Shouldn't the t-statistic-based confidence interval be wider than the z-statistic-based one?!
Using the exercise file for calculating the confidence interval (CI) when the population variance is known, I have tried to ignore this provided information, and I have calculated the confidence interval based on the sample variance. Surprisingly, the t-statistic-based CI was narrower than the z-statistic-based one; shouldn't it be the reverse?

2 answers ( 0 marked as helpful)
Hi Kyrillos,
Good to hear from you.
Why is the t-statistic CI narrower? The t-statistic value is higher than the z-statistic one, right?
Sure there isn't a calculation error?
Best,
Ned
Hi Ned,
Thanks for your response.
Yes, I am sure that there are no calculation errors. I have calculated the CIs using the 365 CI online calculator one more time, and the results are the same as when I calculated them before. I have attached two explanatory screenshots from the pre-mentioned calculator.
Kind regards,
Kyrillos

