Jake Porway - Data Scientists at Work - page 320

Porway: I think science is going to be democratized. Here we've been talking
about data and data science, but I think what we're really talking about is a
new age of science. It's no longer just the experts in academia or government
labs who have the technology or the capacity to draw in information and
create conclusions that they release in papers, findings, and actions. We're all
going to become scientists.
Our kids are going to laugh at us about the old days. “How did you pick your
plumber? You just read about him in the yellow pages? You just went down the
list calling people? What? Why didn't you use any data or evidence to make
your decision?” Or, “How did you guys eat? Oh, you just followed some diet?
Didn't you know your endocrine levels? Were you guys doing something like
an experiment on yourselves where you just followed your gut?” I actually
think that's absolutely coming and I think that's really exciting. That's defi-
nitely something that I don't know if people really grasp yet. The new age of
science—of citizen science—in which we're all empowered with the data col-
lection and analysis tools to study our world and learn about ourselves.
Gutierrez: What does it take to do great data science work?
Porway: I think a strong statistical background is a prerequisite, because you
need to know what you're doing, and understand the guts of the model you
build. Additionally, my statistics program also taught a lot about ethics, which
is something that we think a lot about at DataKind. You always want to think
about how your work is going to be applied.You can give anybody an algorithm.
You can give someone a model for using stop-and-frisk data, where the police
are going to make arrests, but why and to what end? It's really like building any
new technology. You've got to think about the risks as well as the benefits and
really weigh that because you are responsible for what you create.
No matter where you come from, as long as you understand the tools that
you're using to draw conclusions, that is the best thing you can do. We are
all scientists now, and I'm not just talking about designing products. We are
all drawing conclusions about the world we live in. That's what statistics is—
collecting data to prove a hypothesis or to create a model of the way the
world works. If you just trust the results of that model blindly, that's danger-
ous because that's your interpretation of the world, and as flawed as it is, your
understanding is how flawed the result is going to be.
In short, learn statistics and be thoughtful.
Gutierrez: Should data scientists have an ethical responsibility for their
work?
Porway: Data scientists should and do have a very strong ethical responsibil-
ity in what they do. As a data scientist, you may find yourself running a version
of what are essentially psychological experiments on users. That's something