The 'party line' is to take the first 30 or so points, calculate limits, throw out any outside ones & add more at the end, until you have 30 points, all of which are inside the control limits. For both X-bar and R, if you are doing this sort of chart.
Based on (approximate) probabilities, there 3 chances in 1000 (p=2*0.0017) that a point from a well behaved population will fall outside. so for 30 points, there are 90 chances in 1000, or 9 in 100 or about 1 in 10 that one of the points you threw out in fact comes from a well behaved population, just like the other points in your 30. Which one? Can't tell. Selecting points to include in the 'base line' group is a trade-off, and Rich Ulrich is right. An early outsider point is Nature's way of telling you to look closely at your data and process. Maybe it is less stable than you thought. Since your objective is to make the process more stable and consistent, you will be doing this a good deal anyway, until the assignable cause issues are removed. Have a good trip... Jay MrTequila wrote: > Hi all, hope this is the right place. > > i was just wondering what you should do when you establish some > control limits but some of the data points you've just used are > outside of the limits you just established? > > should you write them off as bad, leave them or go back and see/fix > the problem? > > any help appreciated, hope the above makes some sense. > > -steve- > > ================================================================= > 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/ =================================================================
