Statistics is About People, Even if You Can’t See the Tears

  

You will find Video 24: Statistics is About People, Even if You Can’t See the Tears by navigating to the MSL Tools for Success link under Course Home.

Transcript 

00:04>> ANDREW VICKERS: Do you consider yourself a person or

00:05>> a statistic?

00:07>> SPEAKER 1: Seeing as I know nothing about

00:08>> statistics, I’m a person.

00:10>> ANDREW VICKERS: Do you consider yourself a person or

00:11>> a statistic?

00:12>> SPEAKER 2: Well, a person, certainly.

00:14>> ANDREW VICKERS: Oh yeah, well do you think anyone might

00:15>> consider you a statistic?

00:16>> SPEAKER 2: Sure.

00:17>> ANDREW VICKERS: Yeah?

00:18>> Like who?

00:20>> SPEAKER 2: I don’t know, anybody that

00:21>> wants to sell me anything.

00:22>> ANDREW VICKERS: Would you consider yourself a person or

00:25>> a statistic?

00:25>> SPEAKER 3: A stadeestic?

00:26>> ANDREW VICKERS: A statistic.

00:28>> SPEAKER 3: A statistic?

00:29>> ANDREW VICKERS: Yeah.

00:30>> SPEAKER 3: About a people?

00:31>> ANDREW VICKERS: Do you consider yourself a person or

00:34>> a statistic?

00:36>> SPEAKER 4: Well, I like to think of myself as a person,

00:38>> but I feel like I’m often referred to as a statistic.

00:41>> ANDREW VICKERS: Of course, no one wants to be a statistic.

00:44>> It sounds so cold, so devoid of life, and somewhat harsh.

00:49>> You sometimes read something like on the basis of where you

00:52>> live, we can tell that 75% of you are going to like a

00:55>> certain movie.

00:56>> And it sounds like you’re just some kind of robot, as if you

00:59>> don’t have any real feelings about movies.

01:02>> Now, I work in a cancer hospital.

01:04>> And unfortunately, every single day doctors at my

01:07>> hospital have to say something like for every 100 patients

01:11>> with a cancer like yours, only 10 are going to be alive in

01:15>> five years’ time.

01:17>> And of course the patients think, I’m a person not a

01:20>> statistic, that doesn’t apply to me.

01:23>> But let’s imagine that the doctor said OK, you’ve got

01:26>> this 10% chance.

01:28>> But if you have surgery, you have a 90% probability of

01:32>> being alive in five years’ time?

01:35>> Now in that case, who’s going to say hey, I’m a

01:38>> person not a statistic.

01:40>> That doesn’t apply to me, I’m not going to have the surgery.

01:43>> The point is, of course you’re not a statistic.

01:46>> And of course numbers don’t represent

01:48>> everything about you.

01:49>> It just so happens though that sometimes you’re better off

01:52>> acting as if you’re just a statistic.

01:56>> A few years ago I was giving a talk at this

01:58>> small town in the Midwest.

02:00>> And some people I met there were asking me, what’s it like

02:03>> to live in a dangerous big city like Brooklyn?

02:06>> Hey Eric, what’s up?

02:07>> ERIC: Hey, what’s up?

02:08>> ANDREW VICKERS: Good to see you.

02:09>> ERIC: How are you?

02:10>> ANDREW VICKERS: Alright.

02:10>> ERIC: Good.

02:11>> ANDREW VICKERS: Yeah, that’s my friend Eric.

02:11>> Our kids go to the same school.

02:13>> The point is that appearances can be deceiving and sometimes

02:17>> you’ve got to look up the numbers.

02:18>> So it turns out that the small Midwestern town has almost

02:22>> exactly the same population as my neighborhood in Brooklyn.

02:25>> And when I went to the numbers, I found out that it

02:28>> was actually their crime rate that was higher, way higher.

02:31>> Four times as many crimes in that small Midwestern town as

02:35>> in the dangerous big city of Brooklyn.

02:37>> In the 1800s, a British doctor called John Snow did a very

02:41>> simple statistical analysis to show that cholera cases were

02:45>> clustered around one particular water pump in the

02:49>> city of London.

02:50>> Now, he had the handle of the pump removed and the cholera

02:54>> epidemic abated.

02:55>> It’s because of John Snow’s work that we now know that

02:59>> germs are associated with disease, and why we have good

03:02>> sewage systems and clean water.

03:04>> And before John Snow, about one in five children would not

03:08>> live to see their 10th birthday.

03:10>> The number today is closer to one in 100.

03:14>> So that means that in this playground, probably four or

03:18>> five children owe their lives to being treated as a

03:21>> statistic not a person.

03:23>> But this is an issue that cuts both ways.

03:26>> What I have here is what’s called a survival graph, and I

03:30>> use those the whole time in my work.

03:32>> So you start here with a group of cancer patients who are all

03:35>> alive, and they die over time.

03:38>> And the proportion alive, the probability goes down.

03:42>> Now when I discuss these type of graphs with my statistician

03:47>> colleagues, we talk about all these esoteric statistical

03:50>> issues such as the proportional hazards

03:52>> assumption, or competing risks or bootstrapping.

03:55>> And it’s very easy to forget that each one of those steps

03:59>> down is when a patient died.

04:01>> And that’s somebody’s son or daughter.

04:04>> Perhaps also a mother or a father, or a husband or wife.

04:08>> So just as people have to remember that they are also

04:11>> statistics, statisticians like me have to remember that

04:15>> statistics are also people.

04:17>> The point is, we want to live our lives better, and to do

04:20>> that we have to make good decisions.

04:22>> And it turns out that sometimes analyzing numerical

04:26>> data with statistics can help us to do that.

The video makes the point that though nobody wants to think that they are a statistic, acting as if you are a statistic can help you make better decisions. That said, statisticians should never forget that the numbers they analyze correspond to real people, who have friends, relatives, and stories to tell.
 

Respond to one of the following questions in your initial post:

· The reason why so many of us now live long, healthy lives is due to statistical analysis of health data. What other statistical analyses have had a large impact on how we live our lives?

Your initial post should be 250 words in length

Respond to at least two of your classmates who choose different questions;

· Why do you think people often feel that “the statistics don’t apply to me”?

· Why do you think statistics often has a bad name? The reason why so many of us now live long, healthy lives is due to statistical analysis of health data. What other statistical analyses have had a large impact on how we live our lives?