Friday, 14 August 2009

More about Paula Rothermel's work

In 1998, when she had been studying for two years at Durham University, Paula Rothermel conducted some research into home education. The results of that research have been quoted endlessly ever since by home educators, including the present author. In Deborah Durbin's book, Teach Yourself Home Education, is the standard account of this work, written in collaboration with Rothermel herself. Rothermel says that she was "overwhelmed with over a thousand responses, of which a smaller sample were analyzed". Elsewhere in the book it is mentioned that she questioned four hundred and nineteen home educating families. However, the important facts, the PIPS baseline assessments and the literacy tests were in fact restricted to fewer than fifty children, a very small number indeed. The actual figures are thirty five PIPS baselines assessments and five literacy tests administered by Rothermel. Another forty four literacy tests were conducted by the children's parents.

It is important to bear in mind when claiming that, "Paula Rothermel's research has shown", or proved or demonstrated, that we are usually talking about these same thirty five children and that this all took place eleven years ago. It is also perhaps worth bearing in mind that this work was carried out not by a professor but by a second year student. True, questions were asked of another four hundred or so parents via questionnaires. These discovered many things such as that parents found home educating fun and so on, but they do not shed any light on educational attainment. The facts, the meat of the matter, are to be found in those thirty five children.

Of course, the fact that it was a small sample is not in itself reason to disregard the findings of this particular research. I have to say though, that most home educators seem to be under the impression that the research is far more extensive than is actually the case. Most are astounded when told that the claims about academic achievement mostly boil down to these same thirty five children. Compared with work in the USA which looked at the academic achievement of over twenty thousand home educated children, this really is a tiny sample from which to draw any meaningful conclusions. We must also remember that all these children were under eleven, making the sample even more restricted. We must finally keep in mind that it was a self selected sample, limited to those who chose to become involved in the first place.

An awful lot of the home educating dogma in Britain today is founded upon this one sample. We routinely see that, "Paula Rothermel showed that working class parents make good home educators" or that "Paula Rothermel proved that home educated children do better than children at school." The next time you read such an extravagant claim, remember that it is almost certainly based upon one tiny, self selected sample from over a decade ago.


  1. In 1998 Paula Rothermel was a lecturer at Durham university with two degrees, a postgraduate diploma and an post-graduate education qualification (Rothermel's CV). It is rather misleading to describe her as a "second-year student", which gives the impression that she was an undergraduate.

    Actually, it is difficult to see why her status at the time is at all relevant.

    (For what it's worth, I agree with some of your remarks about the sample size. It's a shame that the occasional valid points you make tend to be buried in vile rants. If you were to improve the ratio of content to bilge ratio in your writing then people might be more inclined to listen when you have something worthwhile to say.)

  2. I agree that the Rothermel study is small but I think that the fact that it is consistent with much larger studies in the US and Canada and Alan Thomas' research in Australia and the UK gives it greater significance and strength.

  3. Anonymous points out quite correctly that Paula Rothermel had other academic qualifications when she undertook her study of home education in 1998. She neglects to mention that Rothermel's background was in art and languages at least until 1994, when she started studying with the Open University. Her status at the time is extremely relevant, for obvious purposes.

    If somebody makes a surprising or controversial claim, then we first look at why they are making this claim and what evidence there is for it. Let us suppose, for the sake of argument that somebody claims to have discovered that black people are less intelligent than white people. If this claim is made by a long established Professor of Genetics at Oxford University who, with the aid of a an expert in statistics has analysed the educational records of twenty thousand students, well we might take the claim seriously and look closely at it. If on the other hand, the person making the claim is a second, or forth, year student of psychology who has looked at thirty five students as part of her thesis, perhaps we might view the claim in a different light.

    I am sorry that you see this as a vile rant, Anonymous. Perhaps you might profitably read an earlier post of mine in which I compare home education to a cult and remark on how angry devotees become when questions are asked about their beliefs.....

  4. The problem I have with the other research in the UK, Sharon, is this. Alan Thomas has looked at a couple of hundred home educated children in this country. He did not, as far as I know, administer any formal tests to them. From the tone of his books and his remarks about late reading, I suspect that if he had done the NLP assessment of reading, that it would have shown lower scores than school educated children. In this sense the evidence can be said to be contradictory. As far as small samples go, his book "How Children Learn at Home" was based upon only twenty six children. His work is fascinating and I agree with most of what he says, but it is a little anecdotal.

    The reason that I am focusing upon Paual Rothermels work is that it is the only research carried out in the United Kingdom which is produced as objective evidence that home education works at least as well as school. Since almost every claim made about the efficacy of UK home education is based on this work, I consider it well worth looking very hard at it.

  5. Do you think that Rothermel's thesis committee took into account that her background was in art and language until 1994? Do you think that the reviewers who decided to accept her work to conferences and journals took this fact into account? The answer is "no" in both cases, and you might do well to consider why.

    Scientific journals often use "double-blind reviewing", which conceals the identity of the author of a paper from the reviewers of the paper. They do so because the author's identity, status and background are not relevant to the validity of their claims.

    Finally, I don't see this particular post as a vile rant (although some of it is silly smearing, as usual), neither am I angry. Your nonsense about
    cults is just a device that helps you to avoid considering the real reasons why people respond to you in the way that they do.

  6. Simon said,
    "If somebody makes a surprising or controversial claim, then we first look at why they are making this claim and what evidence there is for it."

    How can Rothermel's results be classed as surprising or controversial when they agree with all the research carried out so far?

  7. Simon said,
    "He did not, as far as I know, administer any formal tests to them. From the tone of his books and his remarks about late reading, I suspect that if he had done the NLP assessment of reading, that it would have shown lower scores than school educated children."

    Well yes, of course a child who doesn't learn to read until they are 11 will have a lower score than school educated children at age 10, 11 or even 12, but Alan Thomas was very surprised about this aspect of informal education. He says:

    "In light of this [the usual effect of late reading, that the child does not catch up and the deficit in reading age widens as they get older], a wholly unexpected finding was the number of children who learned to read 'late', even as late as 10 or 11 years of age. Even more surprising was that starting to read late had, as far as could be ascertained, no adverse effect on intellectual development, self-worth, or even subsequent attainment in literacy. In general these 'late' readers soon caught up with and passed the reading level commensurate with their ages and, in common with most other home educated children, went on to thoroughly enjoy reading."

    This seems to contradict your recollection of his tone.

  8. I've just read some more of the chapter on late reading in Alan Thomas' book and found this section particularly interesting:

    "In the final analysis of 100 families, a total of 19 children (12 male, seven female) could be described as 'late' readers out of 105 who had never been to school and were aged at least 8 years of age at the time of interview. It is fair to conclude that learning to read 'late' is a feature of home education, at least for those children who have never been to school"

    19 out of 105. That's roughly 1 in 5 which is the proportion of adults in the UK who are functionally illiterate (according to an article in the Times). Now maybe this is a coincidence, but it seems possible to me that if 1 in 5 people are naturally best suited to learning to read late, this may explain the 1 in 5 who fail to learn to read sufficiently well in school, because school is just not suitable for late readers. As I've mentioned before, I know a school educated child who didn't read until they were 12 and learnt from computer games at home in the end. Until that point he had been set the same work and tests as his peers in school and, of course, this severely limited the amount he could learn and he failed repeatedly. Contrast this to my late readers who were able to learn in a variety of ways before they could read without failing, and have gone on to learn to read at adult levels without problems.

    Maybe, for roughly 1 in 5 of the population, the brain development from concrete to abstract thinking during adolescence is more significant than for the 4 in 5 who are able to learn to read early, it could possibly even be a requirement.

    Brain Development in Young Adolescents

  9. That is quite an interesting point, Sharon, and I don't doubt for a moment that some people's brains are wired up so the part handling readding does not work very well until they are alittle older.