Now assume you wanted to screen for an outcome where there was no gold standard.
ID: 3209485 • Letter: N
Question
Now assume you wanted to screen for an outcome where there was no gold standard. Instead of assessing the validity of the test by sensitivity and specificity, you can measure the reliability of two observers. Assume you have 100 samples and 2 observers both rate these samples as positive 5% of the times and both rate as negative 90% of the time. In addition Observer 1 identifies 3 cases as positive that Observer 2 did not find. Set up a 2x2 table to evaluate the reliability between the two observers Observer 1 Observer 2 Pos Neg Pos A B n1. Neg C D n2. n.1 n.2 N A B C D N n1. n.1 n2. n.2
A. 2
B. 90
C. 6
D. 93
E. 92
F. 3
G. 100
H. 7
I. 94
J. 8
Explanation / Answer
Result:
Now assume you wanted to screen for an outcome where there was no gold standard. Instead of assessing the validity of the test by sensitivity and specificity, you can measure the reliability of two observers. Assume you have 100 samples and 2 observers both rate these samples as positive 5% of the times and both rate as negative 90% of the time. In addition Observer 1 identifies 3 cases as positive that Observer 2 did not find. Set up a 2x2 table to evaluate the reliability between the two observers Observer 1 Observer 2 Pos Neg Pos A B n1. Neg C D n2. n.1 n.2 N A B C D N n1. n.1 n2.
Observer2
Observer1
+ve
-ve
Total
+ve
5
3
8
-ve
2
90
92
Total
7
93
100
Observer2
Observer1
+ve
-ve
Total
+ve
5
3
8
-ve
2
90
92
Total
7
93
100
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.