This is an introduction to SPSS for first-time users.
The following topics will be covered:
SPSS Window Layout
Data View vs. Variable View
Part 1 – The SPSS Window
Part 2 – Data View
As previously mentioned, Data View is where you can see raw data.
Let’s say that we are interested in pain treatment for patients with Migraines.
We want to see if a popular over-the-counter pain management, Excedrin Migraine,
is as effective as a commonly used physician-administered triptan called Zomig.
We take 24 young adults who suffer from migraines and randomly assign them to receive Excedrin, Zomig, or a placebo. Participants then self-report their pain level after receiving treatment for their migraines.
This is what this data set might look like in Data View.
We have four variables for each participant: age, gender, treatment, and pain.
Each of these variables are represented by a column (see below).
As mentioned before, each row is a participant. So for example,
Participant 1: 23 years old, gender = 0, treatment #1, and reported pain of 18.57.
Participant 2: 22 years old, gender = 1, treatment #2, and reported pain of 17.32.
Part 3 – Variable View & Variable Details
This is what the Variable View looks like when there is actually data.
As you can see below, each row is a variable in the data set. So our 4 variables (age, gender, treatment, and pain) are in rows 1, 2, 3, and 4, respectively.
One of the variable details that you will specify often are the width/decimal.
Width is generally set at 8 or 10. You don’t need to worry about this much, unless you have a variable such as “population” which would be very large (e.g., 1,000,000), and then you would have to set the width to at least 10 to accommodate such a large number.
Decimal is typically more relevant. For example, our variable of age is a whole number, so we can set the decimal to 0. On the other hand, our pain variable is measured to the hundredths spot, so we need two decimal places (see below).
You will also specify “type” for most variables. If you click on the “type” box for any variable, the box below will appear.
Here, you can specify what kind of variable you are using. For very basic uses of SPSS, “numeric” will almost always work. However, if your variable is a date or a dollar amount, you need to specify that or else your analyses will not work appropriately.
You can also use this textbox to designate variable width and decimal places.
Labels are useful when you have complicated datasets or variable names that are similar.
Basically, the label is a brief description of what the variable is. For example, below I labeled the treatment variable as “what kind of treatment patient received.”
If for some reason I ran an analysis and wasn’t sure what that “treatment” variable meant in the output, the label I added would show up instead and make things more easy to comprehend.
Personally, the values designation is the one I find most useful.
This is a tool you can use to label what different values of a variable represent in categorical variables.
Below, you can see the small “…” box that appears if you click on this field. In order to enter labels, you have to click on that symbol.
The box below will pop open, and you can specify your value labels.
For example, earlier I said “gender=0” when describing the data for a participant. But that’s not very intuitive and definitely not very useful for interpreting your data.
Instead, we can add labels so that we know what each dummy-code for gender represents. We have no designated that 0 represents “male” and 1 represents “female” (see below).
Another example we could use is assigning values to our treatment variable that has been dummy coded in our raw data. So if 1 represents the Excedrin treatment and 2 represents Zomig, we would enter it as below.
We can designate missing values as well. This isn’t valid for our example because we are not missing any data, but for example…
If we had entered any “missing” values for self-reported pain as being “999”, we would have to tell SPSS that 999 = missing…Otherwise, when we run an analysis, SPSS will think that a patient reported a pain level of 999, which would seriously throw off our results.
Again, you click the little “…” icon to open the dialogue box…
And the field below is where you enter the information.
Note: You can specify more than one value as being “missing”
Finally, you can also specify what measure each variable uses.
If it’s a continuous variable, you would select scale.
For ordinal or nominal variables, you can designate the same (see below).
And that’s it for a basic intro to using SPSS.
For more detailed tutorials on other topics,
see the SPSS section of this site.