SPSS Statistics

 View Only
  • 1.  SPSS analyzing survey data: composite variable

    Posted Fri January 28, 2022 03:06 PM
    I am analyzing survey data in SPSS. Specifically, I am creating a composite variable for a question that is structured as "Choose your level of agreement with the following statements.." and then I list 4-5 statements in the question. My question is: does SPSS automatically exclude entire responses to a question if they didn't select a response for each of the statements? Or does it include those who only responded to some of the statements in the composite variable? I only want to include responses who answered every statement in the question in my composite variable, which is why I ask. I hope this makes sense.

    ------------------------------
    Nicki Nimlos
    ------------------------------


  • 2.  RE: SPSS analyzing survey data: composite variable

    Posted Fri January 28, 2022 03:18 PM
    I suppose that each response is a variable.  What you do with these is up to you.  If nonsponse is code as missing, you can use NMISS function to count the number of missing values and use that in case selection via FILTER, SELECT IF, or Data > Select Cases.

    --





  • 3.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 07:36 AM
    But do you have to calculate missing responses or does SPSS automatically exclude those who didn't respond to every statement variable when creating composite variables?

    ------------------------------
    Nicki Nimlos
    ------------------------------



  • 4.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 07:49 AM
    Hi, 

    n_miss is the number of missing in s1 to s5, if n_miss = 5 it means s1 to s5 are all missings ,  n_miss < 5 means at least one of s1 to s5 is not missing
    with select if command spss automatically exclude those who didn't respond to every statement variable 

    regards,


    ------------------------------
    xiaoqin Oezener-Wan
    ------------------------------



  • 5.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 08:34 AM
    This really depends on what you are doing.  Some casewise transformation functions like sum(x1, x2, ,,,) will automatically ignore system and user missing values.  In general, though, you would need to code what you want to do in transformation code.

    Procedures will generally provide for listwise or pairwise treatment of missing values, but, again, you should be specific about what procedure you are running and what treatment you want.

    --





  • 6.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 09:34 AM
    an alternative to jon's suggestion would to use:

    compute score = var1 + var2 +var3.

    or better yet:

    compute score =sum.3(var1 to var3)

    in both cases, the result will be set to missing if any of the values are missing for var1 to var3.

    ------------------------------
    Eugenio Gonzalez
    ------------------------------



  • 7.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 04:22 AM
    Hello,

    suppose s1, s2, s3, s4, s5 are your 5 statements variable.

    get file = myfile.

    compute n_miss = NMISS(s1, s2, s3, s4, s5). 

    select if (n_miss < 5) . 

    regards

    ------------------------------
    xiaoqin Oezener-Wan
    ------------------------------



  • 8.  RE: SPSS analyzing survey data: composite variable

    Posted Mon January 31, 2022 07:35 AM
    But do you have to calculate missing responses or does SPSS automatically exclude those who didn't respond to every statement variable when creating composite variables?

    ------------------------------
    Nicki Nimlos
    ------------------------------



  • 9.  RE: SPSS analyzing survey data: composite variable

    Posted Tue February 01, 2022 05:22 PM
    Yet, do you need to work out missing reactions, or does that SPSS consequently avoid the people who didn't react to each assertion variable while making composite factors?

    ------------------------------
    clint ford
    ------------------------------