WEBVTT

00:00:04.500 --> 00:00:08.280
So we just made this
plot in ggplot2.

00:00:08.280 --> 00:00:10.070
When we compare it
back to the pie graph,

00:00:10.070 --> 00:00:14.050
the first thing I notice is that
now all the data is visible.

00:00:14.050 --> 00:00:15.840
We haven't lost
the small regions

00:00:15.840 --> 00:00:17.840
and we can read out
the exact share that

00:00:17.840 --> 00:00:20.940
comes from Africa, Oceania,
and the unknown or stateless

00:00:20.940 --> 00:00:23.400
column.

00:00:23.400 --> 00:00:26.660
I believe it is also easier
to compare the relative sizes

00:00:26.660 --> 00:00:29.530
of each region
because they're all

00:00:29.530 --> 00:00:31.520
put side by side
on a similar scale.

00:00:31.520 --> 00:00:33.150
There's no tricks,
or three dimensions,

00:00:33.150 --> 00:00:37.350
or colors to create
a perception issue.

00:00:37.350 --> 00:00:39.520
But, I will say that
something to consider

00:00:39.520 --> 00:00:43.020
is, for some people
and some applications,

00:00:43.020 --> 00:00:46.190
being not as visually
exciting is a negative.

00:00:46.190 --> 00:00:49.620
This plot, while very
readable and correct,

00:00:49.620 --> 00:00:51.620
is certainly a little bit dull.

00:00:51.620 --> 00:00:54.000
In some applications, this is
an important consideration.

00:00:56.790 --> 00:00:59.220
Now, wouldn't it be
interesting if we

00:00:59.220 --> 00:01:02.130
could plot this
data on a world map?

00:01:02.130 --> 00:01:04.000
It would be possible,
but a bit tedious

00:01:04.000 --> 00:01:05.950
to create because
we need to determine

00:01:05.950 --> 00:01:09.000
which country lies
in which region.

00:01:09.000 --> 00:01:11.530
Shading all countries in
a region of the same color

00:01:11.530 --> 00:01:13.560
might be misleading though.

00:01:13.560 --> 00:01:16.039
For example, countries
in Latin America

00:01:16.039 --> 00:01:19.440
will send students at
different rates, naturally.

00:01:19.440 --> 00:01:21.270
But, if we color them
all the same color,

00:01:21.270 --> 00:01:24.810
it kind of gives a
false impression.

00:01:24.810 --> 00:01:26.600
As it turns out, we
actually have access

00:01:26.600 --> 00:01:28.780
to per country data.

00:01:28.780 --> 00:01:30.560
So we will plot
that on a world map

00:01:30.560 --> 00:01:33.300
instead and see if it is an
effective way of communicating

00:01:33.300 --> 00:01:36.100
where students come from to MIT.