Question

Researchers at the University of Mississippi wanted to determine whether the reaction time (in seconds) to go/no go stimulus of males differed from that of females. The researchers randomly selected 20 females and 15 males to participate. The go/no go stimulus required the student to respond to a particular stimulus and not to respond to other stimuli. Use this information to answer questions 2 and 3.

2. The researchers decided to construct a confidence interval to determine if the difference of the means is significant. To determine whether the confidence interval would be reliable they create Q-Q plots of both random samples and see that they are approximately normal. Are the requirements for constructing a confidence interval satisfied? Why or why not?

a. Yes. The distribution of the mean for each sample is normal since the researchers can apply the Central Limit theorem.

b. Yes. The distribution of the mean of each sample is normal since the data have been determined to be normal.

c. No. The distribution of the mean of each sample is not normal since the sample size is not large enough.

d. No. The company needs to check that all the data combined is normal.

3. Of the four types of confidence intervals listed below, which one is appropriate for this experiment?

a. Confidence interval for μμ when σσ is known.

b. Confidence interval for μμ when σσ is unknown.

c. Confidence interval for the mean of differences, using dependent samples.

d. Confidence interval for the difference of means, using independent samples.

Answer 1