WEBVTT

00:00:00.000 --> 00:00:03.060
>> So here we are in SPSS.

00:00:03.060 --> 00:00:04.900
I have created a junk data

00:00:04.900 --> 00:00:07.220
set just to help
illustrate how we can

00:00:07.220 --> 00:00:08.840
compute variables
or compute

00:00:08.840 --> 00:00:09.980
new variables using

00:00:09.980 --> 00:00:12.010
the values from
existing ones.

00:00:12.010 --> 00:00:14.140
This is a pretty
straightforward process.

00:00:14.140 --> 00:00:15.280
It looks a little bit

00:00:15.280 --> 00:00:16.680
different than things
that we've done

00:00:16.680 --> 00:00:18.580
before or other
ways that we're

00:00:18.580 --> 00:00:20.640
going to be using
SPSS in this course.

00:00:20.640 --> 00:00:22.440
So it might look a
little bit different,

00:00:22.440 --> 00:00:23.980
but it is pretty
straightforward and

00:00:23.980 --> 00:00:26.375
pretty easy once you
know how to do this.

00:00:26.375 --> 00:00:27.740
So the first
thing that we'll

00:00:27.740 --> 00:00:29.345
do is go to Transform,

00:00:29.345 --> 00:00:30.680
and then the
first option we

00:00:30.680 --> 00:00:32.120
have there is
compute variable.

00:00:32.120 --> 00:00:34.440
So click on that. We get

00:00:34.440 --> 00:00:36.165
this new window
that pops up.

00:00:36.165 --> 00:00:37.800
The first thing
that we really want

00:00:37.800 --> 00:00:39.340
to look at is in
the top left,

00:00:39.340 --> 00:00:42.580
and that's this target
variable window.

00:00:42.580 --> 00:00:44.130
Basically what
the computer

00:00:44.130 --> 00:00:45.610
is asking you is to create

00:00:45.610 --> 00:00:47.630
a name for the
new variable

00:00:47.630 --> 00:00:49.070
that you're going
to be computing.

00:00:49.070 --> 00:00:50.860
This name needs
to be unique,

00:00:50.860 --> 00:00:53.170
it's one that can't
have spaces in it,

00:00:53.170 --> 00:00:55.315
and it should avoid
special characters.

00:00:55.315 --> 00:00:56.710
But, in this case,

00:00:56.710 --> 00:00:58.630
what I like to
do is include

00:00:58.630 --> 00:01:01.330
some piece of
information in there

00:01:01.330 --> 00:01:03.550
in the naming
convention I use that

00:01:03.550 --> 00:01:04.810
helps me know what this is

00:01:04.810 --> 00:01:06.250
just by looking at it.

00:01:06.250 --> 00:01:07.890
So what we're going
to be doing is

00:01:07.890 --> 00:01:09.650
we're going to be
summing items 1,

00:01:09.650 --> 00:01:10.970
2, 3, 4, 5,

00:01:10.970 --> 00:01:12.740
and 6 together, adding
those together.

00:01:12.740 --> 00:01:14.330
So an easy way
to do that is

00:01:14.330 --> 00:01:17.480
just to put sumofitems.

00:01:17.480 --> 00:01:20.040
That way, when I look at

00:01:20.040 --> 00:01:21.850
this new variable
once it's generated,

00:01:21.850 --> 00:01:23.350
I know that that's
the variable

00:01:23.350 --> 00:01:24.930
that I really want
to be working

00:01:24.930 --> 00:01:26.720
with or running
my tests on

00:01:26.720 --> 00:01:28.030
because that contains

00:01:28.030 --> 00:01:29.865
the information from
the other ones.

00:01:29.865 --> 00:01:32.970
You can click on
Type & Label.

00:01:32.970 --> 00:01:36.390
The label is the more
drawn-out explanation.

00:01:36.390 --> 00:01:37.870
That's one that
you can include

00:01:37.870 --> 00:01:39.390
additional characters or

00:01:39.390 --> 00:01:41.070
spaces and special
characters.

00:01:41.070 --> 00:01:43.150
In our case, I'm
going to put this is

00:01:43.150 --> 00:01:48.975
the sum of items
1 through 6.

00:01:48.975 --> 00:01:50.790
And that way I know that

00:01:50.790 --> 00:01:52.550
that is going to give me

00:01:52.550 --> 00:01:55.150
a more drawn-out
explanation

00:01:55.150 --> 00:01:57.010
of what this
variable includes.

00:01:57.010 --> 00:01:59.325
We know this is
going to be numeric.

00:01:59.325 --> 00:02:01.175
So click Continue.

00:02:01.175 --> 00:02:03.110
The next thing we're
going to look at

00:02:03.110 --> 00:02:05.045
is this numeric
expression box.

00:02:05.045 --> 00:02:06.530
And will actually
come back

00:02:06.530 --> 00:02:07.730
to this in just a minute,

00:02:07.730 --> 00:02:08.390
but this is where you're

00:02:08.390 --> 00:02:09.330
going to be doing most

00:02:09.330 --> 00:02:11.390
of the work when you're
computing a variable.

00:02:11.390 --> 00:02:13.190
What you'll want
to go to next

00:02:13.190 --> 00:02:14.880
is this function
group area.

00:02:14.880 --> 00:02:16.665
So I'm going to
click on All.

00:02:16.665 --> 00:02:18.630
And you notice right
below it we've

00:02:18.630 --> 00:02:20.580
got these functions
and special variables,

00:02:20.580 --> 00:02:22.815
and this popped up
when I clicked on All.

00:02:22.815 --> 00:02:24.550
This has a whole bunch of

00:02:24.550 --> 00:02:26.200
different pre-existing
functions,

00:02:26.200 --> 00:02:27.640
so you don't need to
reinvent the wheel,

00:02:27.640 --> 00:02:28.870
you can basically use

00:02:28.870 --> 00:02:30.950
these pre-existing
functions to

00:02:30.950 --> 00:02:33.135
help you compute
this new variable.

00:02:33.135 --> 00:02:34.710
I'm going to go
down to sum,

00:02:34.710 --> 00:02:36.250
and these are in
alphabetical order,

00:02:36.250 --> 00:02:37.890
so it's pretty
easy to find.

00:02:37.890 --> 00:02:40.180
Here we have sum, and
when I click on sum,

00:02:40.180 --> 00:02:41.490
you'll notice
that this box on

00:02:41.490 --> 00:02:43.955
the left populated
with some text.

00:02:43.955 --> 00:02:46.005
This box is actually
pretty useful.

00:02:46.005 --> 00:02:47.650
So the first thing
it shows you is

00:02:47.650 --> 00:02:48.570
an example of what

00:02:48.570 --> 00:02:50.210
this expression
is going to be.

00:02:50.210 --> 00:02:51.540
And our expression
sounds like

00:02:51.540 --> 00:02:52.635
a really fancy term.

00:02:52.635 --> 00:02:54.050
If you've ever
used a formula

00:02:54.050 --> 00:02:55.670
in Excel, it's
basically that.

00:02:55.670 --> 00:02:56.990
It's telling the computer

00:02:56.990 --> 00:02:58.465
what it needs to do.

00:02:58.465 --> 00:03:00.750
In this case,
just like Excel,

00:03:00.750 --> 00:03:02.030
we've got the word sum,

00:03:02.030 --> 00:03:03.570
which tells the computer
it's going to be

00:03:03.570 --> 00:03:05.365
summing different
values together,

00:03:05.365 --> 00:03:06.630
and then in parentheses,

00:03:06.630 --> 00:03:07.570
immediately after that,

00:03:07.570 --> 00:03:09.070
we have different
variables

00:03:09.070 --> 00:03:10.705
that are separated
by a comma,

00:03:10.705 --> 00:03:11.930
and that's what tells the

00:03:11.930 --> 00:03:13.390
computer to add x, y,

00:03:13.390 --> 00:03:15.150
and z things
together to create

00:03:15.150 --> 00:03:17.800
a new variable or
compute a new variable.

00:03:17.800 --> 00:03:18.870
This also provides

00:03:18.870 --> 00:03:20.750
additional extra
content here,

00:03:20.750 --> 00:03:22.190
but we don't need to
worry about that.

00:03:22.190 --> 00:03:23.930
I found this is
helpful if you

00:03:23.930 --> 00:03:26.000
are stuck or something
isn't working right.

00:03:26.000 --> 00:03:27.930
So we've got sum selected,

00:03:27.930 --> 00:03:29.780
we're going to be
summing the items of 1,

00:03:29.780 --> 00:03:31.760
2, 3, 4, 5, and
6 together.

00:03:31.760 --> 00:03:33.055
We've got that selected.

00:03:33.055 --> 00:03:34.110
So the next thing you do

00:03:34.110 --> 00:03:35.190
is click this up arrow,

00:03:35.190 --> 00:03:36.270
which basically takes

00:03:36.270 --> 00:03:38.030
that pre-existing function

00:03:38.030 --> 00:03:39.720
and it kicks it up
to this numeric

00:03:39.720 --> 00:03:41.550
and expression box.

00:03:41.550 --> 00:03:43.870
The next thing that
we're going to do,

00:03:43.870 --> 00:03:44.710
we've already told

00:03:44.710 --> 00:03:45.810
the computer
that we're going

00:03:45.810 --> 00:03:47.740
to sum different
variables together,

00:03:47.740 --> 00:03:48.910
we need to tell it which

00:03:48.910 --> 00:03:50.480
variables to sum together.

00:03:50.480 --> 00:03:52.790
So the easy way
to do this is to

00:03:52.790 --> 00:03:54.110
click on the first

00:03:54.110 --> 00:03:55.445
item that you
want to include,

00:03:55.445 --> 00:03:57.850
and you'll notice
that this arrow

00:03:57.850 --> 00:03:59.290
switch sides from pointing

00:03:59.290 --> 00:04:00.785
to the left to
pointing to the right,

00:04:00.785 --> 00:04:02.060
click over 1,

00:04:02.060 --> 00:04:04.350
and we've already
moved over item 1.

00:04:04.350 --> 00:04:07.020
What you need to be
careful of is that you

00:04:07.020 --> 00:04:09.640
always have a comma
after each variable,

00:04:09.640 --> 00:04:11.085
except for the last one.

00:04:11.085 --> 00:04:12.600
So a comma is
already here,

00:04:12.600 --> 00:04:14.815
but what I usually do
is I just hit a comma,

00:04:14.815 --> 00:04:16.675
I go to item 2,

00:04:16.675 --> 00:04:19.010
kick that over, comma,

00:04:19.010 --> 00:04:21.030
item 3, kick that over,

00:04:21.030 --> 00:04:23.550
comma, item 4,
kick that over,

00:04:23.550 --> 00:04:25.260
comma, item 5,

00:04:25.260 --> 00:04:26.880
kick that over, comma,

00:04:26.880 --> 00:04:28.935
item 6, kick that over.

00:04:28.935 --> 00:04:30.420
Now, we know that item 6

00:04:30.420 --> 00:04:31.920
is the last one we have,

00:04:31.920 --> 00:04:33.690
the last variable
we're including,

00:04:33.690 --> 00:04:35.100
so what we're
going to do is

00:04:35.100 --> 00:04:36.700
just delete out
that question mark,

00:04:36.700 --> 00:04:38.515
we're going to delete
out that last comma,

00:04:38.515 --> 00:04:40.260
and then if we read
through this real quick

00:04:40.260 --> 00:04:40.980
just to make sure we

00:04:40.980 --> 00:04:42.190
don't have any problems,

00:04:42.190 --> 00:04:45.095
we've got creating
sumofitems,

00:04:45.095 --> 00:04:46.880
this is going to
be the sum of

00:04:46.880 --> 00:04:48.435
item 1, 2, 3,

00:04:48.435 --> 00:04:51.280
4, 5, and 6,

00:04:51.280 --> 00:04:54.095
and what we're going
to do now is click OK.

00:04:54.095 --> 00:04:56.620
An output window pops

00:04:56.620 --> 00:04:57.720
up and it basically tells

00:04:57.720 --> 00:04:58.860
you what the computer did.

00:04:58.860 --> 00:05:00.500
It computed the
sum of items,

00:05:00.500 --> 00:05:01.830
which was the
sum of item 1,

00:05:01.830 --> 00:05:03.315
2, 3, 4, 5, and 6,

00:05:03.315 --> 00:05:05.110
it gives us the
variable label,

00:05:05.110 --> 00:05:08.620
it gives us that
descriptive text I put in.

00:05:08.620 --> 00:05:10.080
I'm just going to
minimize this window,

00:05:10.080 --> 00:05:11.280
and if we look back here,

00:05:11.280 --> 00:05:12.200
you might notice that

00:05:12.200 --> 00:05:13.620
there's this new column,

00:05:13.620 --> 00:05:15.510
this new variable,
that sumofitems,

00:05:15.510 --> 00:05:16.935
that wasn't there
when we started,

00:05:16.935 --> 00:05:19.155
and that's because
SPSS computed that.

00:05:19.155 --> 00:05:20.780
And if we add
these together,

00:05:20.780 --> 00:05:22.235
if you add each row,

00:05:22.235 --> 00:05:24.320
we've got 1, 2, 1, 3, 5,

00:05:24.320 --> 00:05:26.565
1, that adds to 13,

00:05:26.565 --> 00:05:28.005
5, 1, 2,

00:05:28.005 --> 00:05:29.940
3, 5, 1 adds to 17.

00:05:29.940 --> 00:05:31.540
So it's done that
math for us.

00:05:31.540 --> 00:05:33.425
We didn't have to
manually do it.

00:05:33.425 --> 00:05:35.720
I know, yes, it'd
be pretty easy to

00:05:35.720 --> 00:05:38.280
manually do this
for 10 cases,

00:05:38.280 --> 00:05:41.060
but if you have a dataset
that has 500 cases,

00:05:41.060 --> 00:05:43.000
that would be very
time-consuming to do.

00:05:43.000 --> 00:05:45.460
And you could see
that this function of

00:05:45.460 --> 00:05:47.360
computing this
variable can

00:05:47.360 --> 00:05:48.460
save you a lot of time.

00:05:48.460 --> 00:05:49.820
It's also more accurate.

00:05:49.820 --> 00:05:51.385
It's going to
take any error

00:05:51.385 --> 00:05:53.790
out of the equation.

00:05:53.790 --> 00:05:54.940
And we know that often

00:05:54.940 --> 00:05:56.120
when we do things by hand,

00:05:56.120 --> 00:05:57.900
we can introduce
error into things,

00:05:57.900 --> 00:05:59.550
so we want to avoid that.

00:05:59.550 --> 00:06:01.845
That's one way to
computer a variable.

00:06:01.845 --> 00:06:04.760
I'm going to show
you a different type

00:06:04.760 --> 00:06:06.160
of variable that
we can compute.

00:06:06.160 --> 00:06:07.845
I'm going to go
back to Transform,

00:06:07.845 --> 00:06:09.570
I'm going to go back
to Compute Variable.

00:06:09.570 --> 00:06:11.420
I'm actually going
to clear this out.

00:06:11.420 --> 00:06:12.860
We're done with
someofitems,

00:06:12.860 --> 00:06:13.840
we don't need to
worry about it,

00:06:13.840 --> 00:06:15.280
we're not going to use
the same expression,

00:06:15.280 --> 00:06:16.140
so I'm going to click

00:06:16.140 --> 00:06:17.835
Reset in the bottom here.

00:06:17.835 --> 00:06:20.240
So what I want
to do is create

00:06:20.240 --> 00:06:22.540
the mean of the different
items that we have.

00:06:22.540 --> 00:06:24.000
Sometimes if
you're working

00:06:24.000 --> 00:06:26.100
with a pre-existing scale,

00:06:26.100 --> 00:06:28.320
it might tell you to add

00:06:28.320 --> 00:06:29.740
all the items
together and then

00:06:29.740 --> 00:06:31.340
that's what represents

00:06:31.340 --> 00:06:34.225
that theoretical concept
or that measure,

00:06:34.225 --> 00:06:36.380
sometimes it might tell
you to use the mean.

00:06:36.380 --> 00:06:38.600
So what I want to do
is compute both of

00:06:38.600 --> 00:06:39.720
these things because they

00:06:39.720 --> 00:06:41.295
both might be
helpful to me.

00:06:41.295 --> 00:06:42.860
So what I'm going to do is

00:06:42.860 --> 00:06:44.990
I'm going to type
meanofitems.

00:06:44.990 --> 00:06:47.240
And again, that very
quickly lets me

00:06:47.240 --> 00:06:48.230
differentiate between

00:06:48.230 --> 00:06:49.970
sumofitems and
meanofitems.

00:06:49.970 --> 00:06:51.700
Type & label, and again,

00:06:51.700 --> 00:06:54.720
I'm just going to put
mean of items 1 through

00:06:54.720 --> 00:07:01.905
6, click Continue.

00:07:01.905 --> 00:07:04.900
Just like we did
before, click on All,

00:07:04.900 --> 00:07:06.380
and then we're
going to go down

00:07:06.380 --> 00:07:10.335
to mean, here's mean.

00:07:10.335 --> 00:07:11.790
And when I click
on that, again,

00:07:11.790 --> 00:07:14.090
it gives me that
sample text here.

00:07:14.090 --> 00:07:16.975
This is very similar
to what we did before.

00:07:16.975 --> 00:07:19.330
I'm going to click up, and

00:07:19.330 --> 00:07:21.505
we're going to
kick over 1,

00:07:21.505 --> 00:07:29.610
2, 3, 4, 5, and then 6.

00:07:29.610 --> 00:07:31.070
I'm making sure to not

00:07:31.070 --> 00:07:33.835
include this last
one, this sumofitems.

00:07:33.835 --> 00:07:35.250
And then just
like last time,

00:07:35.250 --> 00:07:36.150
we're going to delete out

00:07:36.150 --> 00:07:37.280
that question mark,
that last comma.

00:07:37.280 --> 00:07:38.790
We're going to
read through

00:07:38.790 --> 00:07:40.505
it really quickly just
to avoid an error.

00:07:40.505 --> 00:07:42.820
We've got a new
variable, meanofitems.

00:07:42.820 --> 00:07:44.835
It's going to be
the mean of item 1,

00:07:44.835 --> 00:07:47.265
2, 3, 4, 5, and 6.

00:07:47.265 --> 00:07:51.130
I'm going to
click OK. SPSS

00:07:51.130 --> 00:07:53.895
again showed me the
syntax of what it did.

00:07:53.895 --> 00:07:55.695
I'm going to
minimize that.

00:07:55.695 --> 00:07:58.080
And then here we have
the mean of items.

00:07:58.080 --> 00:08:00.010
And if we really
quickly look at this,

00:08:00.010 --> 00:08:01.990
I'm going to pick this
one, for example,

00:08:01.990 --> 00:08:03.700
we've got 1,

00:08:03.700 --> 00:08:06.110
1, 3, 3, 1, 2,

00:08:06.110 --> 00:08:07.710
and a mean of 1.83 would,

00:08:07.710 --> 00:08:09.975
just eyeballing
it, make sense.

00:08:09.975 --> 00:08:11.580
If we look at this one,

00:08:11.580 --> 00:08:13.590
3.67, 2,

00:08:13.590 --> 00:08:15.645
3, 5, 4, 5, 3,

00:08:15.645 --> 00:08:16.990
that would roughly
make sense

00:08:16.990 --> 00:08:18.650
without having to
do the math of it,

00:08:18.650 --> 00:08:21.310
but this has computed
the average score,

00:08:21.310 --> 00:08:24.450
the mean score for all
six of those items.

00:08:24.450 --> 00:08:26.590
So that is very
quickly how to

00:08:26.590 --> 00:08:28.810
compute a new
variable in SPSS.

00:08:28.810 --> 00:08:30.410
This is
particularly useful

00:08:30.410 --> 00:08:32.190
when you have
collected data using

00:08:32.190 --> 00:08:34.610
a pre-existing measure
and that measure has

00:08:34.610 --> 00:08:36.310
instructed you to compute

00:08:36.310 --> 00:08:38.510
the sum of items and
run your tests on that,

00:08:38.510 --> 00:08:39.930
or maybe the mean of items

00:08:39.930 --> 00:08:41.450
and run your
tests on that.

00:08:41.450 --> 00:08:43.130
But this is a useful way

00:08:43.130 --> 00:08:44.450
of understanding
how to create

00:08:44.450 --> 00:08:48.190
new variables using
existing ones in SPSS.