ABC Narrative Recording

How to take Antecedent–Behavior–Consequence notes that the team can actually use

Why this brief

ABC narratives are the simplest, most-used form of behavior data — and the most often poorly written. A well-taken ABC sample is the BCBA's or supervising teacher's best raw material for understanding why a behavior is happening; a poorly-taken one is descriptive paperwork that no one can analyze. The difference is small in effort and large in usefulness.

This brief covers what ABC means, what each column should and should not contain, formats for taking notes in real time, common errors, and what happens to the data after you take it.

| |

| :-: |

| ABC is not the only form of behavior dataCross-reference brief 06.01 for the broader landscape (frequency, duration, latency, prompt level, permanent product, interval recording). ABC is best for puzzling, escalated, or low-frequency events where context matters and counts don't. For high-frequency known-pattern behaviors, frequency or interval recording is usually more useful. |

1\. What ABC means

Three columns:

Antecedent (A) — what was happening immediately before the behavior. The setting, who was there, what was being asked, what the student had been doing. The trigger, in plain language.

Behavior (B) — what the student did. Observable, descriptive, specific. Not interpretation.

Consequence (C) — what happened immediately after. Adult response, peer response, environmental change, what the student got or got out of.

ABC notes work because of the function-based logic underneath them: behavior is shaped by what comes before and what comes after. A pattern of A's that reliably precede the behavior, and C's that reliably follow it, points to function. (See brief 05.01 on function-based thinking.)

2\. What each column should — and should not — contain

2.1 Antecedent

Should contain

Time and date.

Setting (location, period, activity).

Who was present (which adults, which peers).

What was being asked of the student or what they were doing.

Anything in the immediate environment that might be relevant (noise, transitions, schedule changes).

Setting events you know about (slept poorly, missed meds, fight at home) — note these even if they happened earlier in the day.

Should NOT contain

Interpretation. "He was getting frustrated" is not antecedent — it's already a hypothesis. The antecedent is what set up the frustration: "asked to start writing."

Mind-reading. "She didn't want to do the work" — not observable. "Was given a math worksheet at 9:32; turned away; pushed paper to the floor" — observable.

Generalities. "Earlier today" is not enough; pin down what the student was doing 30 seconds before.

2.2 Behavior

Should contain

What the student actually did. Words used (in quotes), motor behavior, voice, body.

Topography — the form of the behavior. "Pushed worksheet off the desk and onto the floor; pulled hood up over head; said 'leave me alone' in a quiet voice; head down on desk."

Duration — how long it lasted, if relevant.

Intensity — for behaviors that vary in intensity. "Voice rose to a yell. Stomped feet. Threw chair across the room" reads very differently than "voice rose to a yell. Stomped feet. Walked to the cubby."

Should NOT contain

Inference about why. The B column describes; the A and C columns are where the why-pattern emerges across many entries.

Labels. "Tantrum" or "meltdown" or "aggression" can be a useful tag, but the description has to be specific enough that someone reading it can picture the moment.

Adult emotion or judgment. "Was very disrespectful" tells me how you felt; not what the student did.

2.3 Consequence

Should contain

What adults did, exactly. Words used (in quotes), procedures invoked, who came over, what was offered, what was withheld, where the student ended up.

What peers did. "Class went quiet for 10 seconds, then resumed." "Two students laughed."

Environmental changes. "Bell rang." "Lunch arrived." "Lights flickered."

What the student got, or got out of. The escape ("sent to hallway, no longer doing math"), the attention ("adult sat down beside her"), the tangible ("received iPad"), the sensory ("removed jacket and stretched"). This is the heart of the function analysis.

Should NOT contain

What you wish had happened. Document what did.

Justifications. "I had to send him out because…" — note the action, not the rationale.

Future plans. "Going to talk to mom" — that goes in a separate field, not in the C column.

3\. Practical formats

ABC narratives can be in written sentences or in tables. Tables are usually faster to scan and easier to keep up with in real time. Apps and digital systems exist (BehaviorSnap, Catalyst, Rethink); paper still works fine.

3.1 A simple ABC table format

| Field | Example entry |

| :-: | :-: |

| Date / time | 5/8 9:32 a.m. |

| Setting | 3rd grade math, room 207, whole-group instruction |

| Antecedent | Teacher distributed worksheet on multi-digit subtraction. Marcus sighed audibly. Teacher said 'You can do this, just try the first one.' |

| Behavior | Marcus pushed worksheet off desk, said 'No,' put head down on desk, hood up. Lasted 4 minutes. |

| Consequence | Para approached, sat in adjacent chair, said 'Take a minute, I'll be right here.' Did not require worksheet. After 4 min, Marcus picked up worksheet and started problem 1 with verbal prompts. Worksheet completed at 9:48 with 50% accuracy. |

| Notes / hypothesis | Possibly escape from non-preferred task. Setting event: parent reported limited sleep last night. |

3.2 Quick narrative format for short entries

"5/8, 9:32, math. Worksheet given; M sighed; teacher said 'try it.' M pushed paper, head down 4 min. Para sat nearby, did not require work. M started after break with verbal prompts."

This compresses the same information into one paragraph. Useful when you have to capture an incident on the fly and will fill in detail later.

3.3 Sticky-note method for paras in motion

Some paras carry a small notebook or sticky pad in a pocket. Capture A/B/C as bullet points or three short lines on a sticky in the moment, then transfer to the formal sheet at the end of the period or end of day. Memory fades within hours; even a few words written in the moment beats a polished sentence written from memory.

4\. Taking ABC in real time

ABC notes are most useful when written within minutes of the incident. Writing them at end of day from memory introduces drift — antecedents get blurry, sequences get reshuffled, words students said become words you remember them saying.

4.1 In the moment

Note the time.

Note who was present and what was happening before.

Note exactly what the student did, in their words and movements.

Note what happened next.

Don't try to write a paragraph; write notes you can expand later.

4.2 When you can't write in the moment

During a crisis, your job is the student's safety, not the paperwork. Capture the time on a clock when you can; reconstruct as soon as you have a moment. Some teams have one para focused on the student and another writing notes — the team's agreed-upon division of labor matters more than always being able to write while supporting.

5\. Common errors

| Error | Why it matters / how to fix |

| :-: | :-: |

| Interpretation in the B column | "Got frustrated" is interpretation. "Pushed worksheet, head down, hood up" is observation. Save the interpretation for a hypothesis line; keep B observable. |

| Vague time markers | "Earlier this morning" can't be analyzed. "9:32" can. Use clock times. |

| Missing the C | Many ABCs trail off in the C column. The C is the most analytically valuable column — it's the function clue. Write it fully. |

| Bias toward dramatic incidents | If you only write ABC for crisis episodes, you miss the everyday moments where the same function shows up at lower intensity. Sample some calm-but-difficult moments too. |

| Bias against student | Adult words in quotes; student words paraphrased. Reverse: student words in quotes; adult actions specified. |

| Not noting setting events | If the student didn't sleep last night, that belongs in the record. Setting events explain why a normally-tolerated trigger went sideways today. |

| Writing one ABC and stopping | ABC is a sampling tool. Patterns emerge across many entries. One incident is a story; ten incidents are data. |

| Editing for kindness | Softening the student's words or actions to spare their feelings makes the record useless. Be accurate; a sympathetic team will still read it sympathetically. |

| Writing what the BIP says will happen rather than what did happen | If the BIP says "give a break card" and you actually said "sit back down," record what you did. The team can't fix fidelity issues you don't surface. |

6\. What happens to the data

ABC notes feed the team's analysis. They are most useful when:

There is enough volume to see patterns (typically 10+ incidents over 1–2 weeks).

Multiple writers contribute (different paras, the supervising teacher, the gen-ed teacher) — different perspectives catch different things.

Someone is reviewing them. ABC notes that no one reads might as well not exist. Confirm with the supervising teacher who reviews and how often.

The team uses them to refine hypotheses, antecedent strategies, or BIP procedures — not to assign blame.

6.1 When to take ABC vs. when to take other data

| Use ABC when… | Use other formats when… |

| :-: | :-: |

| You're trying to identify function (FBA-prep). | Function is already established and you're tracking frequency. |

| A behavior is low-frequency or high-stakes (rare aggression, elopement, severe self-injury). | A behavior is high-frequency and you need counts (frequency, interval recording). |

| A pattern is unclear and you need context. | You're tracking prompt level for an instructional skill (prompt-level data, see 06.03). |

| A new behavior has appeared and you don't know why. | You're documenting permanent products of student work. |

| The team is considering whether to do an FBA. | The student is on a stable BIP and you're monitoring fidelity. |

7\. A worked example

Three poorly-written entries

"Marcus had a meltdown in math today. He just refused everything. Got him calmed down eventually."

"Maria was being defiant during reading. She wouldn't do the assignment. I had to send her out."

"Aggressive episode at lunch. Talked to him about better choices."

None of these is usable. No times. No specific antecedents. Behavior is labeled, not described. Consequences are vague. The team can read these for a year and learn nothing about function.

Three well-written entries

"5/8 9:32. 3rd grade math, whole group. Teacher distributed double-digit subtraction worksheet. Marcus sighed audibly. Teacher said, 'You can do this, just try the first one.' Marcus said, 'No,' pushed worksheet to floor, head down, hood up. Para sat in adjacent chair, said, 'Take a minute, I'll be here.' Did not require worksheet. After \~4 min, Marcus picked up worksheet, started problem 1 with verbal prompts. Worksheet completed by 9:48, \~50% accuracy. Setting event noted: parent reported only 5 hours sleep."

"5/8 11:14. 3rd grade reading, small group of 4. Maria asked to read aloud. Maria looked at adjacent peer, looked down, said nothing. Teacher said, 'I'll wait.' 12-second silence. Teacher said, 'OK, let's try together.' Maria said, 'I'm not doing it,' stood up, walked toward door. Para said, 'Maria, you can sit with me at the back table.' Maria walked to back table, sat with head on table. Did not return to group. Read with para 1:1 at 11:30 with no resistance, completing assigned passage at instructional level."

"5/8 12:18. Cafeteria, lunch. Marcus seated next to Diego at usual table. Diego reached toward Marcus's lunchbox. Marcus shouted 'no\!' and shoved Diego's hand away. Diego said 'sorry' and pulled hand back. Marcus continued eating. No further incident. Para reminded both about asking before reaching."

These are usable. A reader can picture the incident. Antecedents are specific. Behavior is described, not labeled. Consequences include both the adult response and what happened next. The team can analyze function across multiple entries and refine the BIP.

8\. Common pitfalls

Writing ABCs only for the worst behaviors. The same function often shows up in lower-intensity moments and is easier to learn from.

Writing from memory at end of day rather than noting in the moment.

Treating ABC as a punitive tool rather than an analytic one. The records exist to help the team understand, not to build a case.

Letting the form constrain the substance — if the form has too few lines for what you saw, write more.

Different paras using different conventions. Calibrate as a team.

Not noting your own response in the C column. Adult behavior is data.

Forgetting setting events. The night-before sleep, the morning conflict, the missed meds — these belong in the record.

Recording but not reviewing. ABC piles that no one reads are dead weight.

9\. Resources

PBIS — FBA and Behavior Documentation — pbis.org — Free national-level technical resources.

IRIS Center — Functional Behavior Assessment — iris.peabody.vanderbilt.edu — Free self-paced module.

O'Neill, Albin, Storey, Horner, Sprague — Functional Assessment and Program Development — Cengage — Foundational text on FBA, including ABC narrative methods.

BehaviorSnap, Catalyst, Rethink — vendor sites — Common digital ABC/data platforms; check FERPA and district approval before use.

Brief 05.01 — Function-Based Thinking — this library

Brief 05.02 — Functional Behavior Assessment — this library

Brief 06.01 — Data Types Overview — this library

Page of