Lies, damned lies and statistics: misinterpreting data on child sexual abuse
Anyone who has been wrongfully accused of child abuse will know how difficult it is to convince others, whether police, social services or their employers, that they are innocent.
One reason for this may be that everyone is subject to cognitive biases. These are very common thinking errors which influence the way we make decisions. One type which is especially relevant is confirmation bias. This is the tendency for people to search for or interpret information in a way that confirms their preconceptions, and ignore information that doesn’t support their view. Unfortunately we may not be aware that we are subject to these biases, and may not know where they came from.
Faulty preconceptions could arise from a misunderstanding of statistical information. One of our members who had been wrongfully accused had to have a risk assessment to return to his organisation. He was dismayed to hear the assessor state that “only 2% of those who are abused get justice”. The context of this statement was such that it seemed the assessor was thinking ‘98% are guilty but get away with it, and so it’s highly likely that even though you haven’t been charged you are in fact guilty’. Obviously, this was not a good starting point for an open, unbiased assessment of the facts.
I decided to look into the statistics around child sexual abuse (CSA) to find out if there was any truth in the statement that only 2% of those abused get justice. I used ‘Child abuse and neglect in the UK today’ commissioned by the NSPCC in 2009, the Children’s Commissioner’s report ‘Protecting children from harm’ from November 2015, and the Office of National Statistics Crime Survey of 2016. It was a long and difficult process to extract all the necessary data. Different surveys used different definitions of CSA and even different definitions of childhood.
I was able to estimate figures for the various subgroups in the following diagrams. The details of how all these numbers were calculated can be found here. I’ve made some graphics to illustrate the figures more clearly.
My aim is to show why the statement “only 2% of those who are abused get justice” can give a false impression of what is really happening. I also want to explain how this statement doesn’t exclude the possibility that a substantial proportion of accusations are false. During my exploration of these reports I discovered other interesting information which I’ll share with you.
The first graphic illustrates the number of new cases of child sexual abuse in a year (in this case 2009) in England, and the annual number of convictions for those offences (average between 2012 -2014). These figures are mainly extrapolated from an NSPCC survey of self-reported abuse in a sample of 1,761 young adults aged 18 – 24 in 2009. The total of 167,000 is broadly in line with the figure of around 200,000 calculated using a different method, in the Inquiry into Sexual Abuse within the Family by the Children’s Commissioner in 2015.
The first impression is that there is an appalling injustice here, with only 2% of abuse ever being resolved by a conviction.
The next graphic shows how many people were abused by an adult. A child is anyone under the age of 18 and of course an adult is 18 or older. So it is clear that most of the self-reported abuse was perpetrated by siblings, or other children or intimate partners below the age of 18.
Next, I’ve added those who have suffered contact sexual abuse, that is, abuse where there has been some physical contact with the perpetrator. Although all abuse is clearly wrong and can cause long term harm, most people would agree that contact abuse is more severe than non-contact abuse. Although approximately 26,000 cases of contact abuse were perpetrated by adults the majority (52,000) were perpetrated by other children such as siblings and ‘intimate partners’ (presumably boyfriends).
And now I’ve added an estimation of how many were abused by rape or penetration including attempted rape and penetration. This estimation uses data from the ONS Crime Survey 2016. Note this this figure is for the under 16s only, so would be larger if we had data that included 16 and 17 year olds.
In this image I’ve added those that were reported to the police in a year (average for 2012 – 2014).
I don’t have any information about the age of those reported or whether they were accused of non-contact or contact sexual abuse, so it is difficult to know exactly where to place the yellow circle. In fact, the yellow circle should really be placed partly outside the blue one, to include the wrongfully accused.
What is surprising is that such a small proportion (11%) of cases is reported to the police. So why weren’t they reported? According to the Children’s Commissioner’s ‘Inquiry into Child Abuse in the Family Environment’ there are many reasons why children abused in the family environment didn’t tell anyone about it. This is a list of the most common reasons.
• I felt scared or afraid
• I felt ashamed/guilty
• Didn’t want to upset other family members/to protect others
• Didn’t know how to explain it
• I thought I would get into trouble
• Didn’t think anyone would believe me/I wasn’t believed/nothing happened after telling
• I didn’t know it was abuse/was confused
• Didn’t trust anyone
• I was being threatened/told not to tell
• I was afraid of dishonouring my family
The inquiry found that survivors spoke about their desperation to be asked about what was wrong with them. Teachers were the professionals that children preferred to talk to about their abuse. It found that the ‘Savile effect’ – whereby victims and survivors feel encouraged to report abuse to the authorities with greater confidence in the ability of the Police and other services to respond effectively – was evident for adult survivors, but had not extended to children and young people. The inquiry recommended better training for teachers to recognize and respond to abuse and more effective Achieving Best Evidence (ABE) interviews by the police.
Here I’ve added the brown circle representing the number of convictions in a year. Again, it is difficult to know exactly where to place it, but it should probably be partly outside the blue circle. The conviction rate for those actually reported to the police was about 17%.
This is only part of the story because nobody really knows how many cases reported to the police are wrongful allegations. This next graphic illustrates that there may be people who have genuinely been abused and those that believe or suspect that they have been abused. There are also those who don’t know that they have been abused, because they may not realise that what happened was wrong.
The difficulty is in knowing what size these circles should be and how much they overlap. It may be something like this.
How can someone come to believe that they have been abused when they haven’t? Sometimes genuine mistakes occur. A person may misinterpret something that happened to them as abuse when in fact it wasn’t. People can also come to develop false memories either as a result of some kinds of psychotherapy or from associating with others who have been abused. False beliefs can even be induced by poor interviewing techniques. You can read more about false memories here.
Here I’ve added a red circle representing those who have been reported to the police. Again, it’s not possible to know exactly where to place it.
If we zoom in on this image, the graphic looks like this.
Unfortunately nobody knows for sure how many of those accused of abuse are actually guilty. There are serious problems in working out how many allegations are false, however the experience of FACT and other support groups for the wrongfully accused is that false allegations certainly occur and can cause serious harm, (See the Oxford University Paper ‘The Impact of Being Wrongly Accused of Abuse in Occupations of Trust: Victims’ Voices’) and even wrongful imprisonment. Herman in ‘Wrongful Allegations of Sexual and Child Abuse’ estimated that as many as 15% of allegations of CSA made by children are false. This doesn’t mean that children deliberately lie; in fact Herman states that most false allegations made by younger children are the result of repetitive suggestive questioning by adults.
So what can be concluded after looking into the data on child sexual abuse? It goes without saying that all CSA is wrong. However the public perception of a child abuser is probably that of an adult having penetrative sex with a child. The data show that this perception is wrong. I was surprised to find that most contact and non-contact CSA as defined by the NSPCC was committed by other children. I had to dig deep into the data to find this information and could not help wondering why it was not stated more openly in the report. Most sexual abuse was “non-contact”. Most contact CSA probably does not involve penetration or rape. Misuse of the figures – such as stating that there are 200,000 cases of CSA a year – without qualifying the data, could grossly inflate the public perception of the severity of the problem. This in turn may influence those sitting on a jury, those working in safeguarding and those responsible for making government policy.
The main reason why abusers are not brought to justice is because about 90% of victims never report their abuse to the police. The Children’s Commissioner’s report added useful information about the reasons why children don’t report their abuse. The report did not comment on the police’s policy of ‘believe the victim’ but it stated that teachers were the preferred professional for children to talk to about their abuse. It seems unlikely that younger children would approach the police rather than their teacher, so the ‘believe the victim policy’ probably wouldn’t help them.
The fact that CSA is greatly under-reported, and that only a minority of those reported are convicted does not rule out the possibility of a significant number of allegations of CSA being false. All suspects need to be investigated by a detective who is open to the possibility that they may be innocent and who will seek out evidence that supports their version of events and not only that of the complainant. The ‘believe the victim’ policy must surely be a cause of bias in the investigation, and instead the public should be told ‘if you make a complaint we will treat it very seriously and investigate it thoroughly without fear or favour’. Henriques Report October 2016.
Child abuse and neglect in the UK today, Radford, L et al. ,2011, NSPCC
Children’s Commissioner: Inquiry into Child Abuse in the Family Environment, 2015
Office of National Statistics, 2016 Crime Survey and demographic data for 2009
Wrongful Allegations of Child and Sexual Abuse, Ed. Ros Burnett, 2016, Oxford University Press