Question

According to shopper data, 97% of households purchase

toilet paper. An analyst believes this is too low. To
investigate, the analyst selects a random sample of 500
households and finds that 98% of them purchase toilet
paper. We found that we do not have convincing
evidence at the a= 0.05 level to conclude that the true
proportion of all households that purchase toilet paper
differs from 0.97.
It can be difficult to detect such a small difference, such
as from 0.97 to 0.98, even if the difference exists.
What can be done to increase the power of this test to
reject 0.97 if 0.98 is true? Check all that apply.
increase a
decrease a
O increase n
decrease n
discard first results and select a separate random
sample of 500 households

Answer 1

To increase the power of this test to reject 0.97 if 0.98 is true, the following can be done:
- Increase n (sample size)
- Decrease the level of significance (a)
- Discard the first results and select a separate random sample of 500 households

Therefore, the correct options are:
- Increase n
- Decrease a
- Discard first results and select a separate random sample of 500 households