How to take and analyze ABC data
By the IEP Atlas editorial teamLast updated
Short answer
To take ABC data, write down three things each time the behavior happens: the antecedent (what happened just before), the behavior (exactly what the student did), and the consequence (what happened just after). Record facts, not feelings, within a few minutes. After 15 to 30 entries, tally the antecedents and consequences to see what keeps the behavior going.
ABC data is a running record of behavior in context. A stands for antecedent, B for behavior and C for consequence. The point is not to judge the student. The point is to find the pattern that tells the team what the behavior does for the student, so they can teach a better way to get the same thing.
Paraprofessionals often take most of the ABC data, because they are closest to the student. This page is written for that job. The data feeds the team's functional behavior assessment, so what you write matters.
What ABC data is
Each entry has the same three parts, plus the date, time and setting.
| Part | What to write | Example |
|---|---|---|
| Antecedent | What happened in the 1 to 2 minutes before | "Teacher handed out the math quiz and said, 'Start now.'" |
| Behavior | What the student did, in words a stranger could picture | "Put head on desk, then pushed the quiz onto the floor." |
| Consequence | What adults and peers did right after, and what changed | "Para picked up the quiz and said 'Let's try two problems.' Student did 0 problems." |
The antecedent is the trigger. The consequence is what the student got or avoided. Together they point to the function.
Step table: how long each step takes
| Step | What you do | Typical time |
|---|---|---|
| 1. Learn the definition | Read the team's definition of the behavior | 5 minutes |
| 2. Set up the sheet | Fill in the student, setting and codes | 10 minutes, once |
| 3. Record each incident | Write A, B and C within a few minutes | 1 to 2 minutes per incident |
| 4. Keep going | Collect across different days and times | 2 weeks, 15 to 30 incidents |
| 5. Tally | Count each type of antecedent and consequence | 30 to 45 minutes |
| 6. Share | Bring the tally to the teacher or team | 10 minutes |
Step 1: Know exactly which behavior to record
Ask the teacher or behavior specialist for the operational definition of the behavior. It should say what counts and what does not. "Aggression" is too broad. "Hits, kicks or pushes another person with enough force to move them" is clear.
If you are unsure whether something counts, write it down and flag it. Then ask. The guide to operational definitions of behavior has examples you can compare against.
Step 2: Set up the sheet
Use a sheet with columns for date, time, setting, antecedent, behavior and consequence. Many teams add check-box codes for common antecedents, such as "task given," "told no," "transition" and "adult attention elsewhere." Codes make the tally faster. Keep a blank line for notes, because codes never cover everything.
Step 3: Record each incident as soon as you can
Write the entry within a few minutes of the behavior, once the student is safe. Memory fades fast. By the end of the day, "teacher said 'Start now'" turns into "transition to math," and that detail is lost.
Write what a video camera would show. Use quotes for what people said. Note how long the behavior lasted if it went on for more than a minute.
Step 4: Collect across days and times
One bad Monday does not make a pattern. Aim for 15 to 30 incidents across at least 6 different days. Try to cover each part of the day where the behavior happens, and note the times when it does not happen at all. The quiet times are data too.
Step 5: Tally the pattern
When you have enough entries, count. Make two lists: the types of antecedents and the types of consequences. Put a tally mark next to each one for every incident. The biggest numbers usually show the trigger and the payoff.
Then match the pattern to a function:
| If the consequence is usually... | The function may be... |
|---|---|
| Adult or peer attention, even scolding | Getting attention |
| The task is removed, delayed or made shorter | Escaping a task |
| The student gets an item, food or a turn | Getting an item or activity |
| Nothing changes around the student | Sensory or automatic |
The team, not one person, decides the function. Your tally is the evidence they use. To see how the function of behavior becomes a plan, read the BIP examples by function.
Step 6: Share the data with the team
Bring the sheet and your tally to the teacher or the next team meeting. Point out anything that surprised you. Paras often notice things no one else sees, such as the student being fine with one adult and not another.
Worked example: Lena, grade 2
Lena is fictional. Her team asked the para to take ABC data on "screaming": a yell louder than classroom voice, lasting 2 seconds or more.
Here are 3 of the 18 entries the para recorded over 8 school days:
| Date and time | Antecedent | Behavior | Consequence |
|---|---|---|---|
| Oct 6, 9:40 | Teacher said, "Put the blocks away, it's reading time." | Screamed for about 20 seconds, held onto the blocks | Para said, "Two more minutes." Lena played 2 more minutes, then put blocks away. |
| Oct 7, 12:15 | Recess bell rang while Lena was on the swing | Screamed and sat down in the mulch | Aide let her swing until the line was gone (about 3 minutes) |
| Oct 9, 10:05 | Teacher handed out a worksheet | No screaming; asked "Do I have to?" and started | Teacher said "Yes, thank you for starting." |
The tally across all 18 entries:
| Antecedent | Count |
|---|---|
| Told to stop a preferred activity | 14 |
| Given a new task with no activity to stop | 2 |
| Peer took a toy | 2 |
| Consequence | Count |
|---|---|
| Got more time with the activity | 12 |
| Activity ended anyway, with an adult beside her | 4 |
| Toy returned | 2 |
Fourteen of 18 incidents happened when Lena had to stop something she liked. Twelve ended with more time. The team's hypothesis: Lena screams to keep a preferred activity going. Her plan will teach her to ask for "one more minute" with a visual timer, and give the extra minute when she asks calmly.
Common mistakes when taking ABC data
- Weak: "Lena had a meltdown." Better: "Screamed for about 20 seconds and held onto the blocks."
- Weak: Antecedent: "Nothing." Better: "Teacher was helping another student; Lena had been waiting about 4 minutes." Something almost always happened.
- Weak: Consequence: "Redirected." Better: "Para said 'Two more minutes' and Lena kept playing." The detail is what shows the payoff.
- Weak: Writing entries at 3:15 for the whole day. Better: A short note within a few minutes, filled in at the next break.
- Weak: "She was angry because she's stubborn." Better: Leave out guesses about feelings and motives. Record what you saw and heard.
- Weak: Recording only the worst incidents. Better: Record every incident that meets the definition, small ones included.
Data sheets that pair with ABC data
ABC data shows the pattern. A count shows how much. Many teams use both at once.
- Count each incident on the frequency data sheet, so the team has a baseline.
- If episodes vary from seconds to many minutes, time them on the duration data sheet.
- The measurement guide on frequency recording explains how to turn counts into a rate when session lengths differ.
The full set of steps, from ABC data to the hypothesis, is in how to conduct an FBA.
The Center on PBIS publishes free tools for collecting and using behavior data in schools.[1] The IRIS Center's FBA module includes practice with ABC recording.[2]
Frequently asked questions
How many ABC entries do I need?
Aim for 15 to 30 incidents across at least 6 different days and times. Fewer than 10 entries rarely show a clear pattern. If the behavior is rare, keep collecting longer rather than stopping at a small number.
What if the behavior happens too often to write everything?
Ask the team to pick sample periods, such as two 20-minute blocks a day, and record every incident in those blocks. Keep a separate tally count for the rest of the day so the total is still known.
Is ABC data the same as an FBA?
No. ABC data is one part of an FBA. A full FBA also includes interviews, a record review, a baseline count, a hypothesis statement and often a test of that hypothesis.
Use it in IEP Atlas
- ABC data in the IEP glossary
- Antecedent in the IEP glossary
- Frequency data sheet
- Frequency recording measurement guide
- IEP goal checker
Related guides
- FBA and BIP: a practical guide for IEP teams
- How to conduct an FBA, step by step
- Operational definitions of behavior: examples
- Behavior IEP goal examples with data methods
- Choosing a behavior data method
Sources
- Center on PBIS, Center on Positive Behavioral Interventions and Supports
- Functional Behavioral Assessment: Identifying the Reasons for Problem Behavior and Developing a Behavior Plan, IRIS Center, Vanderbilt University
Written and checked by the IEP Atlas editorial team under our editorial policy. This guide describes general practice and federal rules. It is not legal advice; your state may add requirements, and the IEP team makes the decisions for each student.