None of the above


As background for an article complaining about falling standards in education, The Times scanned and posted the 2006 Edexcel GCSE Science: Physics P1b exam paper. This is a multiple choice paper covering topics of waves, electromagnetic radiation, astronomy, cosmology, and seismology.

I don’t necessarily concur with the judgment of the article (not without knowing more about the subject). The range of material covered by the syllabus seems quite good to me: certainly I never studied cosmology or seismology in school, even at A level. There is a trade-off between breadth and depth of coverage, and it’s unfair for commenters to pick on the loss of the latter without acknowledging the gain in the former. Also, this is just one exam out of a large number, and it not necessarily representative of what the pupils actually learn in the class.

Having said all that, I’m going to nitpick the paper mercilessly because it contains some really, truly, horribly poor questions.

Some ground rules. Multiple-choice questions need to be, among other things, (a) based on correct assumptions; (b) not subject to ambiguity; and (c) admit only one correct answer.

Having one answer that is correct if you read the question naïvely, and a different answer that is correct if you read the question pedantically, is a disaster. In a written paper a candidate who spots this kind of ambiguity can point it out in their answer, and have, I think, a good chance of persuading the marker. But in a multiple choice exam there’s no such chance: you have to guess just how dumb the examiner is. And the evidence from this paper is that the examiner can be very dumb indeed.

Anyway, onto the questions.

1. Which of these shows the orbit of a moon?

Which of these shows the orbit of a moon?

A looks like a representation of the orbit of a planet, and the circle at C is a conventional representation of the orbit of a moon around that planet (in a rotating reference frame in which the star and the planet are both stationary). So C is a good answer. On the other hand, with respect to a non-rotating frame of reference, the moon follows essentially the same orbit as its parent planet, plus a small cycloidal wobble, which would very likely be too small to be visible in a diagram like this if it were to scale (1 part in 375 in the case of Earth’s Moon). So A is a good answer too.

But there are more possibilities. If C can represent the orbit of a moon by a convention about reference frames, then surely so can D, except that in this case the parent orbit has been omitted. And since A is a good answer, then so is B: it’s the orbit taken by a moon whose parent planet has a highly eccentric orbit.

A couple of the commenters to the Times article were a bit faux-naïve over this question: the circle at C is a perfectly conventional representation of the orbit of a moon, and I’m sure they know it.

5. Our Moon seems to ‘disappear’ during an eclipse. Some people say that this is because an old lady covers the Moon with her cloak. She does this so that thieves cannot steal the shiny coins on the surface.

Which of these would help scientists to prove or disprove the idea?

A. collect evidence from people who believe the lady sees the thieves
B. shout to the lady that the thieves are coming
C. send a probe to the Moon to search for coins
D. look for fingerprints

The commenters at The Times are uniformly scathing about this one but I think it is a reasonable question to ask. An important aspect of science is to ask and answer, “how could we test this idea?”, even if the idea is stupid. The point being that science is about evidence, not just authority, and that it is a coherent set of theories which attempt to explain all physical phenomena, so even stupid ideas have scientific refutations.

So the idea for this question needs to be ridiculous without actually being too similar to an idea that anyone actually believes (for fear of causing offence which would spoil the paper for some exam takers): so it can’t be a dragon that eats the moon, for example.

But I can’t say I like the answer. Only answer C has any bearing on the actual question, but it would be utterly ridiculous to go to the expense of sending a probe when much simpler and cheaper observations would do a better job. This is also an important point: do the simplest, quickest and cheapest experiments first so that you can evaluate the worth of more expensive tests.

Better answers would be, “look at the moon during an eclipse.” Or even more simply and cheaply, “look at photographs of the moon previously taken during eclipses.”

8. The type of radiation which damages eyes and can cause skin cancer is

A. X-ray
B. microwave
C. ultraviolet
D. gamma

Fails (c). A, C, and D are all correct answers.1

This is especially miserable because science ought not to be just about learning isolated facts like “ultraviolet radiation can cause skin cancer”, but about linking facts together with the theories that explain them, so that you can deduce new things that you haven’t learned. In this case, if you knew that one of the several mechanisms by which ultraviolet radiation causes cancer is by ionization, and if you knew that the chances of ionization increases with the energy of each photon, and if you knew that photon energy increases with frequency, and if you knew that X-rays are higher frequency than ultraviolet, then you could deduce that X-rays are carcinogenic without having to learn it as a separate fact. (Which one of those links in the chain of deduction was the examiner unaware of, I wonder?)

10. Jupiter is too cold to support life. […]

Fails (a). Jupiter is cold at its surface: its black-body temperature is 110 K. But the temperature rises steadily as you descend through its atmosphere, reaching many thousands of K at the solid core. So the question makes no sense: Jupiter may conceivably have the wrong composition of gases in its atmosphere, or have the wrong range of temperature/pressure environments, to support life, but it’s hardly too cold.

I note also that although there’s certainly no environment on Jupiter that would support Earthlike life, the question doesn’t restrict itself to life-as-we-know-it. (Maybe a nitpick too far.)

19. Which of these is an advantage of using digital signals in radio broadcasts?

A. digital signals travel quicker than analogue signals
B. digital signals can carry more information than analogue signals
C. analogue signals travel quicker than digital signals
D. analogue signals can carry more information than digital signals

None of the above. Digital signals have to be encoded as analogue waveforms in order to be transmitted, so neither type of signal can carry more information than the other. The maximum bandwidth of the channel is given by the Shannon–Hartley theorem in either case.

The advantage of digital signals is that they can be processed by digital computers, which can process signals more flexibly and cheaply than analogue devices can. In particular, operations like error correction, compression, statistical multiplexing, and encryption, are more convenient to perform with digital computers on digital signals.

I guess that the examiners want the answer “B” because they are comparing analogue and digital radio transmission standards. Digital radio standards such as DAB squeeze more information into the same part of the spectrum than the old analogue standards such as FM, but they do this by using more sophisticated modulation (in the case of DAB, quadrature phase-shift keying) of the underlying analogue signal, taking advantage of improvements in electronics in the last half-century. Broadcasters could have chosen to transmit analogue radio using new modulation schemes, getting a similar increase in the information transmitted. But then we wouldn’t have got the other advantages of digital transmission listed above.

So I think this fails criterion (b): the examiner is using “signal” to mean something like “broadcast transmission standard”.

Now it may be that this is resolved by material presented in the coursework. For example, maybe the GCSE course always uses “signal” to mean “broadcast transmission standard” and never the usual meaning in electronic engineering. But if that’s the case, I think it is very unwise as it puts one more barrier in the way of communication between specialists and non-specialists.

20. Digital technologies, such as CD and DVD players, have increased

A. the speed at which sound travels
B. the quality of the sound you can hear
C. the range of frequencies you can hear
D. the loudness of sound which can be produced

Obviously CD and DVD players can make no difference to any of these: the speed of sound (in some medium) is a physical constant; the quality of sound and the range of frequencies you can hear are facts about the physiology of hearing, and the loudness of sound that can be produced is a fact about whatever it is that is making the sounds (e.g. an amplifier and a set of speakers).

I think the examiner means something like “Compared to analogue sound reproduction technologies such as vinyl records, digital technologies, such as CD and DVD players, have increased … B. the quality of the sound you can hear coming from typical consumer audio equipment playing typical recorded media [etc]”.

But if you make all the necessary assumptions, then multiple answers are possible: digital audio is not prone to defects such as the “hiss” and “pop” of vinyl, so B seems like a good answer. Compact disks have a higher dynamic range than standard vinyl records so D is a good answer; on the other hand there were noise reduction (and volume compression) standards like Dolby’s for analogue media so maybe this is only arguable. Similarly, even cheap, often-played, compact disks support frequencies up to 22 kHz, but only the highest quality, best-cared-for, vinyl records achieve this range, so maybe C is a good answer too.

23. Assume the orbits of the Earth and Pluto are circular. […]

Fails (a). Surely about the second or third most important fact to know about Pluto is that its orbit is eccentric, by comparison to the other planets?

There was nothing else special about Pluto in this question, so why not pick a planet with a low eccentricity, so as to avoid a false assumption? Neptune would have done just as well.

30. Which of these do scientists think is moving away from the origin of the Big Bang? […]

Fails (a). The Big Bang has no “origin”. This is an unfortunately common misconception that I would have hoped that a science syllabus would do something to dispel.

31. […] the Big Bang theory is supported by evidence from

A. neutron stars
B. nebulae
C. ultrasound radiation
D. microwave radiation

Fails (c). X-ray bursts from neutron stars are standard candles used to determine the distance of remote galaxies, which together with their redshift provides evidence for Hubble’s law and so for the Big Bang. The planetary nebula luminosity function is another distance indicator. And of course there’s the cosmic microwave background radiation.

40. The diagram shows some waves passing through the Earth from an earthquake at Q.

Some waves passing through the Earth from an earthquake at Q.

The diagram is rather unfortunately drawn. The idea behind the question is that R is in the “shadow zone” in which seismic waves are not detected. The lines on the diagram make this reasonably clear. But the shadow zone does not start until about 104° away from the epicenter, and R appears to be only about 95° away from Q. A “not to scale” label is needed, or maybe the examiners should have drawn the diagram correctly in the first place.

But hey, it’s only a science exam, who cares about accuracy? This seems to be the attitude throughout and I don’t like it.

  1. ^ I thought about microwaves but wasn’t sure: repeated burning of tissue is another cause of skin cancer, and microwaves of sufficient power can heat skin, so I think it’s medically plausible. But I couldn’t find any evidence of it actually happening in real situations (just a lot of studies finding no evidence that mobile phones/Wi-Fi/microwave ovens cause cancer), whereas I could find cases for the other kinds of radiation. The closest I got was Szmigielski et al. (2005), a study showing that microwaves can accelerate the development of cancer in cancer-prone strains of mice.