Saturday, October 20, 2007

The Trouble With Surveys

James Forsyth has stirred up a hornets nest with his post on the Coffee House blog that elder children are smarter than their younger siblings. He refers to the TIME story on the subject.
In June, for example, a group of Norwegian researchers released a study showing
that firstborns are generally smarter than any siblings who come along later,
enjoying on average a three-point IQ advantage over the next eldest—probably a
result of the intellectual boost that comes from mentoring younger siblings and
helping them in day-to-day tasks. The second child, in turn, is a point ahead of
the third.

I usually look at surveys and reports like these and look at them from a personal perspective. With respect, I think this one is rubbish. I have two sisters and I readily admit that one of them (I'm not saying which one!) probably has a far higher IQ than me. Indeed, when I think of my friends and relations it is often the second child who is cleverer. Maybe things are different in Norway!

9 comments:

  1. I read a couple of similar studies years ago and I think by and large - not in every case, obviously, it is true. First, the first born gets the advantage of its parents undivided attention, They are ALWAYS available to answer questions and explain things.

    Second, as you note, they then reinforce their own knowledge and learning, by the time they have a sibling who can talk and communicate fairly well, by teaching siblings and eventually becoming regarded as an authority.

    I think it makes sense, although clearly families are different and it's not a hard and fast rule.

    (On the other hand, I think Harry is about 20 times smarter than Prince William.)

    ReplyDelete
  2. Which just goes to show the problem with citing anecdotes as a supposed legitimate repudiation of something which in theory at least would have looked at hundreds of children and the variances which were identified. That I know a 5 foot tall Dutchman does not make the statement that the Dutch are the tallest people in Europe false.

    ReplyDelete
  3. Forget it Iain. A 3 point difference on a reputable IQ test is about the same as the Standard Error of Measurement SEM. For example,the Weschler Adult Intelligence Test (WAIS) is one of the most reputable & widely used measures of IQ and has an SEM close to 3. This means for someone who scores an IQ of 100 (ie is of average IQ, there is a 68% probability that the true IQ could be anywhere between 97 & 103 and a 95% probability it is between 94 & 106. I am in no way implying your IQ is 100....judging from your blog your IQ is way higher than this (creep creep). Apologies for the geeky statistics but it's part of my trade as a psychometrician

    Leatherhead Matters at
    http://leatherheadblog.com

    ReplyDelete
  4. For what it's worth, children who start school in September (because they are already five), generally perform better at school than their younger classmates, who might join in January. The younger ones would be better off waiting for the following September.

    ReplyDelete
  5. Speaking as the second child in the family I'm sure you're quite right, Iain - this theory is rubbish!

    ReplyDelete
  6. Here's one 'survey' that's not rubbish, Iain, detailed in the following letter I've just posted to David Miliband's on his official website.

    Dear David Miliband,

    Re: the EU Treaty and Harlow's People's Referendum on the Treaty.

    In the past you have emphasised the demand for bottom-up rather than top-down approaches to government and PM Brown has promised more open government. Are such promises anything more than meaningless soundbites designed to win votes?

    For the report of the parliamentary EU scrutiny committee on the new EU Treaty criticises, among other things, the essential secrecy surrounding the new EU Treaty and the rushed and secretive manner with which it has been drafted.

    It is surely scandalous, undemocratic and of huge concern that Britain, government Ministers and government departments were given just 48 hours’ notice of the details in the European Union reform treaty before being asked to agree to it.

    The scrutiny committee has asked for clarification of the 'Red Lines' - the treaty opt-outs which the Prime Minister claims make a popular vote unnecessary - and have questioned whether these red lines will be effective.

    In addition, the committee has suggested that a requirement for national parliaments to contribute to "the good functioning of the Union" may contradict the Bill of Rights, which protects the UK's Parliament from being placed under legal obligations by any outside body, the MPs said.

    They added: "In our view, the imposition of such a legal duty on the Parliament of this country is objectionable as a matter of principle and must be resisted."
    Senior Labour MP, Gisela, who helped draft the constitution, has dismissed the "red lines" - the treaty opt-outs which the Prime Minister claims make a popular vote unnecessary - as "red herrings", saying that they were also present in the earlier constitution on which Labour promised a referendum. Ms Stuart stated: "It's a matter of trust and integrity. A referendum was promised. It should be delivered"..."If Labour can't trust the people, why should the people trust Labour?"..."The path adopted by the Government is neither honest not coherent." (BBC news website)


    Former Labour government Minister, Tony Benn has said this Treaty will result in the "death of democracy" in our country.

    Here in Harlow many of us are so deeply concerned that this Treaty will create a distant, undemocratic and top down EU superstate and so angry at the government's failure to honour it's manifesto pledge to hold a Referendum on this issue of fundamental importance which will have a major influence on all of our lives that we organised a public meeting on this.

    The public meeting was attended by people who support a broad spectrum of political parties, all united by our deep concern about this Treaty. Those at the meeting voted unanimously to hold our own People's Referendum on the Treaty.

    Harlow's People's Referendum is staffed entirely by local volunteers at no cost to our town or council. We rely on a shoe string budget of voluntary contributions. Volunteers, some of whom are pensioners, pay our own expenses. We volunteers are working our socks off to do give the people of Harlow a say and we are heartened by the overwhelmingly positive response of the people of Harlow.

    People and politicians all over our country are calling for a national Referendum.

    Why does the Government refuse to listen, Mr Miliband?

    Auntie Flo'

    ReplyDelete
  7. Maybe you're right Iain -- truly smart people would never ever draw a straight line through one data point. ;-D

    ReplyDelete
  8. Iain, jokes aside, whenever you read any survey like that, once you ascertained that it's been conducted properly, the way to understand it is that the likelihood of occurrence in the target group is higher, but not a givens in a particular individual' -- ie, the mean of the group shifts. It could be that it has a small effect on every member of the group, or that it simply produces outliers that push up the means.

    Visit www.numberwatch.co.uk and purchase the man's outstanding books on how to read statistics and not get bamboozled by the scaremongers. It should be compulsory reading for anyone who needs to be able to critically take on statistics, even if they are not a mathematician.

    Surveys, statistics and forecasts can be great tools, but never forget that they are only a very crude model based on our limited understanding. That's all -- they don't foretell the future, and they don't speak of single events, but just present a range of outcomes and assume the likelihood thereof.

    As my favorite lecturer use to say:

    ASSuME maketh an ASS outta U and ME.

    ;)

    ReplyDelete
  9. Welcome back Auntie Flo.

    ReplyDelete