Question 5: What is meant by ‘the repeatability coefficient was 2.1’? In what units would this coefficient be measured?
The repeatability coefficient is an upper limit for the size of difference we might get between two measurements on the same individual, whose true value has not changed. It is calculated from the within subject standard deviation, sw, by 1.96 root 2 times sw.
It is measured in the same units as the observations. Hence we expect that two measurements of femur length made on the same fetus would be closer than 2.1 mm for the 95% of pairs.
Back to question.
To Martin Bland's home page.
This page maintained by Martin Bland.
Last updated: 23 February, 2007.
Back to top.