Multiple Response Output

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

Hopeful Analyst
Posts: 2
Joined: Thu May 10, 2012 1:23 pm

Multiple Response Output

Postby Hopeful Analyst » Fri May 11, 2012 7:50 am

Hi all. I've been reviewing my multiple response output (MULT RESPONSE syntax) and when I use the counts and the base from the case summary to calculate my own percentages of people responding to each item, I get slightly different results than that provided in the output table - does anyone know why this might be? What base would SPSS use?
Thanks in advance for your help.
Penguin_Knight
Posts: 473
Joined: Thu Apr 05, 2012 5:58 pm

Re: Multiple Response Output

Postby Penguin_Knight » Fri May 11, 2012 11:22 am

The base is whoever have at least being counted once. You can find that N at the first table. For instance, if you are count "1" from two binary variables, a and b:

a b
1 1
1 0
1 1
0 0
0 1

Then the base is 4 because the fourth case with 0 in both a and b is considered "missing".
Hopeful Analyst
Posts: 2
Joined: Thu May 10, 2012 1:23 pm

Re: Multiple Response Output

Postby Hopeful Analyst » Sat May 12, 2012 1:20 pm

Thanks, that's helpful.

So, the table/percentages SPSS produces includes those respondents who answered 0 to each response option, whereas my manual calculation using the base from the case summary only includes those who responded 1 to at least one variable/response option. This means both percentages are valid, but it depends on what you are trying to present.

Please correct me if I've misinterpretated this.

Cheers!
Penguin_Knight
Posts: 473
Joined: Thu Apr 05, 2012 5:58 pm

Re: Multiple Response Output

Postby Penguin_Knight » Sat May 12, 2012 9:33 pm

Yes, I believe that's the difference.

Who is online

Users browsing this forum: No registered users and 1 guest

cron