How can I export into SPSS?
SPSS, originally named in 1968 as the Statistical Package for the Social Sciences, is a popular computer program used for statistical analysis. It is widely used in areas such as market research, health research, government, and education research. With SurveyMonkey, you can export our SPSS option to download a .sav file of the survey responses.
- On the My Surveys page, click the Analyze Icon , or click the Analyze Results tab in your survey.
- In the upper right corner, click the Export All button.
- Choose All responses data.
- Select the SPSS format option.
- Name the file and click Export.
- Once the export is complete, click the Download button in the prompt that appears in the upper right corner. Or, in the EXPORTS section of the left sidebar, click the large down arrow to the left of the file name. This will deliver a compressed folder containing the .sav file.
NOTE: You will need to have SPSS software in order to open this file type (or any program that supports .sav files). Our export format supports SPSS 17.0.1 and onwards.
A. SPSS Views:
- The Variable View defines what each question is and its possible responses.
- The Data View is the spreadsheet-style view with all the responses.
- We label the question name/columns as q0001, q0002, q0003 and so on. You can edit these labels in the cells of the Variable View so they update into the column headers of the Data View.
- The answer choices are assigned both a numeric value and a variable label name. You can confirm the values/names while in the Variable View. Once in the Data View, click View > Value Labels in the toolbar to show the variable label names instead of the numeric values.
This is defined by the type of question used in the survey:
- Scale is used for Rating Scale and Numerical Textbox questions
- Nominal is used for all other questions.
C. Multiple Choice Questions - One Answer Allowed vs. Multiple Answers Allowed:
1. Multiple Choice (One Answer Allowed):
You can only select one choice: CEO, supervisor or peer. SPSS codes them as the following in the Variable View:
CEO = 1
Supervisor = 2
Peer = 3
If this is Q1 in the survey, then the question column will be labeled as q0001 in the export. You will then see the values as 1, 2 or 3 in the rows associated with the respondents' answers in the Data View.
2. Multiple Choice (Multiple Answers Allowed):
With this question type, the data will be spread out over multiple columns in the Data View. This is because you can select more than one answer . SPSS codes them as the following in the Variable View:
CEO = 1
Supervisor = 1
Peer = 1
If this is Q1 in the survey, then the question will be split out into multiple columns in the Data View like:
q0001_0001: 1 = CEO
q0001_0002: 1 = Supervisor
q0001_0003: 1 = Peer
NOTE: Multiple answers cannot exist in one cell. If the question allows multiple responses, they are coded per the example above because:
- if CEO = 1 and Peer = 2
- and you pick both CEO and Peer as answer choices
- then if they are combined into one cell the number would be "12"
- this number is completely different than the value of "1" or "2".
D. Rating and Matrix Questions:
When using these question types, the data will also be separated into multiple columns. If you have a 5 column Matrix or Rating scale, then there will be 5 columns in the SPSS export.
Rating Scale: On this question type, SPSS shows the data in the Variable View in the order of the weighted values.
Ex: You have a reverse scale meaning you have three columns where column one = 3, column two = 2, and column three = 1. When you export into SPSS, then in the Variable View under the Value Labels they are presented in this order:
1 = column three (because this was assigned the weighted value of 1)
2 = column two (because this was assigned the weighted value of 2)
3 = column one (because this was assigned the weighted value of 3)
NOTE - N/A column: When using the default N/A column, this data will be coded as a 0 in SPSS. Do not assign a weighted value of 0 to any other column rating if you are using the N/A default option. If so, SPSS will show two values of .00 and you will need to manually manipulate the data to recode the values to differentiate between the two.
E. Skipped Responses:
Skipped answers are assigned the value of "." in the Data View. This is defined as a SYSMIS in the export.