Why you should love statistics | Alan Smith

Translator: Niels Justus Reviewer: Sune Vilsted Østergaard

Back in 2003,

Tilbage i 2003,

the UK government carried out a survey.

gennemførte den britiske regering en undersøgelse,

And it was a survey that measured levels of numeracy

der målte regnefærdigheden

in the population.

i befolkningen.

And they were shocked to find out

Og de var rystet over at se,

that for every 100 working age adults in the country,

at for hver 100 erhvervsaktive voksne,

47 of them lacked Level 1 numeracy skills.

manglede 47 af dem niveau 1-regnefærdigheder.

Now, Level 1 numeracy skills -- that's low-end GCSE score.

Niveau 1-regnefærdighed -- det er en lav GCSE-score.

It's the ability to deal with fractions, percentages and decimals.

Det er evnen til at regne med brøker, procenter og decimaler.

So this figure prompted a lot of hand-wringing in Whitehall.

Dette tal skabte en masse hændervridning i Whitehall.

Policies were changed,

Politikker blev ændret,

investments were made,

investeringer blev foretaget.

and then they ran the survey again in 2011.

Derefter udførte man undersøgelsen igen i 2011.

So can you guess what happened to this number?

Kan I gætte, hvad der skete med dette tal?

It went up to 49.

Det gik op til 49.

(Laughter)

(Latter)

And in fact, when I reported this figure in the FT,

Og da jeg skrev om dette tal i Financial Times,

one of our readers joked and said,

spurgte en af vore læsere og sagde:

"This figure is only shocking to 51 percent of the population."

"Dette tal er kun chokerende for 51 procent af befolkningen."

(Laughter)

(Latter)

But I preferred, actually, the reaction of a schoolchild

Men jeg foretrak reaktionen fra en skoleelev,

when I presented at a school this information,

da jeg præsenterede denne information på en skole.

who raised their hand and said,

Han rejste sin hånd og spurgte:

"How do we know that the person who made that number

"Hvordan kan vi vide, at ham, der regnede tallet ud,

isn't one of the 49 percent either?"

ikke også var en af de 49 procent?"

(Laughter)

(Latter)

So clearly, there's a numeracy issue,

Det er klart, at der er et problem med regnefærdigheden,

because these are important skills for life,

fordi dette er en vigtig færdighed i livet.

and a lot of the changes that we want to introduce in this century

Mange af de forandringer, som vi gerne vil gennemføre i dette århundrede

involve us becoming more comfortable with numbers.

involverer, at vi bliver mere komfortable med tal.

Now, it's not just an English problem.

Det er ikke bare et engelsk problem.

OECD this year released some figures looking at numeracy in young people,

OECD udgav i år nogle tal om regnefærdighed hos unge,

and leading the way, the USA --

og forrest lå USA

nearly 40 percent of young people in the US have low numeracy.

med næsten 40 procent af unge, som har ringe regnefærdigheder.

Now, England is there too,

England er der også,

but there are seven OECD countries with figures above 20 percent.

men der er syv OECD-lande med værdier over 20 procent.

That is a problem, because it doesn't have to be that way.

Det er et problem, fordi det ikke behøver at være sådan.

If you look at the far end of this graph,

Ser man på den anden ende af denne graf,

you can see the Netherlands and Korea are in single figures.

kan man se Holland og Korea med ét ciffer-størrelser.

So there's definitely a numeracy problem that we want to address.

Der er et problem med regnefærdigheden, som der skal gøres noget ved.

Now, as useful as studies like these are,

Selvom disse undersøgelser er nyttige,

I think we risk herding people inadvertently into one of two categories;

tror jeg, at vi risikerer uforvarende at putte folk i én af to kategorier;

that there are two kinds of people:

at der er to slags mennesker:

those people that are comfortable with numbers, that can do numbers,

de mennesker, der er komfortable med tal, som kan regne med tal,

and the people who can't.

og de mennesker, der ikke kan.

And what I'm trying to talk about here today

Og det, jeg forsøger at sige,

is to say that I believe that is a false dichotomy.

er, at jeg tror, at det er en falsk modsætning.

It's not an immutable pairing.

Det er ikke en uforanderlig parring.

I think you don't have to have tremendously high levels of numeracy

Jeg tror ikke, man behøver en voldsom god talforståelse

to be inspired by numbers,

for at interessere sig for tal,

and that should be the starting point to the journey ahead.

og det bør være udgangspunktet for, hvad vi gør fremover.

And one of the ways in which we can begin that journey, for me,

Og en af måderne, hvorpå vi kan gøre tal interessante, er, for mig,

is looking at statistics.

at se på statistikken.

Now, I am the first to acknowledge that statistics has got somewhat

Nu er jeg den første til at erkende, at statistikken

of an image problem.

har lidt af et imageproblem.

(Laughter)

(Latter)

It's the part of mathematics

Det er den del af matematikken,

that even mathematicians don't particularly like,

som selv matematikere ikke holder ret meget af,

because whereas the rest of maths is all about precision and certainty,

fordi, mens resten af matematikken handler om præcision og sikkerhed,

statistics is almost the reverse of that.

er statistik næsten det modsatte.

But actually, I was a late convert to the world of statistics myself.

Jeg var faktisk sen til at konvertere til statistikken.

If you'd asked my undergraduate professors

Hvis man havde spurgt mine professorer,

what two subjects would I be least likely to excel in after university,

om hvilke to emner jeg næppe ville udmærke mig i efter universitetet,

they'd have told you statistics and computer programming,

ville de have sagt statistik og programmering af computere,

and yet here I am, about to show you some statistical graphics

og alligevel er jeg ved at vise jer statistiske grafikker,

that I programmed.

som jeg har programmeret.

So what inspired that change in me?

Hvad fik mig til den forandring?

What made me think that statistics was actually an interesting thing?

Hvad fik mig til at tænke, at statistikken faktisk er interessant?

It's really because statistics are about us.

Det var fordi statistik handler om os.

If you look at the etymology of the word statistics,

Hvis man ser på etymologien for ordet statistik,

it's the science of dealing with data

er det videnskaben om data,

about the state or the community that we live in.

om staten eller det samfund, som vi lever i.

So statistics are about us as a group,

Så statistik drejer sig om os som en gruppe,

not us as individuals.

ikke os som individer.

And I think as social animals,

Og jeg tror, som sociale dyr,

we share this fascination about how we as individuals relate to our groups,

vi deler denne interesse i, hvordan vi er, i forhold til vore grupper,

to our peers.

til vore jævnaldrende.

And statistics in this way are at their most powerful

Og statistik er på denne måde mest kraftfuldt,

when they surprise us.

når den overrasker os.

And there's been some really wonderful surveys carried out recently

Der er blevet lavet nogle vidunderlige undersøgelser for nyligt

by Ipsos MORI in the last few years.

af Ipsos MORI i de sidste par år.

They did a survey of over 1,000 adults in the UK,

De undersøgte over 1.000 voksne i Storbritannien.

and said, for every 100 people in England and Wales,

Man spurgte, "For hver 100 mennesker i England og Wales,

how many of them are Muslim?

hvor mange af dem er muslimer?"

Now the average answer from this survey,

Det gennemsnitlige svar fra denne undersøgelse,

which was supposed to be representative of the total population, was 24.

som skulle repræsentere den samlede befolkning, var 24.

That's what people thought.

Det er, hvad folk tænker.

British people think 24 out of every 100 people in the country are Muslim.

Britter tror at 24 ud af 100 mennesker er muslimer.

Now, official figures reveal that figure to be about five.

Det officielle tal viser, at tallet er omkring fem.

So there's this big variation between what we think, our perception,

Der er en stor forskel mellem, hvad vi tror,

and the reality as given by statistics.

og den statistiske virkelighed.

And I think that's interesting.

Jeg synes, det er interessant.

What could possibly be causing that misperception?

Hvad kunne være årsagen til denne fejlopfattelse?

And I was so thrilled with this study,

Jeg var så begejstret over det,

I started to take questions out in presentations. I was referring to it.

at jeg begyndte at stille spørgsmål ved mine foredrag. Jeg fortalte om det.

Now, I did a presentation

Jeg gav et foredrag

at St. Paul's School for Girls in Hammersmith,

på St. Pauls School for Girls i Hammersmith,

and I had an audience rather like this,

og jeg havde et publikum magen til jer,

except it was comprised entirely of sixth-form girls.

bortset fra, at det udelukkende bestod af piger i 10. klasse.

And I said, "Girls,

Og jeg spurgte, "Piger,

how many teenage girls do you think the British public think

hvor mange teenagepiger tror I, at den britiske offentlighed mener

get pregnant every year?"

bliver gravid hvert år?"

And the girls were apoplectic when I said

Og pigerne blev rasende, da jeg sagde,

the British public think that 15 out of every 100 teenage girls

den britiske offentlighed mener, at 15 ud af hver 100 teenagepiger

get pregnant in the year.

bliver gravide hvert år.

And they had every right to be angry,

Og de havde god grund til at være vred,

because in fact, I'd have to have closer to 200 dots

fordi jeg skulle bruge flere end 200 prikker,

before I could color one in,

før jeg kunne farve én,

in terms of what the official figures tell us.

til at vise, hvad de officielle tal siger.

And rather like numeracy, this is not just an English problem.

Ligesom regnefærdighed, er dette ikke bare et engelsk problem.

Ipsos MORI expanded the survey in recent years to go across the world.

Ipsos MORI udvidede undersøgelsen med mange lande.

And so, they asked Saudi Arabians,

De spurgte de saudiske arabere:

for every 100 adults in your country,

"For hver 100 voksne i dit land,

how many of them are overweight or obese?

hvor mange af dem er overvægtige eller fede?"

And the average answer from the Saudis was just over a quarter.

Det gennemsnitlige svar fra saudierne var lidt over en fjerdedel.

That's what they thought.

Det er, hvad de troede.

Just over a quarter of adults are overweight or obese.

Lidt over en fjerdedel var overvægtige eller fede.

The official figures show, actually, it's nearer to three-quarters.

De officielle tal viser faktisk, at det er tættere på tre fjerdedele.

(Laughter)

(Latter)

So again, a big variation.

Så igen, en stor forskel.

And I love this one: they asked in Japan, they asked the Japanese,

Jeg elsker denne her. De spurgte i Japan:

for every 100 Japanese people,

"For hver 100 japanere,

how many of them live in rural areas?

hvor mange af dem bor i landdistrikterne?"

The average was about a 50-50 split, just over halfway.

Den gennemsnitlige svar var omkring 50-50, lidt over halvvejs.

They thought 56 out of every 100 Japanese people lived in rural areas.

De troede, at 56 ud af hver 100 japanere levede i landdistrikterne.

The official figure is seven.

Det officielle tal er syv.

So extraordinary variations, and surprising to some,

Ekstraordinære forskelle, og overraskende for nogle,

but not surprising to people who have read the work

men ikke overraskende for folk, som har læst

of Daniel Kahneman, for example, the Nobel-winning economist.

Daniel Kahnemans arbejde: Nobel-prismodtager i økonomi.

He and his colleague, Amos Tversky, spent years researching this disjoint

Han og hans kollega, Amos Tversky, brugte år på undersøgelser af denne

between what people perceive and the reality,

uoverensstemmelse mellem, hvad folk oplever og virkelighededen.

the fact that people are actually pretty poor intuitive statisticians.

Det faktum, at folk faktisk er temmelig dårlige intuitive statistikere.

And there are many reasons for this.

Og der er mange grunde til dette.

Individual experiences, certainly, can influence our perceptions,

Individuelle erfaringer kan i hvert fald påvirke vore opfattelser.

but so, too, can things like the media reporting things by exception,

Men også det, at medierne skriver om det usædvanlige,

rather than what's normal.

snarere end hvad der er normalt.

Kahneman had a nice way of referring to that.

Kahneman udtrykte det på en pæn måde.

He said, "We can be blind to the obvious" --

Han sagde: "Vi kan være blinde overfor det indlysende" --

so we've got the numbers wrong --

så vi tror på forkerte tal --

"but we can be blind to our blindness about it."

"men vi kan være blinde overfor vores blindhed."

And that has enormous repercussions for decision making.

Det har enorme konsekvenser for beslutningstagning.

So at the statistics office while this was all going on,

Dengang syntes jeg,

I thought this was really interesting.

at dette var virkeligt interessant.

I said, this is clearly a global problem,

Jeg sagde, at dette er et globalt problem,

but maybe geography is the issue here.

men måske er det geografien, der er problemet her.

These were questions that were all about, how well do you know your country?

Spørgsmålene drejede sig om, hvor godt kender du dit land?

So in this case, it's how well do you know 64 million people?

Sagt på en anden måde: Hvor godt kender du 64 millioner mennesker?

Not very well, it turns out. I can't do that.

Ikke særlig godt viste det sig. Jeg kunne ikke gøre det.

So I had an idea,

Så jeg fik en idé,

which was to think about this same sort of approach

som var, at have samme slags tilgang,

but to think about it in a very local sense.

men at sætte det i en meget lokal sammenhæng.

Is this a local?

Er det om det lokale?

If we reframe the questions and say,

Hvis vi omformulerer og spørger,

how well do you know your local area,

hvor godt kender du dit lokalområde,

would your answers be any more accurate?

ville svarene så blive mere præcise?

So I devised a quiz:

Så jeg udtænkte en quiz:

How well do you know your area?

Hvor godt kender du dit område?

It's a simple Web app.

Det er en simpel web-app.

You put in a post code

Man skriver et postnummer,

and then it will ask you questions based on census data

og så stiller den spørgsmål baseret på folketællingsdata

for your local area.

for lokalområdet.

And I was very conscious in designing this.

Jeg var meget bevidst om designet.

I wanted to make it open to the widest possible range of people,

Jeg ønskede at gøre det åbent for det bredeste udvalg af befolkningen.

not just the 49 percent who can get the numbers.

Ikke bare de 49 procent, der kan regne.

I wanted everyone to engage with it.

Jeg ønskede at alle deltog i det.

So for the design of the quiz,

Så til udformningen af quizzen,

I was inspired by the isotypes

blev jeg inspireret af isotyperne

of Otto Neurath from the 1920s and '30s.

af Otto Neurath fra 1920'erne og 30'erne.

Now, these are methods for representing numbers

Det er metoder til repræsentere tal

using repeating icons.

ved hjælp gentagne ikoner.

And the numbers are there, but they sit in the background.

Tallene er der, men de findes i baggrunden.

So it's a great way of representing quantity

Det er en fantastisk måde at repræsentere en mængde

without resorting to using terms like "percentage,"

uden at ty til at bruge udtryk som "procent",

"fractions" and "ratios."

"fraktioner" og "nøgletal".

So here's the quiz.

Her er quizzen.

The layout of the quiz is,

Layoutet af quizzen er,

you have your repeating icons on the left-hand side there,

man har de gentage ikoner på venstre side,

and a map showing you the area we're asking you questions about

og et kort, der viser dig det område, vi stiller spørgsmål til

on the right-hand side.

på højre side.

There are seven questions.

Der er syv spørgsmål.

Each question, there's a possible answer between zero and a hundred,

For hvert spørgsmål, er der et muligt svar mellem nul og hundrede,

and at the end of the quiz,

og i slutningen af quizzen

you get an overall score between zero and a hundred.

får du en samlet score mellem nul og hundrede.

And so because this is TEDxExeter,

Og så fordi dette er TEDxExeter,

I thought we would have a quick look at the quiz

tænkte jeg, vi tager et hurtigt kig på quizzen

for the first few questions of Exeter.

for de første par spørgsmål af Exeter.

And so the first question is:

Så det første spørgsmål er:

For every 100 people, how many are aged under 16?

"For hver 100 mennesker, hvor mange er under 16 år?"

Now, I don't know Exeter very well at all, so I had a guess at this,

Nu kender jeg ikke Exeter ret godt, så jeg gættede,

but it gives you an idea of how this quiz works.

men det giver en idé om, hvordan denne quiz fungerer.

You drag the slider to highlight your icons,

Man trækker skyderen for at fremhæve ikonerne,

and then just click "Submit" to answer,

og så bare klikke på "Send" for at svare,

and we animate away the difference between your answer and reality.

og vi animerer forskellen mellem dit svar og virkeligheden.

And it turns out, I was a pretty terrible guess: five.

Og det viste sig, at jeg havde et ret forfærdeligt gæt: fem.

How about the next question?

Hvad med det næste spørgsmål?

This is asking about what the average age is,

Det spørger hvad den gennemsnitlige alder er,

so the age at which half the population are younger

den alder, hvor halvdelen af befolkningen er yngre

and half the population are older.

og halvdelen er ældre.

And I thought 35 -- that sounds middle-aged to me.

Og jeg troede 35, det lød midaldrende for mig.

(Laughter)

(Latter)

Actually, in Exeter, it's incredibly young,

Det er utroligt ungt i Exeter.

and I had underestimated the impact of the university in this area.

Jeg havde undervurderet virkningen af universitetet her i området.

The questions get harder as you go through.

Spørgsmålene bliver sværere efterhånden.

So this one's now asking about homeownership:

Denne spørger om eget hus:

For every 100 households, how many are owned with a mortgage or loan?

"For hver 100 husstande, hvor mange har et banklån?"

And I hedged my bets here,

Jeg afdækkede mine gæt her,

because I didn't want to be more than 50 out on the answer.

for jeg ikke ønskede, at have flere end 50 forkert.

(Laughter)

(Latter)

And actually, these get harder, these questions,

Disse spørgsmål bliver sværere og sværere,

because when you're in an area, when you're in a community,

for når du bor i et område, når du bor i et samfund,

things like age -- there are clues to whether a population is old or young.

er der fingerpeg om, hvorvidt befolkningen er gammel eller ung.

Just by looking around the area, you can see it.

Bare ved at kigge rundt i området, kan man se det.

Something like homeownership is much more difficult to see,

Noget som det at eje hus er langt vanskeligere at se,

so we revert to our own heuristics,

så vi vender tilbage til vore egen heuristik,

our own biases about how many people we think own their own homes.

vore egne fordomme om, hvor mange vi tror ejer deres eget hjem.

Now the truth is, when we published this quiz,

Nu er sandheden den, at da vi offentliggjorde denne quiz,

the census data that it's based on was already a few years old.

var de data, som det er baseret på, allerede et par år gamle.

We've had online applications that allow you to put in a post code

Vi har haft online-applikationer, der tillader en at skrive et postnummer

and get statistics back for years.

og få statistik fra årevis tilbage.

So in some senses,

Så på en måde,

this was all a little bit old and not necessarily new.

var dette altsammen lidt gammelt og ikke nødvendigvis nyt.

But I was interested to see what reaction we might get

Men jeg var interesseret i at se, hvilken reaktion man kan få

by gamifying the data in the way that we have,

ved at gamificere det

by using animation

ved hjælp af animation

and playing on the fact that people have their own preconceptions.

og lege med det faktum, at folk har deres egne fordomme.

It turns out, the reaction was, um ...

Det viste sig, at reaktionen

was more than I could have hoped for.

var mere end jeg kunne have håbet på.

It was a long-held ambition of mine to bring down a statistics website

Det var en af mine ambitioner at bringe en statistik-hjemmeside i knæ

due to public demand.

grundet offentlig efterspørgsel.

(Laughter)

(Latter)

This URL contains the words "statistics," "gov" and "UK,"

Denne URL indeholder ordene "statistik", "gov" og "UK",

which are three of people's least favorite words in a URL.

som er tre af folks mindst foretrukne ord i en URL.

And the amazing thing about this was that the website came down

Og det fantastiske ved dette var, at hjemmesiden gik i knæ

at quarter to 10 at night,

kvarter i 10 om aftenen,

because people were actually engaging with this data

fordi folk faktisk legede med disse data

of their own free will,

af egen fri vilje,

using their own personal time.

og i deres egen tid.

I was very interested to see

Jeg fandt det meget interessant,

that we got something like a quarter of a million people

at vi fik noget i retning af en kvart million mennesker,

playing the quiz within the space of 48 hours of launching it.

der spillede quizzen gennem 48 timer efter den blev lanceret.

And it sparked an enormous discussion online, on social media,

Og det udløste en enorm diskussion online, på sociale medier,

which was largely dominated

som i vid udstrækning var domineret

by people having fun with their misconceptions,

af folk, der har det sjovt med deres misforståelser,

which is something that I couldn't have hoped for any better,

hvilket er noget, jeg ikke har kunnet håbe på noget bedre

in some respects.

på en måde.

I also liked the fact that people started sending it to politicians.

Jeg kunne godt lide, at folk begyndte at sende det til politikerne.

How well do you know the area you claim to represent?

Hvor godt kender du området, du hævder at repræsentere?

(Laughter)

(Latter)

And then just to finish,

Og så bare til afslutning,

going back to the two kinds of people,

lad os gå tilbage til de to slags mennesker.

I thought it would be really interesting to see

Jeg syntes det kunne være interessant at se,

how people who are good with numbers would do on this quiz.

hvordan mennesker, der er gode til tal ville klare quizzen.

The national statistician of England and Wales, John Pullinger,

Den nationale statistiker i England og Wales, John Pullinger,

you would expect he would be pretty good.

skulle man tro ville klare sig ret godt.

He got 44 for his own area.

Han fik 44 for sit eget område.

(Laughter)

(Latter)

Jeremy Paxman -- admittedly, after a glass of wine -- 36.

Jeremy Paxman -- ganske vist efter et glas vin -- 36.

Even worse.

Endnu værre.

It just shows you that the numbers can inspire us all.

Det viser bare, at tallene kan inspirere os alle.

They can surprise us all.

De kan overraske os alle.

So very often, we talk about statistics

Ofte taler vi om statistik,

as being the science of uncertainty.

som værende videnskaben om usikkerhed.

My parting thought for today is:

Mine afskedsord i dag er:

actually, statistics is the science of us.

I virkeligheden er statistik videnskaben om os.

And that's why we should be fascinated by numbers.

Og det er derfor, vi bør være fascineret af tal.

Thank you very much.

Mange tak.

(Applause)

(Bifald)