Tip:
Highlight text to annotate it
X
(male narrator) So sometimes we need to compare two different groups of data.
And so, for example, this data comes from a task
in which the goal is to move a computer mouse,
uh...on the...to-to, uh...to a target on the screen
as fast as possible,
so you have to move the mouse and click on a target.
Uh...and it was done twice,
uh...with a small rectangle and a large rectangle,
so two different targets.
Uh...and we've collected frequency data,
uh...and we've created class intervals,
um...each a width a 100 milliseconds.
And so this is our data.
Uh...and so to create a graphical representation,
one option would be to create a...histograms
of each of these data on top of each other.
This is called a "comparative histogram."
Uh...and it would look something like this.
Now this particular graph is done as a bar graph,
uh...where the, uh... the class intervals
are the, uh...bar labels,
and-and that's mainly because of the limitations
of the software I was using.
But you can see that the-the lighter blue here
is representing the times for the small target,
and the darker reddish color or purple color
is showing the times for the... for the large target.
And we can start analyzing the-the-the behaviors here
using a graph like this.
For example, we can tell that the large target scores seem
to sort of, on average, be smaller,
and-and that they're... they tend to be more clustered,
whereas the small target scores tend to be more...
appear to be more spread out.
Now this is still not a great representation of the data,
and so some... another option here
would be something called a-a "frequency polygram."
Uh...sorry, polygon... frequency polygon.
Uh...and the idea is basically the same,
except, uh...we put a dot at each interval--
uh...basically at the same height
that the histogram, uh...would be--
Uh...and then we sort of connect the dots.
Now this is a little misleading,
'cause there's... so we're not suggesting
that this actually, you know, increases that way.
Uh...but it gets the idea of...of trend,
uh...perhaps a little bit more
than-than-than the histogram does.
Uh...and this is what that ends up looking like.
Now this again sort of gets that idea
that the large target values are a little smaller,
uh...and a little more compact, whereas the small target scores
are a little, uh...more spread out
and appear to be slightly larger.