12:28
TED@IBM

Susan Etlinger: What do we do with all this big data?

Filmed:

Does a set of data make you feel more comfortable? More successful? Then your interpretation of it is likely wrong. In a surprisingly moving talk, Susan Etlinger explains why, as we receive more and more data, we need to deepen our critical thinking skills. Because it's hard to move beyond counting things to really understanding them.

- Data analyst
Susan Etlinger promotes the smart, well-considered and ethical use of data. Full bio

Technology has brought us so much:
00:13
the moon landing, the Internet,
00:16
the ability to sequence the human genome.
00:18
But it also taps into a lot of our deepest fears,
00:21
and about 30 years ago,
00:24
the culture critic Neil Postman wrote a book
00:26
called "Amusing Ourselves to Death,"
00:29
which lays this out really brilliantly.
00:31
And here's what he said,
00:34
comparing the dystopian visions
00:35
of George Orwell and Aldous Huxley.
00:38
He said, Orwell feared we would become
00:41
a captive culture.
00:44
Huxley feared we would become a trivial culture.
00:47
Orwell feared the truth would be
00:50
concealed from us,
00:52
and Huxley feared we would be drowned
00:54
in a sea of irrelevance.
00:57
In a nutshell, it's a choice between
00:59
Big Brother watching you
01:01
and you watching Big Brother.
01:04
(Laughter)
01:06
But it doesn't have to be this way.
01:08
We are not passive consumers
of data and technology.
01:10
We shape the role it plays in our lives
01:13
and the way we make meaning from it,
01:16
but to do that,
01:18
we have to pay as much attention to how we think
01:20
as how we code.
01:23
We have to ask questions, and hard questions,
01:25
to move past counting things
01:28
to understanding them.
01:30
We're constantly bombarded with stories
01:33
about how much data there is in the world,
01:35
but when it comes to big data
01:38
and the challenges of interpreting it,
01:39
size isn't everything.
01:42
There's also the speed at which it moves,
01:44
and the many varieties of data types,
01:47
and here are just a few examples:
01:49
images,
01:51
text,
01:53
video,
01:57
audio.
01:59
And what unites this disparate types of data
02:01
is that they're created by people
02:04
and they require context.
02:06
Now, there's a group of data scientists
02:09
out of the University of Illinois-Chicago,
02:12
and they're called the Health Media Collaboratory,
02:14
and they've been working with
the Centers for Disease Control
02:16
to better understand
02:19
how people talk about quitting smoking,
02:21
how they talk about electronic cigarettes,
02:23
and what they can do collectively
02:26
to help them quit.
02:28
The interesting thing is, if you want to understand
02:30
how people talk about smoking,
02:32
first you have to understand
02:34
what they mean when they say "smoking."
02:36
And on Twitter, there are four main categories:
02:39
number one, smoking cigarettes;
02:43
number two, smoking marijuana;
02:46
number three, smoking ribs;
02:48
and number four, smoking hot women.
02:51
(Laughter)
02:55
So then you have to think about, well,
02:58
how do people talk about electronic cigarettes?
03:00
And there are so many different ways
03:02
that people do this, and you can see from the slide
03:04
it's a complex kind of a query.
03:07
And what it reminds us is that
03:09
language is created by people,
03:13
and people are messy and we're complex
03:15
and we use metaphors and slang and jargon
03:17
and we do this 24/7 in many, many languages,
03:20
and then as soon as we figure it out, we change it up.
03:23
So did these ads that the CDC put on,
03:27
these television ads that featured a woman
03:32
with a hole in her throat and that were very graphic
03:34
and very disturbing,
03:36
did they actually have an impact
03:38
on whether people quit?
03:40
And the Health Media Collaboratory
respected the limits of their data,
03:43
but they were able to conclude
03:46
that those advertisements —
and you may have seen them —
03:48
that they had the effect of jolting people
03:51
into a thought process
03:54
that may have an impact on future behavior.
03:56
And what I admire and
appreciate about this project,
03:59
aside from the fact, including the fact
04:03
that it's based on real human need,
04:05
is that it's a fantastic example of courage
04:09
in the face of a sea of irrelevance.
04:12
And so it's not just big data that causes
04:16
challenges of interpretation, because let's face it,
04:19
we human beings have a very rich history
04:22
of taking any amount of data, no matter how small,
04:25
and screwing it up.
04:27
So many years ago, you may remember
04:29
that former President Ronald Reagan
04:33
was very criticized for making a statement
04:35
that facts are stupid things.
04:37
And it was a slip of the tongue, let's be fair.
04:40
He actually meant to quote John Adams' defense
04:43
of British soldiers in the Boston Massacre trials
04:45
that facts are stubborn things.
04:48
But I actually think there's
04:51
a bit of accidental wisdom in what he said,
04:54
because facts are stubborn things,
04:57
but sometimes they're stupid, too.
05:00
I want to tell you a personal story
05:03
about why this matters a lot to me.
05:05
I need to take a breath.
05:08
My son Isaac, when he was two,
05:11
was diagnosed with autism,
05:13
and he was this happy, hilarious,
05:16
loving, affectionate little guy,
05:18
but the metrics on his developmental evaluations,
05:20
which looked at things like
the number of words —
05:23
at that point, none —
05:25
communicative gestures and minimal eye contact,
05:29
put his developmental level
05:33
at that of a nine-month-old baby.
05:35
And the diagnosis was factually correct,
05:39
but it didn't tell the whole story.
05:42
And about a year and a half later,
05:45
when he was almost four,
05:46
I found him in front of the computer one day
05:48
running a Google image search on women,
05:51
spelled "w-i-m-e-n."
05:56
And I did what any obsessed parent would do,
06:00
which is immediately started
hitting the "back" button
06:02
to see what else he'd been searching for.
06:04
And they were, in order: men,
06:08
school, bus and computer.
06:10
And I was stunned,
06:17
because we didn't know that he could spell,
06:19
much less read, and so I asked him,
06:21
"Isaac, how did you do this?"
06:23
And he looked at me very seriously and said,
06:25
"Typed in the box."
06:28
He was teaching himself to communicate,
06:31
but we were looking in the wrong place,
06:35
and this is what happens when assessments
06:38
and analytics overvalue one metric —
06:40
in this case, verbal communication —
06:43
and undervalue others, such
as creative problem-solving.
06:45
Communication was hard for Isaac,
06:51
and so he found a workaround
06:53
to find out what he needed to know.
06:55
And when you think about it, it makes a lot of sense,
06:58
because forming a question
07:00
is a really complex process,
07:02
but he could get himself a lot of the way there
07:05
by putting a word in a search box.
07:07
And so this little moment
07:11
had a really profound impact on me
07:14
and our family
07:17
because it helped us change our frame of reference
07:18
for what was going on with him,
07:21
and worry a little bit less and appreciate
07:24
his resourcefulness more.
07:27
Facts are stupid things.
07:29
And they're vulnerable to misuse,
07:32
willful or otherwise.
07:34
I have a friend, Emily Willingham, who's a scientist,
07:36
and she wrote a piece for Forbes not long ago
07:39
entitled "The 10 Weirdest Things
07:42
Ever Linked to Autism."
07:44
It's quite a list.
07:45
The Internet, blamed for everything, right?
07:48
And of course mothers, because.
07:52
And actually, wait, there's more,
07:56
there's a whole bunch in
the "mother" category here.
07:57
And you can see it's a pretty
rich and interesting list.
08:01
I'm a big fan of
08:05
being pregnant near freeways, personally.
08:08
The final one is interesting,
08:11
because the term "refrigerator mother"
08:13
was actually the original hypothesis
08:16
for the cause of autism,
08:19
and that meant somebody
who was cold and unloving.
08:20
And at this point, you might be thinking,
08:23
"Okay, Susan, we get it,
08:24
you can take data, you can
make it mean anything."
08:26
And this is true, it's absolutely true,
08:28
but the challenge is that
08:32
we have this opportunity
08:38
to try to make meaning out of it ourselves,
08:40
because frankly, data doesn't
create meaning. We do.
08:43
So as businesspeople, as consumers,
08:48
as patients, as citizens,
08:51
we have a responsibility, I think,
08:54
to spend more time
08:56
focusing on our critical thinking skills.
08:58
Why?
09:01
Because at this point in our history, as we've heard
09:02
many times over,
09:06
we can process exabytes of data
09:07
at lightning speed,
09:09
and we have the potential to make bad decisions
09:11
far more quickly, efficiently,
09:15
and with far greater impact than we did in the past.
09:17
Great, right?
09:22
And so what we need to do instead
09:23
is spend a little bit more time
09:26
on things like the humanities
09:29
and sociology, and the social sciences,
09:31
rhetoric, philosophy, ethics,
09:35
because they give us context that is so important
09:37
for big data, and because
09:40
they help us become better critical thinkers.
09:42
Because after all, if I can spot
09:45
a problem in an argument, it doesn't much matter
09:49
whether it's expressed in words or in numbers.
09:52
And this means
09:54
teaching ourselves to find
those confirmation biases
09:57
and false correlations
10:02
and being able to spot a naked emotional appeal
10:03
from 30 yards,
10:05
because something that happens after something
10:07
doesn't mean it happened
because of it, necessarily,
10:10
and if you'll let me geek out on you for a second,
10:13
the Romans called this
"post hoc ergo propter hoc,"
10:15
after which therefore because of which.
10:19
And it means questioning
disciplines like demographics.
10:22
Why? Because they're based on assumptions
10:26
about who we all are based on our gender
10:29
and our age and where we live
10:31
as opposed to data on what
we actually think and do.
10:32
And since we have this data,
10:36
we need to treat it with appropriate privacy controls
10:38
and consumer opt-in,
10:41
and beyond that, we need to be clear
10:44
about our hypotheses,
10:47
the methodologies that we use,
10:49
and our confidence in the result.
10:52
As my high school algebra teacher used to say,
10:55
show your math,
10:57
because if I don't know what steps you took,
10:59
I don't know what steps you didn't take,
11:02
and if I don't know what questions you asked,
11:04
I don't know what questions you didn't ask.
11:07
And it means asking ourselves, really,
11:10
the hardest question of all:
11:11
Did the data really show us this,
11:13
or does the result make us feel
11:16
more successful and more comfortable?
11:19
So the Health Media Collaboratory,
11:23
at the end of their project, they were able
11:25
to find that 87 percent of tweets
11:27
about those very graphic and disturbing
11:30
anti-smoking ads expressed fear,
11:32
but did they conclude
11:36
that they actually made people stop smoking?
11:38
No. It's science, not magic.
11:41
So if we are to unlock
11:44
the power of data,
11:47
we don't have to go blindly into
11:50
Orwell's vision of a totalitarian future,
11:54
or Huxley's vision of a trivial one,
11:57
or some horrible cocktail of both.
12:00
What we have to do
12:03
is treat critical thinking with respect
12:05
and be inspired by examples
12:08
like the Health Media Collaboratory,
12:10
and as they say in the superhero movies,
12:13
let's use our powers for good.
12:15
Thank you.
12:17
(Applause)
12:19

▲Back to top

About the Speaker:

Susan Etlinger - Data analyst
Susan Etlinger promotes the smart, well-considered and ethical use of data.

Why you should listen

Susan Etlinger is an industry analyst with Altimeter Group, where she focuses on data and analytics. She conducts independent research and has authored two intriguing reports: “The Social Media ROI Cookbook” and “A Framework for Social Analytics.” She also advises global clients on how to work measurement into their organizational structure and how to extract insights from the social web which can lead to tangible actions. In addition, she works with technology innovators to help them refine their roadmaps and strategies. 

Etlinger is on the board of The Big Boulder Initiative, an industry organization dedicated to promoting the successful and ethical use of social data. She is regularly interviewed and asked to speak on data strategy and best practices, and has been quoted in media outlets like The Wall Street Journal, The New York Times, and the BBC.

More profile about the speaker
Susan Etlinger | Speaker | TED.com