Computing a new variable in SPSS

>> So here we are in SPSS.

I have created a junk data

set just to help
illustrate how we can

compute variables
or compute

new variables using

the values from
existing ones.

This is a pretty
straightforward process.

It looks a little bit

different than things
that we've done

before or other
ways that we're

going to be using
SPSS in this course.

So it might look a
little bit different,

but it is pretty
straightforward and

pretty easy once you
know how to do this.

So the first
thing that we'll

do is go to Transform,

and then the
first option we

have there is
compute variable.

So click on that. We get

this new window
that pops up.

The first thing
that we really want

to look at is in
the top left,

and that's this target
variable window.

Basically what
the computer

is asking you is to create

a name for the
new variable

that you're going
to be computing.

This name needs
to be unique,

it's one that can't
have spaces in it,

and it should avoid
special characters.

But, in this case,

what I like to
do is include

some piece of
information in there

in the naming
convention I use that

helps me know what this is

just by looking at it.

So what we're going
to be doing is

we're going to be
summing items 1,

2, 3, 4, 5,

and 6 together, adding
those together.

So an easy way
to do that is

just to put sumofitems.

That way, when I look at

this new variable
once it's generated,

I know that that's
the variable

that I really want
to be working

with or running
my tests on

because that contains

the information from
the other ones.

You can click on
Type & Label.

The label is the more
drawn-out explanation.

That's one that
you can include

additional characters or

spaces and special
characters.

In our case, I'm
going to put this is

the sum of items
1 through 6.

And that way I know that

that is going to give me

a more drawn-out
explanation

of what this
variable includes.

We know this is
going to be numeric.

So click Continue.

The next thing we're
going to look at

is this numeric
expression box.

And will actually
come back

to this in just a minute,

but this is where you're

going to be doing most

of the work when you're
computing a variable.

What you'll want
to go to next

is this function
group area.

So I'm going to
click on All.

And you notice right
below it we've

got these functions
and special variables,

and this popped up
when I clicked on All.

This has a whole bunch of

different pre-existing
functions,

so you don't need to
reinvent the wheel,

you can basically use

these pre-existing
functions to

help you compute
this new variable.

I'm going to go
down to sum,

and these are in
alphabetical order,

so it's pretty
easy to find.

Here we have sum, and
when I click on sum,

you'll notice
that this box on

the left populated
with some text.

This box is actually
pretty useful.

So the first thing
it shows you is

an example of what

this expression
is going to be.

And our expression
sounds like

a really fancy term.

If you've ever
used a formula

in Excel, it's
basically that.

It's telling the computer

what it needs to do.

In this case,
just like Excel,

we've got the word sum,

which tells the computer
it's going to be

summing different
values together,

and then in parentheses,

immediately after that,

we have different
variables

that are separated
by a comma,

and that's what tells the

computer to add x, y,

and z things
together to create

a new variable or
compute a new variable.

This also provides

additional extra
content here,

but we don't need to
worry about that.

I found this is
helpful if you

are stuck or something
isn't working right.

So we've got sum selected,

we're going to be
summing the items of 1,

2, 3, 4, 5, and
6 together.

We've got that selected.

So the next thing you do

is click this up arrow,

which basically takes

that pre-existing function

and it kicks it up
to this numeric

and expression box.

The next thing that
we're going to do,

we've already told

the computer
that we're going

to sum different
variables together,

we need to tell it which

variables to sum together.

So the easy way
to do this is to

click on the first

item that you
want to include,

and you'll notice
that this arrow

switch sides from pointing

to the left to
pointing to the right,

click over 1,

and we've already
moved over item 1.

What you need to be
careful of is that you

always have a comma
after each variable,

except for the last one.

So a comma is
already here,

but what I usually do
is I just hit a comma,

I go to item 2,

kick that over, comma,

item 3, kick that over,

comma, item 4,
kick that over,

comma, item 5,

kick that over, comma,

item 6, kick that over.

Now, we know that item 6

is the last one we have,

the last variable
we're including,

so what we're
going to do is

just delete out
that question mark,

we're going to delete
out that last comma,

and then if we read
through this real quick

just to make sure we

don't have any problems,

we've got creating
sumofitems,

this is going to
be the sum of

item 1, 2, 3,

4, 5, and 6,

and what we're going
to do now is click OK.

An output window pops

up and it basically tells

you what the computer did.

It computed the
sum of items,

which was the
sum of item 1,

2, 3, 4, 5, and 6,

it gives us the
variable label,

it gives us that
descriptive text I put in.

I'm just going to
minimize this window,

and if we look back here,

you might notice that

there's this new column,

this new variable,
that sumofitems,

that wasn't there
when we started,

and that's because
SPSS computed that.

And if we add
these together,

if you add each row,

we've got 1, 2, 1, 3, 5,

1, that adds to 13,

5, 1, 2,

3, 5, 1 adds to 17.

So it's done that
math for us.

We didn't have to
manually do it.

I know, yes, it'd
be pretty easy to

manually do this
for 10 cases,

but if you have a dataset
that has 500 cases,

that would be very
time-consuming to do.

And you could see
that this function of

computing this
variable can

save you a lot of time.

It's also more accurate.

It's going to
take any error

out of the equation.

And we know that often

when we do things by hand,

we can introduce
error into things,

so we want to avoid that.

That's one way to
computer a variable.

I'm going to show
you a different type

of variable that
we can compute.

I'm going to go
back to Transform,

I'm going to go back
to Compute Variable.

I'm actually going
to clear this out.

We're done with
someofitems,

we don't need to
worry about it,

we're not going to use
the same expression,

so I'm going to click

Reset in the bottom here.

So what I want
to do is create

the mean of the different
items that we have.

Sometimes if
you're working

with a pre-existing scale,

it might tell you to add

all the items
together and then

that's what represents

that theoretical concept
or that measure,

sometimes it might tell
you to use the mean.

So what I want to do
is compute both of

these things because they

both might be
helpful to me.

So what I'm going to do is

I'm going to type
meanofitems.

And again, that very
quickly lets me

differentiate between

sumofitems and
meanofitems.

Type & label, and again,

I'm just going to put
mean of items 1 through

6, click Continue.

Just like we did
before, click on All,

and then we're
going to go down

to mean, here's mean.

And when I click
on that, again,

it gives me that
sample text here.

This is very similar
to what we did before.

I'm going to click up, and

we're going to
kick over 1,

2, 3, 4, 5, and then 6.

I'm making sure to not

include this last
one, this sumofitems.

And then just
like last time,

we're going to delete out

that question mark,
that last comma.

We're going to
read through

it really quickly just
to avoid an error.

We've got a new
variable, meanofitems.

It's going to be
the mean of item 1,

2, 3, 4, 5, and 6.

I'm going to
click OK. SPSS

again showed me the
syntax of what it did.

I'm going to
minimize that.

And then here we have
the mean of items.

And if we really
quickly look at this,

I'm going to pick this
one, for example,

we've got 1,

1, 3, 3, 1, 2,

and a mean of 1.83 would,

just eyeballing
it, make sense.

If we look at this one,

3.67, 2,

3, 5, 4, 5, 3,

that would roughly
make sense

without having to
do the math of it,

but this has computed
the average score,

the mean score for all
six of those items.

So that is very
quickly how to

compute a new
variable in SPSS.

This is
particularly useful

when you have
collected data using

a pre-existing measure
and that measure has

instructed you to compute

the sum of items and
run your tests on that,

or maybe the mean of items

and run your
tests on that.

But this is a useful way

of understanding
how to create

new variables using
existing ones in SPSS.