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
- :
- Is there a way to calculate 1 sigma shift?

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

Feb 10, 2016 1:45 PM
(3341 views)

I have a data set that includes 20 part and each part is measured 20 times for its thickness. I also have a reference mean and std.I would like to compare each part's mean to the reference mean and to see if their difference is more than my reference point's std. Is there a way to do this in JMP? I am using JMP 11.2. Thanks.

1 ACCEPTED SOLUTION

Accepted Solutions

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

There are a variety of ways one could approach this. My proposal would be to go to the putlldown menus and select Tables==>Summary. There you select the 20 different collumns of data and specify to calculate the mean of each column. It will create a new table with one row, which will have one column for each input column. Then, a new column can be added for each of the columns, and then specify a formula for each column, which makes the comparison to see if the reference std is greater than the absolute difference between the calculated mean and the reference mean.

Rather than doing separate calculations I would add a step that would allow the calculation to only have to be specified once. What I would do, is once the summary data table is created, I would then go to Tables==>Stack and stack all of the summary columns. Once this is done, only a single additional column would have to be created to determine if the reference std was greater or less than the difference .

Jim

2 REPLIES

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

There are a variety of ways one could approach this. My proposal would be to go to the putlldown menus and select Tables==>Summary. There you select the 20 different collumns of data and specify to calculate the mean of each column. It will create a new table with one row, which will have one column for each input column. Then, a new column can be added for each of the columns, and then specify a formula for each column, which makes the comparison to see if the reference std is greater than the absolute difference between the calculated mean and the reference mean.

Rather than doing separate calculations I would add a step that would allow the calculation to only have to be specified once. What I would do, is once the summary data table is created, I would then go to Tables==>Stack and stack all of the summary columns. Once this is done, only a single additional column would have to be created to determine if the reference std was greater or less than the difference .

Jim

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

Re: Is there a way to calculate 1 sigma shift?

Hi Jim,

Thanks for your response.

My data is in two columns. Below is an example of how my data looks like.

Part Value

1 1.2

1 1.3

1 1.5

1 1.4

1 1.6

2 1.3

2 1.2

2 1.5

2 1.6

2 1.6

I think the Summary works. I can summarize the mean and std from each part, then create a column with a formula to calculate the sigma shift. Thanks a lot.