How to get descriptive statistics on questionnaire elements by a group using SPSS?

Question

I checked the product rating using the questionnaire on the questionnaire scale and imported the date into SPSS. My columns are arranged as follows:

Identifier, Group, Q1, Q2, Q3, Q4

I have two different groups filling out the questionnaire, with each person a different digital identifier. In the Q columns, I have a rating for this person (1 to 5) from the Likert scale.

In total there are more than 300 answers.

I start the analysis using "descriptive statistics / frequencies" from the menu and do not get the tables I'm looking for. In principle, it includes all respondents together, while I would like to compare these two groups in the tables.

How to get descriptive statistics on the elements of a questionnaire by a group using SPSS?

In addition, if you have further advice on what kind of analysis I could perform on this data type in SPSS, I would be very grateful. I would like to show that there is no significant difference in opinion between the groups, and looking at the data, it seems that this is so.

Answer 1

One option

• split file into groups
• run descriptive statistics as usual

See this UCLA SPSS FAQ for how to analyze data by category .

Answer 2

A short answer to your question: crosstabs Q1 to Q4 by group. will create the table you need. If you have the ctables package, a more compact table will be created

 variable level group_id Q1 to Q4 (nominal). ctables /table Q1 + Q2 + Q3 + Q4 by group_id. 

You can either design to get other statistics if you want. It seems to me that the chi-square criterion will be sufficient for your question.

As for further analysis, this is a slightly open question, which requires more attention in order to be able to respond effectively. I often offer visual research for such an exploratory analysis, and therefore I propose to study this question on the website Visualize Likert Answers with R or SPSS for potential insights on how to visualize the answers. Another motivating post might be How to visualize a three-dimensional matrix of unforeseen circumstances? .

There are many other questions related to the analysis of answers to this site, but it is difficult to give more specific recommendations without more specific motivation for analysis.

Answer 3

While the above answers have their good points, I usually prefer this procedure (enter the following into the syntax window and run):

 means q1 to q4 by group/stat anova. 

This will give you group means, sample sizes and standard deviations, as well as tests of the difference in means between groups for each of the variables Q1-Q4. Of course, tests will only give real results to the extent that your data matches anova's standard assumptions. Some may say that variables measured on an ordinal scale of 1-5 are not suitable for anova, and in academic contexts this is often true, but in a business setting, most people are willing to sacrifice some rigor for convenience. It is much more convenient to compare the value 4x2 = 8 than to compare the distributions 4x5x2 = 40 categories of answers.

Answer 4

This can be easily done using the "Crosstabs" feature in SPSS for Windows:

Analysis → Descriptive Statistics → Crosstabs. Move the dependent variable in the Rows field, then move the grouping variable in the Column field, then click OK.