The dataset "textmessages.rds" contains variables measured from a survey conducted of Harvard students enrolled in a recent Statistics class: female: an indicator variable which is 1 for females and 0 for males; underclassman: an indicator variable which takes on the value 1 for a freshman or a sophomore, or 0 for a junior, senior, or a grad student; and texts: a variable that measures the number of text messages the respondent says he or she sends in a typical day. An investigator wants to know whether male or female students send more text messages.

Note 1: All of the below computations must be conducted in R.

a) Obtain the mean and standard deviations of the number of text messages by male and female students.

b) Perform an appropriate t-test to address the investigator's scientific question (without doing any transformations).

c) Is your conclusion in (b) applicable to the population (i.e., all the Harvard's students)? Explain.

d) Graphically investigate the assumption of Normality in the "response" variable to compare these groups. Perform an appropriate transformation to make the Normality assumption more valid. Be sure to include important graphic(s) to substantiate your choice of transformation.

e) Obtain the mean and standard deviations of transformed variable by male and female students.

f) Perform an appropriate t-test using the transformed variable to address the investigator's scientific question.

g) What is the female-to-male median text messages ratio?

h) Without doing any calculation, do you think a 95% confidence interval for the female-tomale median text messages ratio includes 1? Explain.

i) Find a 95% confidence interval for the female-to-male median text messages ratio.