In: Statistics and Probability
Which should be larger: a 90\% confidence interval based on a sample of size n=50n=50 or a 95\% confidence interval based on a sample of size n=50n=50? Explain.
The 95% CI is wider than that of 90% CI.
I assume we are are dealingf with the Confidence interval for population mean. Since the sample size, n-50 is fairly large we can assume a normal distribution for sample mesan.
Hence the
CI for the population mean is given by
where
is the standard deviation and
is the Z-value obtained from standard normal curve.
Now since the same sample is used to estimate 90% and 95% CI,
clearly from above the only value that changes is the Z-value. From
the standard normal tables we can see that as
reduces or as
increases the value of Z statistic increases. It is beacuse
as
reduces Z moves towards the tails of the standard normal curve.
The Z-value for 90% CI is 1.645 and for 95% is 1.96.
Hence the 95% CI is wider than that of 90% CI.