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
- :
- How evaluate the agreement of: real diagnosis (1-true,0-false) and diagnosis pro...

Topic Options

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

Highlighted

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

Apr 1, 2016 11:12 AM
(3975 views)

I have three var:

ID patients

D (real Diagnosis): 1-true, 0-false

and

P (Prognosis of diagnosis): 1-true, 0-false.

How evaluate D and P agreement?

I know about cohen's kappa ("Analyze > Quality and Process > Variability / Attribute Gauge Chart").

Do I correctly organized data?

Gr Y1 Y2

1 0 D

2 1 D

3 1 D

4 0 D

5 0 D

6 1 P

7 0 P

8 1 P

9 1 P

10 1 P

1 ACCEPTED SOLUTION

Accepted Solutions

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

Your three variables should be your three columns: ID, Diagnosis, Prognosis.

In the diagnosis and prognosis column you would have your 1/0 or true/false, value for each subject.

ID Diagnosis Prognosis

1 True False

2 Ture True

etc....

Then you can start with fit y-by-x to compare the diagnosis to the prognosis (both columns should be categorical so that you get a mosaic plot and a cross-tabs table). From there look at the JMP documentation on fit y-by-x for agreement statistics. You may also wish to google "FDA guidance on reporting diagnostic tests" for further insight to the types of analyses you might want to run with regard to diagnostic tests.

2 REPLIES 2

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

Your three variables should be your three columns: ID, Diagnosis, Prognosis.

In the diagnosis and prognosis column you would have your 1/0 or true/false, value for each subject.

ID Diagnosis Prognosis

1 True False

2 Ture True

etc....

Then you can start with fit y-by-x to compare the diagnosis to the prognosis (both columns should be categorical so that you get a mosaic plot and a cross-tabs table). From there look at the JMP documentation on fit y-by-x for agreement statistics. You may also wish to google "FDA guidance on reporting diagnostic tests" for further insight to the types of analyses you might want to run with regard to diagnostic tests.

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

Oh, it turns out y-by-x has "Agreement statistics"! But that is not cross-tabulation analysis, is not it? Thanks, Karen! P.S. "FDA guidance..." Oh, it's interesting!