"Functional Analysis is the most precise, rigorous, and controlled method of conducting a functional assessment"- O’Neill, Horner, Albin, Sprague, Storey, and Newton (1997, p. 7)
This is a very basic outline of a functional analysis and is completely hypothetical – please do not emulate it. It is provided to give a general outline of how an analysis might be carried out but you should never attempt to do anything like this without a professional supervising the entire assessment.
Please be aware that prior descriptive assessments would generally be made before doing a functional analysis i.e. direct observation of the client as well as interviews and questionnaires completed by parents, staff members etc.
The information gathered through these initial methods would allow practitioners to develop a hypothesis about the function of the behaviour and guide the development of the functional analysis.
In addition, all relevant stakeholders (e.g. parents, staff, ethics boards etc.) would be consulted about the goals of the analysis and informed consent obtained from all parties.
For this hypothetical functional analysis, the client is an 8 year old boy with autism who is engaging in "head banging" (self-injury) which we will define using Iwata et al’s (1994, p. 219) definition: ‘audible or forceful contact of the head against a stationary object’.
This analysis will use an "alternating design" which is sometimes called a "multi-element design". This basically means that during a given session/day, only one type of manipulation will be used, and then on the next session/day a completely different manipulation will be used.
Each type of manipulation would require its own session/day and so each manipulation will “alternate” and thus it is called an “alternating design”.
Although we are alternating each manipulation over successive days, a functional analysis might do this over hourly sessions, two hourly sessions, half-day sessions etc.
For the first day, the practitioner gives the boy attention every time he bangs his head and records the frequency of head bangs throughout the entire day.
This would be called the "contingent attention" condition and if the behaviour was frequent during this condition it would suggest the function of the behaviour is to gain social attention.
During this condition, head-bangs occurred nine times.
The boy would be observed while on his own; there would be low levels of “stimulation” available i.e. no practitioner, teacher, parent, toys etc. present. The frequency of head bangs could be recorded via CCTV or a one-way mirror.
This would be called the "alone" condition and if the behaviour was frequent during this condition it would suggest that the behaviour is occurring because of "automatic" reinforcement.
The easiest way of explaining automatic reinforcement is to think of it as something that is "self-pleasing" or “self-stimulating” and does not involve another person (e.g. reading a book). You might wonder how head-banging could be self-pleasing but this can occur because, for example, there may be some form of internal self-stimulation that is obtained from the head-banging.
During this condition there is one head-bang.
On the third day, the frequency of head banging is recorded while the child is in free play - which basically means the child plays games or with toys of their choice with the practitioner.
This is called the "free play" condition and it is used as a control for the other conditions as it basically gives the client what he wants, places no demands on him and gives him attention even when he is not head-banging.
As a result, it would be expected that the frequency of head bangs would be low (Cooper et al, 2007) but this is not always the case; e.g. see The functions of self-injurious behavior: An experimental epidemiological analysis by Iwata et al (1994).
During this condition there is one head-bang.
On the fourth day, the practitioner would stop running any academic programmes each time the boy engaged in head-banging.
This is called the “contingent escape” condition and if the behaviour was frequent during this condition it would suggest the function of the behaviour was to get away from having to complete academic tasks. Basically, he is using the head-banging behaviour to get the teacher to stop making him complete any academic tasks.
During this condition there are no head-bangs.
These four conditions would be alternated over a number of days and data would be continuously recorded on the frequency of head-bangs throughout.
This data would then be plotted on a graph so it can be visually analysed in order to identify a possible causal link between the manipulations and the frequency of the behaviour.
We have plotted hypothetical data from this functional analysis below; showing data recorded over a 12 day period. From this data, can you identify what the cause or function of this boy’s head-banging is?
As can be seen, the behaviour occurs more frequently when the practitioner gives the boy attention in comparison to when the client is on their own (alone condition), is playing (free play condition) or when teaching sessions are stopped (escape condition).
As you can probably tell, this data would strongly suggest that the function of the behaviour is to get attention from the practitioner.