Who survived the sinking of the Titanic?
On April 15, 1912, the largest passenger ship of the first half of the twentieth century, the Titanic, sank. There were 2240 passengers on board. As a result of the iceberg accident, more than one and a half thousand people died, and only about 700 were saved. Historians, journalists and analysts around the world are still asking the question: what was the cause of the disaster, who is to blame, who was able to escape in the catastrophe of the century, how did everything really happen? Today, you can use modern data analytics tools to help you find the answer to this historical question. The purpose of such a study is to determine exactly what factors contributed to the survival of the passenger in that tragic situation: age, gender, cabin class, place of boarding the liner?
Features of Titanic passenger data
Analysts studied a dataset about the passengers of the Titanic, which was posted on Kaggle, an international platform for competitions in the field of Big Data and Machine Learning. It contains data for 1309 people. This is everyone who was on the ship and left a documentary trail about it. Data about the remaining passengers was lost over time. And there is no information about the crew of the liner, because they were simply not included in this list. However, the vast majority of passengers are described and the set can be worked with. The data was entered upon boarding the ship at the port of departure, there are 3 ports in total: Southampton (UK), Cherbourg (Fr), Queenstown (Ir). And in case of their absence, they were supplemented, where possible, on the basis of additional materials obtained during the investigation of the disaster.
The lists were published in a number of major media outlets in the West, and the fate of the passengers was monitored by all the media in the world; people in the morning resorted to newspaper offices to find out news about relatives, friends and acquaintances who sailed on the ship, to find out who escaped.
Let's denote some terms from the dataset: Survived = Survivors (0 = No, 1 = Yes)), Pclass = Passenger class (1st, 2nd, 3rd), Fare = Ticket price, Parch = Presence of relatives on board, Age = Age. The passenger's name, ticket number and a number of other variables in the dataset do not have predictive power. That is, they can be neglected. This conclusion can also be made using visualization. For example, a “heat map” – a diagram built in Python that shows in thermal colors (red – warmer, blue – colder) the correlation of dataset variables. The legend on the right indicates the color of the matching fields; the closer the pantone is to dark red (maximum value), the higher the correlation.
As a result of EDA (preliminary or “exploratory” data analysis), it was concluded that such initial variables as gender, age, boarding location, and passenger class will be prioritized. We use the Dataprep library with Python under the hood, as well as the Pandas, Matplotlib and Seaborn visualization libraries. With their help, it was possible to identify a number of interesting trends. Thus, the variable Embarked (landing location) correlates with the Survived variable. More passengers with Embarked = S (i.e. Southampton) survived. It was probably at Southampton that the richest category of passengers loaded onto the ship. In Ireland (the city of Queenstown, later renamed Kobh – “Kov” in Russian), the village has the most poor people. They most likely went to work in the USA.
The fact that so many Irish poor people boarded at Cove indirectly confirms one of the hypotheses about the death of third-class passengers: many simply could not understand the commands. Any linguist will tell you that Irish and British English are quite different. To such an extent that the Irish and the English sometimes cannot understand each other. It's a matter of phonetics, consonant omission, and word order in a sentence. Of course, the language factor could hardly have become the main reason for the death of a large number of 3rd class passengers. But in any case, he did not help save people: the Irish did not understand the cries of the Titanic crew well. Another important point: young people, under the age of 32-33, were in all classes; they made up the majority of passengers. It is likely that many of them sailed to America to stay there forever. In Ireland at that time the economy was very bad, millions of young people dreamed of leaving the country overseas.
What else can this say, from a common sense point of view? For example, that so many young people had excellent chances to save themselves. They are stronger and could push the crew, all the weaker ones, away from the lifeboats. After all, at some point it was clear that there would be no witnesses among those who would not get a place in the boat. But they didn't. And if it is true that when the third class passengers got out on deck, there were no longer boats there, then why did so few young people from the second class survive (see bar chart No. 3)? After all, they were among the first to evacuate to the upper deck. However, it cannot be ruled out that the ship’s crew forcibly distributed places in the boats in favor of the weak. But the survivors don’t remember this… It seems that the truth, as always, is somewhere in the middle: the team led the evacuation, saving the weak, and no one interfered with it. Thus, in my opinion, there can only be one logical explanation: they were not saved because they preferred to give their place in the boats to women and children. And the tragic scene of Jack and Rose's farewell (he freezes in the water, she is nearby on a raft) is not just a beautiful romantic episode of the movie “Titanic”, but a fairly truthful description of what happened: men died saving women.
Passenger gender was found to be an important survival factor. Histogram No. 1 in the upper gallery clearly reflects the ratio of surviving men and women. From this interdependence, analysts have drawn a clear conclusion that it was women, and not the rich, who received priority places on lifeboats. The men acted nobly. They all accepted the disaster with honor. Only 25% of them survived. And a little less, 24% of the team members survived. 75% of women and 52% of children were saved. The percentage of children is lower because some children died on the way home on board the Carpathia, from complications caused by severe colds.
The men acted nobly. They all accepted the disaster with honor. Only 25% of them survived. And a little less, 24% of the team members survived. 75% of women and 52% of children were saved. The percentage of children is lower because some children died on the way home on board the Carpathia, from complications caused by severe colds.
Money decides who lives and who dies?
Many believe that most of the dead were third class passengers. Data analysis only partially confirmed this hypothesis. The smallest number of deaths, respectively, the largest number of survivors is observed in the first class. Thus, the largest part of the dead were men from the third class. The same Jackie and Fabrizio, whom we will now probably forever associate with the appearance of DiCaprio and his Hollywood colleague Danny Nucci.
However, the comparison of ticket fare and age. Perhaps it was mostly rich old people who sailed in first class? We saw in Cameron's film these rich men in tailcoats and women in furs, who were put on boats first. In fact, as reflected in the scatter plot below, the number of high-priced tickets is evenly distributed, and mostly among young and middle-aged people. Young people were sailing to America. Both poor and wealthy. All with one goal – to take a closer look at the new country. Segregation on financial grounds when the Titanic crew carried out a rescue operation… These accusations were heard quite often and even led to criminal prosecution of the Titanic crew. But people have thought of a lot. Suffice it to recall the episode of Cameron’s film with the closed gates for the inhabitants of the III class hold cabins: people were breaking up to the boats. But the evil rich people and their hirelings from the ship’s crew did not allow them to do this. However, there are serious questions about this version of events. Firstly, why is the ratio of fatalities to survivors in the second class and first class approximately the same? (see Gallery – chart No. 3). Testimonies from surviving eyewitnesses also indicate that such openly criminal things as blocking the corridors of a sinking ship did not actually happen.
Then what's the matter? But the fact is that the number of boats was simply insufficient for more than 2 thousand people. It never occurred to anyone that such an advanced giant airliner would damage five compartments at once, passing sideways along the sharp edge of an iceberg. That the iceberg will methodically, compartment by compartment, tear off the plating sheets. That not one or two will be filled with water (as would be the case in the case of a strike at one point), but five sealed compartments at once, equipped with steel valves from floor to ceiling, which block the access of water from the punctured place to other parts of the hold.
Timeline of the disaster: there was a lot of time, but few boats.
So, all the women and children were put on the lifeboats, then all the elderly passengers from first and second class. There simply weren’t enough seats for the rest, and they were left on the sinking liner to wait for how it would all end… The ship sank for 2 hours and 40 minutes. And it’s unlikely that anyone could have held back a crowd of thousands of young men from the 3rd class in the corridors of the hold for so long.
The timeline graph shows that after the collision with the iceberg, almost an hour passed before the physical evacuation began, that is, directly boarding the boats and lowering them into the water. This is the so-called “preparatory time,” and it lasted so long because the team hoped until the last that everything would work out and rescue would come.
And, perhaps, it would not be a mistake to say that all this time those remaining on the sinking Titanic first hoped that it would not sink; then that help would arrive, and in the end… that they would survive in the water at a water temperature of minus 2 degrees Celsius, which was in this part of the Atlantic at that time. Truly, hope dies last… Survivors later said that they would never forget the cries of “Lord, why me?!” that stood over the water.
The design of the airliner, the financial model, insufficient preparation – the culprits in the death of more than 1,500 people
All the people on the Titanic behaved nobly and looked death in the face with dignity. Most of the dead are young men. And from different classes. Yes, most of them were from the third class, but there were also the most people floating there. The men did not fight for a place in the boat, but gave the chance to women, children and old people to board first. It also becomes clear that no one deliberately “drowned” 3rd class passengers. However, initially the design of the huge ship, the life-saving devices (and most importantly their number) and imperfect rules, the lack of people on the crew who spoke different languages (while many foreigners were sailing) and a number of other factors doomed the passengers to death in the event of a disaster. The third-class cabins were located in the hold, and the path to the saving upper deck lay for most of the poor passengers through a complex system of narrow corridors and stairs. And first class could go up to the deck via the main staircase and even 2 elevators. And yes, just like today: “…First class passengers leave the liner first, followed by business, and then economy…” Nothing has fundamentally changed since that time.