Wednesday, 28 August 2013

Birmingham nursery rapist – confirmation bias?

Nursery worker Paul Wilson was jailed for life this week for among other things raping a small child at the Birmingham nursery where he worked.

Jane Held of the Birmingham Safeguarding Children Board provides a clear account of what went wrong in a BBC TV interview. 

She explains that staff at the nursery had expressed concerns about Wilson’s forming an inappropriately close relationship with the child, but that the local authority's designated officers made a judgement on the basis of what they were told that this was “… a matter of professional conduct not a matter of child abuse”.

Subsequent investigations by Ofsted and the local authority lacked rigour and depth, she said.

Crucially we are told that nobody talked to the child.

It seems to me that this case demonstrates what human factors experts call confirmation bias. This is a type of loss of situation awareness in which facts are selected or distorted to support or confirm a wrong mental model of the situation. ‘Bending the facts to fit the theory’ is another way of saying it.

So once a decision was taken – an issue of professional conduct, not a matter of child abuse – those investigating simply clung to this 'fact' and whatever evidence they found seemed to support it.

Confirmation bias is something that happens to all of us. Those who work in safety critical industries and those who have special safety responsibilities – such as Ofsted inspectors and local authority designated officers – should always be aware of the possibility that evidence can be tailored to support the status quo. They need to make sure that they double-check any theory or belief from an alternative standpoint.

Speaking to, and listening to, the child in this case would probably have made people think twice. Being aware of the risks of confirmation bias would have disposed those conducting investigations to review their findings more critically.