
- Keep the window fixed. Pick the part of the day when the behavior will be counted, such as 9 to 12, and keep it. Days outside the window are not added to the graph.
- Know where the records live, and how they will be graphed.
Utah's guide agrees on the daily habit and adds a warning: collecting data on an as-needed basis, or changing how often data are collected informally, can produce serious errors in interpretation.
Four layouts
| Measure | Use it when | What the sheet needs | Colorado's caution |
|---|---|---|---|
| Frequency | The behavior is brief and separate from other incidents | Date, window, tally, total | Not best for long or rapid bursts |
| Duration | The behavior has a clear start and stop and length matters | Start, stop, minutes per event, daily total | Per event or total, choose one |
| Latency | The behavior follows a directive | Time of direction, time of start, gap | Define how long counts as starting |
| Intensity | Severity matters | A defined scale, with the level each time | Define each level in words |
A daily frequency sheet
| Date | Window | Behavior count | Replacement skill count | Notes |
|---|---|---|---|---|
| Oct 12 | 9:00 to 12:00 | 7 | 1 | Substitute teacher |
| Oct 13 | 9:00 to 12:00 | 6 | 2 | |
| Oct 14 | 9:00 to 12:00 | 5 | 3 | Used the card twice before the worksheet |
The replacement skill column is what lets a team see the plan working before the problem behavior falls. Colorado notes that skills for increase might not progress as quickly, but data often show the behavior decreasing.
Duration, latency, and intensity
For duration, Colorado describes per event recording (start, stop, record, reset for each event) and total recording (adding all events in the day). Ask which gives the clearer picture: two behaviors today, or two behaviors that lasted 215 minutes.
For latency, start a timer when the direction is given, stop it when the student begins, and record the gap. The definition must say what counts as beginning.
For intensity, Colorado's example is a five-level scale from minor to critical, with each level described in words. Write the descriptions before you start, so two people rate alike.
One page, one purpose
Put at the top of the sheet: the student's initials or code, the behavior definition, the replacement skill definition, the window, and who counts. Put at the bottom: where to send the sheet and when. If a sheet needs more than a page, simplify it. A form people do not finish is worse than a short form they do.
Look at it within three weeks
Colorado says progress monitoring data should be analyzed within three weeks to judge the effectiveness of the intervention. That is a good habit even if your district does not require it. Put the numbers on a graph: baseline on the left, a line where the plan began, and each day after. NCII says to always include a goal line on the graph. For how to build graphs, see the data guides on this site, starting with baseline before a behavior plan.
Three things to look at on the graph
NCII's behavior guide says to interpret a student's response to an intervention by analyzing graphed data, comparing the period after the plan starts with the baseline for variability, level, and trend. In plain terms:
- Level. Is the typical number lower than at baseline for a behavior to decrease, or higher for a skill to increase?
- Trend. Is the line moving in the right direction, and how fast?
- Variability. Are the points close together, or scattered? NCII's glossary says highly variable data may mean the tool is not measuring accurately, that fidelity is an issue, or that the behavior chosen is not appropriate.
If level, trend, and variability all look good for several weeks, the plan is probably doing its job. If they do not, check fidelity first, then the definition, then the hypothesis. That is the order in BIP review and revision.
A fidelity check for staff
A sheet of numbers tells you what happened. It does not tell you whether the plan was carried out. Colorado suggests a fidelity checklist: a list of the strategies in the plan, completed for each adult in the room, twice a week for the first three weeks and weekly after that. It is not a "gotcha," the guide says, but a way for staff to see how things are going.
| Strategy in the plan | All steps done as written? | Student responded as intended? | Done as often as planned? |
|---|---|---|---|
| Give a two-minute warning before transitions | Yes or no | Yes or no | Yes or no |
| Respond to a help card within 10 seconds | Yes or no | Yes or no | Yes or no |
Consider changes for strategies that get a "no" again and again. The staff member's feedback is part of it: a strategy that is hard to do may need training or a simpler form.
Where sheets go wrong
- Counting on memory. Use a physical counter.
- Changing the window. It breaks the graph.
- Tracking only the problem behavior. Add the replacement skill.
- Skipping the notes column. A substitute day or a schedule change explains a spike.
- Filing and forgetting. Review within three weeks. See BIP review and revision.
What to confirm locally
Your district may require a particular sheet or a data system. If so, use it, and add what it lacks: the definition, the window, and the replacement skill. If staff will be recording, see the paraprofessional's role in a BIP. Terms are in the glossary.