Description:
Researchers are planning a study to determine whether two assessment methods agree. The current method of assessment takes a heavy toll on each patient, while the new method is much less invasive. For the study, each patient will be assessed using both methods, and a score will be recorded for each assessment method. The average difference of the paired scores will then be determined.
The maximum average difference that the researchers will still consider the methods to be in agreement, is 6. That is, the two methods will be said to be ‘in agreement’ if the upper confidence limit of the upper limit of agreement is less than 6, and the lower confidence limit of the lower limit of agreement is greater than -6. 95% confidence levels will be used for both the ‘limits of agreement’ and the ‘confidence intervals for the limits of agreement’. The researchers will consider a range of presumed mean differences between 0.6 and 2. They will also look at a range of values for the standard deviation of the paired differences. The needed power is 90%.
The numeric report at the beginning gives the number of pairs needed for each combination of mean difference and standard deviation. The charts at the end show the increased sample size required for a larger assumed difference or larger standard deviation.