Cooper, Heron, and Heward (2014) state:
Measurement (applying quantitative labels to describe and differentiate natural events) provides the basis for all scientific discoveries and for the development and successful application of technologies derived from those discoveries. Direct and frequent measurement provides the foundation for applied behavior analysis. Applied behavior analysts use measurement to detect and compare the effects of various environmental arrangements on the acquisition, maintenance, and generalization of socially significant behaviors. (p.93)
According to Cooper, et. al. (2014), practitioners need measurement for the following reasons:
- Measurement helps practitioners optimize their effectiveness.
- Measurement enables practitioners to verify the legitimacy of treatments touted as evidence based.
- Measurement helps practitioners identify and end the use of treatments based on pseudoscience, fad, fashion or ideology.
- Measurement enables practitioners to be accountable to clients, consumers, employers, and society.
- Measurement helps practitioners achieve ethical standards.
Behavior is the focus of applied behavior analysis. Behavior analysts and those working in the field identify behaviors and then seek to measure those specific behaviors. Behaviors can be measured by three fundamental properties which include repeatability, temporal extent, and temporal locus.
Repeatability refers to how a behavior can be counted or how it can occur repeatedly through time. For example, if the behavior being measured is the behavior of throwing objects, repeatability refers to the fact that you can count how many times the individual throws objects throughout the day or the session.
Temporal extent refers to how much time a behavior takes up. For instance, if you are interested in measuring the behavior of crying, you can measure the duration of crying by starting a timer at the first sound of crying and ending the timer when the crying stops.
Temporal locus refers to at what point in time does the behavior occur. For instance, when measuring throwing objects, you can indicate the time that the behavior occurs, such as at 8:30 am, 10:00 am, and 11:00 am. This might inform you that the behavior only occurs in the mornings (if you see the same pattern over multiple days).
Research in applied behavior analysis can occur in a single case study or group design. For further research information and detailed measurement and data collection strategies, consider the book, Research Methods in ABA.
TYPES OF MEASUREMENT
Based on the three fundamental properties, there are multiple types of measurement that can be used in applied behavior analysis. Here are some of them:
Based on Repeatability:
- Count/Frequency: The number of occurrences of a behavior
- Rate: The number of occurrences of a behavior per a set amount of time
- Celeration: how rate of responding changes over time
Based on temporal extent:
- Duration: how long a behavior occurs (how much time)
Based on temporal locus:
- Response latency: how long it takes from the SD (direction or provided stimulus) occuring to the behavior beginning to occur (For example, how long does it take from the time you give a child a direction for them to start following the direction.)
- Interresponse time: how much time in between responses
- Percentage: a ratio, how many times out of 100 did the response occur
- Trials-to-criterion: how many responses did it take to reach a predetermined criteria
- Topography: the physical form or shape of a behavior
- Magnitude: the force or intensity with which a response is emitted
As you can see, there are numerous types of measurements that can be taken on the behaviors of interest to behavior analysts.
You can utilize event recording, which is a method of measurement that covers a variety of procedures which are used to identify the number of times a behavior occurs.
You can also use timing procedures which involve identifying various aspects of a behavior that relate to time, such as duration, response latency, and interresponse time.
Time sampling is another type of measurement which covers a range of procedures which allow you to measure behavior based on various samples of time.
Additionally, you can measure behavior by permanent products. This means that you dont actually have to observe the behavior taking place. You can know that it took place because the behavior results in some sort of product that is left for others to observe. An example of this is homework. Assuming that kids arent allowing someone else to do it for them, you can tell that a child completed homework without actually watching them complete the homework because you will see the homework completed after the behavior occurs.
See the videos below to learn more about measurement in ABA.
All information referenced from: Cooper, Heron, and Heward (2014). Applied Behavior Analysis. 2nd edition. Pearson Education Limited.
Image credit: CyberHades via Flickr