https:/.../2020-06-29_gov355m_14.g_conclusion.mp4

Alright, this is the last segment
for the module but also the

last segment for the course.

We have gone through a tremendous
amount in what feels like a very short

period of time.

This is all recorded in a summer session
so you know that everyone has been

working overtime to go to learn a lot
of this and for those without some

data or statistical background.

There's a lot of material here
so if you're here and you're

watching this still, great job
with everything and recognize

the incredible progress.

If you're at the point where
you can even do some of this

basic webscraping, text analysis
and mapping, you are in a much

better position than so many
other students especially that

come out of the liberal arts in terms
of [inaudible] and hard skills.

Now, the hard skills, I just wanted
to comment a few things but you

know, the hard skills like this
are incredibly important but

it's not to say that the broader
liberal arts education is not --

as someone who teaches courses
that are very much about coming

at things with this broad kind of
liberal arts mindset, very analytical

and critical synthetic, that type
of thing, I'm a huge believer in

all of that but I also think that
getting these hard skills is

incredibly important and it's
not really kind of a 'you do this

or you do that' approach,
it's really one in which you should

develop that broad, strong liberal
arts mindset by getting some of

these hard skills will hopefully
help you to apply that and

apply it in a more sophisticated way,
maybe not more sophisticated, in

an asophisticated way or one
sophisticated way, there's

lots of sophisticated ways to
get at things.

This is one sophisticated way
to help enhance your liberal

arts education and be able to do
a lot of things hopefully that you

couldn't have done otherwise
and that will hopefully set you

apart from someone else that
comes out with a strong liberal

arts education but you know
just doesn't have the ability to

take on a lot of these difficult
political, social, and economic

challenges of the world armed
with data and some ability to

program and manipulate data,
clean data, get it in some format

that you can then analyze it
either through basic statistics and/or

visualization, mapping, text analysis
and so forth and so our hope here

and what we set out to do originally
was where we started on the opening

day was to say this course is not
about making you a computer

science expert.

We can't replace computer science
department, not going to try, the

goal here was to hopefully
introduce a bunch of substantive

challenges in the world, again
political, social, economic challenges

or problems at least in brief
and then use those as examples

for a lot of the r and excel skills
that you learned in the course.

So you came for data science,
and a lot of it has to focus on

the code and the data and so
we hope that the political, social

economic side didn't get lost in
all of that.

We certainly tried to keep it there
but again, we kind of had to delve

into data early and often and
deeply in order to develop enough

skills so we could competently
engage with these substantive

issues and so that was at least
the goal and the hope with trying

to in each session work with
real data, work with real problems

but then in the process try to
teach you enough code that

you can get around various things.

So, to get really not just proficient
but to excel in data science, it's

going to require a lot more work
as with most courses you take,

probably all courses you take,
you've got to continue to apply

things in the future and so I'm not
going to give you a big pep talk here

but what I will say is just to the extent
that you think that a lot of this can

be useful in the future, I just
strongly encourage you to

keep applying this stuff early
enough and to give you some

context on this, when I go, I've
been doing this stuff for quite

a long time now, I think I revealed
my age on webscrape when I entered

my birthday to see what the billboard
hot 100 said.

At any rate, I've been in this for
15-20 years now and when I

do a statistical or when I do any
sort of like data work, I'm always

looking up stuff still and told you
here that when I was trying to

parse that time stamp on the twitter
data, it took me a little bit of time

and so someday you might get
to the point where you never have to

look up anything but I've met very
few people in that situation.

I've got to look it up a lot.

I know Mike Denly does, others
do, the key though is to get enough

of the understanding of the
infrastructure, the infrastructure

in terms of like r and our studio,
the way it approaches, conceptualizes

how it deals with data, which is
thinking about things like different

types of data, their classes, their types,
all of that, how many [inaudible] stuff

kind of like etched in is really important
like it, that you deal with the character

variable different that you deal
with the numeric, I mean those

things, you'll get to the point
where that stuff is super basic

in time and you just always
kind of know when you can

work through that stuff very
quickly, and then undersanding

enough the basics on each topic
that you can jump in and do a lot

of it on your own but then you
know, you know when you need

to look when you need a little bit
more help with something and

that's where most people are at
that point and so should you go

on and apply this, guaranteed
you're going to spend time googling,

you're going to spend time on
various different web clearing

houses and otherwise, but that's
part of the game to the extent

that some of this could be useful
in the future, just encourage you

to keep going on some of this stuff.

To the extent that you don't end up
doing a lot of this and you just sort

of say that it's useful to have
some basic literacy in terms

of like especially if you're in
an organization where maybe

you're not the data scientist
in your organization but

someone else is, if you have
enough literacy of this that

you can engage competently
with that person, goes so far,

it absolutely goes so far
in terms of helping you

be able to be the type of person
on the team that can both engage

with, whatever the data scientists,
or the non-data scientists or whatever

like being that person that can engage
fully in all of that can be super useful

as well so just having some of that basic
literacy and trying to keep up some of

that you know can be important
in various ways so at the end of

the day though, the goal was to
have some of these substantive

problems that shape things teach
you enough to be able to learn

some new things about those
substantive problems and

at the end of the day, hopefully
have a few tips and tricks up

your sleeve like being able to do
some maps which is like very

cool, maybe basic webscraping and text
analysis, which can often be a really nice

way to go above and beyond what someone
could just standardly do with the

[inaudible] file in Excel or something.

And so hopefully, you know have some
higher order skills to go with this as

well, so finishing up then, and moving
to the final exam, just a note on this

which is just plan for, understand
and you should understand this

by now of course but the material
course is fairly cumulative, right,

it's tough to do a text analysis
without doing material from

module 7, 8, and 9.

We use [inaudible] here in
text analysis that was brought

up in module 7 and like the spread
and gather which come up earlier

in modules 8 and 9 or filtering
and mutating, stuff like that

so when it comes to the final
exam, just a final note that

it's somewhat cumulative
but we'll also have a weighted

towards like the third, the last
third of this course, so you're

going to want to give special
emphasis to the last third

but remember, again, a lot that
earlier material is important to

pay attention to and be prepared to
work on and illustrate for the final exam

so with that we'll go ahead and close
and just say, just let us know if any

questions, concerns, challenges
and then just a note to please

keep in touch, if not in office hours
in the next little bit in coming years

especially if some of the data
science ends up coming in handy

for future courses or jobs or
internships.

I would absolutely love to hear about
it, if it turns out that it's absolutely

not useful in any way, shape, or form,
we'd love to hear about that too,

honestly truly, it would be really
nice to just get feedback down the

road of what you guys find useful
and not useful, and so that we can

make sure that this course stays
optimally beneficial for students

in the future.

So with that, thanks everyone
for being in the class and we

will be in touch.