I would add to Jay's comment that the mean may not be the best value to replace an outlier that you have good reason to replace. As with missing values, it is likely that outliers are from atypical rather than typical cases. Something like expectation maximization would be far better than substituting the mean. Alan Acock ----- Original Message ----- From: "Jay Warner" <[EMAIL PROTECTED]> To: "Dennis" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Wednesday, July 09, 2003 3:33 PM Subject: Re: oultliers : replacing with the mean
| Replacing an outlier is acceptable, if an only if you have a sound reason, and | the reason is NOT because it doesn't "fit." | | When you do have to replace a mis- measured value, replacing with the average | of a sample is sometimes done, PROVIDED the sample is truly from the identical | population. | | If you think about it for a time, I suspect you will begin to understand why | replacing measured values is done with such care - you don't want to fudge in | a number that changes the results. | | Jay | | Dennis wrote: | | > Hi all | > | > I would like to remove outliers from my repetitive measures design, however | > making it missing removes the whole case of the subject. | > I've heard that it's possible to replace outliers with the mean of the | > group. I am wondering if it's a standard practice (to use for my thesis), | > and are there any good references? | > If it is acceptable, how should I compute the mean if there are several | > outliers in one group/DV, (or variable in SPSS)? | > | > Dennis | > | > . | > . | > ================================================================= | > Instructions for joining and leaving this list, remarks about the | > problem of INAPPROPRIATE MESSAGES, and archives are available at: | > . http://jse.stat.ncsu.edu/ . | > ================================================================= | | -- | Jay Warner | Principal Scientist | Warner Consulting, Inc. | 4444 North Green Bay Road | Racine, WI 53404-1216 | USA | | Ph: (262) 634-9100 | FAX: (262) 681-1133 | email: [EMAIL PROTECTED] | web: http://www.a2q.com | | The A2Q Method (tm) -- What do you want to improve today? | | | | | . | . | ================================================================= | Instructions for joining and leaving this list, remarks about the | problem of INAPPROPRIATE MESSAGES, and archives are available at: | . http://jse.stat.ncsu.edu/ . | ================================================================= . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
