Using Behavior Data to Make Intervention Decisions With Confidence

At a November team meeting, three adults describe the same student in three different ways. His teacher says things are "a little better." The paraeducator says "about the same." The counselor says last week was "rough." Everyone is sincere, and no one can say with confidence whether the intervention that started in September is working.

Behavior data turns that conversation around. When a team has a clear measure, a baseline and a simple graph, the question shifts from "How does it feel?" to "What does the data show, and what should we do next?" Good data doesn't replace professional judgment, but it gives judgment something solid to work with.

This article explains how to choose a behavior measure, collect a baseline, set a goal, graph progress, read the graph and use decision rules to decide whether to continue, adjust, fade or intensify support. It builds on ABC data, which describes why behavior happens; this article focuses on measuring how much it is changing.

Understanding behavior data for decisions

Descriptive data and progress data

Two kinds of data do different jobs. Descriptive data, such as ABC records, helps explain when and why a behavior happens. Progress-monitoring data measures the behavior repeatedly over time so the team can see whether it is changing in response to support. Decisions about an intervention rest mainly on progress data.

Common behavior measures

Measure	What it records	Works well when
Frequency or rate	How many times the behavior happens in a set period	The behavior has a clear start and end and doesn't last long (call-outs, hitting, hand raises)
Duration	How long the behavior lasts	Length matters more than count (tantrums, time out of seat, time on task)
Latency	How long between a direction and the start of the behavior	The concern is delay (starting work, following a direction)
Interval recording or time sampling	Whether the behavior occurred during, or at the end of, short intervals	The behavior is frequent or continuous and hard to count (engagement)
Rating scales, such as Direct Behavior Rating	A brief rating of the behavior after an observation period	Teachers need a quick, repeatable measure during instruction
Daily progress report points	Percentage of possible points earned, as in Check-In/Check-Out	The student is in a Tier 2 intervention that already generates ratings

Direct Behavior Rating is a method developed and studied by university researchers specifically for efficient classroom use; the National Center on Intensive Intervention includes information about behavior progress monitoring tools.

Reading a graph: level, trend and variability

• Level: roughly where the data points sit, higher or lower.
• Trend: the direction the data is moving over time.
• Variability: how much the data bounces around from point to point.

Comparing level, trend and variability before and after an intervention starts is the basis of visual analysis, a standard way of reviewing single-student behavior data.

Why data-based decisions matter

• They prevent staying too long with what isn't working, and abandoning something too early that is.
• They make meetings shorter and more focused. A graph answers in seconds what discussion takes many minutes to circle.
• They reduce bias. Memory tends to overweight recent or dramatic incidents.
• They show students their progress. Many students are encouraged by seeing their own graph improve.
• They support decisions about intensity. Teams need evidence to decide whether a student needs more support or is ready for less.

Practical strategies

1. Choose one measure that fits the behavior

What to do: Pick the measure from the table that best matches what matters about the behavior. Define the behavior precisely first.

Why it helps: A mismatched measure, such as counting a behavior that lasts twenty minutes, can hide real change.

2. Make it feasible

What to do: Choose a measurement period you can sustain, such as one class period a day, and a method that fits teaching, such as a tally counter, a clipboard or a rating at the end of the period.

Why it helps: Data that's collected inconsistently is hard to interpret. A modest measure collected reliably is more useful than an ambitious one abandoned in week two.

3. Collect a baseline

What to do: Measure the behavior before the intervention starts, under typical conditions, for several sessions.

Why it helps: Without a baseline, you can't tell whether change happened. If the behavior is unsafe, the team may begin support right away and use existing records as a baseline.

4. Set a clear goal

What to do: With the team, write a goal that names the behavior, the measure, the target level and the time frame. For example: "During math, Jordan will begin work within two minutes of the direction on four of five days by the end of the grading period."

Why it helps: A goal lets you draw an aim line on the graph and see whether progress is on track.

5. Graph the data

What to do: Plot each data point on a simple line graph, with time on the horizontal axis and the measure on the vertical. Draw a vertical line where the intervention began or changed.

How to implement: A spreadsheet works well. Many schools also use data systems for discipline or intervention records; ask what your school uses.

6. Check implementation before judging the intervention

What to do: Before deciding an intervention isn't working, check whether it was delivered as planned. A brief fidelity checklist helps.

Why it helps: If an intervention was delivered only half the time, the data tells you about implementation, not about the intervention.

7. Use decision rules set in advance

What to do: Agree as a team on what patterns will lead to which decisions, and how often you'll review. Your school or district may already have decision rules; use them.

Why it helps: Rules set in advance make decisions consistent and less influenced by a single bad week.

Tool: a decision guide

What the graph shows	Questions to ask	Possible team decision
Steady improvement toward the goal	Is progress on track to meet the goal on time?	Continue as planned
Goal met and sustained over time	Is the student ready for less support? What is the fading plan?	Begin fading gradually; keep monitoring
Some improvement, but too slow	Was the intervention delivered with fidelity? Does it match the function?	Adjust (dose, reinforcer, timing, skill focus)
No change or worsening	Was it delivered as planned? Is the hypothesis right? Are there new setting events?	Address fidelity, revise the plan or consider more intensive support
Highly variable data	What differs between good and difficult days?	Look for setting events; gather ABC data; check consistency across adults

A sample weekly summary

Week	Phase	Average minutes to start work (math)	Days intervention delivered (of 5)
1	Baseline	8	n/a
2	Baseline	9	n/a
3	Intervention	6	5
4	Intervention	7	2
5	Intervention	4	5
6	Intervention	3	5

These numbers are an invented illustration, not results from a real student.

Notice week 4: the latency rose when the intervention was delivered only two days. Tracking fidelity alongside the behavior makes that pattern visible.

Step-by-step implementation

1. Introduce: Agree with your team on the behavior, its definition and the measure.
2. Baseline: Collect data for several sessions before starting the intervention.
3. Set the goal and the review schedule, and confirm decision rules.
4. Teach and implement the intervention, tracking fidelity.
5. Monitor: Collect data at the same time and in the same way, and graph it.
6. Review at scheduled meetings, using level, trend and variability.
7. Adjust: Continue, adjust, fade or intensify based on the data and decision rules, changing one main thing at a time where possible.

Illustrative scenarios

These are invented examples, not real case studies.

Jordan, sixth grade

Jordan's team measures how long it takes him to start math work. Baseline shows long delays. After an intervention with a quick-start card and an initial check-in, the graph shows improvement, except during a week when a substitute didn't use the card. The team decides to continue and adds the card to the substitute plan.

Nia, kindergarten

Nia's teacher records the duration of crying episodes at drop-off. After several weeks of a predictable goodbye routine worked out with her family, the durations shorten and episodes become less frequent. When the goal is met for several weeks, the team begins fading the extra support.

A middle school grade level

A grade-level team reviews office referral data by location and time and sees a cluster in the cafeteria at lunch. Instead of looking only at individual students, they adjust supervision and reteach cafeteria expectations, then review the next month's data.

Differentiation and adaptations

Early elementary

Short measurement periods and simple counts work best, because young children's behavior varies across the day. Visual charts can be shared with students in a positive way, such as coloring in a bar.

Upper elementary

Students can help graph their own data, which can build motivation and self-awareness.

Secondary

Data often comes from several teachers. Agree on a common measure and a simple shared method, such as one rating per class period. Respect privacy when sharing data with the student.

Students with disabilities

IEP goals often specify how progress is measured and reported. Use the measurement method in the IEP, and share concerns about it with the case manager. The IEP team, not an individual teacher, makes decisions about changes to IEP services.

English learners

Consider whether language demands affect the behavior being measured. A change in a student's English proficiency or a shift to more language-heavy tasks may show up in behavior data.

Common challenges

Missing data

Simplify the method or shorten the period. Some data, collected consistently, is better than ambitious data collected occasionally.

Different adults measure differently

Practice together, compare ratings for the same period and agree on definitions.

Changing everything at once

If several changes happen together, the team can't tell what helped. Change one main thing where possible, and mark it on the graph.

A bad week causes panic

Look at the trend, not a single point. Follow the decision rules.

Key takeaways

• Progress-monitoring data shows whether behavior is changing in response to support.
• Choose a measure that fits the behavior and is feasible to collect consistently.
• Baselines, goals and simple graphs make progress visible.
• Check implementation fidelity before concluding that an intervention isn't working.
• Decision rules agreed in advance help teams continue, adjust, fade or intensify support consistently.

Next step: Choose one student or one class routine and set up a simple graph this week. For how this data is used at higher tiers, read Developing Tier 2 Interventions, and find more in Positive Behavior Support.

This article offers general classroom guidance. It does not replace your school's policies, a student's IEP or 504 plan, or advice from your school's specialists.

Sources & Further Reading

• National Center on Intensive Intervention, Behavior progress monitoring and data-based individualization resources. https://intensiveintervention.org/. Accessed November 2026.
• Center on Positive Behavioral Interventions and Supports, Center on PBIS (data-based decision making resources). https://www.pbis.org/. Accessed November 2026.
• IRIS Center, Vanderbilt University, behavior and progress monitoring modules. https://iris.peabody.vanderbilt.edu/. Accessed November 2026.
• Cooper, J. O., Heron, T. E., & Heward, W. L., Applied Behavior Analysis (Pearson). Accessed November 2026.