Question about Multiple Response Sets in SPSS

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Bungalows
Posts: 2
Joined: Mon Dec 15, 2014 3:27 pm

Question about Multiple Response Sets in SPSS

Postby Bungalows » Mon Dec 15, 2014 3:39 pm

I've recently been using SPSS to handle some data I've collected through a survey and a number of the questions use checklists. I.e., they are something like this:
Question: Which of the following emotions did you experience after leaving your last job?

Happy
Relieved
Depressed
Sad
Angry
Excited
When inputting the data into SPSS I treated each item on the checklist as its own variable, and now I've grouped all of these individual variables into a multiple response set and produced a frequency count on that basis.

The issue I have is with the percentages in the frequency count. At present, they represent the number of times each item appears as a percentage of the sum total of all of the responses given, but what I want it to tell me is the number of times each item appears as a percentage of the total number of respondents to the survey.

In other words, I want to be able to say: "??% of the respondents to my survey said that they felt 'happy' after leaving their last job, while ??% said that they felt relieved'. I also want to be able to represent each item visually in a bar chart, and the way it currently is, the bar chart wouldn't make sense.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Question about Multiple Response Sets in SPSS

Postby GerineL » Mon Dec 15, 2014 3:51 pm

could you indicate what your data is set up like?
i.e., rows represent individuals... etc
Bungalows
Posts: 2
Joined: Mon Dec 15, 2014 3:27 pm

Re: Question about Multiple Response Sets in SPSS

Postby Bungalows » Mon Dec 15, 2014 3:54 pm

GerineL wrote:could you indicate what your data is set up like?
i.e., rows represent individuals... etc
The rows represent individuals and all of my questions are in the columns.

But I think I've sort of found the answer I need. The 'Percent of cases' column in the SPSS output of the frequency count of this multiple response set is basically what I need. However, I didn't realize that at first because the percentages were different to what I was getting when I was doing each column individually, and the reason why -- it turns out -- is because when I analyse those same variables as a multiple response set I only get 49 valid cases, when I should be getting 55 (i.e., I have 55 respondents in my survey and all of them gave valid answers).

What am I doing wrong?

Edit: I think I've found the issue, but not the solution: I told my respondents that if they felt 'neutral' to not select any of the items in the checklist, which meant that for that respondent in SPSS I put a '2' for 'No' in each of the relevant boxes. I think what's happened is that I've got 6 'neutral' respondents and because there are no '1's in their rows, SPSS is treating them as invalid cases. It seems like SPSS assumes your respondents will always pick at least one of the items in the multiple response set.
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Question about Multiple Response Sets in SPSS

Postby GerineL » Tue Dec 16, 2014 9:19 am

you can specify user missings for each variable.
In your case, it sounds like you might want to add another variable "neutral".

Who is online

Users browsing this forum: No registered users and 3 guests

cron