What is CABAS®?

The Comprehensive Application of Behaviour Analysis to Schooling (CABAS®) is a research-driven system-wide approach providing individualised programmes for children and young people with and without disabilities. It was developed by R. Douglas Greer, PhD and colleagues of Teachers College, Columbia University. Several schools in the U. S. currently use this approach to teach children with autism, pervasive developmental disorder, communication disabilities, and other delays in learning, using an ABA (Applied Behaviour Analysis) methodology in a school setting. A profile of Dr. Greer describing some of his current research projects in the U.S is detailed in J Levine (2007).

CABAS® uses a scripted curricula measured through "learn units", tactics from the scientific literature and PSI (Personalised System of Instruction) for staff training. Research has shown strong relationships between the number of learning objectives achieved by pupils and (a) the number of learn units received by pupils, (b) the number of weekly teacher observations, and (c) the number of accurate reactions to pupil responses.

ABA is a science and has been around for over 30 years. So, the CABAS® approach is based on 30 years worth of data. It provides us with an individualised approach to education. Research on CABAS® has shown 4-7 times more learning compared to more traditional approaches to teaching.


Students in a CABAS® ABA Programme are initially assessed to see what behaviours are present and which require instruction. An inventory of repertoires to determine the student's current skill level is used, and are categorised as follows:

• Communication: This is based on the verbal behaviour model and focuses on listener repertoires, such as compliance to verbal commands, and speaker repertoires, such as responding verbally to either verbal or non-verbal stimuli.

• Academic Literacy: This category includes responding either by classifying relationships between phenomena or by following steps involved in simple or complex operations. Examples include reader and writer behaviours, matching stimuli and discriminating between stimuli.

• Social skills: This refers to the behaviours and the controlling stimuli between the student and the teacher or the student and his or her peers. It includes attentional control, sharing, appropriate touching and appropriate demeanour.

• Community of Reinforcers: This involves expanding each pupil’s community of reinforcers, e.g. playing with toys independently, looking at books independently.

• Self-sufficiency: Behaviours listed in this repertoire are those that are likely to determine whether or not the pupil can succeed in a mainstream classroom.

• Physical Development: This includes the fine motor skills that are important for school, such as holding and using scissors, and also gross motor skills, such as throwing a ball and jumping.

Learn Units

The basic pupil to teacher interaction is referred to as a Learn Unit in CABAS® instruction (McDonough & Greer, 1999; Greer, 2002). A learn unit includes teacher presentations, pupil responses, and how the teacher should respond depending on whether the pupil’s response is accurate or inaccurate.

A learn unit is not complete unless all of these components for the pupil and the teacher are present. They are termed learn units because they are the basic units of teaching. That is, unless the appropriate antecedent is learned together with the appropriate response and consequence then the function of the behaviour is not learned.
A learn unit can also take the form of permanent products. The teacher might present the pupil with a worksheet to complete which the teacher would mark later and then show to the pupil. The antecedent would be the written question or instruction on the worksheet. The behaviour would be the pupil's written response. The consequence would be the teacher's tick or cross on the work. The learn unit would not be complete until the pupil had had feedback on their work.
Another example of a learn unit is intervals of on-task behaviour. If the long-term objective is for the pupil to remain on-task looking at books for 5 minutes then a learn unit could be a 5 second interval of on-task behaviour. The antecedent would be the book itself. The behaviour would be whether the pupil was looking at the book or not. The consequence would be verbal praise or some kind of reinforcement from the teacher. If the pupil was not looking at the book then the behaviour would be corrected, with a gestural prompt from the teacher.

Learn units are presented to both mastery and fluency criteria. All pupil responses and all objectives achieved are measured and graphed. The direct measurement of each of the pupil's responses during instruction is the most sound measure of their learning.

Decision Protocol

We use the decision protocol, as described by Keohane and Greer (2005) and Greer (2002), to analyse the graphs and make empirically driven instructional decisions, and to gauge the effectiveness of the teaching.

The decision protocol criteria is counted as 90% correct over 2 consecutive sessions. When the short-term objective is mastered, we move the pupil onto the next step towards meeting the long-term objective. If 3 data paths occur all with an ascending trend then the decision is made to continue with that short-term objective. If after 3 data paths the trend is descending, or there is no trend, then the decision is made to change the instruction. Tactics from the scientific literature are used to aid the pupil in meeting that short-term objective or it is checked whether the pupil has the prerequisite skills to meet the current objective. If after 5 data paths the overall trend is ascending then, again, the decision is made to continue with that level of instruction. If after 5 data paths the overall trend is descending or no trend then the decision is made to change the instruction.


An observational procedure has been designed by CABAS® to collect data on pupil and teacher responding, and to convert these responses to rates of teacher and pupil behaviour. This procedure is termed Teacher Performance Rate and Accuracy (TPRA).

The supervisor records data on the teacher's antecedent, whether it was accurate or inaccurate. She then records inter-observer reliability on the pupil's response, again whether it was correct or incorrect. Then she records data on the teacher's consequence; whether it was accurately reinforced or corrected.

Basically, the supervisor is collecting data on whether the learn unit presentation as a whole is accurate. She also times the 20-learn unit programme to convert the responses to rates of teacher and pupil behaviour.

Ingham and Greer (1992) found that use of the CABAS® teacher-performance observation procedures by a supervisor resulted in significant increases in total learn units taught, and correct responses by pupils in the observed and other settings.


To be system-wide means that the approach is applied to the whole organisation. The pupil and teachers are trained using an ABA approach. Teachers are taught to be scientists, decision-makers and leaders. All are working through a set of individualised modules set up by Columbia University. BES teaching staff receive intensive training, and their skills are assessed on an ongoing basis to ensure that standards are maintained.


• Greer, R.D., & McDonough, S. H. (1999). Is the learn unit the fundamental unit of pedagogy? The Behavior Analyst, 20, 5-16.
Click here to view book details

• Greer, R.D. (2002). Designing Teaching Strategies: An Applied Behavior Analysis System Approach. San Diego, CA: Academic Press.

• Greer, R.D., & Ross, D.E. (2008).  Verbal Behavior Analysis - Inducing and Expanding New Verbal Capabilities in Children with Language Delays.  Boston: Allyn & Bacon

• Ingham, P., & Greer, R.D. (1992). Changes in student and teacher responses in observed and generalized settings as a function of supervisor observations. Journal of Applied Behavior Analysis, 25, 153-164.
Click here to view article

• Keohane, D, & Greer, R. D. (2005). Teachers use of verbally governed algorithm and student  learning. Journal of Behavioral and Consultation Therapy, 1 (3), 249-259.

• Levine, J (2007).  The Unorthodox Behaviorist - a profile of Doug Greer.   TC Today, The Magazine for Teachers College, Columbia University 24-31, 41-43. 
Click here to view article