+ Reply to Thread
Results 1 to 3 of 3

Thread: Matched Groups Statistics

  1. #1
    Points: 48, Level: 1
    Level completed: 96%, Points required for next Level: 2

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Matched Groups Statistics




    Hello all

    hope this is the right thread for the question and that someone will be able to help me out

    I have created a propensity matched control for my case group and am unsure now which test to use for statistical comparision of my outcome variables (most categorical, some ordinal and some continuous).

    Primarily I would use McNemar's test for the categorical variables and paired t tests for the continuous, but maybe I should/could use also some kind of regression (with or without control for some other variables not included in the matching procedure?)

    And how should I proceed to compare the groups also after including the unmatched cases (giving respective group sizes of ~400 cases vs ~350 controls)?? I don't think I can use McNemars or paired t tests there, as they compare only the matched cases...

    I would be very grateful if someone could answer my probably pretty basic questions.. .
    Thank you very much!

    Cheers

    summit cross

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Matched Groups Statistics

    Just look up "conditional regression". It has a option for strata typically where you account for the unique identifier that matches the units.
    Stop cowardice, ban guns!

  3. #3
    Points: 48, Level: 1
    Level completed: 96%, Points required for next Level: 2

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Matched Groups Statistics


    Hi
    thanks for the tip! I came across conditional regression but don't know how to do it in r or spss... could you tell me or do you maybe have a link to a tutorial or step-by-step walkthrough or something like that ?

    and (it's probably a silly question) why should conditional regression be preferred over McNemar's test?

    and does it make a difference for the analysis if it is not a case-control but a matched cohort study?

    Sorry for the many questions, I've been trying to find it out all day long but just don't have enough knowledge of the matter to make sense of the partially conflicting advises I found...

    Cheers

+ Reply to Thread

           




Tags for this 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