When trying to determine how many samples are needed to perform an experiment, the desired accuracy of the results are often a determining factor. Dr. Smith has designed a new scanner capable of determining blood bilirubin levels using a cell phone. Because this is a diagnostic device, Dr. Smith is concerned not just about having small standard errors, but also having good estimates for population percentiles. If he needs to determine the 95th percentile to an accuracy of 5%, what is the number of samples needed to ensure that the percentile as determined from his data (using a t distribution) is within 5% of the percentile as calculated as if the standard deviation was taken from the entire population (using a z distribution)?
When trying to determine how many samples are needed to perform an experiment, the desired accuracy of the results are often a determining factor. Dr. Smith has designed a new scanner capable of determining blood bilirubin levels using a cell phone. Because this is a diagnostic device, Dr. Smith is concerned not just about having small standard errors, but also having good estimates for population percentiles. If he needs to determine the 95th percentile to an accuracy of 5%, what is the number of samples needed to ensure that the percentile as determined from his data (using a t distribution) is within 5% of the percentile as calculated as if the standard deviation was taken from the entire population (using a z distribution)?
Trending now
This is a popular solution!
Step by step
Solved in 2 steps with 2 images