Spilt File: My Favorite SPSS Trick
Learn SPSS
A simple way to temporarily “split” your data into subsets…so you can create easy-to-read results by groups
A lot of my clients’ research (and my own) compares values by groups— pre-test vs. post-test groups, gender groups, etc. We often summarize demographic or outcome data by groups. Meaning, you need to know the same descriptives statistics (mean, median, mode, standard deviation, skew, kurtosis, plus good ole histograms) for each group, not just the total sample.
What the Split File Command will do, is give you a side-by-side view of the descriptives or whatever analysis you want by your grouping variable. It’s that simple!
I love this command so much because it makes creating tables easy. You can combine Split File with Filters too, for your extra fancy folks.
My favorite way to do that is with Split File! Here’s what you click on (pictures below):
1. Data > Split File
2. Click the “Compare groups” button
3. Move over your grouping variable into the “Groups Based on” box (only use nominal or ordinal variables)
4. Press PASTE so the command goes into syntax (if you don’t know what syntax is, press OK and sign up for my SPSS course!)
After you run the Split File using the above steps, conduct your “analysis” as usual. Typically, you’ll want to use Analyze > Descriptives > Frequencies and then the specific variables you’re interested in summarizing.
When you’re done, don’t forget to “turn off” the Split File.
1. Data > Split File
2. Click the “Analyze all cases, do not create groups” button
3. Press PASTE (or OK if you haven’t taken my SPSS course)
You can also type the following into your SPSS syntax file:
SORT CASES BY <group_variable>.
SPLIT FILE BY <group_variable>.
<<Insert your analysis here>>
SPLIT FILE OFF.
Would you like to get better at SPSS? Check out Courses by Rogue Scholar! Or contact us today for one-on-one help with your analysis.
