Domain C: Measurement, Data Display, & Interpretation

Operational definition

A clear, objective description of a behavior that allows observers to identify and measure it consistently.

Topography-based definition

A definition describing what a behavior physically looks or sounds like.

Function-based definition

A definition identifying behavior by its effect on the environment.

Direct measurement

Measuring behavior as it occurs.

Indirect measurement

Measuring behavior through reports, interviews, or rating scales rather than observing it directly.

Permanent-product measurement

Measuring lasting changes in the environment produced by behavior. Confirm that the product accurately represents the target behavior.

Continuous measurement

Recording every occurrence of a response, or its relevant temporal dimensions, throughout an observation period.

Discontinuous measurement

Observing or recording behavior during samples of time rather than measuring every occurrence completely.

Repeatability

The property that behavior can occur repeatedly and be counted.

Temporal extent

The property that each instance of behavior occupies time and has a duration.

Temporal locus

The property that behavior occurs at a particular point in time relative to other events.

Count

The number of times a behavior occurs during an observation period.

Frequency

Often used for the count of responses during an observation period. Some sources use frequency to mean rate, so check whether a time unit is specified.

Rate

The number of responses divided by observation time, expressed as responses per unit of time.

Celeration

The change in response rate over time, expressed as a multiplicative increase or decrease per unit of time.

Duration

The amount of time a behavior lasts.

Total duration

The sum of the time spent engaging in a behavior during an observation period.

Duration per occurrence

The length of time each individual instance of behavior lasts. Mean duration is total duration divided by the number of instances.

Latency

The time from a specified antecedent event to the beginning of a response.

Interresponse time (IRT)

The time between the end of one response and the beginning of the next response.

Percentage

A part divided by its total, multiplied by 100. For performance, divide correct responses by total opportunities and multiply by 100.

Trials to criterion

The number of response opportunities needed to reach a predetermined level of performance.

Trials per unit of time

The number of teaching trials or response opportunities delivered during a specified amount of time.

Training duration

The amount of training time required to achieve a specified outcome.

Cost–benefit analysis

Comparing an intervention's expected or actual benefits with its costs, including time, effort, money, and resources.

Measurement of magnitude/intensity

Measuring the strength or force of a response using defined units or clearly anchored ratings.

Event recording

Counting each occurrence of a behavior during an observation period. Best for responses with clear beginnings and endings.

Timing

Using a clock or timer to measure duration, latency, or interresponse time.

Partial-interval recording

Recording whether behavior occurs at any point during each interval. It tends to overestimate the proportion of time behavior occurs.

Whole-interval recording

Recording whether behavior occurs throughout each entire interval. It tends to underestimate the proportion of time behavior occurs.

Momentary time sampling

Recording whether behavior is occurring at a specified instant, usually the end of each interval. It can overestimate or underestimate occurrence.

Planned activity check (PLACHECK)

A group observation method recording how many individuals are engaged in a specified activity at a particular moment, often expressed as a percentage.

Interval length

The amount of time assigned to each recording interval in a measurement system.

Observation period

The specified span of time during which behavior is observed and measured.

Measurement artifact

A feature of the data caused by the measurement procedure rather than by the behavior itself.

Overestimation

A measurement result that represents more behavior than actually occurred.

Underestimation

A measurement result that represents less behavior than actually occurred.

Accuracy

The extent to which an observed measurement matches the true value.

Validity

The extent to which measurement captures the relevant behavior and dimension needed to answer the question being studied.

Reliability

The consistency of measurement when the same event is measured repeatedly under comparable conditions.

True value

The actual value of the measured event, established using a dependable reference or measurement method.

Measurement bias

A systematic measurement error that consistently shifts results in a particular direction.

Observer drift

A gradual change in how an observer applies the measurement definition or recording procedure.

Observer reactivity

A change in an observer's recording behavior because the observer knows their measurement is being evaluated.

Participant reactivity

A change in a person's behavior because the person knows they are being observed or measured.

Observer training

Teaching observers to identify and record behavior accurately and consistently using the measurement rules.

Calibration

Checking and adjusting a measurement system against a known standard to improve accuracy.

Representative sampling

Selecting observation times and conditions that adequately reflect the behavior and settings of interest.

Total count IOA

Divide the smaller total count by the larger total count and multiply by 100. It does not show agreement on individual responses.

Mean count-per-interval IOA

For each interval, divide the smaller count by the larger count. Average the interval agreement scores and multiply by 100.

Exact count-per-interval IOA

Divide the number of intervals with exactly matching counts by the total number of intervals and multiply by 100.

Trial-by-trial IOA

Divide the number of trials with agreement by the total number of trials and multiply by 100.

Total duration IOA

Divide the shorter total duration by the longer total duration and multiply by 100.

Mean duration-per-occurrence IOA

For each matched occurrence, divide the shorter duration by the longer duration. Average those ratios and multiply by 100.

Interval-by-interval IOA

Divide the intervals with agreement on occurrence or nonoccurrence by the total intervals and multiply by 100.

Scored-interval IOA

Divide intervals in which both observers recorded occurrence by intervals in which at least one recorded occurrence, then multiply by 100. Useful for low-rate behavior.

Unscored-interval IOA

Divide intervals in which both observers recorded nonoccurrence by intervals in which at least one recorded nonoccurrence, then multiply by 100. Useful for high-rate behavior.

Equal-interval graph

A graph with equal numerical differences represented by equal distances on each axis.

Line graph

A graph plotting behavior measures across time or another ordered sequence, with data points connected within conditions.

Bar graph

A graph using separate bars to compare amounts across categories or conditions.

Cumulative record

A graph showing the running total of responses over time. The slope shows response rate, and the record does not decrease unless reset.

Standard Celeration Chart

A standardized semilogarithmic chart used to display behavior frequency and changes in frequency over time.

Semilogarithmic graph

A graph with one equal-interval axis and one logarithmic axis. Equal distances on the logarithmic axis represent equal ratios.

Scatterplot

A display of paired observations. In descriptive behavior assessment, a time-grid scatterplot shows when behavior occurs across days and time blocks.

Horizontal axis/x-axis

The left-to-right axis of a graph, usually representing time, sessions, or trials in behavior analysis.

Vertical axis/y-axis

The up-and-down axis of a graph, usually representing the measured dimension of behavior.

Data point

A plotted symbol representing a measurement at a specified observation or time.

Data path

The lines connecting successive data points within a condition.

Condition-change line

A vertical line showing where an experimental or intervention condition changes.

Phase label

A label identifying the condition in effect during a section of a graph.

Axis break

A mark showing that part of an axis's numerical scale has been omitted.

Level

The value around which data points cluster within a condition, often described by their mean, median, or overall position.

Trend

The overall direction and rate of change in data across time.

Variability

The degree to which data points differ from one another or fluctuate around a level or trend.

Stability

A relatively consistent pattern in the data that permits a dependable prediction of future responding under unchanged conditions.

Immediacy of effect

How quickly behavior changes after an intervention or condition change, assessed using data near the transition.

Overlap

The extent to which data values in one condition fall within the range of values in another condition.

Consistency across similar phases

The extent to which comparable conditions show similar data patterns across repetitions.

Visual analysis

Systematically examining graphed data for level, trend, variability, immediacy, overlap, and consistency to evaluate behavior change and experimental control.

Procedural integrity

The extent to which a procedure is implemented as planned, including relevant dimensions such as accuracy and dosage.

Treatment integrity

The extent to which a treatment is implemented as designed. Often used interchangeably with procedural integrity and treatment fidelity.

Treatment fidelity

The extent to which intervention delivery matches the planned treatment. Often used interchangeably with treatment integrity.

Implementation accuracy

The extent to which intervention steps are performed correctly according to the plan.

Treatment dosage

The amount of intervention delivered, such as its frequency, duration, or number of teaching opportunities.

Adherence

The extent to which an implementer follows the prescribed intervention procedures.

Integrity checklist

A list of required procedural steps used to record whether the intervention was implemented correctly.

3 Characteristics of a Good Operational Definition

Objective

Clear

Complete

3 Types of Repeatability / Occurence Measures

Count

Rate

Celeration

3 Time-Sampling / Interval Recording Procedures

Whole

Partial

Momentary

3 Indicators of Trustwrothy Measurement

Validity

Accuracy

Reliability

5 Types of ABA Graphs

Line

Bar

Cumulative

Scatterplot

Standard Celeration

Split-Middle Line Steps

Count

Divide

Mid-rate

Mid-date

Intersect

Quarterly

Split-middle