Uncorrected Evidence 44

Memorandum submitted by Ms Dani Ahrens

1. Executive Summary

1.1.1. This submission sets out serious flaws in the research which informed the Badman Review report.

1.1.2. How data was gathered:

· The literature review was partial and contained factual errors

· The online questionnaires for the public and local authorities were based on misrepresentations of the law

· A second questionnaire for some local authorities contained leading questions and resulted in a small, unrepresentative sample

· Graham Badman has now contacted local authorities for a third time, explicitly asking for data to support the conclusions already published.

1.1.3. How data was analysed:

· Inconvenient data, such as the responses to the public questionnaire, were ignored by the Review team

· Conclusions were instead based on unrepresentative and unsafe data from a few local authorities

· Illogical conclusions were drawn from this data - the proportion of registered home educated children known to social care does not imply a similar picture among unregistered home educators.

1.1.4. The case made in the Review report:

· The report does not present any data to support its conclusions

· The central notion of a balance between parents' and children's rights is illogical and misrepresents the law.

1.1.5. Conclusion and recommendations:

· Government action and new legislation based on the conclusions of this flawed report should be stopped

· Local authorities should use their existing powers where they have concerns.

2.1.1. I am a home educating parent. I have no particular expertise in research or statistical analysis, but as an active citizen I am extremely concerned about public policy being formed on the basis of inadequate and flawed research.

3. How data was gathered

3.1.1. At every stage, the information gathering process was flawed, giving cause for concern about bias.

3.2. Literature Review

3.2.1. The literature review was not published with the Review report, but was released on 30^th July in response to a Freedom of Information Act request[1]. It is unacceptable that so much crucial information has had to be extracted from the DCSF and the Review team by members of the public in this way.

3.2.2. Though the literature review includes US research, it omits several published studies that give an insight into the practice and outcomes of home education in the UK. For example:

· Mike Fortune-Wood (2005), The face of Home-based Education 1: Who, Why and How, Educational Heretics Press

· Mike Fortune-Wood (2006), The Face of Home-based Education 2: Numbers, Support, Special Needs, Educational Heretics Press

· Julie Webb (1997), Those Unschooled Minds: Home educated children grow up, Educational Heretics Press

3.2.3. Most worryingly, references to Paula Rothermel's 2002 research for her doctoral thesis[2] are inaccurate and misleading.

3.2.4. Dr. Rothermel's research is described as "a UK-based study of 35 home-educated four year olds from diverse socio-economic backgrounds". In fact, the PIPS baseline assessments Dr. Rothermel carried out with 35 four year olds were part of a larger study, involving questionnaires completed by 419 home educating families (with a total of 1099 children), 100 interviews and 238 assessments of children aged between 3 and 10.

3.2.5. The home educated children in Dr. Rothermel's study achieved significantly higher scores than school children of comparable ages, in all the attainment tests she performed.

3.2.6. However, a key conclusion from this important research was that assessment tools designed for use in schools are inappropriate for home educated children. This issue was not mentioned in the literature review.

3.2.7. It has been reported that Graham Badman repeated the error about the number of children involved in Dr. Rothermel's research, while conducting the Review[3]. This suggests he had not read Dr. Rothermel's research.

3.2.8. The literature review is dated January 2009 and headed "Schools Analysis and Research Division".

3.2.9. The bibliography includes a reference for the 2006 report by Arthur Ivatts, with the note "accessed on 26/6/08".

3.2.10. It seems the literature review was written by DCSF officers, not by the 'independent' Review team, and was in fact compiled before the Review was announced in January 2009.

3.3. Online questionnaire for Local Authorities

3.3.1. The Review team sought information from Local Authorities, initially, by inviting responses to an online questionnaire, published as Annex D of the Review report.[4]

3.3.2. Questions 12 to 21 of the questionnaire ask for details of how home educated children are monitored and assessed by the local authority. The assumption is that all local authorities have a system of visiting all home educating families registered with them, to assess suitability of education.

3.3.3. While many local authorities do have systems like this, they are not required to do so; DCSF guidelines (issued November 2007) state that local authorities have no duty to routinely monitor home education, and only need contact home educators if there is reason to believe a suitable education is not being provided.

3.3.4. The existence of systems for routine monitoring should not therefore have been assumed and implicitly approved by the Review team.

3.3.5. For the rare local authorities who are not operating an ultra vires system of routine monitoring through home visits, the questionnaire would have been difficult to complete. Thus the insights gained by local authorities like Milton Keynes, into the advantages of building a positive, respectful relationship with the home educating community, will have been missed by the Review team.

3.4. Public questionnaire

3.4.1. Members of the public were also invited to complete an online questionnaire.

3.4.2. The bulk of the questionnaire was based around the Every Child Matters 'five outcomes'. These outcomes are a way to assess the impact of public services, not to measure the success of individual children's upbringing. Most of the questions were therefore nonsensical.

3.4.3. The questionnaire included confusing references to "the current system for safeguarding children educated at home", "the current system for supporting home educating families" and "the current system for monitoring home educating families". As there are no specific systems in existence for any of these things, answers to these questions will be based on a wide range of perceptions and assumptions on the part of the respondents.

3.5. Second questionnaire sent to 90 Local Authorities

3.5.1. Although not mentioned in the Review report, there was a second questionnaire sent to the 90 local authorities who responded to the first one. The existence of this questionnaire was discovered through FOI requests.[5]

3.5.2. According to a briefing paper[6] (also released following a FOI request), local authorities did not answer question 22 of the initial questionnaire (about the number of cases where home education had been used as a cover for abuse), but instead reported the numbers of EHE children known to social care. It seems the Review team therefore decided to shift the focus of their evidence gathering to this wider measure.

3.5.3. Only 25 local authorities[7] responded to the second questionnaire[8], and not all were able to answer all the questions.

3.5.4. Question 1 (how many EHE children are known to children's social care, broken down into categories) is complex and difficult to answer. In order to exclude "children who are disabled where there is no concern about parenting or quality of EHE" from the number of section 17 enquiries, it would have been necessary to look through each case file individually. Of the seven responses so far released, five do not include detailed figures in response to this question[9].

3.5.5. Questions 2, 4 and 5 invite local authorities to speculate about a range of poorly defined "concerns" they may have about home educating families in their area, and about the motivation of home educating parents (are they home educating to avoid prosecution for truancy?). The categories may well overlap, with more than one type of concern applying to particular families, but there is no indication in the responses so far released whether this is the case.

3.5.6. Question 3 (asking for case studies) is highly leading, and clearly indicates that the Review team is specifically interested in issues of local authority access to children.

3.5.7. Overall, the choice of a written questionnaire to gather this complex and sensitive information was inappropriate, and undoubtedly contributed to the low response rate.

3.6. Third bite of the cherry: Request for supplementary data

3.6.1. On 17^th September 2009, over three months after the publication of the Review report, the DCSF circulated a letter from Graham Badman to all local authorities, asking for more data to support his conclusions[10].

3.6.2. This is a striking illustration of the unprofessional conduct of the whole Review. To begin with conclusions and then gather evidence to back them up is unethical and contrary to the aims of the DCSF's Analysis and Evidence Strategy[11], which states that the evidence base for policy making should be "objective, reliable, relevant and timely".

3.6.3. Table 2 of the supplementary data request invites local authorities to add together several figures, including one for those "not yet assessed", in order to reach a total number of "EHE children not known to be receiving a suitable education". This is a prime example of the presumption of guilt that underpins the Review's recommendations as a whole.

4. How data was analysed

4.1.1. Despite the existence of an "expert panel" tasked with ensuring this, the Review was clearly not "rigorous and firmly rooted in evidence".[12]

4.2. The decision to ignore inconvenient data

4.2.1. The summary of the responses to the public questionnaire[13], shows that a large majority of respondents were not in favour of the changes proposed in the Review report.

4.2.2. 80% considered that current arrangements for safeguarding home educated children were adequate.

4.2.3. Over 90% of respondents believed it is possible for home educated children to achieve the five Every Child Matters outcomes.

4.2.4. Only 10% of respondents called for "Regular checks/on the spot checks/monitoring or registration" in response to speculation that home education could be used as a cover for abuse.

4.2.5. People who did not accept the premise of some of the questions and refused to give a yes/no answer were recorded as having answered Not Sure. This misrepresents these responses, which should have been recorded as a separate category.

4.2.6. None of these figures was mentioned in the Review report. Almost 1400 home educating parents and almost 200 home educated children and young people responded to the questionnaire. Just two of these responses were quoted in the final report.

4.3. The decision to analyse inadequate data

4.3.1. When asked for the factual evidence for Graham Badman's assertion that the number of home educated children known to social care was "disproportionately high" in some local authorities, the DCSF supplied an extract from the briefing paper mentioned above, in para. 3.5.2[14].

4.3.2. After the full briefing paper was released, DCSF officials and ministers stated that this analysis was not, after all, the evidence for the assertion in para. 8.12 of the report, and that it did not meet the DCSF's standards for statistical analysis.

4.3.3. Nevertheless, it is now acknowledged that the claim about the proportion of home educated children known to social care was based on information from less than 25 local authorities.

4.3.4. Only 13 authorities responded to the question about how many EHE children were subject to a child protection plan, and of these, only 5 reported rates of over 4%.

4.3.5. With such a low return rate, it cannot be assumed that the authorities who responded are representative of the overall picture. Given the limited distribution of the second questionnaire and the difficulty of responding, it is likely that local authorities with strong views or exceptional circumstances are disproportionately represented in the sample.

4.3.6. As a direct consequence of the Review team's decision to base their recommendations on this inadequate data, home educators have been falsely maligned in the press (for example, The Independent's news story on 12^th June, headed "Children educated at home more at risk of abuse"[15]).

4.3.7. A similar misrepresentation has even been made by Ed Balls on a popular public website, where he said "There have been high profile cases of 'home educated' children who have been very badly neglected. Graham makes clear that this is a small minority, though disproportionately larger among home educated children."[16]

4.4. Errors of logic 1: proportion known to social care

4.4.1. The actual statement made in the Review report is that "on the basis of local authority evidence and case studies presented, even acknowledging the variation between authorities, the number of children known to children's social care in some local authorities is disproportionately high relative to the size of their home educating population"[17].

4.4.2. The report suggests that this finding is relevant to the question "if there is abuse of children within the home education community, is it disproportionally (sic) high, relative to the general population?"[18]

4.4.3. However, the figures gathered about children known to social care were used to calculate a proportion of the electively home educated children already registered with local authorities, not a proportion of the home education community as a whole.

4.4.4. As the size of the whole home education community in any local authority is not known, it is not possible to calculate the proportion known to social care. The figures are therefore nonsensical.

4.4.5. What is certain is that the proportion of unregistered home educated children who are known to social care is zero, or almost zero. If home educating families are referred to social services, an inevitable result is that the family is henceforth known to the local education authority as home educators, regardless of the outcome of the social services enquiry.

4.4.6. Graham Badman's estimate of the total number of home educated children is that it is at least twice as large as the number registered with local authorities. Mike Fortune-Wood's 2006 research is in accord with this, putting the figure at approximately 45,000, or half of one percent of all school age children[19].

4.4.7. It is therefore not at all unexpected to find that the proportion of registered home educated children known to social services is higher than the proportion of children who are known to social services in the general population. The section of the home educated population that is registered (less than half of the total home educated population) contains, by definition, all the home educated children who are known to social services.

4.4.8. In fact, this figure (if substantiated by additional research) could indicate that existing mechanisms for identifying children and families in need of support are working perfectly well for home educated children, and that home educated children are no less visible in their communities than children who attend school.

5. The case made in the Review report

5.1.1. Ed Balls wrote to Graham Badman on 11^th June that "You make a compelling case for immediate and urgent reforms to ensure all home educated children are known to and monitored by local authorities, and that there is a proportionate focus on ensuring these children are safe."[20]

5.1.2. I suggest that the case made in the Review report is far from compelling. It is based on insufficient evidence, faulty logic, and unsubstantiated opinions.

5.2. Failure to present the data

5.2.1. None of the statistical data gathered as part of the Review was published with the Review report. Notes from face to face interviews have also not been published.

5.2.2. There was no discussion in the report about the reasons for choosing the research methodology used in the Review, nor about whether these methods had proved effective.

5.3. Errors of logic 2: balance between parents' rights and children's rights

5.3.1. At the heart of the argument presented in the Review report is the idea that there is a need to balance the right of parents to choose home education, and the right of children to receive a suitable education.

5.3.2. Altogether there are nine references in the report to the rights of parents, including, in para. 3.11, a reference to "the rights of parents as set out in Section 7 of the Education Act 1996".

5.3.3. However, s.7 does not confer any rights on parents. It merely allows parents to choose the means by which they carry out their duty to educate their children.

5.3.4. Parents do not have a right to do anything that contravenes their children's rights. There is no need to change the law to balance the rights of parents and children, because the primacy of children's rights is already clearly written into the law.

5.3.5. As with the fallacy about home educated children being at greater risk of abuse, the notion of the need to balance the rights of parents and children has been taken up with enthusiasm by government ministers and repeated in public pronouncements[21] and letters to MPs representing their constituents.

5.3.6. This frequent and glib repetition serves only to emphasise the weakness of the Review report's argument, and to add weight to the theory that the outcomes of the Review were decided in advance.

6. Conclusion and recommendations

6.1.1. An overwhelmingly law-abiding community of parents has been misrepresented and maligned by this Review.

6.1.2. The Review report does not make a compelling case for change, and does not present any evidence that there is even a problem to be solved.

6.1.3. Given the many flaws in Graham Badman's research, the current consultation should be ceased, and the proposals withdrawn.

6.1.4. Arrangements for monitoring home education should not be included in the government's planned Improving Schools and Safeguarding Children Bill.

6.1.5. Instead, local authorities should be advised to make use of their existing powers under the Education Act 1996 and the Children Act 1989 to take action in cases where they have reason to believe children are not receiving a suitable education or are at risk of harm.