1
$\begingroup$

Has anyone been able to find any package/function to run FIML in R for multilevel models? I know the lavaan package has a function for it, but it doesn't support MLM. I've only been able to find it in MPlus, but I do not know how to work the program (so I am trying to find a way to avoid using MPlus to do this). Initially, the lmer function I used simply applied listwise deletion to my data, but I ended up with lots of missing data (my study was a daily diary study with two respondents), so reviewers have asked me to do FIML instead.

I used the md.pattern function of the mice package to look at my missing data and right now it looks like I have 11 variables with missing data (with a total of 427 missing cases, 1328 missing cases).

If FIML is not available, is there another free-to-use program (that is beginner friendly) that allows me to run my multilevel models with FIML?

Improve this question
$\endgroup$
3
  • $\begingroup$ can't you use mice instead? $\endgroup$
    – rep_ho
    Commented Jul 15, 2022 at 12:06
  • $\begingroup$ @rep_ho From my search, I haven't been able to find a function that does FIML in mice, is there one? From my understanding, mice does multiple imputations rather than FIML, but my reviewers are very persistent about using FIML. $\endgroup$
    – FFFStudy
    Commented Jul 15, 2022 at 19:19
  • 1
    $\begingroup$ I meant using mice instead of FIML. Usually you should just do what reviewers suggested only in spirit but not literally, but thats for another discussion. Good luck with that, sorry I can't help with FIML $\endgroup$
    – rep_ho
    Commented Jul 15, 2022 at 19:39

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.