Overview
Objective
The goal of this project is to determine if Taylor Swift would have survived the sinking of the R.M.S Titanic. Along the way we will explore the effects of being an adult, woman and first class passenger had on the chances of survival.
Hypothesis
Out of the 891 passengers on board only 342 survived. Because the most likely outcome for any passenger was unfortunately death I will setup our null hypothesis to assume death. :(
- H0 = Taylor swift will not survive the sinking of the R.M.S Titanic
- HA = Taylor swift will survive the sinking of the R.M.S Titanic
Exploratory Data Analysis
Dataset
We will predict Taylors outcome using a combination of the passenger's attributes. I am working with the Titanic dataset hosted by Vanderbilt University*, and modifying project I worked on at udacity.com
- Pclass: Passenger Class (1 = 1st; 2 = 2nd; 3 = 3rd)
- Name: Name
- Sex: Sex
- Age: Age
- SibSp: Number of Siblings/Spouses Aboard
- Parch: Number of Parents/Children Aboard
- Ticket: Ticket Number
- Fare: Passenger Fare
- Cabin: Cabin
- Embarked: Port of Embarkation (C = Cherbourg; Q = Queenstown; S = Southampton)
Heuristic Analysis
Women
She is definitely off to a good start. Not only did a larger total amount of women survive but also a larger percent of the female survival ratio.
Age
Taylors age as of this writing is 26 years old. This puts her in the most interesting bin on the histogram. The titanic had more than 140, 20-30 year olds but yet only a little less than half survived. Her odds of survival are still looking good as almost 80 survived but it's curious how survivorship in the other ranges have much better ratios.
Ticket Price
This is where things get a bit questionable. As a celebrity I assume she paid nothing for her ticket. White Star Line probably comped her ticket for the marketing. So in this case I don't think we can look at price so let's just ignore this one and break down the numbers by 1st, 2nd and 3rd class survivorship instead.
Class
Taylor would certainly be first class, which definitely helped a lot. It's the only class where surviving was actually in your favor.
Let's Put It Together
Accuracy
Predictions using similar attributes similar to Taylor Swift will result in a survival rate of 62.85%, where as a randomly selected passenger would have had a 38.38% change of survival.
Conclusion
Reject the null hypothesis
.3838=342/891
Referenced
- Original Titanic dataset hosted by Vanderbilt University
- Cleaned up data set provided by Udacity, through the Data Analyst nanodegree program
- Some interpretation influenced by a posting on The Analysis Factor
- All python code for analysis available on git-hub
- Python modules used were pandas, numpy and matplotlib