Thursday, October 08, 2009

Graham Badman's Letter to the Select Committee

Below is the letter from Graham Badman to the Select Committee about his recent data collection from LAs, which is also available, in PDF form, via this link:

(For blog commentary see Maire and Bruce).


Dear Mr Sheerman,


In June, the Secretary of State for Children, Schools and Families published the Government's initial response to the findings of the report of my review of home education in England.

The full response to the individual recommendations was published today and I understand that Diana Johnson Parliament Under Secretary of State of Schools has written to you with a copy of that response. Both Diana and I are due to appear before your committee on Monday 12 October.

I thought it would be helpful to write you about some further evidence I have collected to support my recommendations. In my report I said that the number of children known to children's social care in some local authorities is disproportionately high relative to the size of the home education population. The basis for this statement is sound and supported by information provided to me by a number of local authorities both through data collection and through discussions with front line local authority officers and others in organisations with relevent experience.

Given that your inquiry was focussing on the conduct of the revew and my recommendations, I thought it would be helpful to provide the Committee with further evidence on safeguarding and educational issues at that affect home educated children. With the agreement of the department, I therefore wrote to all local authorities in September to seek additional information from a wider range of local authorities. This new data provides additional evidence that supports the conclusions I reached in my review. I would like to add that in providing this additional information a number of local authoroities took the opportunity to remind me that they can only provide figures for those children that they are aware of and they continue to believe that there are a substantial number of home educated children who are not known to their local authority.

I am pleased to be able to enclose a copy of findings from the survey. These findings will be published on the DCSF webside alongside the full response to my report.

Yours sincerely

Graham Badman.

Elective Home Education (EHE) - results of September data collection from local authorites. (LAs).

1. The Review of Elective Home Education in England was underpinned by two data collections from local authorites (LAs). LAs responded on a voluntary basis. The results formed part of the evidence that was taken into account in shaping the conclusions set out in the report.

2. Since the Review was published there has been consderable interest from some parts of the EHE community in its conclusions, in particular concerning the data collected about safeguarding issues affecting home educated children and the quality of education they receive. The sample size of the second survey undertaken during the Review was relatively small (25) LAs and it was therefore decided that additional information be collected from more local authorities.

3. The format of the new questionnaire was discussed with a small group of local authorities and they recommended that an additional question be included in relation to missing children (runaways) as some said that they had data suggesting that home educated children might be overrepresented in this population.

4. On 17th September 2009 all English LAs were invited to provide further information on safeguarding and quality of education, using the proforma attached at Annex 1. The return date was 1 October 2009. The results are set out below.

5. 74 LAs (49% of the total sample) responded to the questionnaire provided a representative sample of all LAs in the country. Not all LAs answered all of the questions. The number of responses is noted in each section of the paper. A number of reasons were given for non responses to specific questons, as noted below:

Unable to provide
Unknown/not known
Not captured
Data not available in time

The figures used in this paper have been quality assured by a DCSF Statistician.

6. The total number of EHE children who were known to the 74 LAs that provided returns was 11,700. (1) As the total mid-year population estimates for 2008 for children of statutory school age in the 74 LAs is 4,303,700 (2) this gives an estimate of the total number of EHE children in England of 20,000 (3) which is consistent with previous estimates.

Information collected.

7. These questions were asked to establish:

  • the proportion of child protection plans relating to EHE children comparied to overall child population;
  • the proportion of EHE children who, in the LAs opinion, are thought not to be receiving a suitable education;
  • the use of school attendance orders amongst the EHE population
  • the proportion of EHE youngsters who became NEET (not in education, employement or training at age 16, compared with the general population;
  • the proportion of missing children (runaways) who were EHE prior to their disappearance, as a proportion of the total number of missing children.

8. The results from this survey are shown below, mostly in tabular form and confirm the findings of the first survey.

Data from 74 sample authorities
Number of child protection plansPopulationChildren with child protection plans
EHE children5111,7000.4%
All children10,0254,712,200 (up to age 17)0.2%

9. This table shows that the vast majority of both home educated children and children in the "general population" do not have a CPP. In the sample of 74 LAs that responded to this survey, 54 said they had no EHE childen with CPPs. The remaining 20 LAs reported 51 CPPs for their EHE children. Annex 2 shows the extent of variation. The proportion for EHE children is approximately double that found in the population of children as a whole.


10. The survey asked a number of questions about the standard of education provided to EHE children. 69 LAs completed this seciton of the survey and the results are summarsied in the following table:

Extent of Education ReceivedNumber of Children% of total
Not receiving any education2101.8%
Receiving education but not full time or suitable6095.3%
Not co-operating with monitoring - no assessment made6565.8%
Not yet assessed(5)10639.3%

11. 68 (92%) LAs responded to this question and reported that 73 SAOs were issued to EHE children over the past 12 months. This suggests relatively low usage of SAOs and the graph at ANnex 3 shows that there is considerable variation between LAs.

12. Each autumn the Connexions Service carries out a survey of children who left Year 11 the previevious summer to establish whether they are in education, employment or training. The lastest available information is for the cohort of young people who had attained 16 by the end of August 2008. 47 LAs were able to provide information which is set out in the table below:

Number NEETPopulation%NEET
EHE Children270122022%1
National Figure


The graph at Annex 4 shows the spread of results from different LAs.

Missing Children (Runaways)

13. Missing children are defined as children who are reported to have been absent for 24 hours or more. 38 LAs reponded to this section of the questionnaire: in total 125 children whose last known educational seeting was EHE were missing. These figures appear high in some LAs, suggesting the need to further scrutinise this data. The distribution is shown in Annex 5.


For Annexes see link here.


Comments upon above stats:


From Maire:

"According to Badman 0.4% of home educating children have child protection plans, meaning that 99.6% are not unsafe. Does such a small proportion as this make a multimillion pound intervention proportionate or desirable, especially as it’s only point two on one percentage point lower than for all children in the 74 local authorities?"


"The top LA for issuing School Attendance Orders issued 26 of them. The second-placed LA issued only 7.

Questions that arise include:

1) given that, statistically, this datum really sticks out (it's an extreme 'outlier', in the jargon), shouldn't we ask whether another variable mightn't be involved, which might make it sensible to exclude that datum from any general conclusions?

3) might a similar argument to 2) not also be applicable to the stats for child protection plans issued to EHE children?

4) how hard did Badman try to get non-responding LAs, or LAs which didn't have any data, to respond? You would have thought that *every* LA would know the number of CPPs or SAOs issued to EHE children. "


A self-selecting sample of 49% of a population is just as self-selecting as a self-selecting sample of 25%, and many or most of our previous criticisms of Badman's statistical incompetence still apply.


The listing of stated reasons for "non responses to specific questions". Why aren't actual figures given? Might the reason just possibly be because they would be inconvenient?


The survey question (Annex 1, Table 2) asks whether figures are actual or estimated. But the totals quoted by Badman (section 10) are all presented as if they are actual.


When a home educator enquired of the LA about the reason for the high number of children known to social care from the indepth questionnaire, the LA replied:

"The high percentage of children arose from all the data we have ever had and therefore that includes people that are now adults. This could have included involvements for other members of the family, referrals where the outcome was no further action etc. The 9% includes all children known to social care so if they are known to social care because they are disabled, then yes, it does include disabled children."


No comments: