Answer:
(A) Since Sample 1 comes from the population of Metal songs only and Sample 2 comes from the population of Blue songs only,
The 90% confidence interval for mean valence for Metal songs is (0.2224 , 0.6797)
The 90% confidence interval for mean valence for Blue songs is (0.3063 , 0.8557)
All intermittent and final answers were rounded up to four decimal places.
(B) The option (b) is correct. This is because the question says that the research group suspects a difference between both means, not that one mean is greater or less than the other.
Explanation:
(A) Using a 90% confidence level, the true mean is within 1.645 standard deviations of the sample mean
For METAL SONGS,
Lower limit = 0.451 - (1.645)(0.139)
= 0.451 - 0.2287 = 0.2224
Upper limit = 0.451 + (1.645)(0.139)
= 0.451 + 0.2287 = 0.6797
For BLUE SONGS,
Lower limit = 0.581 - (1.645)(0.167)
= 0.581 - 0.2747 = 0.3063
Upper limit = 0.581 + (1.645)(0.167)
= 0.581 + 0.2747 = 0.8557
(B) The null hypothesis is:
Mean valence of Metal songs is equal to the mean valence of Blue songs
Alternative hypothesis:
Mean valence of Metal songs is NOT equal to the mean valence of Blue songs.