SPSS users who want to speed up their work by using Python will encounter some surprises. This tutorial walks you through the 5 major pitfalls and shows how to avoid them.
1. Python is Fully Case Sensitive
SPSS is mostly case insensitive; if we have a variable “gender”, we can address it in syntax as gender or GENDER or anything in between. On top of that, we can’t have two variables gender and GENDER in SPSS because they’d be seen as the same variable.
In Python, none of the above holds. As it’s fully case sensitive, we must always use the exact right casing for all objects. This is especially tricky when importing modules or using methods: they don’t seem to exist if we don’t use the correct casing.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
*Wrong casing for module. begin program. import spss,spssclient # ImportError: No module named spssclient end program. *Correct casing. *Wrong casing for attribute. *Correct casing. |
Result
2. Indentation Matters in Python
In many computer languages -SPSS syntax, Javascript, CSS, PhP, HTML and more- indentation is optional and mainly used for making code more readable. In Python, however, indentation indicates where Python loops and Python-if clauses end. The very simple examples below illustrate how it works.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
*Print “hello” and “bye” 5 times in loop. begin program. for i in range(5): print “hello” print “bye” #Indented so still in loop end program. *Print “hello” ” 5 times in loop and “bye just once. |
Python Indentation in SPSS
We could indent lines with 1 or 2 spaces but for some reason, 4 spaces is most common. We can set this in SPSS by navigating to 
and setting it to 5.
Pressing tab in the Syntax Editor window now results in 4 spaces.
Note that Notepad++ -recommended for writing larger block of Python code- has a similar setting and some very handy shortkeys for indenting or outdenting entire sections.
3. Comment Your Code
For SPSS syntax as well as Python, adding comments to your code is a great idea. In Python, all code between a hash tag (#) and the end of a line is seen as a comment. By default, Notepad++ shows Python comments in green as shown below.
4. Print Objects and their Types
If you create Python-objects such as strings or Python lists yourself, you’ll probably know their object types and what they contain. This allows you to work with them in a goal directed manner.
However, if we retrieve objects from SPSS such as value labels or data values, we’re not always sure how they end up in Python. The solution is to run print object and print type(object). The example below -using employees.sav– illustrates how this helps us out.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
*Look up value labels for job satisfaction. begin program. import spssaux sDict = spssaux.VariableDict() vallabs = sDict[‘job_satisfaction’].ValueLabels print type(vallabs) # <type ‘dict’> end program. *Since vallabs = Python dict object, we can retrieve key-value pairs with iteritems() method. |
5. Be Careful with Backslashes
Python uses the backslash (\) as an escape character in strings. This may yield unexpected results if we’re not aware of it. For example, we can’t specify the path to an SPSS file assomepath = ‘c:\newdata\data.sav’because “\n” inserts a line break into our string rather than “\n”. One solution is to prefix the entire string with “r”, which is short for raw string.
Therefore,somepath = r‘c:\newdata\data.sav’will work as we intended.
The examples below show some wrong and correct uses of backslashes in Python.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
*Wrong way: \n indicates new line. begin program. somepath = ‘c:\newdata\data.sav’ print somepath end program. *Right way: \n in raw string is just \n. *Wrong way: second quote ends string prematurely. *Right way: \ escapes second quote. |
Right, I guess these are the major pitfalls in Python. If you agree or disagree or if I forgot to mention something, please let me know by dropping a comment below or otherwise, ok?
