Saturday 13 March 2010

New research commissioned by the Department for Children, Schools and Families is due to start on May 12th

In the last few years there have been several attempts to make a serious study of home educated children in this country. None have succeeded. Home educators seem resolutely opposed to allowing complete strangers to talk to them about their children's achievements! Even when sympathetic people or organisations such as Education Otherwise or Paula Rothermel distribute questionnaires, 80% of parents refuse to answer any questions. Ofsted met somewhat of a blank wall last Autumn when they tried to find out about children being educated home and this new piece of research will probably do no better.

It's a pity really. Home educators are often saying that research shows that home education is better than schooling, but as soon as they are offered the chance to demonstrate the truth of this assertion, they run for cover! There are several possible explanations for this shyness. One is that the DCSF and Ofsted are such a tricksy bunch of bastards that nobody can trust them not to twist their findings. They are after all ultimately working for the government and everybody knows that the government is resolutely opposed to home education. There is also the argument that such research would be an invasion of privacy. A third possibility is that many of these children are not doing very well academically and their parents don't want anybody to find out. Let's look at these ideas a little.

It's quite true that the DCSF will be funding this new research, but that does not mean that it will be biased. Whoever gets the job will carry out the work and then publish a report. I have no doubt that the results will be open to exploitation by both sides in order to argue their case, but I can't see what harm having the facts could do anybody. Suppose that it was found that half of all home educated children had no GCSE's. Of course the government could use this to argue that home education needed close monitoring. On the other hand, home educators could point out that fewer than 50% of children in schools gain five good GCSE's. The raw data itself would still be good to have, even if it could be manipulated subsequently by either side.

As far as the idea that this would be an invasion of privacy goes, this could easily be met by meeting parents in the local McDonalds and seeing copies of GCSE certificates and so on. Or the whole thing could be conducted by post, with parents just sending details of the boards which their children sat for examinations. This leaves us with the final possibility; that parents are worried that their children are under-achieving academically. This is perhaps more plausible than the other reasons for objecting to research about educational outcomes. Let's face it, hardly any parent is backward in talking of her child's achievements. They all boast about the early GCSE's , the guitar examinations, the place at university. Even those whose children have special educational needs are pleased to talk about what their children have managed to do; the place at college, the music examinations passed and so on. The only parents I know who keep quiet about all this are those whose kids aren't doing too well at school. Is this the case with home educating parents? Could this be the true explanation for their curious reluctance to participate in research in this subject?

I have to say that home educators are often keen to talk about research from America which shows how well home educated children are doing. They are also happy to go on about Paula Rothermel's findings. One sees this sort of thing mentioned a lot when people are trying to justify their decision to home educate. The one thing they don't want to do is let anybody look at their kids now and see how well home education is doing for them. I would be keen to hear of other explanations for this, besides the obvious one which I outline above.

26 comments:

  1. Part of the problem with the fact that DCSF are funding it is that we all remember the last piece of work they funded about home education. This is the biggest reason for avoiding involvement, failure of government to listen to input that disagreed with what they wanted to do. However, by the time this research starts, there might be a different government and indeed, DCSF may have ceased to exist. The trouble is that we are aware that even if this happens and we get four or five years of relative calm, the current proposals may come back with another change of government.

    Government needs to demonstrate it has our best interests at heart, not its own. Wasting taxpayers' money on a huge bureaucracy that does nothing to improve education is not part of that - providing support and assistance without nasty strings is.

    As for other research, you missed Alan Thomas, who has looked at the effectiveness of autonomous education and the way children learn.

    ReplyDelete
  2. Indeed Dave H.

    Bad Bad timing. If research is to be done then I reckon enough people will feel the desire to help with it only when ( and if ) a new government come in and starts making the necessary reassurances regarding the issues that Home Educators feel are getting in the way of co-operation.Possibly then the opposition to government funded studies be overcome. Once a bit of trust is rebuilt and an independent body is commissioned.
    I am unsure of what response rate is considered 'average' for enrolling in studies. Obviously not 100% of people who get the request will want to participate- so I cannot comment on a 20% response rate.

    ReplyDelete
  3. I didn't mention Alan Thomas, because firstly a lot of his research was slanted towards primary age children and secondly becuase he himself did not really regard the work on home education as an objective piece of research. He talked to families about their methods and did in-depth work with a couple of dozen. This is very interesting, but sheds no light upon the long term consequences for the children; there was no follow up to see how they did as adults. This is what the DCSF work hopes to look into. The York Consulting project was very interesting indeed. I'm not sure that the DCSF record in the work which they have commissioned leads one to suppose that they are prone to distorting or manipulating the results.

    ReplyDelete
  4. Whether a 20% response rate is poor or good has no bearing on the case. The only way to find out how effective home education in this country is is to take a large group of home educated children and follow them all through into adulthood. Otherwise you will, as is usual, end up with a self selected sample. I would of course be happy to take part, somebody whose kid was an illiterate drop-out might not be so keen. This of course distorts the results. All children in a certain group must be tracked, otherwise the research would be worthless.

    ReplyDelete
  5. Then in your view virtually all social research is worthless as very little follows all members of a group.

    ReplyDelete
  6. By no means. In this case though, where we are looking at a group of young people educated in an unusual setting and their achievements, it would be a nonsense to allow the group to be self selected. The ones who select themselves in such a case are almost invariably those whose children are doing well and this tends to distort the findings. It would be unwise to extrapolate from this particular case and claim that no research should ever involve self selected subjects.

    ReplyDelete
  7. Nearly all social research (including medical research) is self selected. The only way it isn't is if the research is compulsory! As far as I know the only research that does not involve self selection is data collection activities such as the census (the 10 yearly one and the annual school census). Unless you know otherwise?

    ReplyDelete
  8. Yes, Michael Rutter's work on a London borough and also in the Isle of Wight.

    ReplyDelete
  9. He carried out social research of a whole population? I can only find medical research (Organic brain disorder in the Isle of Wight, for instance) and even in this field it's extremely rare for a whole population to be studied. What proportion of social research do you think involves the whole population? I would guess (from the research I've read) that it's less than 1%. I've not seen one piece so far (apart from official data collections like the census). If you discount research that does not involve some degree of self selection you will be left with very little research and it could only be carried out by very rich organisations!

    ReplyDelete
  10. "Even when sympathetic people or organisations such as Education Otherwise or Paula Rothermel distribute questionnaires, 80% of parents refuse to answer any questions."

    The questionnaires were sent out with the newsletter. Many EO members are not currently home educating their children or their children would have been the wrong age for the study. The most detailed part of the study involved reception aged children only, for instance. You would need to know how many parents of home educated children of the correct age received the questionnaire to work out an accurate response rate.

    ReplyDelete
  11. I was talking here about Education Otherwise's 2003 survey.

    ReplyDelete
  12. Rutter's work was with all ten year olds, Camber well and the Isle of Wight. There were follow ups at intervals.

    ReplyDelete
  13. What were they studying, was it social or medical? And how many more of the millions of pieces of social research do you accept as valid because they include a whole population? What do you think of academics who consider that valid information can be gained from smaller parts of a population?

    You would still need to know how many of the 2500 EO members were home educating when they posted out the 2003 survey with to work out an accurate response rate. I joined 3 years before we started to HE.

    ReplyDelete
  14. Rutter was studying all background information, ranging from overcrowding at home to parental occupation. The important thing here is that it involved all ten year olds in a certain area. Something of the sort with home educated children would be very interesting. Of course valid information can be gained from a smaller subset of a population. When this relies upon those who choose to take part though, problems can quickly arise. You might care to look at some of the work done on alcohol consumption. People lie about this and the heavier the drinker, the worse the lies. Serious drinkers seldom offer to have their drinking habit examined. All of which skews the findings.

    ReplyDelete
  15. So were these families not able to opt out of the research? How did this pass an ethics committee? How were they compelled to take part?

    ReplyDelete
  16. The Isle of Wight study began in 1964 and looked at all children on the island aged between ten and twelve. there were follow ups over the next fifteen years. If you wish to establish the incidence in the general population of, for example, autism, it is hopelesss to try and do this by looking only at children who present at clinics. You need to look at a whole population. This is what Rutter did. He did similar work in Camberwell in 1967 and Lorna Wing followed this up.

    ReplyDelete
  17. So were these families not able to opt out of the research? How did this pass an ethics committee? How were they compelled to take part?

    ReplyDelete
  18. Nobody was compelled to take part. It was done via schools and GPs. I really don't have time to track down the protocol used; it was, after all, forty five years ago! If you are that interested, you might contact the Maudsley who have all the records

    ReplyDelete
  19. Extract from a research ethics site on informed consent.

    "Do I have to take part?

    You should explain that taking part in the research is entirely voluntary. For example, you could say: -

    'It is up to you to decide whether or not to take part. If you do decide to take part you will be given this information sheet to keep and be asked to sign a consent form. If you decide to take part you are still free to withdraw at any time and without giving a reason'.

    If your study involves the recruitment of students or pupils you must explain that by choosing to either take part or not take part in the study will have no impact on their marks, assessments or future studies."

    http://www.brookes.ac.uk/res/ethics/consent

    If research is ethical (these days - maybe Rutter carried out his research before these standards were set) it will be self selected to some degree.

    ReplyDelete
  20. Not really. The number of registered pupils in maintained schools, the number of GCSEs taken and the grades achieved, the number of people in the country receiving treatment for various diseases, the number of people sectioned: all this information is collected and published without seeking any consent. It is not in the least self selected. The same principle will probably be applied to home educated children.

    ReplyDelete
  21. I've already said except for data collections such as the census. These form a very small part of the body of social research and usually contribute little in the way of detail necessary to reach any conclusions, hence the need for detailed studies that are invariably small and self selected. You appear to be dismissing the bulk of social and education research because it is self selected, even studies that you have recommended as providing useful information in the past.

    ReplyDelete
  22. And this is not the type of study due to begin on May 12th either. This will be self selected.

    ReplyDelete
  23. You are quite right about the study due to begin on May 12th. It is a pilot. I am not dismissing the bulk of social and education research at all. The ultimate aim is to gather similar information about home educated children as that which is known about schooled children. This does not come from the census. It will probably be limited to educational methods, number of GCSE's passed, higher education and things like that. It is hardly a massive invasion of privacy, since we already know all this stuff about 99% of the children and young people in the country.

    ReplyDelete
  24. "The ultimate aim is to gather similar information about home educated children as that which is known about schooled children. This does not come from the census."

    Much of this information comes from the annual school census as mentioned above.

    "I am not dismissing the bulk of social and education research at all."

    You said, "All children in a certain group must be tracked, otherwise the research would be worthless."

    "It will probably be limited to educational methods, number of GCSE's passed, higher education and things like that. It is hardly a massive invasion of privacy, since we already know all this stuff about 99% of the children and young people in the country."

    The study is more than data collection which is how we know what we know about 99% of the state educated children and young people in this country. The study plans to look at "Educational Experience and Attainment of Home-Educated Children", it will be a "longitudinal project investigating the provision of teaching and learning for, and the attainment of, home-educated children"

    A longitudinal study is observational - not a data collecting exercise.

    ReplyDelete
  25. I suppose this depends upon how one defines a datum.

    ReplyDelete
  26. Easily measurable facts such as number of GCSEs as opposed to observations and descriptions of experiences? Objective v. subjective?

    ReplyDelete