Final answer:
The 98% confidence limit will be wider than the 90% confidence limits. Thus, a) (30, 50) have a wider interval that would be expected for a 98% confidence level.
Step-by-step explanation:
The 90% confidence limits for the population mean are 33 and 47. To determine which of the given 98% confidence limits could be possible, we need to consider the relationship between the confidence level and the width of the interval.
We don't know the sample mean or the standard error, but we do know the 90% confidence limits. We can use these to estimate the standard error:
Standard error = (90% confidence limits upper bound - 90% confidence limits lower bound) / (1.645 * 2)
= (47 - 33) / (1.645 * 2)
= 4.73
Now we can use this to estimate the 98% confidence limits:
98% confidence limits = sample mean ± (2.33) * (4.73)
We know that the range of the 90% confidence limits is 14 (47 - 33). The range of the 98% confidence limits will be wider than this, so we can eliminate options b), c), and e). Checking which of the remaining options gives us a range wider than 14 by subtracting the lower limit from the upper limit:
a) (30, 50) -> range = 20
d) (38, 45) -> range = 7