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
- :
- Different values for variance and std. dev in JMP vs. Excel...why?

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

Jun 16, 2014 12:08 AM
(3775 views)

I'm transforming some data into ZScores. A friend transformed the data in excel and the values came up slightly different than those in JMP. When trying to track down why, I found that the variance and standard deviation in excel differed from JMP. I decided to calculate the standard deviation by hand, using excel as a glorified calculator. My variance and standard deviation still came out different than what JMP was giving me. For example, excel (mean = 1.86, var = 0.1482, st. dev. = 0.3850) and JMP (mean = 1.8533, var = 0.1512, st. dev. = 0.38889). You can see they are close, but still off. I'm not exactly sure how JMP got that mean....even with a hand-held calculator the mean is exactly 1.86. Even if I used JMP's mean to calculate the variance and standard deviation in excel, I get different numbers (using mean = 1.8533 give var = 0.1483 and st. dev. = 3.851 in excel). Note that the variance in each case is calculated as a sample variance (n-1).

Does anyone know why JMP is giving me higher values? This is important as it actually makes the difference between a P = 0.05 and P = 0.06.

Thanks for any help!

**Disregard!** I found the error. One value in my data set between excel and putting it to JMP had changed and I completely blanked on updating that value in excel. Dumb mistake!!!

Message was edited by: Jennifer Campbell-Smith

1 ACCEPTED SOLUTION

Accepted Solutions

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

Jun 16, 2014 12:46 AM
(6335 views)

Oh no! Disregard! I found the error! 100% my mistake and I found it in the data set.

3 REPLIES

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

Jun 16, 2014 12:34 AM
(3505 views)

These statistics should be the same in Excel and JMP. Check if any rows are excluded in the JMP data table. Excluded rows are not included in summary statistics obtained by Summary(). Also, results may differ if the Weight role is assigned to a column.

The data table in the picture below has one excluded row and one column with a Weight role.

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

Jun 16, 2014 12:41 AM
(3505 views)

Nothing is excluded. Can you explain what the weight role is?

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

Jun 16, 2014 12:46 AM
(6336 views)

Oh no! Disregard! I found the error! 100% my mistake and I found it in the data set.