Below is the data dictionary for the Relay data set you will use for this assignment. The data set is contained in
the Homework 3 - Relay Data.xlsx file and can be found in the homework assignment item (under the Table of
Contents -> Learning Modules -> Week 3 area of D2L).
The most important variables for this project are: avgorder, paperless, refill, and doorstep
Assignment Guidelines
Your assignment will be completed and submitted using the UG-MKTG-376-HW-3-Submission-Form.xlsx
document, available with the assignment in the Table of Contents -> Learning Modules -> Week 3 area of D2L.
You will be asked to answer 15 questions based upon a t-test analysis of the Relay dataset.
You will answer the following questions.
Descriptive Statistics
- How many people enrolled in the paperless initiative?
- How many people enrolled in the refill initiative?
- How many people enrolled in the doorstep initiative?
- What is the average avgorder size among those who enrolled in the paperless initiative?
- What is the average avgorder size among those who enrolled in the refill initiative?
- What is the mean avgorder size among those who enrolled in the doorstep initiative?
- Which initiative has the largest mean difference between enrolled and unenrolled individuals?
Variable Name Variable Description
custid Computer generated ID to identify customers throughout the database
retained 1, if customer is assumed to be active, 0 = otherwise
created
Date when the contact was created in the database - when the customer
joined
firstorder Date when the customer placed first order
lastorder Date when the customer placed last order
esent Number of emails sent
eopenrate Number of emails opened divided by number of emails sent
eclickrate Number of emails clicked divided by number of emails sent
avgorder Average order size for the customer
ordfreq Number of orders divided by customer tenure
paperless 1 if customer subscribed for paperless communication (only online)
refill 1 if customer subscribed for automatic refill
doorstep 1 if customer subscribed for doorstep cooler
train 1 if customer is in the training database
favday Customer’s favorite day for delivery
city City where the customer resides in
Page 3 of 6
T-test results - What is the t-value for the t-test on the paperless initiative?
- What is the t-value for the t-test on the refill initiative?
- What is the t-value for the t-test on the doorstep initiative?
- Is the paperless initiative having a significant effect on average order size?
- Is the refill initiative having a significant effect on average order size?
- Is the doorstep initiative having a significant effect on average order size?
Analytical Recommendations - Which (if any) of the three initiatives would you recommend to cancel? Based on your analysis, why would
you make this recommendation? You may take data from this analysis other than just the significance tests into
account when justifying your answer. - Which (if any) of the three initiatives would you recommend to continue? Based on your analysis, why
would you make this recommendation? You may take data from this analysis other than just the significance
tests into account when justifying your answer.
The questions you will need to answer fall into three categories. The first 7 questions ask for descriptive
information about the data set. Namely, enrollment frequency (the number of people who enrolled) for each of
the initiatives, mean order size for those enrolled in each initiative, and the mean difference in order size
between those enrolled and those unenrolled. These questions can be answered using either excel or R. In R,
you can answer these questions using the “table()” and “tapply()” functions.
The second set of questions has to do with the results of your t-tests. You will need to conduct a t-test using the
“t.test()” function in R treating each initiative as your predictor variable and “avgorder” as your outcome
variable. When running this set of t-tests, remember to specific “var.equal=TRUE” within your “t.test()” function
to ensure that your degrees of freedom are calculated correctly.
For each test, you are asked to submit the t-value for the test (please submit results with 2 decimal places only)
and to make a conclusion about whether the result is having a significant effect at the p<.05 level.
Lastly, you will make two recommendations based upon your analysis. Which (if any) of the initiatives would
you cancel and which would you keep? You must state which you would and would not keep, and justify why
you make this recommendation. Remember to consider everything you know about these initiatives when
making your recommendations. Do not fall into the trap of relying only on the results of your significance test.
Save all your answers in column B in the submission form and upload only this when you turn in your
assignment.
Evaluation Criteria and Grading Rubric
Grading will be based on the criteria provided in the grading rubric below.
Descriptive Statistics
Question Points Description Points Description Points Description - How many people enrolled
in the paperless initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - How many people enrolled
in the refill initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct
Page 4 of 6 - How many people enrolled
in the doorstep initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - What is the average
avgorder size among those
who enrolled in the
paperless initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - What is the average
avgorder size among those
who enrolled in the refill
initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - What is the mean avgorder
size among those who
enrolled in the doorstep
initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - Which initiative has the
largest mean difference
between enrolled and
unenrolled individuals?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct
T-test results
Question Points Description Points Description Points Description - What is the t-value for the
t-test on the paperless
initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - What is the t-value for the
t-test on the refill initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - What is the t-value for the
t-test on the doorstep
initiative?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - Is the paperless initiative
having a significant effect
on average order size?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - Is the refill initiative
having a significant effect
on average order size?
0 Not Attempted 1 Attempted, but
Incorrect
2 Correct - Is the doorstep initiative
having a significant effect
on average order size?
Sample Solution