Assignment: Introduction to Data Analysis for Digital Marketing

Module Learning Outcomes

1. Describe and apply statistical techniques for forecasting trends in
digital marketing data.

2. Describe and apply statistical techniques for identifying significant
relationships (or lack thereof) between variables and groups
relating to digital marketing.

3. Use statistical software to carry out appropriate statistical
techniques and interpret the output in a digital marketing context.

Late Submission Details:

Late submission of this assignment will be penalised according to the SETU
Carlow policy on late submissions.

Background

The marketing manager of a hotel has carried out a customer survey to understand how satisfied
hotel guests are with various aspects of the hotel. The resulting dataset contains 25 variables:

 Satisfaction variables: relating to cleanliness, dining experience, staff, satisfaction with elite
status perks, etc…

 Information on customers’ stay: # nights stayed, distance travelled, reason for visiting.

 Spending habits: average amount spent on the room, dining and WiFi.

 Customer segmentation: each guests’ membership level, which includes No Status, Silver,
Gold and Platinum.

The marketing manager wants to learn various things from this survey. For example,

 What aspects of the hotel are guests most satisfied with? Least satisfied with?

 How does overall satisfaction with the hotel relate to satisfaction with other aspects of the
hotel such as food related aspects?

 What makes the customers in each elite status segment different?

 What is the most likely reason for guests to visit the hotel? Does this vary by elite segment?

Task

Your task is to carry out a detailed analysis of the Hotel Satisfaction dataset that will allow the
marketing manager to answer the questions listed above, along with several other questions that he
has not even thought to ask.

Tools

Data Analysis Software:
To complete this report, you will need to use SPSS and Microsoft Excel (pivot tables and graphs) to
analyse the data.

Report Writing Software:
You will then write up your report in Microsoft Word, which will contain all relevant graphs and tables
from SPSS/Excel and your commentary/interpretation.

Datasets:
When carrying out analysis in SPSS, please use the hotel_satisfaction.sav file.
When carrying out analysis in Excel, please use the hotel_satisfaction.xlsx file.

Report Structure:
The final Microsoft Word document should be structured according to the following four sections,
each of which are described in more detail below:
1. Descriptive Analysis.
2. Regression.
3. Hypothesis Testing.
4. Exploratory Analysis.
Section 1: Descriptive Analysis Use Excel’s pivot tables and graphical capabilities to carry out a descriptive analysis of the data.

This part of the report should address the questions below. Support your analysis with relevant graphs
and tables, along with your commentary.

1. Profile the overall customer base.
a. How much do customers spend, on average?
b. How far do they have to travel, on average?
c. How long do they typically stay, on average?
d. How satisfied are customers, on average, with aspects related to cleanliness, staff,
price, other and overall?
e. What is the most likely/least likely reason for guests to visit the hotel?

2. Profile each of the Elite Status Segments.
a. What percentage of customers fall into each segment?
b. How much does each segment spend, on average?
c. How long does each segment stay, on average?
d. How satisfied are each of the segments, on average, with aspects related to room,
staff, price, other and overall?
e. How does the most likely reason for visiting the hotel vary across each segment?

3. Write a summary of each segment’s profile.
Section 2: Regression Use SPSS to carry out the following regression analyses. Your report should contain relevant output
from SPSS, along with your interpretation and commentary.

1. Management wants to know how customers’ overall satisfaction with the hotel is impacted
by staff related aspects:

a. Investigate the relationship between satOverall and each of satFrontStaff,
satDiningStaff, satHouseStaff and satValetStaff using scatterplots and correlation
analysis. Write a few sentences on your findings.

b. Carry out a suitable regression analysis, making sure to include the following in your
commentary: the regression line, an interpretation of the regression coefficients and
an explanation of which variables (if any) are important in predicting a customer’s
overall satisfaction.
c. Based on your results, what would you report back to the manager?

2. Management also wants to understand if a customer’s satisfaction with food price and the
dining staff affects their average food spend per night.

a. Investigate the relationship between avgFoodSpendPerNight and each of
satDiningPrice and satDiningStaff using scatterplots and correlation analysis. Write a
few sentences on your findings.

b. Carry out a suitable regression analysis, making sure to include the following in your
commentary: the regression line, an interpretation of the regression coefficients and
an explanation of which variables (if any) are important in predicting a customer’s
average food spend per night.

c. Based on your results, what would you report back to the manager?
Section 3 Hypothesis Testing:
Use SPSS to carry out the following hypothesis tests. Your report should contain relevant output from
SPSS, along with your interpretation and commentary.

1. From your Descriptive Analysis in Section 1, you should have seen that different Elite Status
segments are likely to visit the hotel for different reasons. Use A/B testing (i.e. the Chi-Square
test) to understand if there is a significant relationship between which segment a customer is
in and the reason for visiting the hotel. Make sure to fully interpret your results.
a. Explain how the marketing manager can use the output of the A/B test to personalize
his marketing activity.

2. From your Descriptive Analysis in Section 1, you should have seen that the average overall
satisfaction score for each of the Elite Status segments is different. In particular, the Platinum
and Silver segments appear to have very different average overall satisfaction scores. Use a
suitable t-test to investigate if this difference between Platinum and Silver segments is
statistically significant. Make sure to fully interpret your results.
a. Suggest some actions that the marketing manager could take, based on the results of
your t-test.

Section 4 Exploration & Storytelling with Data Visualisation:
Use Excel’s pivot tables and graphical capabilities to carry out some additional exploratory data
analysis to see if you can find something else of interest for the marketing manager. Your report
should include the following:

– A clear description of what you found with an explanation of why you think this finding is
important, i.e. what can the marketing manager do with this information?

– One suitable table/graph that supports your findings. This table/graph should “tell the story”
of what you found and should adhere to “Data Visualisation Best Practices” (see notes on
Blackboard).

Grading Scheme
The report will be graded according to the following criteria:

– Use of Software:
o Ability to use pivot tables in Excel.
o Ability to carry out regression & hypothesis testing in SPSS.

– Commentary & Interpretation:
o Clear and relevant commentary on graphs & tables.
o Clear interpretation of the output of regression & hypothesis testing.

– Data Visualization:
o Creative use of graphs & charts.
o All graphs & tables should adhere to “Data Visualisation Best Practices”.

– Overall Quality:
o Report should be well formatted. Refer to notes from Information Technology
module.
o Content should be well structured.
o Writing style should be clear and concise.

– Interview:
o If deemed necessary, there will also be a 10-minute one-to-one interview, during
which I will ask questions on your report to assess your understanding.

Submission Details
You must email the following to deirdre.treacy@setu.ie by the deadline indicated on the front page
of this brief:
1. Microsoft Word report.
2. Microsoft Excel file and the SPSS Data Editor & Statistics Viewer files containing your
graphs/tables.