Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Richard Williams <richardwilliams.ndu@gmail.com> |

To |
statalist@hsphsun2.harvard.edu, "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |

Subject |
Re: st: Situation where multiple imputation may be of no use? |

Date |
Thu, 09 Feb 2012 17:24:58 -0500 |

At 05:06 PM 2/9/2012, Clyde B Schechter wrote:

This is a question of a statistical nature about what multipleimputation can accomplish.I have used MI a few times, and I have a general understanding ofhow it works and the underlying theory, but not in great depth.I'm working with a colleague to plan an experiment. Thisdescription is oversimplified but, I believe, provides the essenceof it. Subjects will be enrolled and baseline data obtained. Theywill then be randomly assigned to intervention or placebogroups. After enough time for the intervention to work has elapsed,the outcome, a continuous variable, will be assessed, once and only once.Based on some preliminary studies, we expect that about 15-20% ofthe participants will not return for the outcome assessment. Givenour fairly small anticipated effect size (due mostly to noise in theoutcome assessment that we can't think of any way to reduce withavailable technology), the sample size we need to adequately powerour study is, as it turns out, about 20% greater than we will beable to manage within budget. So, if there were no losses tofollow-up, we'd be just OK. But there will be losses to follow-up,and efforts to reduce that will also eat into the budget. (As wouldgetting two outcome assessments and using the average or doing amixed model.) So my colleague has suggested that when we analyzeour data we use multiple imputation to make up for the missingdata. I'm by no means opposed to doing that, but I don't think itwill help us with regard to statistical power.I understand that MI lets you squeeze all the information that isreally there in the existing data set, and can even correct some ofthe bias that can result using listwise deletion. But in our case,the only missing data will be the outcome measurement. We will havecomplete data on everything else. So it seems to me, that MI inthis context will just amount to carrying out a listwise-deletionanalysis, and multiply extrapolating the results of that to thecases with missing outcome, and the combining the analyses of theimputed data sets in a way that reflects the between-imputed-samplesvariation. If I am thinking about this correctly, the addedvariance from the multiple imputations should pretty much balancethe reduction in standard error that comes from (appearing to) usethe full sample size. If this were not true, then MI would besynthesizing information ex nihilo. So, my instincts tell me thatwe will not solve our statistical power problem by using MI anal!ysis. I have run a few simulations, and they support my opinion,but I wanted to run this by some people who understand MI better than I do.

http://www.ats.ucla.edu/stat/stata/seminars/missing_data/mi_in_stata_pt1.htm

A pre-publication version of the von Hippel paper is at http://www.sociology.ohio-state.edu/people/ptv/publications/Missing%20Y/accepted.pdf ------------------------------------------- Richard Williams, Notre Dame Dept of Sociology OFFICE: (574)631-6668, (574)631-6463 HOME: (574)289-5227 EMAIL: Richard.A.Williams.5@ND.Edu WWW: http://www.nd.edu/~rwilliam * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**st: Situation where multiple imputation may be of no use?***From:*Clyde B Schechter <clyde.schechter@einstein.yu.edu>

- Prev by Date:
**Re: st: MIXLOGIT: marginal effects** - Next by Date:
**st: Re: Merge problem; missing observation** - Previous by thread:
**st: Situation where multiple imputation may be of no use?** - Next by thread:
**Re: Re: st: Situation where multiple imputation may be of no use?** - Index(es):