This is the fourth in a series of posts prompted by the NSPCC briefing Home education: learning from case reviews. The briefing highlights recommendations from seven serious case reviews (SCRs) published between 2008 and 2013 involving home-educated children. [The full briefing has since been replaced with a summary, but the original is still accessible here. Also note that the Serious Case Review for Child S listed in the NSPCC summary is for the wrong Child S.] In my view, the briefing’s perspective on home education legislation is based on a misunderstanding of the legal framework that underpins it. I think the briefing also misunderstands how the law tackles risk.
legislation and education
In a democracy, the primary function of legislation is to protect the liberty of individuals. But there’s a tension inherent in that principle because protecting the liberty of one person usually means limiting the liberty of another. So legislators have to weigh up the costs and benefits of legislation to different people. I think this is what Graham Badman was getting at when he opened his 2009 report on elective home education (EHE) with a quotation from Isaiah Berlin’s Four Essays on Liberty:
“The need to choose, to sacrifice some ultimate values to others, turns out to be a permanent characteristic of the human predicament”
But Badman then goes on to frame UK education legislation relevant to EHE in terms of a “balance between the rights of the parents and the rights of the child” (p.3). Not only is the legislation not about ‘balance’, it’s not framed in terms of ‘rights’ either, although its purpose is to protect children’s liberty; it’s about parents and local authorities having specific duties that they are required to carry out in certain circumstances. Giving people duties doesn’t sound like it protects anyone’s liberty. It does in the case of education because education should help people understand how the world works and understanding how the world works helps people protect their own liberty and that of others. Under UK law, parents have primary responsibility with regard to educating children and local authorities have powers to step in if the parent asks them to or if it appears that the parent is in breach of the law.
What Badman, the NSPCC and many local authorities are most concerned about is the risk of harm to children educated at home. They argue that home-educated children are denied safeguards applied to children who attend school. In their view, home-educated children are by definition at risk of suffering harm that ‘the authorities’ wouldn’t be able to prevent because they wouldn’t know about it. What they are concerned about is a hypothetical risk.
legislation and the risk of harm
The law takes a dim view of harm and much legislation is about protecting people from it. But hypothetical risks pose a problem for the law because there’s an almost infinite number of them and it would be impossible (and sometimes counterproductive) to legislate against them all. As a consequence, legislation is almost invariably based on evidence of real risk.
For example, Mr A might take it into his head to go on a shooting spree and kill and injure several passers-by. To say that there’s substantial evidence that being killed or injured constitutes a real risk of harm is a something of an understatement, so the law prohibits people from killing and injuring each other. The real risks associated with people being killed or injured are pretty clear-cut. Other risks aren’t so straightforward, such as the consequences of not wearing a seat belt when motoring.
In the 1970s Ms B might have argued that she’d been driving in the UK for forty years without a seat belt and had come to no harm – and even if she did it would be her lookout, so making the use of seat belts compulsory was nothing more than an infringement of her liberty. All that Ms B’s evidence tells us is that not wearing a seat belt doesn’t inevitably lead to harm. As for it being only her business, if she’d been thrown through the windscreen she wouldn’t be the one picking up the pieces; personal tragedies have a wide impact.
On the face of it, the campaign for seat belt legislation and motorists’ objections to it look remarkably like calls to tighten up home education legislation and parents’ objections to that. One obvious difference is that they involve very different activities. A less obvious difference is whether there is evidence showing that not wearing a seat belt or educating children at home constitute real risks as distinct from hypothetical ones. Evidence of the real risk involved in not wearing a seat belt could be seen in casualty departments in hospitals up and down the land in the 1970s. Evidence shows that the real risk attached to being home educated is so vanishingly small that it’s non-existent.
major incidents and serious cases
In the BBC documentary Baby P: The Untold Story, Edi Carmi, who led the serious case review, observed that all the agencies involved made mistakes, but it was when they all happened at the same time that tragedy struck. Mistakes happening at the same time is a familiar scenario in risk management. It was a key factor in the Three Mile Island and Bhopal disasters. The people who design safety-critical systems like those in nuclear power stations and factories handling toxic chemicals take it for granted that mistakes will happen. But they design systems to ensure that if and when mistakes happen, they are very unlikely to happen at the same time. The key phrase is ‘very unlikely’. Systems designers can only minimise risk – they can’t eliminate it entirely.
One characteristic of major incidents is several rare events happening simultaneously; a plane crashes because two engines fail at the same time as a pilot is taken ill and a passenger has a heart attack. I expected to see a similar pattern of rare events coinciding in the seven SCRs; a social worker crashes her car on the way to visit a parent who’s just received a diagnosis of cancer and whose eldest child has just been arrested.
Instead, the SCRs paint a picture of long-term challenging circumstances for the families, coinciding with a series of commonplace mistakes on the part of the agencies supposed to be supporting them. The children in the seven SCRs were adopted, chronically sick, disabled, bereaved, had special educational needs or faced housing crises. The mistakes made by the agencies were typical of those made in organisations that are overstretched; staff lacked relevant knowledge or experience, there was high turnover and problems with cover. Information wasn’t passed on promptly, or wasn’t passed on at all. No one had an overview of individual cases and no one had time to check that procedures were being followed and agreed actions carried out.
home-education and risk
According to the NSPCC, there were seven cases of home-educated children coming to serious harm between 2008-2013 – around 1% of the total number of SCRs for the equivalent period. In the University of East Anglia’s biennial surveys of SCRs I could find no mention of any additional cases between 2003-2007 involving home-educated children. If those seven cases are the only ones in the last decade, the 1% proportion would be more or less halved. In other words, only very small proportion of children coming to serious harm were home-educated.
A closer look at the seven SCRs shows that being home educated doesn’t itself constitute a real risk to children – and that monitoring home-educated children doesn’t guarantee that they won’t come to harm. What characterized the part played by home education in the seven case reviews is local authority confusion about it. LA officers were unclear about the law, their statutory duties or about who was responsible for doing what. But the fact that home education posed a problem for ‘the authorities’ doesn’t mean that it put the children at risk. If there is no evidence of home education posing a real risk to children, why do some people working in child safeguarding think there is one?
Child safeguarding, like any other specialist area, is a small world, and people working in a small world tend to use that world, rather than the wider context, as their frame of reference. Child safeguarding is also a difficult and stressful occupation and agencies are clearly overstretched and have been for some time. Professionals working under intense pressure with inadequate resources are likely to perceive unusual family circumstances such as children being educated at home, coupled with an inadequate understanding of the relevant legislation, the ensuing confusion about what procedures to follow, and the risk of serious consequences if they fail to protect a child, as a real risk when in fact it’s a hypothetical one. Research into risk perception suggests that people find it difficult to make a distinction between hypothetical risks and real ones. That’s why basing policy on evidence rather than opinion is so important.
The evidence shows that there are particular groups of children at real increased risk of harm; being a young child or using roads for example. Over 60% of all SCRs involve under-5s and around 2,000 children are killed or seriously injured on UK roads every year. Steps have been taken to reduce those risks through legislation, but legislators are aware that they are risks that are unlikely ever to be eliminated completely. By contrast the evidence indicates that that being home educated isn’t something that poses a real risk to children. Tightening up the legislation might make the people worried about a hypothetical risk feel better, but it can’t reduce a risk that doesn’t exist.
What did put the children in the seven SCRs at risk was a combination of challenging circumstances for their families and mistakes on the part of the agencies whose job is supposed to be to help them. In the next post I look at something the NSPCC briefing and the SCRs mention only in passing – what services were on offer to help the families with the challenges they faced.