I am revising normal distribution, and I am just wondering about the question below.
"NORMAL DISTRIBUTION HYPOTHESIS TESTING"
https://www.madasmaths.com/archive/maths_booklets/statistics/hypothesis_testing_normal_distribution.pdfQuestion 20 (****)
A pharmaceutical company spokesman claims that a certain pill contains 250 mg of
active ingredient. Tests carried out on 120 tablets resulted in a sample mean of
249 mg, with a standard deviation of s mg.
If the pharmaceutical company's spokesman claim was just rejected at the 5% level of
significance, find the largest possible value of s, correct to 1 decimal place.
I know how to do the calculations, however, why is this not a one-tailed test, as the sample mean is 249, which is lower than 250? Are we not checking the claim on the basis that there is insufficient mass in the tablets? And how can I distinguish between a one-tailed and a two-tailed test because for some of these questions, as I got in a bit of a muddle? Another one is below:
Question 9 (**+) "
The battery lifetime of a certain make of laptop is claimed to have a mean of 6.6 hours
and variance of 3.9 hours^2.
The battery lifetimes of forty such laptops were measured and the mean was 6.1 hours.
Determine, at the 10% level of significance, whether the claim on the battery life
of the laptop is justified.
Here are we not testing just the fact that u < 6.6 as our sample mean was lower? What is it that makes these two-tailed test?.
If you need some context of what I'm talking about and the solutions to these questions, then the URL above has the questions posted, and you need to go to pages 30 and 41
Thanks in advance