A few comments for multilevel data:
1. Casewise deletion at level 1 is generally okay for cases with missing DV data in a multilevel context. If there other data which would help impute the missing DV value then multiple imputation ought to be considered.
2. FIML is an *easy* alternative offered in AMOS
3. Multiple imputations may be combined using Rubin's rule
4. LOCF and listwise deletion should generally be avoided for many reasons
Ryan
> On Aug 1, 2014, at 10:11 PM, Bruce Weaver <
[hidden email]> wrote:
>
> It sounds as though you are using the imputed (EM) dataset generated by MVA
> (rather than using the EM covariances & means as input for a statistical
> procedure). This is viewed by writers on missing data as a bad idea. Take
> a look at this excerpt from
>
http://www.stats.ox.ac.uk/~snijders/Graham2009.pdf (p. 556), for example.
>
> "Note that the procedure in SPSS for writing out a single imputed data set
> based on the EM algorithm is
> not recommended unless random error residuals are added after the fact to
> each imputed value; the current implementation of SPSS, up to version 16 at
> least, writes data out without adding error (e.g., see von Hippel 2004).
> This is known to produce important biases in the data set (Graham et al.
> 1996)."
>
> HTH.
>
>
>
> Gabriela Nakamura wrote
>> Hi, I used EM in dealing with missing data, but some of the estimated
>> values came out to be negative, while the score can only range from 1 to
>> 6. I'm quite lost, and I have no idea what' wrong and how to fix it. Does
>> anyone have an idea as to why I'm having this issue, or how to fix it? :(
>>
>> Thank you.
>
>
>
>
>
> -----
> --
> Bruce Weaver
>
[hidden email]
>
http://sites.google.com/a/lakeheadu.ca/bweaver/>
> "When all else fails, RTFM."
>
> NOTE: My Hotmail account is not monitored regularly.
> To send me an e-mail, please use the address shown above.
>
> --
> View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Negative-values-appeared-in-EM-tp5726871p5726872.html> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
>
[hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
