Description

Instructions
This assignment involves generalized linear models. The data can be found on Sakai: go to Resources → Data Analysis Assignment Datasets → Assignment 4. You are required to type your solutions using R Markdown. You will submit 1) a PDF produced from R Markdown with answers to the questions, and 2) your Rmd file. Submissions should be made on gradescope: go to Assignments → Data Analysis Assignment 4 and Assignments → Data Analysis Assignment 4 RMD CODE.
DO NOT INCLUDE R CODE OR OUTPUT IN YOUR SOLUTIONS/REPORTS All R code must be submitted separately, and R outputs should be converted to nicely formatted tables. Feel free to use R packages such as kable, xtable, stargazer, etc. Please consult the example report and corresponding Rmd file for guidance.
Also, you should round up ALL numbers/estimates to 2 decimal places (4 decimal places at the most to avoid exact zeros when possible).
The exercise is based on the airline customer satisfaction dataset found here: https://www.kaggle.com/ datasets/teejmahal20/airline-passenger-satisfaction.
The data for this question can be found in the file “airline_survey” on Sakai. Note that you are analyzing a modified version of the dataset found on kaggle.
Code book
Variable Description
Age Passenger age
Gender Passenger gender
Type.of.Travel Purpose of the passenger’s flight (personal, business)
Class Travel class in the plane (business, eco, eco plus)
Customer.Type Loyal or disloyal customer
Flight.Distance Flight distance
Inflight.wifi.service Satisfaction level of inflight wifi service (0: Not applicable; 1-5 where 5 is completely satisfied)
Ease.of.Online.booking Satisfaction level of online booking
Inflight.service Satisfaction level of inflight service
Online.boarding Satisfaction level of online boarding
Inflight.entertainment Satisfaction level of inflight entertainment
Food.and.drink Satisfaction level of food and drink
Seat.comfort Satisfaction level of seat comfort
On.board.service Satisfaction level of On-board service
Leg.room.service Satisfaction level of leg room
Departure.Arrival.time.convenient Satisfaction level of departure/arrival time convenience
Baggage.handling Satisfaction level of baggage handling
Gate.location Satisfaction level of gate location
Cleanliness Satisfaction level of cleanliness
Checkin.service Satisfaction level of check-in service
Departure.Delay.in.Minutes Departure delay in minutes
Arrival.Delay.in.Minutes Arrival delay in minutes
Satisfaction Airline satisfaction level (satisfied, neutral, dissatisfied)
An airline called LaneAir is seeking a data scientist consultant to better understand drivers of customer satisfaction. The airline distributed a survey to customers who have flown with LaneAir in the last six months. Customers rated their overall satisfaction as dissatisfied, neutral, or satisfied. Then they rated their satisfaction with various aspects of the flight. The airline also has information on the passengers’ flight details, including flight distance and departure delay. LaneAir would like to know which services are worth investing in to improve customer satisfaction. However, they would like the data science consultant to keep in mind that some services are more difficult to improve than others. LaneAir is considering the following investments, ranked from most difficult to least difficult to implement:
• Newer, larger seats to improve seat comfort and leg room. This will reduce the number of seats per plane, which is LaneAir’s last choice
• Newer plane models that improve reliability to minimize delays
• Hire more flight attendants or other staff to improve services including inflight service, cleanliness, or onboarding.
• Technology investment to improve wifi and entertainment service
• Marketing or promotion initiatives to improve customer loyalty or appeal to different customer types (e.g., different age demographic, business/personal travelers)
You will complete this assignment in two parts. First, you will create a 2-page report for the client. This report should be understood by LaneAir executives with very little understanding of statistics. This report should include the following:
• Introduction: Provide an overview of the dataset and the goals of the analysis. Keep in mind that the LaneAir team is familiar with the questions on the survey, but not the results. So you should include, for example, basic summary statistics to show the team the distribution of customers in the different satisfaction categories.
• Methods: Explain the model you used to analyze the data without getting into technical details. Why did you decide to use that model for this dataset and how does it answer the airline’s question?
• Results: What are the key results of the analysis as they relate to the airline’s question? Present at least one figure that effectively communicates a key takeaway of the analysis.
• Conclusion: Keeping in mind LaneAir’s cost considerations outlined above, what are your recommendations for how they can balance impact with cost? Do you have any recommendations that the airline has not considered? Finally, are there any limitations that the client should be aware of? For example, could certain customers be more likely to respond to the survey than others?
• Data overview and analysis plan: You should present details of the data that were not included in part 1, for example an appropriate distributional assumption for the outcome variable. Then, present the type of model you used for the analysis. Which type of generalized linear model is best suited for this problem? What is the link function?
• Model results: Present a table of model results including odds ratios, confidence intervals, and p-values. Interpret the results that you think are most compelling.
• Model assessment: Present the confusion matrix and explain your conclusion for the model’s predictive accuracy. Additionally, the model you should use for this analysis relies on a key assumption that is unique to this model. Using a different, more “precise” model, compare predictions using the predictors Gender and Customer type. Show the confusion matrix for the more precise model and compare the accuracy.
• Conclusion: What do you conclude about the validity of this analysis?
Grading
40 points