Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Comparasion of the different screening test

By Caner KARA | Dec 06, 2017 05:20PM CET

Hi,
I want to compare different screening test for efficiency in a single group of population. Which test should ı use?
1.Comparing sensitivity or specificity values with McNemar test
2. Or Comparing AUC values of the tests. ( Is it possible to make this with XLSTAT?)
3.Or another method?

Which one is correct.
Thank you.

Best Answer
By Sébastien | Dec 07, 2017 10:15AM CET | XLSTAT Agent

Hi there,

Here is a suggestion:
- If you have only 2 screening tests, you could use paired t-tests on sensitivity or specificity data: https://help.xlstat.com/customer/en/portal/articles/2062451-student-s-t-test-on-two-paired-samples-tutorial
- If you have more than 2 screening tests, then mixed models could be more appropriate. Use patient as a random effect, test as fixed effect and sensitivity or specificity as a dependent variable. https://help.xlstat.com/customer/en/portal/articles/2062292-random-components-mixed-model-in-excel-tutorial?b_id=9283

Let us know if we’re able to help you further.

Jean Paul


Up

0

Down

By Sébastien | Dec 07, 2017 10:15AM CET | XLSTAT Agent

Hi there,

Here is a suggestion:
- If you have only 2 screening tests, you could use paired t-tests on sensitivity or specificity data: https://help.xlstat.com/customer/en/portal/articles/2062451-student-s-t-test-on-two-paired-samples-tutorial
- If you have more than 2 screening tests, then mixed models could be more appropriate. Use patient as a random effect, test as fixed effect and sensitivity or specificity as a dependent variable. https://help.xlstat.com/customer/en/portal/articles/2062292-random-components-mixed-model-in-excel-tutorial?b_id=9283

Let us know if we’re able to help you further.

Jean Paul

Post Your Public Answer

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283