Interview Questions
I saw this article - Could you get into Oxford University - which has a series of interview questions from Oxford.
I thought these were interesting questions so I thought I’d have a quick, time-limited stab at them. Googling not allowed, which was insanely frustrating at times.
Q: Why is income per head between 50 and 100 times larger in the US than in countries such as Burundi and Malawi? (Philosophy, Politics and Economics)
This is a massively political question, and answers are likely to reflect the answerer’s political inclinations.
The proximal answer is that the US has massively higher productivity (can produce far more goods and services per capita) thanks to higher capital and resources. But that's almost circular: in economic terms it's simply restating that income per head is much higher. The underlying question is why is it like that?
This question is almost the same as Yali’s question from Jared Diamond’s “Guns, Germs and Steel”: How come the white people ended up with all the cargo?
I buy his broad argument (as I dimly recall it): Eurasia has a great big east-west band where domesticated crops and animals can more-or-less be transplanted over a large area, which Africa and the Americas lack. Also, by luck Eurasia had better candidates for domestication at hand than elsewhere. This made it easier to spread agricultural technologies as they developed across a bigger area. Larger agricultural surpluses made large states possible, and large populations. That meant more scope for division of labour, specialisation, and further technological development. Lots of people and lots of animals meant lots of horrible diseases, to which Eurasians over time developed resistance.
Then there’s why Europe not China explored, conquered and colonised the New World. I forget his theory here, but ISTR it’s because Europe’s geography (and/or happenstance) meant it developed as lots of little states that fought each other, whereas China was a unitary state for most of its history. That meant more chances to try things out: if the unitary Chinese state decides to go isolationist for centuries, it’s no big deal, but if one European state starts gaining resources by foreign expeditions the others are going to notice and try to beat them at it.
So the relatively well off and capable Europeans conquered and colonised the New World and as much of the Old World as they could, using guns, germs and steel.
Then you have the scientific and industrial revolutions, which North America was well placed to exploit, having the guns/germs/steel starting advantage and some pretty good natural resources.
(There are a set of arguments I don’t buy and don’t like the political implications they’re used to promote, around the inherent virtues of different races, which I mention to dismiss.)
That’s the OECD vs rest-of-the-world story. (Ignoring Japan.) But we also have to account for why things are particularly good for the US vs the rest of the OECD, and why they’re particularly bad for Malawi and Burundi vs the rest of the rest-of-the-world. (And they are pretty bad.)
The US vs OECD story is sort-of easy but contentious. There’s the large endowment of the right natural resources at the right time (wood, coal, iron ore, water power), and the large, relatively homogenous domestic market, which is not terribly controversial. There’s stuff about work ethic I don’t buy, that often strays in to the racist stuff I dismissed. There’s arguments around market flexibility that I might buy some of (though not entirely).
Malawi and Burundi are harder. Sub-Saharan Africa generally is badly off compared to the rest of the world. Some of this is climate and disease, particularly around the Great Lakes/Rift Valley area. [(Does Malawi count as Great Lakes? I think it’s a bit south but Lake Malawi itself is surely a Great Lake.) It’s one of those biodiversity hotspots, which is lovely in a touchy-feely way but actually life is much more pleasant for humans in denuded biomes like the UK and Ireland.] Some of this is the ongoing result of losing badly in the C19th/C20th colonial conflicts - although IIRC the Great Lakes region is a long way inland. Which is another big negative factor - lack of easy access to world trade and influence. And some of this is institutions and governance - I do buy the argument that relative stability, a lack of kleptocracy and general respect for things like the rule of law and peaceful disagreement make a big difference. But I also buy some of the Marxist critique that it’s easier to have those things if you’re already living in comfort and security.
Very specifically, Malawi has fairly recently emerged from one-party authoritarianism. The upside of that was a low incidence of civil violence and some modest growth of the economy. But the downside was serious repression, and the lid coming off big time when the President-for-Life (I want to say Banda but worry I’m confusing him with someone else) ceased to be such - with deteriorating institutions and quality of life. Things are bad and getting worse at the moment.
On the other hand, in Burundi things were very bad but were getting better, until this year. They’ve had appalling wars and genocide over the last 40 years or so, but have had some degree of peace this century. However, they’re building from a low base - both in economic and political terms. And there’s been serious violence and political turmoil this year and they’re in the news for it again this week, which suggests it’s Very Bad News since ‘political turmoil in small sub-Saharan African state’ is not exactly a headline-grabbing proposition round these parts unless it’s very bad.
It’s interesting to contrast war affecting Burundians with war affecting the USA. Both countries have been engaged in armed conflict for about as long as I can remember. But the USA uses its technological advantages to fight those wars a long way away from the places where most of its population live, work and make things and provide services, so one could consider American wars as providing Keynesian stimulus for its economy. Burundi lacks those advantages, and the wars have happened right in Burundi and been massively destructive of the economy and people - and the warring sides still live right on top of each other so it’s hard for Burundi to withdraw from conflict when it’s had enough.
Wars are bad, but civil wars are generally very very bad. That was true in the US too - if they’d kept having Civil Wars after 1865 they would certainly not be in the dominant economic position they are now.
Coming back to the question of US wealth vs Burundi and Malawi poverty, the question of ‘how did it come to be’ is interesting, but to my mind the more important question is ‘how can we improve things’, and that’s even harder. Although in general things are improving (see Hans Rosling) so we are not without hope here.
Q: Imagine that 100 people all put £1 into a pot for a prize that will go to the winner of a simple game. Each person has to choose a number between 0 and 100. The prize goes to the person whose number is closest to 2/3 of the average of all of the numbers chosen. What number will you choose, and why? (Experimental Psychology)
This is a fascinating question as a practical matter, but a boring one mathematically.
If you model the participants as a certain sort of rational agent, you end up picking 0. We start off knowing nothing about what the other people choose, so we guess they will be randomly distributed with an average of 50. (I’m assuming that by average they mean mean.) 2/3 of 50 is 33 1/3, so we pick 33. However, the others could easily work that out, and if they all pick 33, then 22 will be a winninger answer. And then 15, and then 10, and so on until you get to 0. In my poor understanding of game-theoretic terms, ‘everyone picks 0’ is a Nash equilibrium: that’s the best decision taking in to account everyone else’s decision.
But as a practical matter amongst actual people, where the strict conditions for Nash equilibria are not always met, if you put 0, you may well not win, because some bozo or anarchist will put in a high number, and a single high number will pull the mean up considerably. And even if you do put 0, all the other clever people will have put 0, so you’ll have to share the prize lots of ways (since there’s only £100 in the pot) and it won’t be worth having. So you might as well put something a bit higher since the expectation value might be more. But of course clever people will have thought of this, so the average probably won’t be that close to 0. It’s turtles all the way down, and up.
If I had to give an answer, I’d probably pick 5. Or maybe roll a d6 and use that. (Mixed strategies can be very powerful in some games, but I don’t think there’s a closed solution to this one. But randomising your answer bit is still a good way to avoid falling in to some sorts of inadvertent bias.) But I would much prefer to look up the research about what people actually do in such experiments before answering. Come to think of it, I would even more prefer not to play at all.
That leads on to a related experimental economics-y question: how much would I pay to play this game? Well, I wouldn’t. I don’t like to take those sorts of gambles on principle. But it would be really interesting to find out what actually happens, and being part of the game is more fun, so I might chuck in £1 anyway for giggles. But not repeatedly.
(At this point I could go off on one about the Kelly criterion and what fraction of one’s wealth one should stake on risky prospects, but I’ll save that one since I’ve done it before.)
It becomes an even messier- and arguably more interesting - problem if the winner gets an amount directly proportional to their guess. This requires someone else to stump up the money, but would allow, say, a payout of an amount in pounds equal to your guess, if you are the winning guess. That gives you a countervailing force pulling the guesses up, and makes it more fun. (Where fun=confusing.)
Q: Place a 30cm ruler on top of one finger from each hand. What happens when you bring your fingers together? (Engineering)
I’ve done this! They come together pretty much right in the middle of the ruler, even if they weren’t evenly spaced from the middle to start with. This is really cool!
Off the top of my head, I think this is to do with there being more friction from more ruler overhanging the nearer-to-the-middle finger, so the more distant one can slide more easily.
To check if that’s true: if that theory is right, if you do it with two identical pens, you’ll get the same result, but if you do it with one pen and one finger (where the finger is much less slidey than the pen) it’ll tip over.
[pause]
Being me, I had to check this out empirically. I was right about the two fingers, as I thought. I was right about the one pen vs one finger as well - the ruler basically goes straight over the pen without stopping. But it kept falling off with the two pens as well - but forward or back, rather than over the ends. Probably because it was a lot harder to keep the pens perfectly straight compared to each other. If I’m really careful I can get the two pens to come together in the middle, though. But I’m not confident I wasn’t cheating: being careful, I can get the pen-and-finger to come together in the middle, but I have to life the pen up higher than the finger to make it happen.
You probably need a much more rigorous setup to check this out properly, but I’m confident that the maths for two identical cylinders perfectly perpendicular to the ruler works out neatly.
Q: Can archaeology prove or disprove the Bible? (Oriental Studies)
No. That’s a deliberately strong answer.
It’s entirely possible for archaeology to make robust findings that contradict specific stories from the Bible. And in fact this is the case: for instance, almost all archaeologists would agree that the archaeological (and geological) evidence strongly suggests that the Noachian flood did not happen as stated in Genesis, including most Christian and Jewish archaeologists (but not all).
But those Christian and Jewish archaeologists (and those who believe them) do not regard this (or any of the many such examples, existing or future) as disproof of the Bible. Their arguments for why it isn’t disproof vary, but most are around there being some degree of flexibility in the literal truth of the Bible, but that the overall message remains true in some broader, more important sense. On this line of argument, the Bible is entirely true in some figurative, metaphorical, moral sense, rather than in the sense of being an accurate representation of the real world. I'd disagree: if a story is morally satisfying and pleasing, but doesn't match what happens in the actual world, that makes a great story, but it is not true.
You do also get literal, fundamentalist Christians and Jews who argue that the Bible is correct and any archaeology that says the contrary is therefore wrong. But I think they are wrong.
The same would apply even if there were some eye-popping archaeological discovery about the sources of the text.
Edit: It was President-for-Life Banda I was thinking of. And I completely forgot about the difference between the coefficients of static and sliding friction, aka friction is more of a problem for gettingget things moving than keeping them moving, which is why the ruler thing is a bit fiddly and also a big part of why Newton's Laws don't seem to apply in everyday life.
This entry crossposted to http://doug.dreamwidth.org/304565.html, where there are
comment(s) not shown here.
I thought these were interesting questions so I thought I’d have a quick, time-limited stab at them. Googling not allowed, which was insanely frustrating at times.
Q: Why is income per head between 50 and 100 times larger in the US than in countries such as Burundi and Malawi? (Philosophy, Politics and Economics)
This is a massively political question, and answers are likely to reflect the answerer’s political inclinations.
The proximal answer is that the US has massively higher productivity (can produce far more goods and services per capita) thanks to higher capital and resources. But that's almost circular: in economic terms it's simply restating that income per head is much higher. The underlying question is why is it like that?
This question is almost the same as Yali’s question from Jared Diamond’s “Guns, Germs and Steel”: How come the white people ended up with all the cargo?
I buy his broad argument (as I dimly recall it): Eurasia has a great big east-west band where domesticated crops and animals can more-or-less be transplanted over a large area, which Africa and the Americas lack. Also, by luck Eurasia had better candidates for domestication at hand than elsewhere. This made it easier to spread agricultural technologies as they developed across a bigger area. Larger agricultural surpluses made large states possible, and large populations. That meant more scope for division of labour, specialisation, and further technological development. Lots of people and lots of animals meant lots of horrible diseases, to which Eurasians over time developed resistance.
Then there’s why Europe not China explored, conquered and colonised the New World. I forget his theory here, but ISTR it’s because Europe’s geography (and/or happenstance) meant it developed as lots of little states that fought each other, whereas China was a unitary state for most of its history. That meant more chances to try things out: if the unitary Chinese state decides to go isolationist for centuries, it’s no big deal, but if one European state starts gaining resources by foreign expeditions the others are going to notice and try to beat them at it.
So the relatively well off and capable Europeans conquered and colonised the New World and as much of the Old World as they could, using guns, germs and steel.
Then you have the scientific and industrial revolutions, which North America was well placed to exploit, having the guns/germs/steel starting advantage and some pretty good natural resources.
(There are a set of arguments I don’t buy and don’t like the political implications they’re used to promote, around the inherent virtues of different races, which I mention to dismiss.)
That’s the OECD vs rest-of-the-world story. (Ignoring Japan.) But we also have to account for why things are particularly good for the US vs the rest of the OECD, and why they’re particularly bad for Malawi and Burundi vs the rest of the rest-of-the-world. (And they are pretty bad.)
The US vs OECD story is sort-of easy but contentious. There’s the large endowment of the right natural resources at the right time (wood, coal, iron ore, water power), and the large, relatively homogenous domestic market, which is not terribly controversial. There’s stuff about work ethic I don’t buy, that often strays in to the racist stuff I dismissed. There’s arguments around market flexibility that I might buy some of (though not entirely).
Malawi and Burundi are harder. Sub-Saharan Africa generally is badly off compared to the rest of the world. Some of this is climate and disease, particularly around the Great Lakes/Rift Valley area. [(Does Malawi count as Great Lakes? I think it’s a bit south but Lake Malawi itself is surely a Great Lake.) It’s one of those biodiversity hotspots, which is lovely in a touchy-feely way but actually life is much more pleasant for humans in denuded biomes like the UK and Ireland.] Some of this is the ongoing result of losing badly in the C19th/C20th colonial conflicts - although IIRC the Great Lakes region is a long way inland. Which is another big negative factor - lack of easy access to world trade and influence. And some of this is institutions and governance - I do buy the argument that relative stability, a lack of kleptocracy and general respect for things like the rule of law and peaceful disagreement make a big difference. But I also buy some of the Marxist critique that it’s easier to have those things if you’re already living in comfort and security.
Very specifically, Malawi has fairly recently emerged from one-party authoritarianism. The upside of that was a low incidence of civil violence and some modest growth of the economy. But the downside was serious repression, and the lid coming off big time when the President-for-Life (I want to say Banda but worry I’m confusing him with someone else) ceased to be such - with deteriorating institutions and quality of life. Things are bad and getting worse at the moment.
On the other hand, in Burundi things were very bad but were getting better, until this year. They’ve had appalling wars and genocide over the last 40 years or so, but have had some degree of peace this century. However, they’re building from a low base - both in economic and political terms. And there’s been serious violence and political turmoil this year and they’re in the news for it again this week, which suggests it’s Very Bad News since ‘political turmoil in small sub-Saharan African state’ is not exactly a headline-grabbing proposition round these parts unless it’s very bad.
It’s interesting to contrast war affecting Burundians with war affecting the USA. Both countries have been engaged in armed conflict for about as long as I can remember. But the USA uses its technological advantages to fight those wars a long way away from the places where most of its population live, work and make things and provide services, so one could consider American wars as providing Keynesian stimulus for its economy. Burundi lacks those advantages, and the wars have happened right in Burundi and been massively destructive of the economy and people - and the warring sides still live right on top of each other so it’s hard for Burundi to withdraw from conflict when it’s had enough.
Wars are bad, but civil wars are generally very very bad. That was true in the US too - if they’d kept having Civil Wars after 1865 they would certainly not be in the dominant economic position they are now.
Coming back to the question of US wealth vs Burundi and Malawi poverty, the question of ‘how did it come to be’ is interesting, but to my mind the more important question is ‘how can we improve things’, and that’s even harder. Although in general things are improving (see Hans Rosling) so we are not without hope here.
Q: Imagine that 100 people all put £1 into a pot for a prize that will go to the winner of a simple game. Each person has to choose a number between 0 and 100. The prize goes to the person whose number is closest to 2/3 of the average of all of the numbers chosen. What number will you choose, and why? (Experimental Psychology)
This is a fascinating question as a practical matter, but a boring one mathematically.
If you model the participants as a certain sort of rational agent, you end up picking 0. We start off knowing nothing about what the other people choose, so we guess they will be randomly distributed with an average of 50. (I’m assuming that by average they mean mean.) 2/3 of 50 is 33 1/3, so we pick 33. However, the others could easily work that out, and if they all pick 33, then 22 will be a winninger answer. And then 15, and then 10, and so on until you get to 0. In my poor understanding of game-theoretic terms, ‘everyone picks 0’ is a Nash equilibrium: that’s the best decision taking in to account everyone else’s decision.
But as a practical matter amongst actual people, where the strict conditions for Nash equilibria are not always met, if you put 0, you may well not win, because some bozo or anarchist will put in a high number, and a single high number will pull the mean up considerably. And even if you do put 0, all the other clever people will have put 0, so you’ll have to share the prize lots of ways (since there’s only £100 in the pot) and it won’t be worth having. So you might as well put something a bit higher since the expectation value might be more. But of course clever people will have thought of this, so the average probably won’t be that close to 0. It’s turtles all the way down, and up.
If I had to give an answer, I’d probably pick 5. Or maybe roll a d6 and use that. (Mixed strategies can be very powerful in some games, but I don’t think there’s a closed solution to this one. But randomising your answer bit is still a good way to avoid falling in to some sorts of inadvertent bias.) But I would much prefer to look up the research about what people actually do in such experiments before answering. Come to think of it, I would even more prefer not to play at all.
That leads on to a related experimental economics-y question: how much would I pay to play this game? Well, I wouldn’t. I don’t like to take those sorts of gambles on principle. But it would be really interesting to find out what actually happens, and being part of the game is more fun, so I might chuck in £1 anyway for giggles. But not repeatedly.
(At this point I could go off on one about the Kelly criterion and what fraction of one’s wealth one should stake on risky prospects, but I’ll save that one since I’ve done it before.)
It becomes an even messier- and arguably more interesting - problem if the winner gets an amount directly proportional to their guess. This requires someone else to stump up the money, but would allow, say, a payout of an amount in pounds equal to your guess, if you are the winning guess. That gives you a countervailing force pulling the guesses up, and makes it more fun. (Where fun=confusing.)
Q: Place a 30cm ruler on top of one finger from each hand. What happens when you bring your fingers together? (Engineering)
I’ve done this! They come together pretty much right in the middle of the ruler, even if they weren’t evenly spaced from the middle to start with. This is really cool!
Off the top of my head, I think this is to do with there being more friction from more ruler overhanging the nearer-to-the-middle finger, so the more distant one can slide more easily.
To check if that’s true: if that theory is right, if you do it with two identical pens, you’ll get the same result, but if you do it with one pen and one finger (where the finger is much less slidey than the pen) it’ll tip over.
[pause]
Being me, I had to check this out empirically. I was right about the two fingers, as I thought. I was right about the one pen vs one finger as well - the ruler basically goes straight over the pen without stopping. But it kept falling off with the two pens as well - but forward or back, rather than over the ends. Probably because it was a lot harder to keep the pens perfectly straight compared to each other. If I’m really careful I can get the two pens to come together in the middle, though. But I’m not confident I wasn’t cheating: being careful, I can get the pen-and-finger to come together in the middle, but I have to life the pen up higher than the finger to make it happen.
You probably need a much more rigorous setup to check this out properly, but I’m confident that the maths for two identical cylinders perfectly perpendicular to the ruler works out neatly.
Q: Can archaeology prove or disprove the Bible? (Oriental Studies)
No. That’s a deliberately strong answer.
It’s entirely possible for archaeology to make robust findings that contradict specific stories from the Bible. And in fact this is the case: for instance, almost all archaeologists would agree that the archaeological (and geological) evidence strongly suggests that the Noachian flood did not happen as stated in Genesis, including most Christian and Jewish archaeologists (but not all).
But those Christian and Jewish archaeologists (and those who believe them) do not regard this (or any of the many such examples, existing or future) as disproof of the Bible. Their arguments for why it isn’t disproof vary, but most are around there being some degree of flexibility in the literal truth of the Bible, but that the overall message remains true in some broader, more important sense. On this line of argument, the Bible is entirely true in some figurative, metaphorical, moral sense, rather than in the sense of being an accurate representation of the real world. I'd disagree: if a story is morally satisfying and pleasing, but doesn't match what happens in the actual world, that makes a great story, but it is not true.
You do also get literal, fundamentalist Christians and Jews who argue that the Bible is correct and any archaeology that says the contrary is therefore wrong. But I think they are wrong.
The same would apply even if there were some eye-popping archaeological discovery about the sources of the text.
Edit: It was President-for-Life Banda I was thinking of. And I completely forgot about the difference between the coefficients of static and sliding friction, aka friction is more of a problem for gettingget things moving than keeping them moving, which is why the ruler thing is a bit fiddly and also a big part of why Newton's Laws don't seem to apply in everyday life.
This entry crossposted to http://doug.dreamwidth.org/304565.html, where there are
comment(s) not shown here.