Tip:
Highlight text to annotate it
X
Hi, this is Bart Poulson and I'm going to be going over, uh, doing some elementary graphics
in StatCrunch for the students in my statistics class. What I'm going to do today is a simple
bar chart in StatCrunch. Ok, the first thing I'm going to do is I'm going to sign into
the program. And then I'm going to use a data set that we've used a few times before. Because
it's a data set that exists on StatCrunch, I'm going to come over here to Explore. And
then where it says "Explore the StatCrunch site," I come down here to Data. Now, I'm
going to need to search for it. The one I'm looking for is called surveyf08.xls. Because
I've used it before, it popped up the first letter that I entered. I press Search. And
it brings up this one with the picture of the barber shop. Uh, and right here. There
is another copy of it right here, but I'll use the original one. So I click on that.
Um… it takes a moment to load. OK. What I'm going to do in here is I'm going to do
a bar chart for gender. See how many people, um, there are in the data set. Actually, there
is a total of 76 people. Ok, to do a bar chart, the first thing I do is I come over here to
Graphics. Click on that. And actually StatCrunch calls it a Bar Plot. I'm going to use it "with
data" because I have a complete data set of one row for each person. If I— I could
also do it "with summary." Um, I could show that in a different situation. But right now
I'm just going to plic—pick: Graphics, Bar plot, "with data." And a dialogue box
will come up. It has the variables listed on the left. And I can click the ones that
I want. Now I can click more than one if I hold down the Control key. Um, However, I,
that causes problems with the, uh, with the titles so I'm not going to do that right now.
I'm not going to do a select statement, for instance, where age is less than this or height
is more than that, and I'm not going to group it by any other variables. However, I am going
to click on "Next." Now, if I wanted to, I could get something called a relative frequency
chart. That's that one. The idea here is that would give me the proportion of people who
were male or female which could be important in certain situations. This one would give
the frequencies. The bars look exactly the same. The only difference is the numbers that're
listed up the side. If I had several categories, for instance, the state that people were from,
and there were a bunch of states that had just one or two people, I could have it group
them into an "Other" category. There's a lot of analyses where that would be a good idea
where, for instance, if there's less than 10% of people in any one category then we
should lump them together. This can change the order of the bars. I'm not going to deal
with that right now. I'm gonna come down here and I'm gonna hit "Next" again. Ok. Something
that you should always do is put descriptive labels and titles on the charts. What I'm
going to do here is I'm gonna put a title for the top. [Typing] Gender of 76 Participants
in, uh, Student Survey on StatCrunch. OK. Actually, I don't think I need the 76 part
so I'm gonna come back and delete that. Alright. I think I can leave the rest of it the way
that it is. Uh, The Graph Layout I would use if I were to have more than one bar chart
or more than one box plot or something I wanted to be on the same page. I'd usually put more
than one row so that they could be stacked above each other. I'm not going to deal with
that here. So, I click "Create Graph" to make the chart. Alright. There's my resulting chart.
It has two bars here: One for women and another one for men. And this is the frequency up
this side. You can tell there's about 54, 55 women and there are exactly 20 men. And
I have the title right here across the top. Now there is one thing; and that is down here
I have a label for the axis across the bottom. It says Gender right there, and Female and
Male right here. Now truthfully, I think that's redundant because I say Gender up here in
the top and because everybody knows that Male and Female refers to gender. So, I'm actually
going to get rid of this one down here. The way to do that, at least the way that I know
of, is I come up to Options, and I edit the chart. Now, I'm going to back up a little
bit. This is the same set of dialogue boxes I had at first. Now I'm going to come up here
to the x-axis label because that's the one across the bottom. Now I think perhaps the
only way to get rid of this is I have to replace the text, but if I just replace it with one
space. So I just hit the space bar once. Is it blank? I press Create Graph and it goes
away. That makes a slightly cleaner chart and I think it works better that way. Um,
what I need to do now is I need to come up here to Options and I need to export it to
save the chart. So I'm going to Export to my Results. [Click.] And I will give it a
title here. I'll say "Bar plot of Gender." And I'll hit Export. And there it is, right
over there. And we'll talk a little later about how to export that into a Word Processing
document or presentation. But, that will do it for now.