+ Reply to Thread
Results 1 to 5 of 5

Thread: How can SPSS recognize a dataset from another package as imputed

  1. #1
    Points: 44, Level: 1
    Level completed: 88%, Points required for next Level: 6

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    How can SPSS recognize a dataset from another package as imputed




    I have imputed my data, which is multilevel, on Mplus and combined all the iterations to form a dataset and transferred to SPSS.
    but I am not sure how to make SPSS recognize this dataset as imputed with multiple iterations.
    I have tried splitting the file, but the problem is I cant get a pooled analysis of the results from all imputations this way.
    anyone got an idea how I can beat this one?

  2. #2
    Omega Contributor
    Points: 39,120, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,082
    Thanks
    402
    Thanked 1,194 Times in 1,155 Posts

    Re: How can SPSS recognize a dataset from another package as imputed

    I haven't used SPSS much, but traditionally what I have done is, say use multiple imputation (m=10), then marked each imputed set with a numeric value (e.g., 1-10). So all observations within an imputed set has the same variable value, example variable name Impute value = 1 if in the first set. Then when pulling or performing other functions, I just reference the variable "Impute" as a strata variable. Not sure why you are unable to do this in SPSS.
    Stop cowardice, ban guns!

  3. #3
    Points: 44, Level: 1
    Level completed: 88%, Points required for next Level: 6

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How can SPSS recognize a dataset from another package as imputed

    Quote Originally Posted by hlsmith View Post
    I haven't used SPSS much, but traditionally what I have done is, say use multiple imputation (m=10), then marked each imputed set with a numeric value (e.g., 1-10). So all observations within an imputed set has the same variable value, example variable name Impute value = 1 if in the first set. Then when pulling or performing other functions, I just reference the variable "Impute" as a strata variable. Not sure why you are unable to do this in SPSS.
    Hi hlsmith.
    Many thanks for your response. I have already computed a new variable on 'imputation' which I have used to mark each imputed set from 0 to 5 (m=5), with 0 for the original dataset with missingness. I referenced the variable 'imputation' as a strata variable and even split the merged dataset based on the 'imputation variable'.
    The persistent issue is that when I run an analysis, it produces output for each of the 6 sets (5ms and 1 original), but not a pool of the results from these 6 sets.

  4. #4
    Points: 44, Level: 1
    Level completed: 88%, Points required for next Level: 6

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How can SPSS recognize a dataset from another package as imputed

    Hi hlsmith. thanks for your earlier comment. I sorted this out by adding an underscore to the variable label 'imputation'.
    Apparently, adding a variable with the label 'imputation_' makes SPSS recognize the dataset as having imputed iterations.

  5. #5
    Omega Contributor
    Points: 39,120, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,082
    Thanks
    402
    Thanked 1,194 Times in 1,155 Posts

    Re: How can SPSS recognize a dataset from another package as imputed


    Great. Yes, it is great if the program can run the pooling process. Otherwise if I remember correctly, pooling by hand is easy enough, but getting the standard errors is a little more difficult by hand.


    Stop cowardice, ban guns!

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats