Scatter Plot Behavior Data Sheet: A Practical Guide

A scatter plot behavior data sheet is a time-based recording grid used to note when a behavior occurs across days, helping a team spot possible patterns in a person’s schedule or environment. To use one, divide the day into consistent time blocks, mark the blocks in which the defined behavior occurs, and compare the pattern across multiple days; the resulting pattern suggests questions to investigate, but does not by itself establish why the behavior happens.

What a Scatter Plot Data Sheet Records

In behavior support and applied behavior analysis, a scatter plot is a structured way to organize observations by time. A typical sheet has rows for time intervals and columns for dates or days of the week. The observer marks a cell when the target behavior occurs during that interval. Looking across the grid can make recurring clusters—such as behavior near lunch, during a particular class, or late in the day—easier to notice than they might be in narrative notes.

The sheet is only as useful as its definitions and recording rules. Before collecting data, describe the target behavior in observable terms: what a person does, and, where relevant, what counts as the beginning and end of an episode. For example, “leaves the assigned area by more than two metres without permission” is more consistently recordable than “is noncompliant.” A team may also define whether repeated actions within a period count as one episode or several. The right rule depends on the behavior and the purpose of the observation.

A scatter plot generally records when behavior occurs, not a complete account of what happened. It may not show the exact number of instances, how long each episode lasted, what immediately preceded it, or how other people responded. Those questions call for additional measurement or observation methods.

When Teams Use Scatter Plotting

Scatter plots are useful when a team suspects behavior varies with time, activities, or routine but does not yet know where to look more closely. A broad schedule view can help identify possible periods for focused observation, discuss whether staffing or activity transitions coincide with events, and decide what additional data to collect. It can also reveal that a suspected pattern is inconsistent, or that behavior is spread across the day rather than concentrated in one period.

For example, suppose a school team records a defined form of leaving the seat during six school days. Marks appear most often in the intervals covering independent writing and the transition immediately afterward. That observation does not show that writing caused the behavior. The periods may also differ in task difficulty, adult availability, fatigue, peer activity, or other conditions. The plot gives the team a practical lead: observe those activities more closely and record relevant events.

Scatter plots can be used in schools, homes, clinics, and other settings, provided the schedule and observation method fit the setting. They are particularly suited to recurring daily routines. They are less informative when the schedule changes substantially from one day to the next unless those differences are recorded and considered during interpretation.

How to Set Up the Sheet

A paper form or a spreadsheet can work. The layout should make it easy to identify the person or observation period, understand each time block, and apply the marking rule consistently. Include enough context to interpret the grid later without exposing unnecessary private information.

Identify the target: write the observable definition and any episode rule in a visible note or on an accompanying instruction page.
Choose the observation window: specify the days and hours to be covered, such as the school day from arrival through dismissal.
Set the intervals: divide the window into equal, manageable blocks that match the question and the setting’s routine. Fifteen- or thirty-minute blocks are common practical choices, not universal requirements.
Label dates and times: use actual dates and clear start and end times. Avoid ambiguous labels such as morning or period three unless the schedule is documented.
Define the mark: state whether a mark means at least one occurrence in the interval, a count, or another specified measure. Do not leave this open to each observer’s interpretation.
Provide a way to record missing observations: distinguish an observed interval with no behavior from an interval that was missed or not observed.

A simple grid might have rows labelled 8:00–8:30, 8:30–9:00, and so on, with each date as a separate column. A symbol such as X can mean that at least one occurrence took place in that block, while a blank means the observer was present and observed no occurrence. A separate code, such as N/O, can indicate that data are unavailable. Include a legend so that a blank is never mistaken for missing data.

Choosing Time Intervals and Recording Rules

Interval length affects what the sheet can show. Shorter intervals can locate patterns more precisely but require observers to check and record more often. Longer intervals are simpler to manage but can hide meaningful differences within a block. If a behavior occurs at 9:02 and 9:27 in a thirty-minute interval, an occurrence-only mark does not tell the reader that there were two episodes or when they happened.

Choose intervals according to the decision the data should inform, the behavior’s frequency and duration, and whether an observer can reliably monitor the person while doing other duties. Keep interval boundaries consistent across the observation period. If a schedule begins at an unusual time or changes, note that rather than silently shifting the grid. For overnight or round-the-clock observations, make clear whether times refer to the calendar date on which an interval starts.

There is no single correct marking method for every purpose. An occurrence mark answers whether the behavior happened at least once in a block. A count records how many times it happened, while duration recording captures how long it lasted. These measures are not interchangeable. A brief event and an event that occupies most of a block could receive the same occurrence mark. If intensity, duration, or frequency matters to a support decision, add a suitable measure rather than treating the scatter plot as if it captured it.

How to Read the Pattern

After collecting observations over enough representative days, look for repeated clusters and compare them with the actual routine. A single marked cell is usually weak evidence of a stable schedule relationship. Ask whether the same time blocks recur across days, whether the person was present and observed in each block, and whether activities or conditions differed on marked and unmarked occasions.

Useful questions include:

Does the behavior appear repeatedly in the same part of the day or only on one date?
Do marked periods correspond to a specific activity, transition, location, or staffing arrangement?
Are there unmarked periods when the same activity occurs, and what might differ then?
Could absences, schedule changes, or missing observations explain an apparent gap?
Is another measure needed to determine frequency, duration, context, or consequences?

Suppose marks cluster around the end of lunch. The plot supports a statement such as “occurrences were recorded more often during the intervals around lunch on the observed days.” It does not support a confident conclusion that lunch, hunger, noise, escape from a task, or any other single factor is the cause. Those are hypotheses to examine with additional information, such as direct observation, interviews, and antecedent-behavior-consequence records where appropriate.

Scatter Plots Compared With Other Data

A scatter plot is a quick overview of timing. An ABC record documents what happened immediately before a behavior, the behavior itself, and what followed. That detail can help a team examine events surrounding an episode, although descriptive records alone do not prove a behavioral function. Frequency or rate data quantify how often behavior occurs; duration data quantify how long it lasts. Interval recording methods answer their own measurement questions and use specific sampling rules.

These approaches can complement one another. A team might first use a scatter plot to identify recurring periods, then collect more detailed observations during those periods. If the decision depends on whether behavior is increasing or decreasing, comparable counts or rates may be more suitable than a grid of occurrence marks. If episodes vary greatly in length, duration may matter. Select data methods based on the question, not on the appearance of a familiar form.

Common Problems and Better Practice

Several avoidable problems can make a plot misleading. Vague definitions lead observers to mark different events. Inconsistent intervals make dates hard to compare. Treating an unobserved block as a no-occurrence block creates false gaps. Recording only difficult days can exaggerate a pattern, while changing routines without noting the change can make unlike days appear equivalent.

Train observers on examples and non-examples of the target definition before data collection.
Use the same time boundaries and marking rules across observers and days.
Record schedule changes, absences, unusual events, and observer coverage in a notes area.
Review a sample of records together to identify disagreements and clarify instructions.
Limit collection to information relevant to the support question and follow applicable privacy and organizational procedures.

Do not use a scatter plot as a diagnosis, a stand-alone functional assessment, or proof that a person is choosing to behave a certain way. A pattern can be affected by many interacting conditions, and a visual cluster can occur by chance, especially with few observations. Decisions about assessment and support should be made by appropriately qualified people with relevant knowledge of the individual and setting. Where safety is an immediate concern, follow the setting’s established safety procedures rather than waiting for a pattern to emerge.

Frequently Asked Questions

What does a scatter plot behavior data sheet show?
A completed sheet shows which observed time intervals contained the defined behavior across the dates recorded. It can highlight recurring time-based clusters, but does not by itself explain the behavior’s cause or provide a precise count unless the form is designed to record counts.

How long should I collect scatter plot data?
There is no universal number of days. Collect observations across enough representative routines to judge whether a pattern recurs, while considering changes in schedule, attendance, and observation coverage. A qualified team should decide whether the available data are sufficient for the question being asked.

What should a blank cell mean?
Define it before recording. A useful convention is that a blank means the interval was observed and no target behavior occurred, while a separate code means the interval was not observed or data are missing. Without that distinction, readers cannot tell a true no-occurrence interval from a gap in observation.

Can a scatter plot identify the function of a behavior?
No. It can show when behavior was recorded and suggest periods for closer assessment. Determining why behavior may occur requires broader assessment and careful interpretation; the time pattern alone cannot establish a function or cause.

Can I make a scatter plot behavior sheet in Excel?
Yes. A basic version can be made as a grid with time intervals in rows, dates in columns, and a legend for occurrence, no occurrence, and not observed. Keep the definition and recording rules with the sheet, and remember that formatting the grid does not replace consistent observation or appropriate interpretation.

Making the Sheet Useful

A good scatter plot behavior data sheet is clear enough that different trained observers can use it in the same way, and modest enough in its conclusions to avoid confusing correlation with cause. Define the behavior, choose intervals that fit the routine, separate no-occurrence from missing data, and collect observations across representative days. Use visible patterns to guide the next question—not as the final answer—and pair the grid with more detailed measurement when the decision requires it.

Try it: the Date Difference Calculator: days, months, years and weekdays between two dates, with the DATEDIF formulas.

More in How-to basics

Sources: Function behaviour and menu paths: Microsoft Support documentation for Excel. Method & sources

Spotted an error in this guide? Let us know.