Alan Dix > HCI Education

real men do it generously

SIGCHI Bulletin HCI Education column Sept/Oct 2002

Toby and Tony were the sons of a farmer. They lived in a small town near the mountain. Only one village lay closer to the mountain and it was about 20 miles away down a long straight road. Toby had a sports car and would demonstrate its speed driving down the road to the mountain at 130 mph (when the speed traps allowed). Tony drove the farm tractor, a respectable 25 mph on a good day. Toby, of course, did not take his car onto the fields.

One day the mountain, a long dormant volcano, began to stir. The rumblings were small at first, but a cataclysmic eruption was expected soon. The people in the town began to pack their belongings and Toby and Tony prepared to leave. Cars and trucks from the small village passed through telling of the acrid smell of sulphur and boiling water courses nearer the mountain. Then came stragglers on foot who hadn't been able to fit aboard the crammed vehicles. And they told the story of an old man left behind who had missed the last truck, but was too sick to walk.

There was little time left. Toby took some friends in his car and sped to safety, but Tony turned his tractor around and drove towards the mountain. History does not relate whether Tony reached the village or whether he and his passenger survived.

We'll leave the moral till the end, but first an apology that I was late getting the additional web links online for the May column they are now there. I try not to put them up too soon, so I don't 'give away' the plot of the coming column, but sometimes wait too long! I'll try and get it right this time.

I recently returned from speaking at a large convention of educators at Singapore Polytechnic (see web above for more info). At one point, and with great trepidation, I talked about cultural differences between 'east' and 'west' and how they interacted with education and creativity. I need not have worried as Singapore is a society where the strengths and difficulties of cultural diversity are common topics. Just as I found in South Africa last year, many important areas of life are openly discussed that would be considered dangerously not 'PC' in the UK and I'm sure the US too.

A year or two back I recall the stunned silence when I said to some colleagues that I'd be amazed in view of the significant physiological differences between genders and ethnic groups that there were not also, on average, cognitive differences. My psychologist friends instantly put me right cognition is fundamentally different from physiology and so, they say, subject to less systematic variation. Much as though I respect their opinion, I strongly suspect that this is not a scientific judgement, but based more on the fact that it is not politically correct to admit that there could be differences.

We run into dire danger in secular societies of attempting to derive ethics from science and then trying to make the science fit the ethics we want to derive. But this is a morally dangerous route. We do not treat people the same because they are identical, but because they are people. When we have laws forbidding discrimination based on gender or race, it is not because we believe there are no differences, it is because we believe it is wrong.

In fact, for a combination of cultural and innate reasons, gender is a good predictor of linguistic ability. But when we are assessing candidates for a job or university course, we choose not to use gender and instead look for other predictors, such as examination results, SAT scores, portfolios of work, or how the candidate 'comes across' at an interview.

Of course, these predictors are themselves not neutral. In UK schools, boys used to outperform girls at age 16 (and beyond), now the opposite is true. This may be due to differences in expectation, society and educational style, but not least there has been a significant move from exam to coursework. You may recall my objection (in the January column) to the implicit assumption that educational achievement at 18 is in some way a measure of innate ability or educational potential. It is perhaps the best we have and we must use it, but education systems that permanently lock-out people based on their past performance are intrinsically sexist, racist and socially elitist.

I was recently examining a PhD about experiments using IT in primary maths education in Libya. One of the results seemed to suggest that although the IT support was helping them all, the more able students were making more effective use. In other words the IT support served to amplify existing differences in achievement. If larger studies bore this out what would this mean? Don't use IT because it is divisive? Focus use of IT on less able students to help them? Focus use of IT on more able students who can use it better and thus free teachers' time for the less able?

Science is not neutral, the studies we choose to do are related to the uses we expect. But neither is education pre-determined, the interpretation of results in praxis transcends the facts.

Well, back to the moral of the story of Toby and Tony. First we may have innate or acquired abilities, but it is societies and structures, including education systems, that turn ability into success or failure cars aren't fast, cars on roads are. But far more important whether through breeding, upbringing, nutrition, education or sheer luck you find you have abilities of a particular kind in a particular society and particular situation what really matters is not what you've got, but what you do with it.



I was very aware as I wrote this column that I was treading on dangerous ground!

As well as being a politically and ethically problematic area I am sure I know only a tiny part of the empirical evidence. So I've summarised below some of the salient issues I know about, mostly from vague memories of articles read long ago!

I would welcome either corroboration and sources for any of the things below, or additional or contradicory evidence.

gender differences

First, to not that there is very little measureable cogbnitive difference between genders, despite large amounts of data from school IQ tests etc. Where ther are differences I do not know of any evidence (which would be very hard to obtain) as to whether the effects are cultural or innate. Also it should be noted that any effects are about the average, individuals vary far far more than the group effects.
Despite extensive data there is NO measurable difference between average female and male 'g' (general IQ) scores.
However, I do recall reading (I think in Eysenk, who I believe has recently been brought back from years of purdah) that the standard deviation of 'g' scores was greater in males - as it was put more male geniuses and more male morons ... my female friends all attest to the latter.
There are documented differences in linguistic and spatial skills. ON average females have greater abilities at the former and males at the latter. Even here the personal differences far outweigh the effect.
Just a thought ... but linguistic abaility is usually associatred with left brain activity and spatial morew with right brain. If the gender differences were actually more abiout left/right brain bias, then it woiuld mean that women were the logical analytic thinkers and men the intuitive ones ... hmm.

nature vs. nurture

It is very difficult to distinguish the effects between innate genetic factors, early development (including pre-birth) and family and societal cultures. As I say in the column, one has to be very careful about the reasons one has for studying these things, however, there has been a lot of research, not least on IQ measures, which have had such a problematic history.
As far as I know all studies depend on measuring variation within groups of different genetic closeness - siblings vs. unrelated people, identical twins vs. birth twins, and - perhaps most well known - identical twins separated at birth.
These experiments lead to results of the form factor X (IQ, height etc.) is 30% due to genetics 70% to environment. But are rarely explicated especially in popular media.
First - figures. The 30%:70% figure will be based on the level of environmental differences. If, for example, every child were brought up in identical environments independent of class, background, affluence etc., then there would be NO measurable environmental differences - 100% genetics? The figures are about the level of effect given the society and circumstances it is measured in.
Second - control. If, for example, IQ were 30% genetic, 70% environment, then by giving a low potential child the best possible environment, we would transform a future underachiever to a high achieving graduate. Of course, this is precisely one of the factors that drive class differences in educational achievement.
In addition, 'genetic' potential and environment are not independent. If a child has a slight predisposition to one activity/skill compared to others, then s/he will be more successful in that skill, so will feel more positive about it, do things that use it etc. Furthermore, parents, teachers, friends will also foster and give positive feedback. Later , friendships, subjects studied etc. will all be affected. A very small predisposition leads to a much larger final effect due to positive environmental feedback.
It is not a simple story and any experimental results should not be ignored, but questioned and understood for more than their face figures.