Module 1: Overview of Applied Behaviour Analysis (ABA)
Module 1: Overview of Applied Behaviour Analysis (ABA)
A Objectives ......................................................................................................................... 4
B Introduction ..................................................................................................................... 5
1 Technology........................................................................................................... ....... 19
I Methods of measurement.............................................................................................. 20
1 Duration..................................................................................................... .................. 21
2 Latency ........................................................................................................................ 21
3 Magnitude.................................................................................................................... 22
4 Self-monitoring ........................................................................................................... 22
2 Observational assessment............................................................................................ 29
M References ...................................................................................................................... 36
• Know about the history of ABA, what it is and where it came from;
• Understand some of the basic principles of ABA;
• Understand measurement observation systems, behavioural objectives and baseline
measurement;
• Understand how to write an operational definition of behaviour;
• Become familiar with some data collection systems;
• Know how to analyse behaviour using both a functional assessment and a functional
analysis;
• Be familiar with PLA Check and scatter plot observation systems.
Behaviour analysis has been around for almost 100 years. What is unique about this approach
to behaviour is the contribution it has made in helping psychology to become more scientific.
Behaviour analysis has achieved this by focusing on collecting information (data) to inform
decisions, and by researching learning trends and patterns based on principles of
reinforcement and punishment.
At first, ABA can seem overwhelming and mysterious. It can seem very far removed from
everyday life. However, we are all behaviour analysts because we all shape behaviour every
day, often without being aware of it. In this course we will give you some of the skills and
knowledge that will enable you to influence and understand other peoples behaviour. When
you understand behaviour, you can change the behaviour.
The primary skill needed to be a good behaviour analyst is observation. We hope that by the
end of this course you will be great observers of behaviour. Many people use behavioural
techniques every day without realising it. We will be putting new language on techniques you
probably already use. Behaviour analysis has its own language. We will try to give concrete
examples to explain all the principles involved as we proceed through the module.
ABA has many applications. Our focus will be on the area of disabilities including autism.
Applied behaviour analysis has been used to improve the lives of people with disabilities for
over 40 years now. Within the field of ABA, a variety of approaches has been developed to
help and support people with disabilities. However, all of these approaches have in common
the fact that they focus on careful observation and recording of behaviour i.e. measurement of
behaviour.
For example:
• When playing dolls with her two friends, Deirdre will share her dolls clothes with her
friends at least three out of four times when they ask for them.
• Every day for five consecutive days, when asked a question by his teacher, Tommy
will answer using a complete sentence that is audible.
• Eileen will have no incidents of hitting other children during break time for at least
three days this week.
• Permanent products (e.g. wet pants, arithmetic problems solved correctly, number of
comprehension questions answered) can be a useful and easy way to record behaviour.
• Direct observations of the frequency (how often behaviour occurs), rate (how often a
behaviour occurs in specific period of time), duration (how long it lasts) or latency
(the time it takes for the behaviour to begin) are often used in ABA. Occasionally we
look at the magnitude of the behaviour, for example, severity of self-injury or assault.
• Self-monitoring, where an individual observes and records their own behaviour (e.g.
counting the number of cigarettes smoked, asking a child to record how many times
they get out of their seat during class).
Pavlovs initial work has since been extended to include the study of conditioned reflexes in
other animals and in humans. This whole area is called classical or Pavlovian conditioning.
Essentially, researchers in this field are interested in how we learn to associate feelings
(emotions, physical reactions, etc.) with particular stimuli (e.g. people, places, events).
If we take a leap forward to our own lives, we can see how this theory might apply. The aunt
who always gives out treats can become associated with the treats and the positive feelings
that go with them. The doctor who gives the child an injection can become strongly associated
with pain so the child may begin to cry as soon as she enters the doctors surgery.
B.F. Skinner is the other pioneer who developed theories of behaviour based on observation
of animals and how they learn. His work is called operant conditioning, because in his studies
animals learned to operate on the environment (e.g. press a lever or peck at a circle) to get a
reinforcer. Based on his laboratory work with animals, Skinner was able to describe some
basic principles about how behaviour is learned.
Skinner believed that we are Skinner was able to systematically produce patterns of
behaviour based on the consequences for the behaviour,
who we are because
and how the behaviour was shaped and conditioned during
contingencies in our
training. The core principles are research-based and the
environment have shaped us
application of the Skinner principles to everyday life is
called applied behaviour analysis. Skinners legacy to us is the formation of the basic
principles of learning theory. He set out schedules of reinforcement, how learning is shaped,
how we increase behaviour, and how we extinguish it or punish it. Skinner believed that we
are who we are because contingencies in our environment have shaped us. We are subject to
reinforcement and punishment and based on this we behave in certain ways.
Behaviourists like Skinner see all of human behaviour in these terms, they live by these
principles and organise their thinking in these terms. They use behavioural terminology,
which is like a language in itself to the outsider. Behaviour analysis is not something that a
child receives for two hours a day for these people it is a framework for looking at life and for
living life.
There have been many variations on these two schools of A good behavioural programme
behaviourism. Applied behaviour analysis takes many
works towards the naturally
shapes and forms. Some people take some aspects of ABA
occurring reinforcers in the
and use them in conjunction with other approaches. Other
environment
people are purists and only use a strict behavioural model.
ABA does not have to be delivered in a one-to-one situation. When dealing with younger
children, the one-to-one approach may be useful initially, but the aim of any good programme
should be to establish the normal environmental conditions. A good behavioural programme
works towards the naturally occurring reinforcers in the environment. If intensive one-on-one
input is required then there must be a plan to fade it out. It is possible to implement a
behavioural programme for specific or defined behaviours. Unfortunately, many people are
discouraged when they are told that there has to be intensive intervention. Of course,
behaviour change is faster in an intensive behavioural environment but in the real world this is
not always possible.
Some of the most common approaches used in children with disabilities are pivotal response
training, direct instruction training, precision teaching, TEACCH, discrete trial training, PSI
(personal system of instruction), positive programming and cognitive behaviour therapy. All
of these are behavioural approaches with a slightly different emphasis. It can be confusing!
Another area of confusion is the labels applied to behaviour analysis. Let us look at some of
these terms.
1 Back to basics
In applied behaviour analysis the focus is on looking at behavioural principles and how they
can be used to solve human problems (Grant and Evans, 1994). There are seven common
factors that all behaviour analysts share, despite the variety of their field of study (Grant and
Evans 1994). These are:
• A focus on behaviour;
• A scientific approach to the study of behaviour;
• A pragmatic approach to changing behaviour;
• A clearly defined specific treatment goal and clear definitions of principles and
procedures;
• An emphasis on environmental causes of behaviour;
• A concentration on the present, rather than delving into the past to unearth
hypothetical reasons for current behaviour;
• A sense of optimism about the possibility of behaviour change.
Children learn to behaviour in certain ways. They can Learning principles can be
learn to adapt to the environment in an appropriate or in extremely effective in
inappropriate ways. For example, a child might learn to modifying maladaptive
say, Go away or, I do not want to do that job, which is an behaviours
appropriate behaviour or response in a particular situation.
Alternatively, the child might learn to scratch or pinch or scream to get out of doing a job,
which is a maladaptive or inappropriate response in the same situation. Maladaptive
behaviours are to a considerable degree acquired through learning, in the same way that any
behaviour may be learned. Learning principles, therefore, can be extremely effective in
modifying maladaptive behaviours.
In a key article Baer et al. (1968) define applied behaviour analysis as follows: applied
relates to the social importance of the behaviour, behaviour is the pragmatic problem of how
to get someone to do something effectively and analysis relates to the believable
demonstration of the events that can be responsible for the occurrence or non-occurrence of
the behaviour.
ABA focuses on teaching components of each skill from very small skills, such as eye
contact, to relatively complex skills, like having a conversation. Problem behaviours are
analysed as a target for intervention and appropriate behaviours are taught and reinforced or
the inappropriate behaviour is ignored. The teaching of each skill begins with an operational
or a working definition of that skill, and criteria for successfully learning or mastering that
skill are defined.
The techniques of applied behaviour analysis have been employed across a huge range of
settings, including:
• Parent training, e.g. training parents how to manage their childrens behaviour;
• Health care, e.g. programmes that encourage people to comply with medical
treatments;
• Developmental disabilities, e.g. training direct-care staff to support people with
disabilities in community settings;
• Environmental protection, e.g. increasing recycling of household waste;
• Workplace safety, e.g. increasing the use of safety equipment in factories.
Since the early 1960s research has demonstrated the efficacy of behavioural interventions for
children and adults with autism (Leaf et al., 1999). It has been shown to be effective in both
decreasing disruptive behaviour and in improving skills. There are a number of very detailed
ABA-based intervention programmes available, particularly for younger children with autism
(Maurice et al., 1996, Leaf et al., 1999) and for children with developmental disabilities
(Lovaas, 2002).
Taking this functional approach to curriculum development is broadly consistent with the
pragmatic approach of ABA. Some of the principles which must be taken into account when
planning a functional curriculum are that it must be useful and age appropriate; that the
criterion of ultimate functioning is applied; that partial completions and adaptations, or means
Brown (1976) poses six questions to help us target the curriculum to the criterion of ultimate
functioning. These include asking ourselves:
Ethical issues frequently arise when we are developing programmes to change behaviour, e.g.
why are we choosing to modify this behaviour? Are we using a negative procedure when a
more positive, less restrictive one is available? The techniques of ABA are powerful tools for
changing behaviour. ABA programmes should always be designed and implemented in a way
that promotes dignity and respect for people with disabilities.
A practical illustration of this approach can be seen in the way that behaviour analysts
distinguish between the child and their behaviour. Behaviour analysts see the behaviour as
occurring in a context, which we study, and by doing so we try to understand the function of
the behaviour for the child. For instance, instead of saying a child is aggressive, we say this is
While looking at the ABA approach we will use these principles to guide our practice and
ensure that we maintain the highest ethical standards.
• We teach our friends what to talk to us about by only paying attention to topics that
interest us.
• We often reward ourselves, or reinforce behaviour, so we buy ourselves a treat if we
study hard all day.
• Often people give up cigarettes, and save the money to go on a holiday. They use the
money saved as an incentive to stay off the cigarettes.
• We teach children to eat dinner by telling them they can have a dessert afterwards, or
we tell them they must do their homework before they can watch TV.
• We may deliberately ignore minor misbehaviour by a child (e.g. complaining about
having to clean up a mess) because we know that if we give the child attention for it,
they will learn to do it more often.
• Supermarkets give us points or tokens that have no value other than the fact that they
can be exchanged for items from a catalogue. The same principle applies to the driving
licence penalty points system. This is just a token system, which is designed to make
people take a long, hard look at their driving habits.
• People are now smoking outside pubs because of the smoking ban.
Think about the smoking ban in pubs introduced into the UK in 2007. How has this shaped
peoples behaviour? A fine was introduced for smoking in pubs so to avoid the penalty people
went outside to smoke. What effect did this have on smoking did people smoke less? Did
people avoid pubs and stay at home where they could smoke? An interesting development,
which was not anticipated, is that the smokers corner became very popular with adults looking
to meet a partner and encouraged some people to take up smoking who had never smoked
before!
We might not use the technical terms for it, but we can guarantee you that you use these
principles every day. So let us begin by looking at ways in which behaviour is measured.
Before taking a baseline we may think that a behaviour occurs all the time. After recording
baseline data, we may realise that it only occurs in the morning. Baseline data also gives us
information on what skills the person has if we are about to teach a new programme. If we are
to teach hand washing and we observe the child on three or four occasions, we may find that
the child can turn on the tap and complete some steps in the hand-washing sequence already.
When we take a baseline, everyone knows where they stand.
If you want to increase a behaviour, it is essential to show that the baseline data has a stable or
a decreasing pattern. If you conduct an intervention to increase a behaviour when the
behaviour is increasing on its own without intervention then you cannot truly say that the
intervention was successful in increasing the behaviour.
Similarly, if you want to decrease a behaviour then you must begin intervention on a baseline
that shows the behaviour is static or is on the increase. If the behaviour was decreasing
anyway it may continue to decrease anyway without your intervention. Whatever your
intentions, the first step in taking a baseline or in measuring a behaviour is to agree what
behaviour we are measuring.
Nail biting is an act the person is performing so this is a behaviour. Having a pleasant
personality is not a behaviour itself, but you could identify behaviours (what you can see and
hear) such as smiling, number of positive comments, number of times that the person agrees
to help, number of times the person does things for others, as behaviours that might constitute
a pleasant personality. Studying is active and is a behaviour. Having brown hair is not a
behaviour it is a fact! Feeling hot is not an observable behaviour but sweating is the result of
feeling hot which can be observed.
Striking out at someone or some object, making contact with that person or object, or
attempting to make contact with that person or object is this comprehensive enough?
Probably, but you would have to observe the behaviour and see if you caught all the incidents
of the behaviour in order to fulfil this definition.
Take another example: a dressing programme. John will dress himself on three out of three
mornings. What does dress mean? Does it involve picking out his clothes? Does it involve
doing up all the snaps and buttons and zips? If he gets help, is this counted? Again, we must
be very specific. We might start from the point where Johns clothes are all laid out on his bed
and ready for him in the morning. In this case, we could say: when asked to get dressed, John
Sulzer-Azaroff and Mayer (1991) identify the steps in writing the behaviour objective or
target. The first step is to identify the problem behaviour, next state the goal, then state the
context or conditions, the criterion level and the behavioural dimensions and from this come
up with the behavioural objective.
Example
Problem what we want to increase or decrease.
Criterion level level of behaviour you want to achieve at the end of the programme.
Let us work through an example. Danielle is six years old and is in a special class. She has
been diagnosed as having a moderate learning disability. She has no disruptive behaviours but
bites her nails, causing them to bleed. This then interferes with her progress at home and in
school, as she finds it difficult to engage in any fine motor activity she says sore when asked
to engage in fine motor activities. Her parents dress her in the morning and are not making
any progress on her dressing programme. In school, her class teacher reports that she has not
done any written exercises for days because she cannot hold a pencil or crayon without
crying.
Goal the goal is that Danielle will not bite her nails.
Context conditions a programme will be implemented in the home and the school setting.
Using praise to reinforce hands down (e.g. "Good hands down" in a positive tone of voice to
increase the hands down behaviour), Danielle will be reinforced for hands down by the
classroom assistant every five minutes. When Danielle is unattended her nails will be painted
with Stop n Grow or alternatively she will be reinforced for wearing gloves (verbal praise,
"Good, wearing gloves Danielle"). Twice a day for 10 minutes Danielle will be reinforced for
rubbing nail cream onto her nails (incompatible behaviour) and reinforced by saying, "Nice,
letting your nails grow".
Criterion level the only acceptable level of this is zero. If Danielle has one bad episode of
nail biting then she could do severe damage to her nails and set off all the difficulties
described above. The only acceptable rate is zero occurrence of the behaviour over three
consecutive weeks.
Behaviour objective Danielle will not bite her nails for three consecutive weeks.
The behavioural definitions are very specific. Take a look at some of these definitions of
behaviour taken from journal articles:
Praise: any form of verbal approval delivered contingent on correct responding to the task
(e.g. "Thats right") or contingent on general co-operative behaviour (e.g. "I like the way you
are working today"). (Carr and Durand, 1985).
Head or knee hitting: audible contact of a hand, knee, shin or foot with any part of the face
or head.
Hand mouthing: any contact of the fingers with the mouth that broke the plane of the lips.
Slapping: any audible contact of one body part against another (other the face or head) or
contact against stationary item such as furniture walls or floor (Mazaleski et al., 1993).
• Objective;
• Clear;
• Complete;
• Concise.
Getting an accurate definition is the first step, now how can we measure the behaviour we
have defined? There are a number of different types of measurement systems. The nature of
the behaviour and the frequency of the behaviour decide which measure to use.
1 Technology
Computers
Computers can be used to count behaviours, such as the number of letters typed or the number
of answers given or the level the child achieves on a programme. In laboratory situations
computers can be used to measure how long it takes to respond to an instruction or the
amount of time it takes to complete a task. Computers can be used to measure or log our
phone calls and the length of a phone call. Computers have also been used to measure height
and weight.
Frequency simply means the number of times that behaviour occurs. Some people do this by
simply keeping a tally on a page during the day. The number of times a child spits during the
day is an example of frequency.
Rate is the number of times the response occurs in a particular period of time, usually per
minute/per hour. So if you have a behaviour that occurs very frequently you would not be able
to count it all day or you would get nothing else done so you might decide to count it at a set
time in the morning and the afternoon. So, for example, you might decide to count head
stimulation (rocking head back and forth from right to left or left to right) from 10 am to
10.30 am and from 3 pm to 3.30 pm. The number of times a child follows a direction when
asked over a half-hour period is another example of rate. Rate is the frequency of behaviour
over a specific time period.
Example
How would you measure out-of-seat behaviour in John, a 10-year-old boy with Down
syndrome and a mild learning disability in a mainstream classroom?
What is an acceptable number of times to be out of your seat in the classroom? Look at the
peers and how often they are out of the seat. In this situation, it is likely that your goal would
not be zero or no out-of-seat behaviour as this is not typical behaviour in a class of ten year
olds. However, when working on a written assignment it is reasonable to expect them all to
stay seated for 10 minutes or until their work is finished.
What measure would you use and how would you decide? Frequency or rate? If the child is
constantly out of his seat, then the behaviour occurs too often to use frequency and so rate
might be more appropriate. You might decide to take count of the behaviour during the
English writing class from 2pm to 2.30pm every day or even do the count on two days per
week (e.g. Monday and Friday or Tuesdays and Thursdays) depending on what suits your
schedule.
Next, you would write a behavioural definition of the target behaviour. Out of seat if the
childs buttocks are off the seat for three seconds or longer it is counted as out of seat.
Next, decide your criterion so you may decide that he has to be in his seat unless he puts up
his hand and is given permission to leave his seat. While you would like him to sit for the
whole half hour, you might allow him to get up on one occasion. John will only get out of his
seat on one occasion during English writing on three out of three consecutive days.
Alternatively, you could measure in-seat behaviour or duration of in-seat behaviour.
Duration is the length of time to complete a response or how long the behaviour lasts when it
occurs. Duration can also be the extent of time behaviour occurs in an observation interval.
The behaviour has to have a discrete onset and offset. When a child has a tamper tantrum it is
often useful to measure the length of time it lasts. The length of time each incident lasts
provides more information then the total time per day. This measure gives you the number of
tantrums plus how long each one lasted, as opposed to simply measuring the minutes spent
having tantrums all day. Sometimes you can impact on the amount of time a temper tantrum
lasts but it still occurs with the same frequency.
Example
Michael is 12 years old and he has been diagnosed as having autism. He is in a special class in
a mainstream school and lives at home with his parents and his older brother. He has severe
temper tantrums, which occur once per day. He begins by rocking his body back and forth in
his chair while seated and making a loud screeching noise. He then stands up and kicks the
chair behind him. He will run around tearing at his clothes (often ripping his shirt jumper or t-
shirt). He will go to the corner of the room opposite the door and throw himself on the ground
face down kicking and punching the ground with his legs and fists. This can last up to 10
minutes and then he sobs for a further 10 minutes before standing up.
In this example, the behaviour is not frequent but it would be important to time the temper
tantrums and note the topography or the shape of them. If, in conducting the analysis, it
became clear that the temper tantrums were associated with a particular activity then teaching
an escape response might impact on the behaviour. If you were only counting frequency, and
Michael just began his tantrum and finished by kicking back the seat without any of the rest
of the cycle of behaviour, you would impact on the duration and the topography, but if you
only tracked frequency you would not know this and think your programme was not working.
2 Latency
In some instances, the key variable is the length of time from some specified cue to the onset
of behaviour. It assumes that once the onset occurs the rest of the behaviour will follow. We
might be interested to note the time it takes from the moment the teacher tells the child to sit
in her chair until the child sits in her chair. We may want to decrease this time. Or the length
of time between each question or request for assistance from the child to the teacher this time
we would want to increase the latency. Or the length of time from taking a drink to urinating
if we were setting up a toilet training programme. In this case we have no control over the
latency but by learning the persons schedule we can assist them to become toilet trained by
developing a toileting schedule around the latency.
Albert is an 18-year-old man with autism. He has good expressive language skills, but poor
comprehension. He has become very dependent on verbal and physical prompts in order to
complete behaviours he is well capable of completing independently. In this situation, you
might look at the time it takes from Alberts alarm clock going off to Albert getting dressed, or
from the time the dinner plate is placed on the table to the time Albert takes his first bite of
food.
Fintan is a five-year-old boy with multiple sensory disabilities and moderate level of
disability. He eats his food very fast, stuffing the food into his mouth without taking a break
between mouthfuls. You might look at the latency from taking one bite to taking the next bite.
3 Magnitude
Sometimes the topography of the behaviour stays the same but may increase in magnitude, for
example, blood pressure or noise levels in a classroom or severity of aggression or self-injury
can change in level of intensity. It is important that we factor this in, because we could start a
programme and the intensity of the behaviour could change, but if we were measuring by just
the number of times it occurred we would miss this crucial data.
Example
Marie is an eight-year-old girl who has been diagnosed as partially deaf and autistic. She
engages in self-injury. She will slap her face on the right side with her open hand. This is
severe enough to leave a red mark and cause a loud noise. She repeats this behaviour five to
six times in rapid succession and then makes a high-pitched screaming noise and finally she
will cry. This crying can last anywhere between three and five minutes. Over time, the nature
of this behaviour changed to a gentle tapping on the face not leaving a mark.
Example
Dean is a three-year-old boy who has been diagnosed with Downs syndrome and is said to be
functioning in the low-average range of ability. Dean picks at any scab on his skin. Over time,
this behaviour has increased and Dean now picks at the scab and makes it bleed. In this
situation, you might want to measure the scab and then measure after Dean has picked at it.
Photographs can help measure this behaviour.
4 Self-monitoring
Self-monitoring can be a very effective way to improve behaviour. Many adults can change
their behaviour by just keeping track of it. For example, if you decide to count the number of
cigarettes you smoke, you will probably smoke less just by virtue of the fact that you are
counting. The same applies to calorie or caffeine intake. If you are working with someone
who can count or track their own behaviour then this is an easy way to monitor behaviour.
Drew is an adult with dual diagnosis of mild mental handicap and schizophrenia. Drew has
difficulty managing his temper and this has actually caused him to assault others resulting in a
prosecution. Drew was asked to monitor his outbursts in the training centre he attends. He had
to write out any incidents where he lost his temper during the day identifying the lead-up to
his behaviour and the consequences of his behaviour. The staff at the training centre also
monitored his behaviour one day per week. Drew did not know when the staff would be
monitoring his behaviour. Drew was highly motivated to work on anger management as he
did not want to end up in prison again. Having him monitor his own behaviour not only
helped him to look at the lead-up to his outburst, but also gave him the responsibility for his
behaviour and involved him in working on the problem.
These methods of measurement are fine, but what if the behaviour occurs very frequently?
What if you require very detailed accurate data for research purposes? There are structured
ways of gathering this information using systems of measurement.
1 Partial interval
With partial interval you tick the box if the behaviour occurred at any time in the interval. If,
for example, you decide to collect information on initiating conversation with a peer at
lunchtime (between 11am and 11.30am) you would set up a data sheet like this one. The
person observing the child would have a timer to signal the five-minute intervals and the
person would record if the child initiated conversation with peer any time in the interval. If
your goal is to decrease a behaviour you would use partial interval because it is overestimates
behaviour and it is better to overestimate a behaviour we want to decrease (Van Houten and
Hall, 2001).
2 Whole interval
Using whole-interval recording, the behaviour has to occur during the whole of the interval to
be scored. In whole interval you underestimate the behaviour and so this is used when the
behaviour is one that you want to increase (Van Houten and Hall, 2001).
The example is as above, except the behaviour has to occur for the whole interval, meaning
we would be looking at initiating and sustaining conversation over the full five-minute period.
You lose a lot of information using these systems and so most people use duration or
frequency count instead.
On-task behaviour = engaged in the appropriate activity with the correct books/equipment and
with his attention focused on relevant material/teacher/student.
This is an example of momentary time sampling hourly intervals between 10am and
2pm.
This data can then be compared to the schedule of activities to see which activities engage the
student.
4 How to choose
The most important thing about a measurement and a recording system is that it is something
you can live with. It does not have to be very elaborate or extensive. Many behaviour
therapists take data on behaviours once or twice per week. They implement a programme
every day but only do the analysis of data on two days per week. Once you start to analyse
behaviour this way and record very specific details of behaviour the benefits reinforce your
behaviour. You can quickly see small improvements in behaviour or small steps towards a
goal. You can set more realistic goals and if something is not working you can act quickly.
The data you are collecting tells you very quickly if a programme is working.
What about very difficult behaviour? It is useful to try to define the behaviour and set up a
measurement system but some challenging behaviour has a long history and can be very
complex. It takes time to understand the function or cause of a particular behaviour. There are
There are five outcomes of functional assessment as identified by ONeill et al. (1997):
It is important to separate out the child and the behaviour. Behaviour analysts are concerned
with the circumstances in which a behaviour occurs as well as the behaviour. A child is not a
biter or a kicker. The child is seen as a child first and then a description of the circumstances
in which the behaviour occurs is analysed. The biting behaviour occurs when the child is
being changed or dressed. This is a child who bites when being changed or dressed. This is a
child who kicks when he is told he cannot leave the classroom. This is a very important
distinction but is critical in the functional analysis.
Functional assessment can be done in a number of ways. It is important to remember that the
cornerstone of the functional assessment is the ABC or Antecedent, Behaviour and
Consequences analysis. The information collected can be through questionnaire assessment,
observational assessment, and experimental functional analysis (Martin and Pear, 1999).
Whatever way the information is collected it is analysed looking at the ABCs (Reese et al.,
1977).
Another method for looking at disruptive behaviour and its function is the Motivation
Assessment Scale (Durand and Crimmins, 1988). This scale helps to classify behaviour into
one of four areas: sensory feedback, escape form aversive conditions, attention from others or
tangible reinforcers. There are problems with this scale as there are some behaviours which do
not fit into these categories and also some behaviours which serve multiple functions. For
example, self-stimulatory behaviour such as hand flapping may provide visual stimulation,
which is reinforcing, but it may also allow the child to avoid social interaction because other
children stay away and it may also be used to gat a teachers attention as the teacher has to stop
teaching to tell the child to stop.
ONeill and his colleagues (1997) suggest using the FAO form (functional assessment
observation) as it maximises the information obtained without going into lengthy summaries.
Also, they suggest that this form is easily summarised. Each incident of behaviour is given the
number of sequence in which it occurs and it is then recorded on the form. The incident is
logged under behaviours, predictors (demand-request, difficult task, transitions, interruptions,
alone/no attention), perceived function (get/obtain, escape/avoid), and the actual
consequences. So it is similar to the ABC but the behaviour or B is first, predictors or A is
next and then a detailed section similar to C (consequences) called perceived function
get/obtain, escape/avoid.
Another variation on the ABC analysis, which combines observation and interview of
informed people in assessing the behaviour, is the STAR system. Zarkowska and Clements
(1988) introduced the STAR (Setting, Trigger, Action and Result) model to analyse and
manage problem behaviours.
Settings are the stable features of the environment or the general context in which actions
occur. Settings can be places, people, times, tension in the environment, hunger, pain,
discomfort, etc.
Triggers are signals that 'set off' or stop actions in a given situation. A whistle was the trigger
for the tantrum on the playground. A swimming costume will suggest to a child that it is time
for swimming. However, if swimming does not follow this trigger, the child may become
confused and anxious and this may lead to challenging behaviour.
Actions are the observable behaviours. They can be challenging behaviours but they can also
be existing skills which need to be encouraged.
Results are the events that follow an action. They provide information to the child about the
appropriateness of an action. They also serve as reinforcers to either encourage or discourage
the child from performing that action again.
The disadvantages of this is that it is time consuming, however, you can be sure of an accurate
analysis at the end of it. Also, it is not possible ethically to set up a situation where self-injury
occurs so it excludes certain severe behaviours.
Click on this link to access the functional assessment There is always a pattern to the
checklist.
behaviour but sometimes we
miss critical information or
The functional assessment is a tool to help understand sometimes there are two
behaviour and put it in context. Any of the above methods patterns contributing to the
will assist you in this. It can be a difficult task. On same behaviour.
occasions it appears as if there is no pattern to a behaviour.
There is always a pattern to the behaviour but sometimes we miss critical information or
sometimes there are two patterns contributing to the same behaviour. Once all the information
is gathered the next step is to put in place a treatment programme. If your analysis is correct
the treatment programme will work. If it is not correct then you have to revise your
assessment and go back and redo it until you come up with the correct analysis and solve the
problem. There are other observation methods, which look not only at challenging behaviour
but also at the students level of interest in a programme or activity.
This assessment was developed to help record temporal factors in challenging behaviour
(Touchette et al., 1985). A scatter plot is simply a graph of times and days where you track
the occurrence or non-occurrence of behaviour (aggression, temper tantrums, self-injury).
This then tells you when the behaviour occurs and when it does not and so by changing the
conditions of the person it is possible to change the behaviour. It is a graphic presentation of
the behaviour and allows for the functional analysis by manipulating the daily schedule to
change the behaviour.
In this example, you can see Saras tantrum behaviour typically does not occur at 10-10.30am,
12-1.30pm, 4-4.30pm and 5-6.30pm. It rarely occurs at snack or mealtimes. By looking at the
activities scheduled, we can identify the preferred tasks or activities for Sara. If it is possible
to increase the preferred tasks throughout the day, this may help to reduce the behaviour.
Also, it may be that the tantrums are related to hunger so by moving the mealtimes forward it
may be possible to improve Saras behaviour. Another alternative is to teach Sara how to
access food or snacks independently when she is hungry. By simply rearranging events or the
schedule, you may be able to decrease the challenging behaviour.
5 PLA Check
PLA Check is a variation of momentary time sampling (remember where you look up at the
end of the time interval and see if you can catch the behaviour) but for using with a group.
PLA Check was developed by Doke and Risley (1972). They compared participation in
before-school group activities. In a further study (Risley and Cataldo, 1973) examined the
behaviour of preschool children using PLA Check. At particular times observers scored the
number of children who were engaged in an activity. By dividing this by the total number of
Example
From this example you can see that during the first activity observed, table work, eight out of
20 children were on task for the first three minutes then the figures dropped to only four
children on task by nine minutes. If advising the teacher in this situation, you would suggest
modifying the task or decreasing the time spent at the task. By comparison, the number of
children engaged in free play was between 14 and 19 out of 20. So this is a high-interest
activity. By using PLA Check you can see what activities hold the childrens interest and what
activities do not. By making some modification in the daily schedule, you may be able to
increase the amount of time the children spend on task.
We have looked at a range of measurement and assessment tools and systems of measurement
There is a tendency for people to want to take this information and, in a rush of enthusiasm,
dash off to set up data collection systems. But be cautious, if you wait you can avoid some
mistakes that are common when beginning a behavioural programme. We have highlighted
some of these below.
Behaviour is complex and often we go for a very simple analysis. The most common analysis
is that the child was looking for attention. If we then decide to implement a programme based
on the child seeking attention and the behaviour has another function such as self-stimulation,
then our programme will not work. The data tells us what to do. If we have the analysis right,
the behaviour changes very quickly. It is worth putting the time in early on and getting the
analysis right before we move in and try to change the behaviour.
Stay positive
Another common problem is setting negative targets and writing programmes with the focus
on negative behaviour. Many children are discouraged and feel incompetent. By setting goals
in negative language we add to this. When writing a goal, ask yourself, “Is there a way I can
phrase this in the positive?” Try these examples. What would be the positive of hand flapping,
rocking, not paying attention, making screeching noises, and stripping off clothes?
Overload
Another common error is to try to work on too many behaviours at the same time. The best
thing to do is to have a meeting to plan the programme and set priorities to work on. Involve
as many people as possible who can contribute information about the child. Once you have
gathered the information then try to decide what behaviours you want to work on. How do
you pick the behaviours? Brown’s six questions provide a good guide. Also, you have to take
the age of the child and the child’s level of ability into account. While Brown talks about the
criterion of ultimate functioning with younger children it is often worth looking at the next
stage, so with a toddler you would prioritise increasing skills needed for preschool and
decreasing any behaviours which would prevent the child from attending preschool.
Goals for preschoolers have been well-documented (Lovaas, 2002, Maurice et al., 1996, Leaf
et al., 1999). With a preschool-aged child, again look at the next environment – primary
school and what skills the child requires and what will prevent the child from going to school.
Once in primary school think about secondary school and what are the skills needed to
progress onto this environment? The new legislation in the United States IDEA (1995/1997)
requires planning for transition from school to start at 12 years. So from the age of 12 years
the individual education plan contains goals towards this transition from school to a work or
after-school college or further training centre.
Keep it simple
Another problem is we try too hard and set ourselves, as teachers and parents, goals that are
too elaborate and data collection systems that are too extensive. The methods selected to
gather information depend on the information required and the resources available. Often we
get too ambitious and set up very elaborate data sheets. It is important that whatever way you
decide to track the behaviour that it is possible in the setting you work in and with the
resources you have. Here are some examples of data collection procedure.
Example
One seven-year-old is out of bed during the early part of the night. This observation system
was set up to track this behaviour.
From this we can see that the child was out of bed more frequently in the early hours of the
night. On Thursday night it was noted that the child was agitated (she was louder then usual
and engaged in more motor activity) – if this pattern continued on Thursdays then you would
look at the activities the child engages in on Thursday and see if this could be part of the
problem. Tracking this data over a number of weeks would help you establish a clear pattern.
Then an intervention could be planned including a bedtime routine, wind-down activities,
perhaps setting the bedtime later if this is an option. The data will guide you in the process. If
you set the bedtime later and find that the behaviour just moves to 9pm then going to bed later
is not working and you revert back to the earlier bedtime.
Example
Shane makes loud noises, which disrupt the class. It was decided to take a count between 2pm
and 2.30pm three days per week in school when the class is working on independent
assignments.
Operational definition
Brown, L., Nietupski, J., & Hamre-Nietupski, S. (1996). The Criterion of Ultimate
Functioning and Public School Services for the Severely Handicapped Students. In
M.A. Thomas (Ed.), Hey Dont Forget About Me: Educations Investment in the
Severely, Profoundly and Multiply Handicapped (pp. 2-15). Retson, VA: Council for
Exceptional Children.
Carr, E.G., & Durand, V.M. (1985). Reducing Behaviour Problems Through Functional
Communication Training. Journal of Applied Behaviour Analysis, 18, 111-126.
Cooper, J.O., Heron, T.E, & Heward, W.L. (1987). Applied Behaviour Analysis. Prentice-
Hall.
Doke, L.A., & Risley, T.R. ( 1972) The organisation of day care environments: required vs
optional activites, Journal of Applied Behaivour Analysis, 5, 405-420.
Durand, V.M., & Crimmins, D.B. (1988). The Motivational Assessment Scale (MAS)
Administration Guide. Topeka, KS: Monaco and Associates.
Grant, L. & Evans, A. (1994). Principles of Behaviour Analysis. Harper Collins College
Publishers.
Iwata, B, Wallace, M.D., Kahng, S.W., Lindberg, J.S., Roscoe, E.M., Conners, J., Hanley,
G.P., Thompson, R.H., & Worsdell, A.S. (2000). Skill Acquisition in the
Implementation of Functional Analysis Methodology. Journal of Applied Behaviour
Analysis, 33, 181-194.
Leaf, R., & McEachin, J. (1999). A Work in Progress Behaviour Management Strategies and
a Curriculum for Intensive Behavioural Treatment of Autism. DRL Books New York.
Lovaas, O.I. (2002) Teaching Individuals with Developmental Delays Basic Intervention
Techniques. Pro-Ed Texas.
Martin, G. & Pear, J. (1999). Behaviour Modification: What it is and How to do it. Pearson
Education.
Maurice, C. Green, G., & Luce, S., (1996). Behavioural Interventions for Young Children
with Autism A Manual for Parents and Professionals. Pro-Ed Inc. Austin Texas.
Mazaleski, J.L. Iwata, B.A., Vollmer, T.R., Zarcone, J.R. & Smith, R. G. (1993). Analysis of
the Reinforcement and Extinction Components in DRO Contingencies with Self
Injury. Journal of Applied Behaviour Analysis 26, 143-156.
ONeill, R.E., Horner, R.H., Albin, R.W., Sprague, J.R., Storey, K., & Newton, J.S. (1997).
Functional Assessment and Program Development for Problem Behaviour A Practical
Handbook. Brookes/Cole Publishing Company USA.
Pavlov, I. (1927). Conditioned Reflexes: An investigation into the physiological activity of the
cerebral cortex. New York Dover.
Reese, E.P., Howard, J.S., & Reese, T.W. (1977). Human behaviour: An experimental
analysis and its applications. Dubuque, IA: Wm. C. Brown.
Risley, T. & Cataldo, M.F. (1973). Planned Activity Check: materials for training observers.
Lawrence KS: Center for Applied Behaviour Analysis.
Skinner, B.F. (1953). Science and Human Behaviour. New York: Macmillan.
Sulzer-Azaroff, B. & Roy Mayer, G. (1991). Behaviour Analysis for Lasting Change
Wadsworth Thompson Learning.
Touchette, P.E., MacDonald, R.F., & Langer, S.N. (1985). A Scatterplot for Identifying
stimulus control of problem behaviour. Journal of Applied Behaviour Analysis, 18,
343-351.
Van Houten R., & Vance Hall, R. (2001) The Measurement of Behaviour: Behaviour
Modification. Pro-Ed Inc.
Will, T.J., LaVigna, G.W., & Donnellan, A. (1987). Behaviour Assessment Guide. Institute
for Applied Behaviour Analysis, Los Angeles CA.
Zarkowska, E. & Clements, J. (1988). Problem Behaviour and people with Severe learning
Disabilities: A Practical Guide to a Contructional Approach. London: Croom Helm.