← More or Less
More or Less
Faith and Charity?
00:28:15
"Religion Makes People More Generous"- according to The Daily Telegraph's of a new BBC...
Loading summary
Transcript86 lines
- [00:00]
A
Thank you for downloading More or less from the BBC.
- [00:03]
B
This is the version first broadcast on Radio 4.
- [00:06]
A
Here's Tim Harford.
- [00:08]
B
Hello, and welcome to the last in this series of More or Less. You can insert your own statistical pun here. This week, we put a famous probability puzzle, the birthday paradox, to the test on a World cup stage and we shine the spotlight on the business craze for big data.
- [00:26]
C
There's this enormous mythology that somehow the larger the data, the closer it is to truth. And I think it's at that level of mythology that we need to be most careful and most critical.
- [00:36]
B
But first, the BBC commissioned a poll recently and one set of figures got some humanists up in arms. Here are the numbers reported on Radio 4's religious affairs show Sunday.
- [00:48]
D
Well, this Cirque Comrades poll asked whether people had given to charity in the last month, and it actually found that seven in 10 adults had done so. So, overall, we are a generous. And then we asked whether people practiced a religion, and we defined that as whether you pray, read a holy book, attend a religious gathering once a month and whether you'd given to charity in the last month. And we found that three quarters of people living in England who practice a religion have given to charity in the last month, and that compares to only two thirds who don't practice a religion.
- [01:18]
B
The presenter, Edward Sturton, wanted to be clear about this. Can you look at the differences thrown up in your poll and legitimately conclude
- [01:27]
A
on the negative side that if you're not religious, you're less likely to give money to charity?
- [01:31]
D
The pollsters tell us that it is a significant difference and it's entirely valid to say that you're less likely to give if you don't practice a religion.
- [01:40]
B
Well, Pavan Dhaliwal from the British Humanist association contacted more or less and asked us to take a look at the figures. Was the survey fair? Well, I've got Charlotte MacDonald here with me. Hello, Charlotte.
- [01:52]
D
Hi. The poll was commissioned by the BBC English Region's religion and ethics unit. Now, far be it from me to defend the BBC.
- [02:00]
B
Please don't defend the BBC, Charlotte. It's much more fun when we put the boot in.
- [02:04]
D
Maybe Tim. But it's fair to say that this was a large sample of over 2,600 people, and the poll gave a lot of consideration to the way the questions were asked. In particular, instead of merely asking people to identify as religious, as we just heard, the question focused on regular religious practices.
- [02:23]
B
But what about the way this was reported?
- [02:25]
D
Well, first off, it must be noted that the poll was not reported in isolation. It was used as a starting point for various reports across BBC local radio about attitudes and opinions to charitable activity. The coverage of the poll findings was pretty carefully phrased. Others who reported the BBC poll were happier to jump to conclusions. Here's the Daily Telegraph's causal leap. Religion makes people more generous.
- [02:53]
B
Hmm. Well, correlation isn't causation, as we're always being told, but what's the alternative? Generosity makes people more religious.
- [03:01]
D
It's not unreasonable to suggest that religiosity is causing generosity, but it's speculation. There's another good explanation. Here's Kimberly Scharf, a professor of economics at Warwick University. She's an expert in charitable giving.
- [03:17]
E
The evidence that you find is actually a correlation. It doesn't establish any kind of causal relationship between religion and charitable contributions. We know that social interactions are important, and we know that almost 60% of giving occurs in social situations. Could be the social situations that are actually driving these things. The social pressure, possibly of being asked to give the weekly collection might be the thing that's actually inducing these people to give to charity within the church.
- [03:48]
D
So it's reasonable to say that the BBC has confirmed earlier research that religious people give more often, but it's more of a guess to say that religion is the cause of the generosity. In fact, the BBC survey also asked about social networks and those around you giving money.
- [04:04]
B
Okay. I have been wondering one thing about the BBC poll, though. Charlotte. So let me ask you a couple of questions. Charlotte, do you believe there's more to life than material possessions? Yes. Were you brought up to believe that it was important to look after people less fortunate than yourself?
- [04:21]
D
Yes, of course.
- [04:22]
B
Charlotte, can you picture a time, perhaps when you were young, perhaps at Christmas when you were sitting in church?
- [04:28]
D
I'm imagining it now.
- [04:29]
B
Charlotte, have you given money to charity in the past month?
- [04:33]
D
Oh, yes.
- [04:35]
B
My point is questions can put people in a particular frame of mind. In other opinion polls, these framing effects have been shown to have a very strong influence on how people respond. You mentioned that the BBC Comrades poll was scrupulous in asking people to reflect on their regular religious practices, and indeed it was. But that may have influenced what they said immediately afterwards about giving to charity.
- [04:59]
D
So you're worried that if you first ask people to think in some detail about their religious practice and then ask them if they've recently given to charity, you'll shape their answers?
- [05:10]
B
Well, I think it's possible. It would have been good to see what would have happened if the questions were asked in the opposite order.
- [05:17]
E
We have what we call framing effects on these kinds of surveys, where the first question stimulates people to start thinking in a particular way. So if somebody identifies themselves as a religious person and they think that religious people are actually more charitable, then it actually might have some effect and influence on the way that they answer the second question. And when academics conduct surveys, what we would do is we would actually switch up the questions for some people. So we would ask the question about charitable giving first and the question about religion second, and then we would actually compare the responses across those different orderings of surveys to see if the orders actually mattered.
- [05:57]
D
That's Kimberly Scharf again. Now, I think it's a fair point, Tim, but speaking to Comres, they point out that the people surveyed were asked a number of questions on all sorts of subjects, from how they feel about commercial products to politicians. And yes, they were asked about their religious practices and then a question about giving to charity. But Comrez point out that the donating question was quite broad. It said, as far as you know, and in the past month only, which of the following people do you think have donated money to charity, if any? So that question's asking about other people, not just about your own giving.
- [06:35]
B
Now, leaving aside the complaints of pro religious bias, what's actually being said here? The BBC poll found that 71% of adults gave to charity in the last month. But what does that mean? Does putting money in the church collection count as charitable giving? I know the secularists wouldn't see it that way. What about giving money to a person begging at a cash point? Pretty sure they're not a registered charity, but the act is charitable and there's no mention of the amounts given either.
- [07:05]
D
The BBC poll simply mentions donating money to charity and we don't know exactly what people were thinking of when they answered yes. So I asked Cathy Farrow of Cass Business School if there were any other national surveys and what they considered to be donating to charity. One of the main surveys in the UK, which is called UK giving, tends to find that around 58 to 60% of people give every month. Everything is included, whether it's a purchase through a charity catalogue, a Christmas card, a raffle ticket, a ticket for a show. All those kinds of purchases do get included as charitable giving. Now, that was a survey conducted by the Charities Aid foundation, and that's a pretty broad definition.
- [07:48]
B
Well, I don't know what to think now. The Charities Aid foundation survey doesn't quite reach 60% of people donating to charity. Even though it includes buying a lottery ticket.
- [07:58]
C
But.
- [07:58]
B
And yet somehow the BBC survey topped 70%.
- [08:01]
D
There's more, Tim. Cathy Farrow and colleagues have been analyzing 30 years worth of data based on long standing household surveys. She and her colleagues found that in fact, the proportion of households donating to charity has fallen slightly over the last three decades, from around 32% to 28%.
- [08:20]
B
But that's a massive difference. That's a fair bit less than one third of households giving to charity. And, and yet the BBC Commerce survey found that it was a fair bit more than 2/3 giving to charity.
- [08:31]
D
Right. In the household surveys used for this research, the definition of charitable giving is much tighter. It really means cash given to a charity. Sending money to a relative overseas doesn't count, neither to raffle tickets or Christmas cards. Plus, participants in this survey are asked to keep a two week diary of all expenditure rather than basing it on memory.
- [08:53]
B
And the other thing we don't know from the BBC poll is how much people were donating or how often they could have given 20p for the only time this year.
- [09:02]
D
Cathy Farrow's research shows that it's actually quite a small pool of people who contribute large sums. Looking at her figures, more than 90% of the total amount donated to charity is given by just half of the households that donate.
- [09:17]
B
So while the majority of people tell pollsters they've recently given to charity, the vast majority of charitable donations are actually coming from just one household in seven. Well, thank you very much, Charlotte. You're listening to more or less all of it. Add all that information together and it's called Big Data. It is immensely valuable to a lot of people for good and possibly for
- [09:40]
A
ill, find out things that are useful. And one area where Big data is about to make quite a big difference.
- [09:47]
B
A lot of people are buzzing with excitement about the promise of what they're calling Big Data. Big Data itself is a vague term. It sometimes refers to the vast data sets produced by scientific instruments such as radio telescopes or the Large Hadron Collider. But another meaning of Big data, the one which interests us for the next few minutes, is the digital information we're constantly producing as a by product of searching online, tweeting, posting to Facebook, paying by credit card, or wandering around with a mobile phone, constantly revealing our location. It looks like computers processing huge data sets are going to give us all the answers that social scientists, marketers and spies could possibly want. But here on More Or Less, we want to make the case for caution.
- [10:33]
C
There's this enormous mythology that somehow the larger the data, the closer it is to truth. And I think it's at that level of mythology that we need to be most careful and most critical.
- [10:44]
B
This is Kate Crawford, an academic and a researcher at Microsoft. We'll hear more from her later. But first, a couple of cautionary tales. Five years ago, a team of researchers from Google announced an impressive discovery. They'd found a way to track the spread of influenza across the United States by analyzing what we search for on the Internet. Google Flu Trends was a lot faster at detecting the spread of the flu than traditional surveillance systems that required monitoring at hospitals. This is David Lazer, professor of political science and computer and information science at Northeastern University. Google Flu Trends could give you flu case figures within 24 hours. The official figures from the US authorities took up to a fortnight. Google Flu Trends was fast, cheap and effective. It was also theory free. Instead of developing some model of what people with flu might search for, the Google team just looked at historical correlations between flu and their top 50 million search terms. Then they let the algorithms do the work. And this created a great deal of attention. There were headlines, and I think it has been held up as one of the exemplars of the potential of big data. But there was a problem. It started going off kilter and systematically so, which was a bit odd. In the season 2011-2012, Google Flu Trends overestimated the flu by 50%. By the following year, it was predicting two cases of flu for every one that actually materialized. If you say that there are more than twice as many cases as there really are, that's a big miss. So what went wrong? Perhaps it was TV coverage about a flu epidemic that scared healthy people into searching online. Or perhaps Google Search itself got too clever for Google Flu Trends, automatically suggesting search terms and changing what people ended up looking for. No one's sure what happened, which is part of the problem. Without a theory for why people were searching for flu terms, Google could only spot patterns, and patterns weren't enough. No doubt Google Flu Trends will bounce back, but unless we learn the lessons of this episode, we will find ourselves repeating it. I've been looking into this as part of my day job at the Financial Times. And what worries me is that for all the genuine promise of these new data sets, we risk forgetting some very old statistical lessons. Google Flu Trends has already shown that finding patterns isn't enough. Knowing what causes those patterns matters too. And every time I hear people boasting about the size of their data sets, it reminds me of an Old statistical story.
- [13:43]
A
The battle is on. The Republican National Convention has nominated Governor Alfred Mossman Landon, the Kansas Coolidge, as its candidate for president.
- [13:53]
B
As Pathe News reminds us, in 1936, the President of the United States States, Franklin Delano Roosevelt, a Democrat, was seeking re election.
- [14:03]
A
The Republicans think Alf Landon is the man who will win in November. He is our next president if he can beat Roosevelt.
- [14:17]
B
A very popular and respected magazine, the Literary Digest, set itself the task of forecasting the result. This was a vast postal opinion poll. They sent ballots to a quarter of the total electorate, 10 million people. A quarter of those contacted. 2.4 million people sent responses. Eventually, the Literary Digest announced its prediction. Alfred Landon would win with a solid margin of 55% to 41%. But it was Roosevelt who crushed Landon by 61% to 37%. The Literary Digest was very wrong. Even worse, a fellow called George Gallup, the opinion poll pioneer, conducted a much smaller survey and was far closer to the eventual result. So what did Gallup understand that the Literary Digest didn't? The Literary Digest went for size, but neglected sampling bias. They got their vast mailing list from the phone book and the list of car registrations. But Americans who owned phones and had cars in 1936 weren't representative of the voting population. George Gallup, on the other hand, carefully selected a representative sample. The Literary Digest thought the bigger the sample, the better the result. But bigger isn't always better, as the authorities in the American city of Boston and recently found out.
- [15:48]
C
One of the things you learn as a resident of Boston is that there's a lot of bad weather.
- [15:53]
B
This is Kate Crawford again. You might remember that she's at Microsoft.
- [15:57]
C
There's actually a big problem with potholes in the road. They end up patching around 20,000 potholes a year. And they're always trying to think of more efficient ways to figure out where the potholes are. So I first heard about this new app that was being released by the city of Boston called Street Bump, that you could download to your smartphone. And what it would do is it would track your accelerometer, which is the way that your phone is moving in space, along with your gps, which gives the coordinates of where you are so that it could actually passively detect every time you would hit a pothole as you were driving around the streets of Boston. And this is actually a very clever idea.
- [16:33]
B
Well, great. Everyone who has the app is sending back data. But Kate asked herself, who's missing?
- [16:41]
C
People in lower income groups and older citizens. That's people over the age of 60 are less likely to have smartphones. Therefore, in the areas where we have those populations living, we're actually getting less data about their roads. And that might mean then that a city could say, well, we're not getting data about any kinds of potholes there. We don't need to send out the road repair crews.
- [17:01]
B
In this particular case, the city of Boston was wise to the bias and took steps to correct for it. But for Kate Crawford, the story represents something bigger.
- [17:09]
C
What's so interesting about this story is that I think it's a kind of parable for Big Data, is that when we start looking to smartphones and apps, we always have to think about who is being left out, who is not in that data set was not being represented
- [17:27]
B
for business. Of course, the exciting success stories for Big Data are the ones about making money. There's a famous story going around. We've even broadcast it on more or less, about the US discount department store Target and its fabulous algorithms, which use data from the store loyalty card to figure out how best to market to different customers. For instance, if Target's algorithms can observe your purchases of vitamin supplements or cocoa butter and figure out that you're pregnant, well, that's an opportunity to acquire a loyal customer at a lucrative stage of life. Perhaps you can guess where the story goes next. A man stormed into a Target store in Minnesota and complained that the company was sending coupons for baby clothes and maternity wear to his teenage daughter. The manager apologised and in fact, felt so guilty that he called back a couple of days later to apologise some more. When he did, he was informed that the teenager was indeed pregnant. Her father didn't know, but Target's algorithms allegedly did. Let's assume for a moment that this oft repeated anecdote is true. Well, then, Target's algorithms seem almost telepathic. But hang on a moment. First, is it really so surprising to hear that a company can deduce you're pregnant when it sees you're buying folic acid on the store loyalty card? That's a pretty easy guess compared to what else companies might hope for from Big Data. And second, when a voucher for maternity wear arrives on the doorstep of a pregnant woman, is that a brilliant algorithmic deduction or a lucky guess? Maybe. Stores like Target mail out loads of these vouchers, and when they arrive on the doorstep of people who aren't pregnant, they simply go in the dustbin. It's not to say that these marketing algorithms are useless. They can work well, but this story picks an easy example, and it doesn't ask about the false positives as well as the surprise hits, so it really doesn't tell us very much. Statisticians are scrambling to develop new methods to seize the opportunity of big data. Now, such new methods are essential, but they'll work by building on the old statistical lessons, not by ignoring them. I spoke to Steve Leavitt, the author of Freakonomics and an economist who's won one of the profession's biggest prizes for his brilliant work with data. Leavitt's been much in demand to do data analysis, and he's not impressed. I've worked with probably 10 of the biggest companies in the world, and they have been completely and totally ineffective at harnessing data to do much of anything that's useful.
- [20:18]
A
And part of the problem is just
- [20:20]
B
coordinating how to integrate the data in these archaic systems. But the second piece is that there
- [20:26]
A
is an incredible lack of talent for analyzing data.
- [20:31]
B
There is currently no educational institution whose aim is to teach people how to understand and interpret data. There's clearly a long way to go before we can understand how to use the information we have. It's dangerous to assume the results of big data analysis are 100% accurate. It's important to understand why we have the results we get, rather than merely picking out patterns. And we need to ask ourselves, who's missing from this data? Big data has arrived, but big answers have not. You're listening to More or less, and I'm Tim Harford. We heard from Steve Levitt just then. He was also on the program last week, talking about an experiment he ran, getting people to toss a coin to make major life decisions. Listener Jeremy emailed to point out he's not the first. The idea of tossing a coin to make a decision and then examining your feelings about the outcome was first thought up and used by Freud. And Margery pointed us to this poem by Danish scientist and mathematician Piet Hein.
- [21:43]
A
Whenever you're called on to make up your mind and you're hampered by not having any, the best way to solve the dilemma you'll find is simply by spinning a penny. No, not so that chance shall decide the affair while you're passively standing there moping. But the moment the penny is up
- [22:02]
B
in the air, you suddenly know what you're hoping you're listening to. More or less with me, Tim Harford. And since the show features poetry by Danish mathematicians, who else would you be listening to? Our email address, as always, is more or lessbc.co.uk. now we've learned this series that if there's one thing more or less listeners love, it's birthdays. We're still getting emails about the problem we posed a few weeks ago about the chances of having a major round number birthday on a weekend. So we thought, why not challenge one of our favourite mathematical authors to find a birthday related twist on the big story that everyone's watching. He introduces himself far better than I ever could.
- [22:45]
A
I'm Alex Bellos, the author of Alex through the Looking Glass, How Life. I'm with a tongue twisting title. I'm Alex Bellos, author of Alex through the Looking How Life Reflects Numbers and Numbers Reflect Life.
- [23:00]
B
Such a good title. You're overwhelmed as you try to say it. So, Alex, you're here to tell us about the birthday paradox. And loyal more or less listeners may know about the birthday paradox, but just remind us.
- [23:14]
A
Well, the birthday paradox is probably one of math's greatest hits. It's something you can say in one line which gives you this kind of wow. And what it is is that you only need 23 people for it to be more likely than not that two of them share the same birthday. And this is not a logical paradox. There's nothing self contradictory about it, but it goes so against sort of intuition because you know there are 365 days of the year, so surely you'd need more than 23 people for it to be more likely than not that two share the same birthday.
- [23:46]
B
What's the most persuasive way you found to convince people that this is at least a plausible statement?
- [23:53]
A
The theoretical way to explain it using probabilities, which is not difficult but quite tricky on the radio you can see it very clearly. It's not difficult, but actually I think you need to just look at the world. People share birthdays all the time and think, oh, it's an amazing coincidence that two people in my class of 30 share the same birthday. Actually, with 30 people, it's 70% chance that two people share the same birthday.
- [24:18]
B
Now, it would be great if we had some convenient real world data set where we gathered together groups of 23 people again and again and again and were able to examine shared birthdays. But I can't think of an example that really works.
- [24:34]
A
Well, I thought long and hard and then all of a sudden I realized that the World cup is a brilliant, brilliant example because the squads for the World cup are each made of 23 players. So you have, you know, 32 examples to test this out.
- [24:53]
B
There are 32 teams. We should expect about half of them to contain a shared birthday. And if we're a long way away from half, either on either side, then something's going on or we've been unlucky.
- [25:06]
A
Expect 50.7% of them. But when you look at the figures, which I did, you will find that there are 19 out of the 32 teams have shared birthdays, including five of the team have two shared birthdays. So four players share birthdays and this is about 60%.
- [25:30]
B
Afraid we're going to have to stop the interview there.
- [25:32]
A
Foul by Alex Bellos for getting his birthdays from Wikipedia, which you won't be surprised here, weren't all correct. Normally that would be an automatic red card and a sending off.
- [25:43]
B
But as we had to record this
- [25:45]
A
interview before the official FIFA squads were released because Alex was about to leave
- [25:49]
B
for Brazil, I'm going to give him a yellow card and the chance to
- [25:53]
A
retake the answer all the way from Sao Paulo. Sorry, ref, my bad. The Revised figures are 16 teams at the World cup have shared birth. And this is actually really rather good because it's mathematically what you would expect. So actually it's all turned out rather well and it's been really good fun just perusing the spread of birthdays because you might have thought that the spread of footballers birthdays, well, the top footballers anyway, would be uniform throughout the year, but that's just not the case. It turns out that footballers are much more likely to be born at the beginning of the year. The average number of footballers that you would expect to be born per month, this is the ones at the World cup, would be 61. But from January to May, every month, much more footballers and the lowest months in terms of the least number of footballers, birthdays are October, November and December. And this is interesting to wonder why. And it's probably because of school cutoff years, or if not school cutoff years, the cutoff for junior team levels.
- [26:58]
B
I should say at this point that most footballers seem to come from countries where the school year begins in January rather than September.
- [27:05]
A
The theory is that those children born just after the cutoff levels are the biggest kids at the team. They get more of the ball they get, they become better players, they're more likely to become professionals anyway. There's so much more stuff to look at. It's really fascinating. But Tim, I've got to go to the game.
- [27:25]
B
See you later, Alex Bellos. This is the last show of the series on Radio 4. We'll be back again in late August, but we do have a special short edition of the programme running all year round. You can subscribe to it@BBC.co.uk more or less. Please keep your emails coming in to more or lessbc.co.uk. we'll be checking them for ideas for our next series. Until then, goodbye. More or Less was presented by Tim
- [27:54]
A
Harford, the Financial Times undercover Economist.
- [27:58]
B
The producer was me, James Fletcher, with
- [28:00]
A
Charlotte MacDonald and Laura Gray and the
- [28:03]
B
programs made in association with the Open University. You can download many more programs for free from Radio 4. You can find them at BBC.co.uk radio4.