spaceship_presentation.rmd

---
title: "spaceship_presentation"
output: html_document
date: "`r Sys.Date()`"
---

Spaceship Titanic Investigation

Goal: to explore trends relevant to the transportation of passengers on Spaceship Titanic

What we knew beforehand:

1.  almost half of these passengers were transported to another dimension during an anomaly
2.  these records are from before the anomaly, and were recovered from the damaged computer system
3.  it is possible to predict who was transported based off the information given

Interesting columns at first glance:

1.  Transported - boolean - TRUE if transported, FALSE if not
2.  HomePlanet - discrete
3.  CryoSleep - boolean - TRUE if they were in cryo sleep in their cabin for the entire voyage, including when the anomaly occurred
4.  Cabin - deck / num / side (port/starboard) - location on ship may have mattered, and all cryo sleepers would be in their cabins -\> mutate into three separate columns
5.  Age - integer - discrete

What I Did:

1.  Investigated transported status vs. each other variable
2.  Aggregated all amenities together to create a total amount spent by each passenger, as well as a boolean of whether a passenger spent a non-zero amount on amenities
3.  Duplicated a dataset of just awake passengers to look at amenities without the confounding factor of cryo passengers who could not spend on amenities.
4.  Separated cabin location into cabin deck, cabin number, and cabin side, then mapped them each onto cryo sleepers and awake passengers spending \$0 on amenities and therefore more likely to be in their cabin during the anomaly

Final Conclusions:

1.  Home planet, destination, and VIP status seemed to have an insignificant relation to whether or not a passenger was transported
2.  Teenage passengers were more likely than older passengers to be transported
3.  Passengers in cryo-sleep in their cabins made up about 60% of the total transported passengers
4.  Awake passengers who spent on amenities were more likely to be transported than those who did not
5.  Passengers who had their cabins on decks E and G were the least affected, which would include cryo-passengers and likely those who were not using the amenities as well.

Graphs:

1.  Import and Transformations

```{r}
# read csv file into variable
spaceship <- read_csv("files/spaceship-titanic/train.csv")

# create new columns for different components of cabin location
spaceship <- spaceship %>% separate(Cabin, c("Deck", "Number", "Side", sep = "/", remove=FALSE))

# create subset for cryo passengers
spaceship_cryo <- spaceship %>%
filter(CryoSleep == TRUE)

# create subset for awake passengers
spaceship_awake <- spaceship %>%
filter(CryoSleep == FALSE)

# create subset with awake passengers and new total amenity variable
spaceship_awake_amenities <- spaceship_awake %>%
  mutate(spaceship_awake,
       amenities_total = RoomService + FoodCourt + ShoppingMall + Spa + VRDeck)

# create subset with awake passengers, amenity variable, and new amenity boolean
spaceship_awake_amenity_bool <- spaceship_awake_amenities %>%
  mutate(amenity_bool = case_when(amenities_total > 0 ~ TRUE, 
                                    amenities_total == 0 ~ FALSE))

# create subset with awake passengers who did not use amenities
spaceship_awake_no_amenities <- spaceship_awake_aminity_bool %>%
+ filter(amenity_bool == FALSE)

# create subset building block for all passengers with total amenity variable
spaceship_transported_1 <- spaceship %>%
  mutate(spaceship, amenities_total = RoomService + FoodCourt + ShoppingMall + Spa + VRDeck)

# all passengers with both amenities variables
spaceship_transported_2 <- spaceship_transported_1 %>%
  mutate(spaceship, amenities_bool = case_when(ameneties_total > 0 ~ TRUE, amenities_total == 0 ~ FALSE))
```

2.  Graph Age vs. Transported in a Histogram

    Teenage passengers had higher rates of transport than all older passengers.

```{r}
ggplot(data = spaceship)+
  geom_histogram(mapping = aes(x = Age, fill = Transported))
```

3.  Graph Cryo-Passengers vs. Cabin Location vs. Transported

    Most cryo-passengers were transported, but those on decks E and G were safer, especially those on the Port side of the G deck.

```{r}
ggplot(data = spaceship_cryo)+
  geom_bar(mapping = aes(x = Deck, fill = Transported, na.rm = TRUE))+
  facet_wrap(~Side, nrow = 2)
```

4.  Graph Awake Passengers; Whether they Spent \>0\$ on Amenities vs. Transported

    Passengers who spent money on amenities were less likely to be transported than those who did not spend.

```{r}
ggplot(data = spaceship_awake_amenity_bool)+
  geom_bar(mapping = aes(x = amenity_bool, fill = Transported, na.rm = TRUE))
```

5.  Graph Awake Passengers who spent \$0 ; Cabin Location vs. Transported

    Less likely to be transported if they had a cabin on decks E and G, similar trend to cryo sleepers

```{r}
ggplot(data = spaceship_awake_no_ameneties)+
  geom_bar(mapping = aes(x = Deck, fill = Transported, na.rm = TRUE))+
  facet_wrap(~Side, nrow = 2)
```

6.  Graph Cryo-Sleep vs. Transported

    Cryo-Sleepers only made up about 1/3 of total passengers

    ```{r}
    ggplot(data = spaceship)+
      geom_bar(mapping = aes(x = CryoSleep))
    ```

    Most cryo-sleepers were transported, at a much higher rate than those who were awake

    ```{r}
    ggplot(data = spaceship) +
      geom_bar(mapping = aes(x = CryoSleep, fill = Transported))
    ```

Cryo-Sleepers made up 50-60% of those who were transported, even though they were only 1/3 of passengers

```{r}
ggplot(data = spaceship)+
  geom_bar(mapping = aes(x = Transported, fill = CryoSleep))
```