Introduction

The objective of this project is to test the social norm which says that “women and children first” theory gives them a survival advantage over men and crew in a maritime disasters, and that the ship capitain and it crew member put passengers lives first before theirs. Our study will be done with collected data from different sources. We will compare the survival rate beween males, females, children, and crew member during a disaster. The informations obtained will be compared to the Titanic which serves as a “the prime example of chilvary at sea.”(Elinder & Oscar, 2012)

Background

The S.S. Vestris was a British passenger and cargo liner built by Workman Clarke & Co. Ltd. of Belfast, Ireland, for the Lamport & Holt Line. It maiden voyage was New York to River Plate , and sank slowly during the storm on November 12, 1928 claiming 115 lives out of 326 passengers. The ship did only two voyages.

My Ship History and Details

Based on collected information from Michael L. Grace article on “Disaster at sea SS Vestris”, the ship was launched on May16,1912 and did it first voyage on September19,1912. When the ship sank it had left New York two days prior with “129 passengers and 196 crew” members on bord. A storm caused the maritime disaste on November 12,1928 . It took “100 hours” for the ship to “fell on her side and sink.”(Grace.Michael,2009"

My Ship Compared to Other Ships in the Disaster Data

With a very close observations made through the semester I was able to identify the different ships involved in the disaster and was able to compare them to my ship, SS Vestris. I was able to see the amount of passengers on the ships and categorize them from female, male, and child, to crew members. According to this data my ship, SS Vestris, has the largest female non survival compared to all other ships. Looking at the amount of males and females who were passengers on my ship. In the code below, we will compare it to the 18 ships we are going to study. In addition, I was able to see why the cause for the ships sinking was. I was also able to tell the difference between.

Lets compare the lenght of voyage between the 18 ship.

plot2<-ggplot(ships, aes(y=`Name of Ship`, x=`Length of voyage` ))

plot2 + geom_point() +
   ggtitle("Compare Lengh_of_voyage") +
    geom_point()

Lets compare the number of passengers the ship had during their voyages.

plot2<-ggplot (ships, aes(y = `Name of Ship`, `No. of passengers`) , x=`No. of passengers` )

plot2 + geom_point() +
  ggtitle("Compare No. of voyage during voyages") +
    geom_segment(aes(yend = `Name of Ship`), xend = 0, color = "grey60")

Compare to the Titanic to SS Vestris by using two variables.

ships$No_of_male_passengers <-ships$`No. of passengers`-ships$`No. of women passengers`
ships$No_of_crew <-ships$`Ship size `- ships$`No. of passengers`

ships$Titanic <- ifelse(ships$`Name of Ship` == 'RMS Titanic', TRUE, FALSE) 
ships$Titanic
##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
## [12] FALSE FALSE FALSE FALSE FALSE FALSE FALSE
ships$SS_Vestris <- ifelse(ships$`Name of Ship` == 'SS Vestris', TRUE, FALSE)
ships$SS_Vestris
##  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [12]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE

Throught the code below, lets see what cause the ships to sink.

plot2<-ggplot(ships, aes(x=`Cause`, y=`Name of Ship` ))
plot2+geom_boxplot()

# Theory and Hypotheses

Since the Titanic sank decade ago, I believe that passengers’ chance of survival have been more different than we are thought to believe. Some researchers also think that survival chance is affect or dependent on sex (Gender) and Social class that our human self-survival side dominates during a disaster. Female and children theory are believed to have a greater survival rate than male and crew members. Our hypotheses will be whether or not the Survival theory is justified and backed by real collected data. With Survival rate as our dependent variable and Gender (Female, Male) as our independent variables, crew and children if our ship has data on it.

With the box plots I was able to find out the length of the voyages and compare then to the amount of people survived. We will be making a boxplots with lines and titles for different Survival (x) as our dependent variable.

Data

First we will create the following variables(Survival, Crew,Gender ) in the ships data set. One of the issues we have is that we want to include Children and class, but there is no data for SS Vestris in our analysis but we don’t want to lose all the data on the crew. Let’s look at the relationship between Gender and Crew, and Survival and Gender as variables using crosstab() Sincs the number of passenger varies by ship , and Based on the graph we can conclude that the Titanic had the most Passenger number, SS Vestris has the lowest number. These result can be self explainatory when it comes to why Titanic also have the highest non survival or survival rate.

let use the crosstab() below to compare “crew and gender”, and “survival and gender” on SS Vestris, compare to the Titanic.

crosstab(SS_Vestris, row.vars = "Crew", col.vars = c("Gender"), type = "c")
##     NA     NA     NA     NA
## 1      Gender      0      1
## 2 Crew                     
## 3 0            28.09  92.68
## 4 1            71.91   7.32
## 5 Sum         100.00 100.00
crosstab(SS_Vestris, row.vars = "Survival", col.vars = c("Gender"), type = "c")
##         NA     NA     NA     NA
## 1          Gender      0      1
## 2 Survival                     
## 3 0                35.21  75.61
## 4 1                64.79  24.39
## 5 Sum             100.00 100.00
crosstab(RMS_Titanic, row.vars = "Crew", col.vars = c("Gender"), type = "c")
##     NA     NA     NA     NA
## 1      Gender      0      1
## 2 Crew                     
## 3 0            49.59  95.27
## 4 1            50.41   4.73
## 5 Sum         100.00 100.00
crosstab(RMS_Titanic, row.vars = "Survival", col.vars = c("Gender"), type = "c")
##         NA     NA     NA     NA
## 1          Gender      0      1
## 2 Survival                     
## 3 0                79.33  26.75
## 4 1                20.67  73.25
## 5 Sum             100.00 100.00

Now let’s try to add “Child” as a new variable since my ship has none. We will compare it to the child data on the RMS Titanic

SS_Vestris$Child <- as.numeric(SS_Vestris$Age) <= 15

RMS_Titanic$Child <- as.numeric(RMS_Titanic$Age) <= 15

Results

The crosstab codes above show us that 48% male did not survived, 81.6% female did not survived, 52% male survived, 18.4% female survived, 70.4% crew member survived. on my ship we realize that gender do not influence your chance of survival. We have a higher male survival rate and higher female non survival rate. My ship SS Vestris did the least voyage before it sank and HMS Birkenhead did the most. Titanic had more No. of passengers and SS Vestris. If during the Titanic the social theory “women and children first” was more implimented, after that disaster, it did nod happen, as data shows genre does not influence the survival outcome.

results<-glm(Survival~Gender + Crew, data=SS_Vestris, family = binomial(link = logit))
coef(results)
## (Intercept)      Gender        Crew 
## -0.01253195 -1.19652668  0.89316300
results<-glm(Survival~Gender + Crew, data=RMS_Titanic, family = binomial(link = logit))
coef(results)
## (Intercept)      Gender        Crew 
##  -1.4541904   2.4520885   0.2107338

Above, we used a linear model also called logistic regression to compare Gender and Crew. As we can see been a femal lower your survival change to a negative value, in constrast you have more chance to survive as a crew member. Compared to the Titanic which female survival was higher and crew or male passenger.

Conclusions and Future Research

On my ship female non survival is higher than male, compare to the Titanic. Which means that the theory about “women and children first”, was just connected to the titanic, which basically was the first massive ship that sank. Same as the crew putting the passengers first was not proving or was denied when it comes to my ship.

the difference in number of carried passengers can in my opinion make us believe that one ship lost more than the other one even with a higher number of passengers. let’s make a code and get the summary of the distribution of No. of passengers and list the actual values and compare.

ships$`No. of passengers`
##  [1]  490  151  257  338  788  801  727 1317 1017 1264  902  113  318  129
## [15]  897  796  739  154
summary(ships$`No. of passengers`)
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   113.0   272.2   733.0   622.1   873.0  1317.0

References

Elinder, Mikael and Oscar Erixson. 2012. “Gender, Social Norms, and Survival in Maritime Disasters.” Proceedings of the National Academy of Sciences 109(33):13220-13224. Retrieved August 7, 2016 (http://www.pnas.org/content/109/33/13220). doi: 10.1073/pnas.1207156109.

Frey, Bruno S., David A. Savage and Benno Torgler.. 2010. “Noblesse Oblige? Determinants of Survival in a Life-and-Death Situation.” Journal of Economic Behavior & Organization 74(1–2):1-11. Retrieved August 7, 2016 (http://www.sciencedirect.com/science/article/pii/S0167268110000259). doi: 10.1016/j.jebo.2010.02.005.

Michael L. Grace November 21, 2009 in “STEAMSHIP LINES”