I use this blog as a soap box to preach (ahem... to talk :-) about subjects that interest me.

Friday, December 28, 2012

Authors' Mistakes #8 - Graham Tattersall

Geekspeak, by Graham Tattersall is a little humorous hard-covered book containing a collection of essays about how to calculate odd things.


Tattersall, in his attempts at making complex calculations easy, cuts too many corners. For the sake of simplicity, he says things that are not right.

For example, at the beginning of Chapter 8, when estimating the number of people who die every year in the UK, he states:

You can estimate the figure quite easily from the average lifespan in this country, which is roughly 75 years — and rising. Imagine for a moment that all births in Britain stopped today, that from now onwards people die off and aren’t replaced.
Each year more die, until, after the time of the average lifespan of 75 years, there will be very few of today’s 60 million people left. So, if ages are evenly spread [my observation: a dubious assumptions, but let’s let it pass], an average of 60 million divided by 75 people will die every year. That’s about 800,000 per year.

He then comments that the actual figure is 500,000 and that the discrepancy can be explained by the fact that our lifespan is growing.

Perhaps. But there is a problem with his estimate due to his erroneous use of statistics: you cannot state that the ages of death are “evenly spread” between 0 and 75 while, at the same time, claim that “very few” live beyond the average. What average is it if there are very few people that live longer?

Tattersall’s assumption that very few people live longer than average is not only conceptually flawed, but also factually wrong, because it turns out that many people happily survive the average age of death. I looked at the US statistics and discovered that almost 60% of people live longer than average.

As an aside, you might wonder how it is possible that more than half survive the average age of death. It seems unreasonable. The reason is that the distribution is highly skewed. I will explain it with an example that, although not completely realistic, proves that for steeply decreasing distributions the number of points above the average can exceed the number of points below the average: Assume that you have a population of 100 people and that 55 of them live 60 years, 30 live 70 years, and 15 live 75 years. The average age of death is (45 x 60 + 35 x 70 + 20 x 75) / 100 = 66.5, with 55 people above average and only 45 below average. Convinced now?

Another example of his “pseudo-scientific” but confusingly imprecise presentations is in the small “Speak Geek” section at the end of Chapter 5. He states that A man who weighs 100kg at the North Pole would weigh only 99.65kg at the equator.

This is roughly correct (it’s actually close to 99.67kg). It is due to the fact that the surface of Earth, which spins at the rate of about 40,000km every 24 hours, is not an inertial system. We are kept in place by the presence of gravity. Otherwise, we would keep moving on a straight line along the tangent. From our local point of you, this appears to us as a force directed away from the axis of the Earth. This is the same apparent force that pushes us away from the centre of a car when we take a curve.

So far so good. Tattersall calls the centrifugal force an “upward ‘flinging’ force”, and that is were I have a problem. After mentioning gravitation, he explains the centrifugal force as follows: The second force is upwards, and is caused by the rotation of the Earth constantly trying to fling your body into space like a stone on a string being swung round your head.

This is misleading in too many ways to be acceptable. First of all, Earth doesn’t try to fling us anywhere. Secondly, the slingshot works because we exercise a pull on it off-centre. Therefore, to compare Earth to a slingshot is not right.

You could dismiss my objections by saying that he explains things in a colourful way and that I am just being my usual hair-splitter and boring self. But when he says that the centrifugal force is upward, he is plainly wrong. The centrifugal force is outward. It is upward at the equator, but to say it in general is badly misleading.

Finally, I would like to report a horrible mistake in Chapter 7. He starts it with the following two sentences: Captain Picard has been sorting out a spot of bother on some planet or other, while the Enterprise orbits at a safe distance. A radio command is sent to the ship: ‘Beam me up, Scotty.’

How can he not know that Scotty was the chief engineer under Captain Kirk? What a shame! And he calls himself a geek? It is painful.

I usually report problems in a book after reading it. In this case, I couldn’t hold back after reading about a third of it. I will probably have to report more...

No comments:

Post a Comment