Please see attached file, question 4 needed to be completed only !
4. In this part, you will examine differences and relationships between customer groups.
(a) Use one of the categorical variables in the data set to define two groups for which you
will estimate the difference in two population means on a numeric variable that would
be of interest to EFH. Explain your reasoning. You do not need to refer to outside
sources for this question. Your reasoning will be sufficient, as long as its not
just something like “because we think the means will be different.” Give a
better, “we-are-a-consulting team-being-paid-to-do-this” reason.
(b) Display a neatly formatted table showing the summary statistics (mean, standard
deviation, minimum, 25th percentile, median, 75th percentile, and maximum) and sideby-side box plots for the numeric variable for the two groups.
(c) Perform a hypothesis test at an appropriate level of α or construct a confidence interval
at an appropriate level for the difference in the means for the two groups. Choose to
assume either equal variances (which I call “Case 1” in the notes and videos) or unequal
variances (which I call “Case 2”) depending on whether the standard deviations for the
two groups are reasonably the same or different (this is your judgment call). Again,
show output or explain all calculations and interpret the interval in context.
(d) Choose two categorical variables for which it would be interesting to see if a relationship
exists. For this part, you do NOT have to justify your choice using a specific outside
source, but you should still explain clearly why examining this relationship might be
useful.
8
i. Create a contingency table and a side-by-side bar chart using the data in these two
variables. Explain what you see.
ii. Perform the chi-squared test of independence at the α = 0.05 level of significance,
either showing output from software or showing the calculations “by hand.” State
the conclusion of the test in the context of the problem.
0 comments