turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- JMP Cpk analysis differences

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 18, 2013 7:14 AM
(1911 views)

I am used JMP 9 and have noticed the "short term" sigma and Cpk result differ depending if the cability study is initiated from Anaylze -> Distributions or from Graph -> X-Bar.

I have data with fixed subgroup size of 6 units.

If I run a capability study from the distribution/histogram report I get a one short term sigma. The short term Cpk result does not match the result when the same data is run on minitab.

If I run a capaility study from the Xbar-R report I get a different short term sigma (control chart sigma). It is lower than the reported sigma from the distribution analsys. The short term Cpk results matches the result when the same data is run on minitab.

Why would I be getting different results for the same analysis? Does JMP estimate sigma differently depending on which report the capability study is initated from? Which results is correct?

Attached is the data set being used.

Thanks

1 REPLY

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

All capability analysis use the same formulas. Options differ in how sigma is computed. See **Help>Books>Basic Analysis and Graphing** where you will find a section on Statistical details for Capability analysis section (p83 in Windows version 9). There is a long term sigma based upon the standard deviation of the entire sample. There is the so called RMSE formula when the distribution capability is chosen and you specify your sample size by a fixed subgroup and finally there is the capability calculated using a moving range.