← More or Less
More or Less
Gay Britain
00:27:35
Tim Harford and the More or Less team examine the micromort measure of risk and on...
Loading summary
Transcript58 lines
- [00:00]
A
Thank you for downloading this week's More or Less podcast. Here's Tim Harford.
- [00:06]
B
Hello, and welcome to the last in the current series of more or less sad news. But you still have our undivided attention for another 1650 seconds. And what a 1650 seconds there'll be. We'll be looking at sexual orientation, Irish bankers, the police and the perks of David Cameron's job. Just to be clear, those are separate items. And we'll examine the micromort, a cheeky little unit of risk.
- [00:31]
C
Every day, just by existing, you risk 30 micromorts, a 30 in a million chance of dying just by existing.
- [00:39]
B
That's Matt Parker, whose show at the Edinburgh Fringe was a raging success, which is exactly what we'd expect of a show all about maths. But first, according to the Office for National Statistics, we don't have very many gay listeners. It's not that More or Less has a particularly heterosexual following, it's that the UK in general has fewer gay, lesbian and bisexual people living in it than we thought. Or so the story went last week. But loyal listener Millie Shaw, also the editor of Lesbilicious, which bills itself as the web's tastiest lesbian magazine, got in touch to ask.
- [01:12]
D
The headlines in the press were all about how 1.5% of the population were gay or bisexual. That seemed very low to us because all the figures we'd seen previously were either 10% from Kinsey or 5 to 7% from the government's own figures from a few years ago. So suddenly to come to 1.5% seemed very low figure. So we're just wondering what the story behind it was.
- [01:34]
B
Well, our reporter Wesley Stevenson's evil twin, Wesley Stevenson has been looking at the numbers. Wes, how did the ons come up with this figure?
- [01:42]
E
What they do is they contact households across Britain and ask them to take part in the survey, where they ask them all kinds of questions and not just about sexuality. And they use two methods. The first is a face to face interview with all members of the household. And the second is a telephone interview.
- [01:57]
B
Isn't it a bit problematic to ask a woman whether she's a lesbian in front of her dad or her husband or her housemates?
- [02:02]
E
Well, yes. One article in the observer suggested that randomly turning up on the doorstep and asking questions might not necessarily get a good response. But this isn't really how it works. People are contacted in advance and they're asked to take part. And then they have to agree and then an interviewer makes an appointment to visit their house. Now as you imply, asking a question in front of other people you live with might not elicit an honest answer. So the Office for National Statistics have devised a method for keeping the answer anonymous. Stephen Hicks from the ONS explains how it works.
- [02:36]
F
We present to them four options. Firstly, heterosexual, straight, gay, lesbian, bisexual or other on a show card. And then we ask the question which of these options best describes how you think of yourself? We use a concealed show card system, which basically means that the card that we show them has a number against each of those options that I mentioned. We ask people then to tell the interviewer which number, rather than actually saying which option out loud in front of other household members. We do the similar method on the phone as well.
- [03:07]
B
So did this method work?
- [03:08]
E
Well, the ONS say that devising this method had brought down the number of refusals to answer. And the don't knows people did seem to be quite frank as well. For instance, nearly 20% of those who said they were gay, lesbian or bisexual were actually married to and still living with a partner of the opposite sex. Stephen Hicks from the ONS thinks they got a pretty good estimate.
- [03:29]
F
The estimate we produced and published last week is in line with other surveys that are similar questions in the UK and also in line with the pilot surveys that we carried out between 2006 and 2008. So we're quite confident actually that the estimate we have is an accurate estimate of self perceived sexual identity.
- [03:47]
E
And they didn't just come up with this question, they did a lot of research into getting the right question. Interestingly, they got more responses by adding the option straight next to the heterosexual category.
- [03:58]
B
So people didn't know whether they were heterosexual, but they knew they were straight?
- [04:02]
E
Well, it does seem that way.
- [04:04]
B
Ok, so the ONS says that their survey compares quite well with other surveys. But what about these figures that Millie mentions?
- [04:10]
E
Yes, other surveys which have asked a similar question have got similar answers, a range of between 0.3% and 3%. The ONS's 1.5 is pretty much bang in the middle of this, but Milley talks about a figure of between 5 and 7%. Now this came from a document produced by the last government when it was looking at the civil partnerships legislation. But as Stephen Hicks points out, even the government wasn't sure how accurate it was.
- [04:35]
F
This estimate, which is commonly used, which is between 5 and 7%, is based on a variety of sources, including surveys and exit polls in the UK and also overseas. And they also measure different concepts and collect by different modes. It is hard to judge the reliability of some of the estimate, given that it's based on a range of different research. And the report, I believe, itself actually acknowledged that there was a lack of reliable data. And that's one of the reasons that ONS set out to produce a measure on our national surveys which asks the whole of the adult population in the UK about their sexual identity.
- [05:09]
B
Stephen Hicks from the ons. Thanks very much, Wes. The other figure that Millie mentioned was that of 10%, which comes from work by Alfred Kinsey. Julia Erickson is a professor of sociology at Temple University in Philadelphia, and she studied Kinsey's work. But first I asked her why she thought the ONS figure was lower than other estimates.
- [05:28]
G
If you're really trying to examine sexual behavior, identity, desire, you have to ask about all three things. And so in the big survey that was done in the United States, they asked three sets of questions. One was a set of questions about desire. Who turns you on? What kind of people do you imagine having sex with? Who would you like to have sex with, what gender those people are? And then there was a second set of questions in a totally different part of the survey that asks, in the last five years, how many times have you had sex, how many of your partners were male and how many of your partners were female? And then there's a third set of questions somewhere else. It's about, what do you call yourself? Which is the identity question that this Office of National Statistics asked. But those three things are not coterminous. And so there are men who've had sex with men who don't think of themselves as gay or homosexual. And there are likewise, there are women who desire other women and have never done anything about it and don't call themselves lesbians. So if you really want to get a good grasp at same gender behavior, then you really have to ask all those questions. And they didn't do that. So it's not surprising that they got a low figure, because when you ask that one question, that figure's much lower than you get when you ask all of them.
- [06:44]
B
A lot of people have a number in their head of 10% of the population being gay, lesbian or bisexual. And this, I understand, comes from Alfred Kinsey's work. Could you just tell us who Kinsey was? And then let's discuss how he actually came up with this number.
- [07:01]
G
Kinsey was a biologist at Indiana University who originally collected gall wasps. And he was interested in variety, which was very common for the kind of scientist he was in his day. So he didn't believe in survey sampling. Survey sampling was only really coming out in the 1930s when Kinsey started collecting his data. And he was quite antagonistic towards it because he thought you should get variety. And if you got enough cases, thousands and thousands of cases, they would represent the population.
- [07:30]
B
So Kinsey in the end published these two famous books, Sexual Behavior in the Human Male and Sexual Behavior in the Human Female. And rather than saying, I want to sample 1% of the population or 0.1% of the population and make it a completely random sample, he was saying, I want to find people who are into and I want to find people who are into that. I want to find people who are into everything and I'll write them all down. So how did he find the people that he interviewed about their sexual behavior?
- [07:56]
G
He started in 1938 collecting data and he only interviewed married college students, which was pretty common in those days on the Indiana campus. So they were the safest, all heterosexual, of course. And then he started spreading out. And so the next thing he started doing, he would go to a city and he would contact clubs like the Kiwanis Club or the Lions Club. And he would get trying to get what he called 100% sample. So he would go and he would, maybe 40% or 50% would agree to an interview. And then he would say to the others, you know, most of your friends in this group have been interviewed. People are going to wonder if you don't do an interview. So he tried to get everybody to line up and agree.
- [08:31]
B
And these are just gentlemen's clubs, social
- [08:33]
G
clubs, young women's groups or all kinds of groups. He went to other universities and he realized he didn't have enough working class people. He couldn't get into working class communities. So he went into the jails. But believed, I think it's not uncommon in his day that most working class men spent some time in prison. Many of them had had same sex sexual experience. Many of them were vagrants because this is a depression, often meant at times these men would get picked up because they didn't have a place to sleep at night. And many of them had had sexual experiences with other men because they were living on the road. Then he started trying gay clubs and he got some gay contacts. His co author Paul Gabbard told me of being taken by Kinsey to the train station lavatory in New York City in the pen. And they stood outside and they watched men go in. And Kinsey said, have you noticed some of them aren't coming out? Gabbard didn't have a clue what was going on. And then Kinsey explained to him that lavatures in America, certain lavatures, were really pickup spots for homosexuals. So he used to go into those lavatures and get men to agree to interviews.
- [09:35]
B
And this famous figure, that 10% of the population is gay or lesbian or bisexual, that's often attributed to Kinsey. Well, we know his methods were not really appropriate for coming up with that figure, but did he actually ever use that figure?
- [09:51]
G
Well, he had a variety of figures. First of all, he had a 40% figure, which was 40% of men have had some sexual contact with other men. Most of them had had it before age 16. He had another set of figures which was, has this person during their life been mostly engaged in homosexual behavior or mostly having homosexual fantasies? Because he conflated fantasies and behavior. And he came up with two figures. One was that 13% of men and 7% of women were mostly homosexual. So then that figure just got. Kinsey never averaged it, but activists in the gay rights movement averaged the figure because it was a really good figure. It suggests that everybody has somebody in their family who's homosexual.
- [10:38]
B
Professor Julia Erickson of Temple University, you're listening to More or Less. In association with the Open University. Yesterday, the Irish Central bank released an estimate of the total cost to the taxpayer of the support it's giving to the country's banking sector. It's about 45 billion euros, or 39 billion pounds, as loyal listeners will know on More or Less. We like to ask a simple question when looking at apparently big numbers. Is it a really big number? Well, let's see. Ireland's population is 4.5 million, €45 billion divided among 4.5 million people is a neat €10,000 per person. So that's quite a whack. And many commentators have compared the sum to Ireland's gdp, gross domestic product. But that's a measure of the productivity of the Irish economy, not the income of the Irish people. Normally, GDP and national income are much the same, but Ireland's GDP is actually over 30 billion euros more than its net national income, because a lot of the GDP is profits of international corporations which have settled in Ireland, but which are actually largely owned by foreigners. 45 billion euros is more than a third of the size of Ireland's expected national income in 2010. You may remember that the headline cost of the UK bank bailout was dramatically exaggerated because it included guarantees that won't be used, ignored shares that look like being quite valuable, and so on. Alas for Ireland, that's not true. Here £39 billion is a sober best guess at the likely total cost to the Irish taxpayer. The last estimate we could find of the total cost of the UK bailout was £6 billion, £100 per citizen. A far cry from the Irish. €10,000. From one depressing number to another. Recently, the Sunday Telegraph reported a disturbing statistic. More than 70 children may have died needlessly after heart surgery. The report was based on research led by David Spiegelhalter, a medical statistician and professor of the Public Understanding of Risk at Cambridge University. And David is no stranger to more or less. I asked him what he and his colleagues had done.
- [12:44]
H
We were looking at the deaths of babies who had heart surgery in 11 English hospitals over a period of nine years. We looked at how many babies died in each hospital and compared it with how many would have been expected in roughly similar patients if they'd been operated on in an average hospital. And so every hospital gets an observed number of deaths and an expected number of deaths. But you can also take the difference between those. So if a hospital was expected to, say, get 50 deaths and they observed 60 deaths, you take the difference, the mortality difference, you get 10 extra deaths. Now, these have been sometimes called excess deaths. I've used that phrase myself. I deeply regret it and I hope never to use it again because it's so easy to misinterpret.
- [13:31]
B
So the Sunday Telegraph picked up on your research. Now, you complained later and had a letter published that they'd got things wrong. What had they got wrong?
- [13:40]
H
They took the hospitals that had more deaths than would be expected. It's a bit like saying that had more than the average number of deaths and they added up all these so called excess deaths and got to a total of 78 and reported these as being possibly needless deaths. But I was arguing later in a letter to the paper that this was inappropriate because half of all hospitals are bound to be worse than average. And to complain about that is like saying, like the famous politician who said that he wanted everybody to earn more than the average wage.
- [14:11]
B
But presumably it does indicate that something might be wrong, that some further investigation is called for. If a hospital is experiencing more children dying than you would expect.
- [14:24]
H
Well, it depends. If that number could be just explainable by random variation, then it doesn't suggest anything's wrong at all. If it's a bit larger than that, then obviously it deserves further investigation. But there is a particular problem with heart surgery on babies, because this idea of trying to get an expected mortality is really tricky with adults with heart surgery. Actually, there's not that much difference between what can go wrong with an adult heart and you can standardise for the risk quite well, but for babies, there's enormous variability in the type of conditions that are trying to solve. And so the calculating the expected is a really imperfect art at the moment. And so any hospital that appears to be an outlier, really, it only suggests taking just a much more careful look at the data and looking for whether that hospital might be treating particular types of patients.
- [15:13]
B
Professor David Spiegelhalter. You can find a longer item on hospital mortality in our online archive@BBC.co.uk more or less. It's also the place to subscribe to our podcast and be sure not to miss the start of the next series in December. And if we pop up on another program in the interim, which we might, then our loyal podcast subscribers are guaranteed to catch every moment. Last week, Charlotte MacDonald noted that the Prime Minister's salary has become a convenient benchmark against which other public sector workers pay is measured. She went further, conjuring up an entire Prime Minister index. Now, many of you wrote in to complain that instating the PM's salary, we'd taken no account of other rewards he receives. A free flat in central London, for example, or his pension. Well, we think you're all wrong. We weren't trying to work out the PM's total reward. We were taking the number that others have used as a benchmark. That's the annual pay he takes home to create an index. However, since so many of you raised the question of the PM's total reward, Charlotte MacDonald has been working it out. Charlotte, you're back. Welcome. Hi, Charlotte. A lot of listeners have told us that the Prime Minister gets a salary for being an MP on top of the £142,500 we quoted as his pay for being Prime Minister. Is that true?
- [16:28]
A
No, that figure is his combined pay for both roles. David Cameron is entitled to claim a total of £198,660 in pay, but he chooses to take £142,500.
- [16:41]
B
OK, so what else does he get? I imagine the pension must be pretty good.
- [16:44]
A
Well, in the past, Prime Ministers were able to claim a pension worth half their salary from the moment they left office without having made any contributions to the scheme. However, David Cameron, like Gordon Brown before him, has opted out of that. Instead, he pays monthly contributions into the ministerial pension, which he'll be able to claim from 65.
- [17:03]
B
OK, so the ministerial pension, I assume that's not really worth anything, then. Right.
- [17:08]
A
Well, we asked Tom McPhail, who's head of pensions research at the financial advisors Hargreaves Lansdowne, to work out how much David Cameron's pension is worth. He says in simple cash terms, David Cameron is benefiting to the tune of £45,651 a year. There are quite a few assumptions, of course, behind that figure.
- [17:28]
B
Ok, so that's a rough guess at what the pension's worth. What about the free accommodation? A lot of listeners said we needed to reflect that.
- [17:35]
A
Well, for a start, the Prime Minister can't claim those infamous accommodation expenses because he gets a home with a job. He and his family are currently living in the four bedroom flat at number 11 Downing Street. They also have chequers at their disposal, the PM's Country Retreat. I called a few letting agents to see how much both properties could command in rent. I was told you'd have to fork out about 2,000 pounds a week for a flat like Downing street and about £4,500 a week for chequers. If you look at it that way, including the pension, you could say the Prime Minister gets an extra £389,651 a year on top of his salary.
- [18:13]
B
Well, that's rather nice for him. But, you know, the thing is, Charlotte, though I hesitate to suggest that our listeners have bullied us into embarking on a ludicrous exercise. And it is a ludicrous exercise, isn't it? I mean, David Cameron wouldn't actually choose to rent out Chequers and Downing Street. He already has homes in Notting Hill and Oxfordshire, and these properties come with a job for a reason. We don't really want the Prime Minister to spend a lot of time commuting. We don't want him to have to rent out a corner of All Bar One to entertain visiting heads of state. These are not straightforward perks, are they?
- [18:43]
A
No, these are good points. And for the record, he does pay council tax on the Downing street flat. That's about £1,400 a year. He also declares it as a benefit in kind and pays tax on it. You could include other benefits in this analysis, though, such as the use of a chauffeur driven car. But as he lives at his place of work and his travel plans are closely related to his work and security arrangement, it's hard to put a sensible number on it.
- [19:08]
B
Still, the overall pay and benefits, they're not too shabby, are they?
- [19:12]
A
I wanted to get a sense of how the Overall package seemed to someone who works with high earners every day. So I spoke to Virginia Eastman, a headhunter, who had this to say about the Prime Minister's deal.
- [19:22]
D
If one was selling this package in its totality to a candidate, you would have to say, yes, the base may be a little bit low going in, but then of course, and you don't have the bonus or the equity component that you would hope to attract a really very savvy candidate with, but all your income that goes basically straight into your bank, let's face it, it's bankable because all the other perks of the job are the things that people would spend money on day to day. Your housing, accommodation, your going out and entertaining. So a large number of ext. Comfortable perks that go with the total package going into the job. You might well take a cut, you might well look at a base that is less your commercial value, but you'll most certainly treble or double your income going out. And frankly, if you don't, it's your own fault.
- [20:08]
B
Virginia Eastman there. Charlotte MacDonald, thank you. You're listening to More or Less with me, Tim Harford and our website is BBC.co.uk more or less. Now, what could be funnier than death or maths or maths and death? Not much, according to the crowds at this summer's Edinburgh fringe who flocked to hear Matt Parker, the stand up mathematician to a gentle comedy show all about statistics and death with the journalist Timandra Harkness. We felt that a well reviewed show all about maths should not go unheeded here at More or less. So we asked Matt to come in and give us his light hearted take on the mathematics of death. But be warned, There is a 0.000003% chance that you will die while listening to Matt's essay. Enjoy.
- [20:54]
C
Whenever you do something, there's a chance it might kill you. Taking ecstasy, for example, might kill you. The numbers tell us that for every million ecstasy pills taken, one person will die as a result. So taking ecstasy exposes you to a one in a million chance of death, which, which is why I've mentioned it. Mathematicians have a name for a one in a million probability of dying. They call it a micromort. So taking ecstasy costs you one micromort. Statistically speaking, if you go canoeing, there is a one in a million chance of dying for every six minutes spent in a canoe. So six minutes canoeing is equivalent to taking ecstasy As a non rigorous form of probability. We can now use the micromort to compare all sorts of risks as a baseline, averaged across your entire life, every day, just by existing, you risk 30 micromorts, a 30 in a million chance of dying just by existing. However, your probability of dying actually goes up dramatically towards the end of your life and is much less earlier on. So an average 30 something in the UK is only exposed to two and a half micromorts on any normal day. About half to one micromort each day is down to random accidents. In 2008, for example, 96 people died because they fell out of bed. This works out to be one micromort risk due to dying from falling out of bed for every 207 nights sleep you get. If you really want to increase your chances of dying though, you need to do something like skydiving or scuba diving. Five micromorts. There are all sorts of ways you can pay good money to increase your chances of dying. You can go hang gliding, that'll be eight micromorts. Or of course, you could pay money to reduce your odds of dying. If you look at what people spend on safety gear such as crash helmets or optional airbags for cars, it works out that people value their own lives at approximately £25 for every micromort they can avoid. This can be compared with Transport for London, who for every micromort avoided by a commuter, will spend on average 1 pound 40. The reason a unit like micromorts can be useful is that as humans we have no real intuitive sense of probability and risk. We are over sensitive to some risks and then blase about something which might be much more dangerous. You can spend your life worrying about what aspects of your diet might give you cancer and then just not think twice about crossing the road unnecessarily, which is actually a much bigger risk and far easier to avoid. By crunching the numbers and calculating the statistical probability of death, we can approach risk in a much more logical manner and make informed choices. Sure, we could instigate a national ban on duvets and enforce tucking in with sheets and blankets just to save those 96 lives a year. But a more reasoned look at the statistics can put everything into perspective. Although to be honest, you probably didn't need statistics to know that it's not a good idea to take ecstasy while
- [24:03]
B
canoeing if you're still alive. That was Matt Parker, the stand up mathematician. The series is nearly over, you know, and we'd like to thank everyone who emailed us while we were on air. As ever, we read all your emails and we reply to as many as we can please keep them coming whether we're on air or off to more or lessbc.co.uk Timothy Treffery did just that. He sent us his analysis of police staffing numbers. Timothy had been looking at headlines generated in July by a report from Her Majesty's Chief Inspector of Constabulary. Here's Jonathan Dimbleby getting up ahead of steam on the subject on last week's Any questions?
- [24:38]
E
How does all this happen? Sir Dennis, in an earlier report in July, observed that only one in 10 police officers at any one time available to deal with crime because either they're often duty or they've got other duties which are not related directly to dealing with tackling crime. And if the budget is falling, isn't that probably going to be worse?
- [25:00]
C
We know there's far too much.
- [25:01]
B
Now just hang on a moment. You might think that if only one officer in 10 is dealing with crime, the other nine are sitting around eating doughnuts. But as Timothy Treffery points out, they may be sleeping or taking time off at weekends or other rest days. Surely nobody thinks that every police officer should be on duty 24 7. But the police force as a whole does need to be available round the clock. As an example, picture if you will, the remote storm lashed island village in which the more or less team reside. When the program is off air, the village police station has to be manned round the clock, though only by a single officer. How many police need to be on the staff to ensure that this is possible? There are nearly 1100 shifts in a year, 365 days times three shifts a day and an officer working a 40 hour week works 260 of them. Over four officers therefore are needed. But wait, perhaps you might want these fine men and women of the constabulary to take occasional holidays. Then you'd need more than five officers. Adjust for eight days sick leave and you'd need five and a half officers. Account for the high rates of days in court training days, rest days and so on at some forces and you get to seven officers required to staff that one man police station. The HMIC report mentioned on any questions actually says 11% of police are available at any one time. Jonathan Dimbleby and in fairness, many others have summarised that as 1 in 10. Actually, as more or less listeners will know, 11% is closer to 1 in 9. Our village police station has got pretty close to that number without making any reference to donuts. And you never know, you might want to recruit a couple of extra officers just in case someone needed to be available to pop out of the station without leaving it unmanned. After all, the More Or Less team can be a rowdy bunch. With that, the series is over. We've scythed through statistics on equality, pay, welfare, life expectancy, disability, Catholicism, immigration, war, death and men's casual wear. And we've antagonized easily the two most powerful and vociferous lobbies in the world cycle helmet deniers and cat fanciers.
- [27:17]
D
Phew.
- [27:19]
B
Thanks for listening. We'll be back in December. Until then, goodbye.
- [27:24]
A
More or Less was presented by Tim Harford, the FT's undercover economist. The producer was Richard Knight and the editor was Richard Varden.
- [27:31]
D
It was made in association with the Open University.