+ Reply to Thread
Results 1 to 2 of 2

Thread: Missing Data

  1. #1
    Points: 4,824, Level: 44
    Level completed: 37%, Points required for next Level: 126

    Thanked 1 Time in 1 Post

    Missing Data

    Hi All,

    I am running a number of mulitple regression analyses and wanted to know how to use Little's MCAR test to assess missing data.

    Specifically, do you run a separate Little MCAR test for each regression (i.e., those variables in each separate regression), or do I run on Little MCAR test on the whole data set?


  2. #2
    Omega Contributor
    Points: 38,289, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,185 Times in 1,146 Posts

    Re: Missing Data

    It has been about four years since I used the test, so accept my post with caution. It probably depends on how the test is set-up, but If I remember correctly, you don't list the dependent variables or independent variables implicitly in the test - it just looks for any general patterns. So if that is so, then I think it provides a global result. Thus meaning no discernible pattern. If so, you probably don't need to run it a bunch of times. An issue with just running it with all variable once, could be that it finds a pattern in variables not in the models you are interested in. Say antecedents to variables of interest, etc. I also, don't know the power of the test. So with sparse data you could (just guessing) get a non-significant pvalue due to small sample size, but there is a possible concern for MCAR.

    Lastly, this below post link covers my other comments and recommendations. Side note, logistic regression is a good approach, though it could also have the same power issues as Little's test.

    Stop cowardice, ban guns!

  3. The Following User Says Thank You to hlsmith For This Useful Post:

    bozatron (08-12-2017)

+ 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