You wrote the plan, trained the staff, and counted for three weeks. The line is flat. It is tempting to start over. A better first move is to find out where the plan and the student are out of step.
This page is a troubleshooting guide. It starts with the cheapest checks and ends with the biggest.
First: is the plan being carried out?
Colorado's guide describes a fidelity checklist for the strategies in the plan, completed for each adult in the room. For each strategy, the adult answers: were all the steps implemented as intended, did the student respond as intended, and was the strategy used as often as intended. It suggests using it twice a week for the first three weeks and weekly after that, as a self-check or in an observation format.
Illinois's list of obstacles includes inconsistent or incorrect implementation across settings and staff, and a lack of training to teach positive behavior strategies and replacement behaviors. Colorado adds that staff who respond in ways that differ from the plan may inadvertently reinforce the behavior.
If the answer is "not really," fix that and wait. Do not change the plan until you know whether the plan was tried.
Next: does the data measure what you think?
Check the definition and the counts. Illinois names vague or subjective definitions and incomplete measurement criteria as obstacles. Have two people count the same period, as in writing an operational definition. If the counts disagree, the data cannot show whether the plan works.
Then: match symptoms to causes
| What you see | Possible cause | First move |
|---|---|---|
| No change, plan not fully used | Implementation gap or training gap | Coach staff, use the fidelity checklist |
| The replacement skill is rarely used | Not taught well, or does not pay off fast enough | Re-teach when calm, respond at once to the skill |
| Behavior gets worse at first | The old behavior is being tested as the response changes | Keep the plan steady, keep the data, check in with the team |
| Works with one adult, not others | Inconsistent responses | Share the plan and the reason with everyone, model the response |
| Works in one class, not another | Different conditions or demands | Compare conditions with a scatterplot |
| The student seems to enjoy the reward and the behavior continues | Reinforcer does not match the function | Revisit the hypothesis, change the reinforcer |
| Behavior improved, then slipped | Fading too fast, or a change in context | Return to the previous step, check for setting events |
Behavior that gets worse first
Illinois's guidance notes that behavior sometimes gets worse before it improves. When a behavior used to pay off and now does not, the student may try harder for a while. That is not a reason to give in or to throw out the plan. It is a reason to keep the response steady, protect safety, keep counting, and talk with the team and the family so no one is surprised. If safety is at risk, follow your school's safety procedures, and call the team together. See crisis plan in a BIP: an overview.
Does the new skill pay off?
Colorado says the demonstration of the replacement skill must pay off faster, better, and more efficiently than the problem behavior, and that if the "when" statements are not reinforced effectively, the "if" returns, because the old behavior generally takes less effort and pays off faster. Ask: when the student uses the card, how quickly does the answer come? If it takes a minute, and leaving takes ten seconds, leaving wins. Shorten the response time. See replacement behaviors examples.
Is the hypothesis right?
If fidelity is good and the skill pays off but the behavior continues, look again at the function. NCII lists revisiting the ABC data to confirm the intervention addresses the correct function as a step when progress is not adequate, along with reviewing data to adapt or intensify, and changing the reinforcer or increasing the schedule. Collect a fresh set of ABC entries and compare them with the original pattern. Illinois says additional FBAs should be conducted if significant modifications or new interventions are necessary. See hypothesis statement examples.
A short example (invented)
A plan for Leo, who leaves his seat during writing, shows no change after three weeks. The team checks. Fidelity: the aide used the break card prompt on 9 of 10 days, the teacher on 4 of 10. Measure: two observers agreed. Skill: Leo used the card only twice, and both times the answer took over a minute. Hypothesis: new ABC entries still point to avoiding long writing tasks.
The finding is not that the plan is wrong. It is that the response is too slow, and the teacher was not using the prompt. The team agrees on a ten-second response to the card, practices it with the teacher for 15 minutes, and sets the fidelity check for twice a week. Two weeks later the card use is up to eight times a week and leaving is down from six to two. One change, aimed at the actual gap, did the work.
What not to do
- Do not add consequences without checking fit. The IES guide says harsh or punitive discipline is not effective in increasing appropriate behavior and tends to elicit resentment and resistance, and Illinois says aversive techniques are typically ineffective.
- Do not change several things at once. You will not know what worked.
- Do not blame the student or the family. The data describe a plan and a setting.
- Do not drop the data. Without them, you cannot tell what changed.
- Do not hide a flat line. Tell the family and bring it to the team.
Check the context
Illinois lists neglecting environmental, cultural, or other issues outside the classroom as an obstacle. Ask what changed: schedule, staff, seating, a home event, sleep, a new medication reported by the family. Colorado's setting event strategies, such as a check-in at arrival, can help. A change in context often explains a sudden slip.
A troubleshooting meeting
Call the team for a short meeting, with the graph, fidelity results, and fresh notes. Follow the order: use, measure, skill, response, hypothesis, context. Decide one change, who will make it, and when you will look again. See BIP review and revision and BIP team meeting agenda. Tell the family what you found and what you will try.
When to ask for outside help
If the plan is carried out, the skill pays off, the hypothesis holds, and nothing changes, ask a school psychologist, behavior specialist, or district team to review the case. That is a reasonable step, not a failure. Terms are in the glossary.
Your district's process
How to request support, how often plans must be reviewed, and what notice is required for changes vary by state and district. Ask your special education director, and follow the procedure for notifying the family.