Help for package survivoR

Type:

Package

Title:

Data from all Seasons of Survivor (US) TV Series in Tidy Format

Version:

2.3.8

Description:

Datasets detailing the results, castaways, and events of each season of Survivor for the US, Australia, South Africa, New Zealand, and the UK. This includes details on the cast, voting history, immunity and reward challenges, jury votes, boot order, advantage details, and episode ratings. Use this for analysis of trends and statistics of the game.

Depends:

R (≥ 4.1.0)

Imports:

tidyr, ggplot2, stringr, magrittr, glue, shiny, purrr, dplyr, crayon, readr, shinycssloaders, lubridate, DT, shinyjs

Suggests:

forcats, testthat (≥ 3.0.0)

License:

MIT + file LICENSE

URL:

https://github.com/doehm/survivoR

BugReports:

https://github.com/doehm/survivoR/issues

Encoding:

UTF-8

LazyData:

true

RoxygenNote:

7.3.2

Config/testthat/edition:

NeedsCompilation:

Packaged:

2025-10-10 22:24:08 UTC; danie

Author:

Daniel Oehm [aut, cre], Carly Levitz [ctb]

Maintainer:

Daniel Oehm <danieloehm@gmail.com>

Repository:

CRAN

Date/Publication:

2025-10-10 22:50:09 UTC

Pipe operator

Description

See magrittr::%>% for details.

Usage

lhs %>% rhs

Arguments

lhs

A value or the magrittr placeholder.

rhs

A function call using the magrittr semantics.

Value

The result of calling rhs(lhs).

Adds alive flag

Description

Adds a logical flag if the castaway is alive at the start or end of an episode

Usage

add_alive(df, .ep, .at = "end")

Arguments

df

Data frame. Must contain version_season and castaway.

.ep

Episode to evaluate the flag.

.at

Either 'start' or 'end'. If 'start' the flag will indicate who is alive at the start of the episode. If 'end' it will indicate who is alive at the end of the episode i.e. after tribal council.

Value

A data frame with a new column alive.

Examples


library(survivoR)
library(dplyr)

df <- confessionals |>
  filter_us(47) |>
  add_alive(12)

df |>
  filter(alive) |>
  group_by(castaway) |>
  summarise(n = sum(confessional_count))

Adds BIPOC

Description

Adds a BIPOC to the data frame. If any African American, Asian American, Latin American, or Native American is TRUE then BIPOC is TRUE.

Usage

add_bipoc(df)

Arguments

df

Data frame. Requires castaway_id.

Value

Data frame with BIPOC added.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_bipoc()

Add castaway

Description

Adds castaway to a data frame. Input data frame must have castaway_id.

Usage

add_castaway(df)

Arguments

df

Data frame. Requires castaway_id.

Value

Data frame with castaway.

Examples


library(survivoR)
library(dplyr)

df_no_castaway <- confessionals |>
  filter_us(47) |>
  group_by(castaway_id) |>
  summarise(n = sum(confessional_count))

df_no_castaway |>
  add_castaway()

Add demographics

Description

Add demographics that includes age, gender, race/ethnicity, and lgbtqia+ status to a data frame with castaway_id.

Usage

add_demogs(df)

Arguments

df

Data frame. Requires castaway_id.

Value

Data frame with castaway added to it.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_demogs()

Add winner

Description

Adds a winner flag to the data set.

Usage

add_finalist(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame with a logical flag for the winner

Examples


library(survivoR)
library(dplyr)

confessionals |>
  add_winner()

Add full name

Description

Adds full name to the data frame. Useful for plotting and making tables.

Usage

add_full_name(df)

Arguments

df

Data frame. Requires castaway_id.

Value

Data frame with full name.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_full_name()

Add gender

Description

Adds gender to a data frame

Usage

add_gender(df)

Arguments

df

Data frame. Requires castaway_id.

Value

Data frame with gender added to it.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_gender()

Add jury member

Description

Adds a jury member flag to the data set.

Usage

add_jury(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame with a logical flag for the jury members

Examples


library(survivoR)
library(dplyr)

confessionals |>
  add_jury()

Add LGBTQIA+ status

Description

Adds the LGBTQIA+ flag to the data frame.

Usage

add_lgbt(df)

Arguments

df

Data frame. Requires castaway_id and version_season.

Value

Data frame with the LGBTQIA+ flag added.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_lgbt()

Add result

Description

Adds the result and place to the data frame.

Usage

add_result(df)

Arguments

df

Data frame. Requires castaway_id and version_season.

Value

Data frame with result and place added.

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_result()

Add tribe

Description

Adds tribe to a data frame for a specified stage of the game e.g. original, swapped, swapped_2, etc.

Usage

add_tribe(df, .tribe_status = "Original")

Arguments

df

Data frame. Requires version_season and castaway_id,

.tribe_status

Tribe status e.g. original, swapped, swapped_2, etc.

Value

Data frame with tribe added.

Examples


library(survivoR)
library(dplyr)

confessionals |>
  add_tribe()

Add tribe colour

Description

Add tribe colour to the data frame. Useful for preparing the data for plotting with ggplot2.

Usage

add_tribe_colour(df, .tribe_status = "Original")

Arguments

df

Data frame. Requires version_season and tribe.

.tribe_status

Tribe status e.g. original, swapped, swapped_2, etc.

Value

Data frame with tribe_colour added

Examples


library(survivoR)
library(dplyr)

get_cast("US47") |>
  add_tribe() |>
  add_tribe_colour()

Add winner

Description

Adds a winner flag to the data set.

Usage

add_winner(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame with a logical flag for the winner

Examples


library(survivoR)
library(dplyr)

confessionals |>
  add_winner()

Advantage Details

Description

A dataset containing the details and characteristics of each idol and advantage. This maps to advantage_movement

Usage

advantage_details

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
advantage_id: The ID / primary key of the advantage
advantage_type: Advantage type e.g. hidden immunity idol, extra vote, steal a vote, etc
clue_details: Details if a clue existed for the advantage and if so where was the clue found
location_found: The location the idol or advantage was found
conditions: Extra details about the unique conditions of the idol or advantage

Details

There are split idols which need to be combined to be played. In these case the first one found is given an ID. The second or subsequent parts are given the same ID with a trailing letter. For example in season 40 Denise found an idol that was split (USHI4002). Later she found the other half (USHI4002b). When played the second half is considered to have 'absorbed' into the first idol. The first idol found is always considered the primary idol.

Advantage Movement

Description

A dataset containing the movement details of each advantage or hidden immunity idol. Each row is considered an event e.g. the idol was found, played, etc. If the advantage changed hands it records who received it. The logical flow is identified by the sequence_id.

Usage

advantage_movement

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
castaway: Name of the castaway involved in the event e.g. found, played, received, etc.
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
advantage_id: The ID / primary key of the advantage
sequence_id: The sequence of events. For example sequence_id == 1 usually means the advantage was found. Each subsequent event follows the sequence_id
day: The day the event occurred
episode: The episode the event occurred
event: The event e.g. the advantage was found, played, received, etc
played_for: If the advantage or idol was played this records who it was played for
played_for_id: the ID for who the advantage or idol was played for
success: If the play was successful or not. Only relevant for advantages since playing a hidden immunity idol is always successful in terms of saving who it was played for.
votes_nullified: In the case of hidden immunity idols this is the count of how many votes were nullified when played
sog_id: Stage of game ID for joining to vote_history and challenge_results

Survivor Auction Details

Description

The details of the items purchased at the Survivor Auction. survivor_auction is at the castaway level and includes all castaways whether or not they purchased an item and auction_details is at the item level.

Usage

auction_details

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
item: Item number
item_description: Item description
category: The item category. See details for more.
castaway: Castaway
castaway_id: Castaway ID
covered: If the item was covered or not
cost: The amount paid for the item
money_remaining: How much money the castaway has remaining
auction_num: If the same item is auctioned for a second time it has a value of 2
participated: The names of castaways that could participate in the purchased item e.g. sharing a tub of peanut butter with the tribe
notes: Additional notes
alternative_offered: If and alternative was offered to the player after purchase
alternative_accepted: If they accepted the alternative offer
other_item: Description of the refused item
other_item_category: Category of the refused item

Details

Each item has been categorised into 5 main categories:

Food and drink: The most common item. It may be simply food or drink, not necessarily both.
Comfort: Things like a shower, toothpaste, etc
Letters from home
Advantage: Could be a clue to a hidden immunity idol, advantage in the next challenge, or in the current auction
Bad item: The not good item, typically one of the covered items. Whether or not it's actually bad is subjective, but where someone is hoping for pizza and gets bat soup I consider it a bad item.

Source

https://survivor.fandom.com/wiki/Main_Page

Boot mapping

Description

A mapping table for easily filtering to the set of castaways that are still in the game after a specified number of boots.

Usage

boot_mapping

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
order: The number of boots that there have been in the game e.g. if order == 2 there have been 2 boots in the game so far and there are N-2 castaways left in the game
final_n: The final number of castaways e.g. you can filter to the final 4 by filter(boot_mapping, final_n == 4). There are missing values where players have returned to the game. This means there are multiple stages of the game where there is a different make up of the final 8, for example. This field just takes the last set so that you can filter for final_n and it will return a single set of castaways.
n_boots: Similar to final_n but the number of boots in the game. This is different to order where order counts if someone has been booted twice. n_boots is simply the number of people in the season minus the final_n.
sog_id: Stage of game ID for joining to vote_history and challenge_results
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
castaway: Name of the castaway
tribe: Name of the tribe the castaway was on
tribe_status: The status of the tribe e.g. original, swapped, merged, etc. See details for more
game_status: Logical flag to identify if the castaway is currently in the game. If FALSE the castaway is on Redemption Island or Edge of Extinction.

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series) https://survivor.fandom.com/wiki/Main_Page

Boot order

Description

Similar to the castaways dataset, boot_order records the order in which castaways left the game. If a player was voted out of the game, returned to the game like seasons such as Redemption Island, and then voted out again, they will have two rows in the table.

Usage

boot_order

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: Season number
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
castaway: Name of castaway. Generally this is the name they were most commonly referred to or nickname e.g. no one called Coach, Benjamin. He was simply Coach
episode: Episode number
day: Number of days the castaway survived. A missing value indicates they later returned to the game that season
order: Boot order. Order in which castaway was voted out e.g. 5 is the 5th person voted of the island
result: Final result

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series); https://survivor.fandom.com/wiki/Main_Page; ack_ features from Matt Stiles https://github.com/stiles/survivor-voteoffs

Examples

library(dplyr)
castaways %>%
  filter(season == 40)

Castaway details

Description

A dataset containing details on the castaways for each season

Usage

castaway_details

Format

This data frame contains the following columns:

castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
full_name: Full name of the castaway
full_name_detailed: A detailed version of full_name for plotting e.g. 'Boston' Rob Mariano
castaway: Short name of the castaway. Name typically used during the season. Sometimes there are multiple people with the same name e.g. Rob C and Rob M in Survivor All-Stars. This field takes the most verbose name used
last_name: Last name
date_of_birth: Date of birth
date_of_death: Date of death
gender: Gender of castaway
african: TRUE if African-American or African-Canadian as per https://survivor.fandom.com/wiki/Main_Page
asian: TRUE if Asian-American or Asian-Canadian as per https://survivor.fandom.com/wiki/Main_Page
latin_american: TRUE if Latin-American as per https://survivor.fandom.com/wiki/Main_Page
native_american: TRUE if Native-American as per https://survivor.fandom.com/wiki/Main_Page
bipoc: Black, Indigenous, or Person of Colour
lgbt: LGBTQIA+ status as listed on the survivor wiki.
personality_type: The Myer-Briggs personality type of the castaway
occupation: Occupation
collar: White Collar, Blue Collar, No Collar, or Unknown. WARNING: this is experimental. The classification has been made using a model and results may be inconsistent.
three_words: Answer to the question "three words to describe you?"
hobbies: Answer to the question "what are you favourite hobbies?"
pet_peeves: Answer to the question "what are your pet peeves?"
race: Race (if known)
ethnicity: Ethnicity (if known)

Details

Race and ethnicity data is included if known and can point to a source, rather than making an assumption about an individual.

poc has been deprecated and replaced with bipoc which is now logical and only for the US. bipoc is TRUE if any of african, asian, latin_american, or native_american is TRUE.

Source

https://survivor.fandom.com/wiki/Main_Page, https://www.personality-database.com/

Examples

library(dplyr)
castaway_details |>
  count(gender)

Castaway scores

Description

The challenge, vote history, and advantage scores are a measure of success or proficiency. Higher the better. See details.

Usage

castaway_scores

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
castaway_id: Castaway ID
castaway: Castaway
score_overall: Overall score for the castaway. Use this to compare players across seasons
score_outwit: Outwit score
score_outplay: Outplay score
score_outlast: Outlast score
score_result: Score based on the placing in the season
score_jury: Jury score based on the proportional number of votes recieved
score_vote: Voting score for the season as a proportion of their potential max score
score_adv: Advantage score. Same as p_score_adv
score_inf: Influence score. Aim at capturing influence in the game e.g. higher the score, the higher their importance to the narrative of the episode/season
r_score_chal_all: Challenge score for all challenges
r_score_chal_immunity: Challenge score for immunity challenges
r_score_chal_reward: Challenge score for reward challenges
r_score_chal_tribal: Challenge score for tribals challenges
r_score_chal_tribal_immunity: Challenge score for tribal immunity
r_score_chal_tribal_reward: Challenge score for tribal reward
r_score_chal_individual: Challenge score for individual challenges
r_score_chal_individual_immunity: Challenge score for individual immunity
r_score_chal_individual_reward: Challenge score for individual reward
r_score_chal_team: Challenge score for team challenges
r_score_chal_team_reward: Challenge score for team reward
r_score_chal_team_immunity: Challenge score for team immunity
r_score_chal_duel: Challenge score for duels
p_score_chal_all: Challenge score for all challenges
p_score_chal_immunity: Challenge score for immunity challenges
p_score_chal_reward: Challenge score for reward challenges
p_score_chal_tribal: Challenge score for tribals challenges
p_score_chal_tribal_immunity: Challenge score for tribal immunity
p_score_chal_tribal_reward: Challenge score for tribal reward
p_score_chal_individual: Challenge score for individual challenges
p_score_chal_individual_immunity: Challenge score for individual immunity
p_score_chal_individual_reward: Challenge score for individual reward
p_score_chal_team: Challenge score for team challenges
p_score_chal_team_reward: Challenge score for team reward
p_score_chal_team_immunity: Challenge score for team immunity
p_score_chal_duel: Challenge score for duels
n_votes_received: Number of votes received
n_successful_boots: Number of successful boots
p_successful_boot: Percentage of successful boots. Tribals where the castaway did not have a vote are removed from the calculation
n_tribals: Number of tribals attended
n_tribals_with_vote: Number of tribals attended where the player had a vote
r_score_vote: Vote history score
p_score_vote: Proportional vote history score for the season
r_score_adv: Advantage scores
p_score_adv: Scaled advantage scores - min max bewtween 0 and 1
n_adv_found: Number of advantages found
n_idols_found: number of idols found
n_adv_played: Number of advantages played
n_adv_not_played: Number of advantages not played
n_voted_out_with_adv: Number of advantages they were voted out with
n_voted_out_with_idol: Number of idols they were voted out with

Details

The difference between the r_ and p_ sores is the r_ is the raw score which is the residual assuming equal probability. Higher the better. p_ is the residual converted to a probability.

Castaways

Description

A dataset containing details on the results for every castaway and season

Usage

castaways

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: Season number
full_name: Full name of the castaway
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
castaway: Name of castaway. Generally this is the name they were most commonly referred to or nickname e.g. no one called Coach, Benjamin. He was simply Coach
age: Age of the castaway during the season they played
city: City of residence during the season they played
state: State of residence during the season they played
episode: Episode number
day: Number of days the castaway survived. A missing value indicates they later returned to the game that season
order: Boot order. Order in which castaway was voted out e.g. 5 is the 5th person voted of the island
result: Final result
place: Place as a number e.g. Sole Survivor is 1, runner-up 2, etc
jury_status: Jury status
original_tribe: Original tribe name
finalist: Logical. TRUE if the castaway was a finalists
jury: Logical. TRUE if the castaway was a jury member
winner: Logical. TRUE if the castaway was the winner
acknowledge: Did the contestant acknowledge their teammates in one of these specific ways after snuffing — or just walk away?
ack_gesture: for any physical gestures towards the tribe after torch snuffing. Types: wave, nod, wink, bow or prayer sign with hands
ack_look: For making eye contact with one or more members of the tribe after torch snuffing
ack_smile: For smiling at the tribe after torch snuffing
ack_speak: For any verbal communication directed at the tribe after torch snuffing
ack_quote: What, if anything, the contestant said. Direct quotes only.
ack_score: The score is derived from the four subcategories of acknowledgment: words, look, gesture, and smile. Each true value in these categories adds 1 to the score.

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series); https://survivor.fandom.com/wiki/Main_Page; ack_ features from Matt Stiles https://github.com/stiles/survivor-voteoffs

Examples

library(dplyr)
castaways %>%
  filter(season == 40)

Challenge Description

Description

A dataset detailing the challenges played and the elements they include over all seasons of Survivor

Usage

challenge_description

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
challenge_id: Primary key
challenge_number
challenge_type
name: The name of the challenge
recurring_name: Challenges can go by different names but are often associated with a particular challenge or element of a challenge. Some challenges use combinations of other challenges so it's not perfect but consistent with the wiki page. Use recurring_name to analyse how often a challenge has been run.
description: Description of the challenge
reward: Description of the reward
additional_stipulation: Some challenges come with various rules or success criteria. This states those conditions.
race: If the challenge is a race between tribes, teams or individuals
endurance: If the challenge is an endurance event e.g. last tribe, team, individual standing
turn_based: If the challenge is turn bases i.e. conducted in rounds
puzzle: If the challenge contains a puzzle element
puzzle_slide: If the challenge contained a slide puzzle
puzzle_word: If the challenge contained a word puzzle
precision: If the challenge contains a precision element e.g. shooting an arrow, hitting a target, etc
precision_catch: If the challenge featured catching a ball or similar
precision_roll_ball: If the challenge featured rolling a ball
precision_slingshot: If the challenge featured a slingshot, either the large version or handheld version
precision_throw_balls: If the challenge featured throwing balls
precision_throw_coconuts: If the challenge featured throwing coconuts
precision_throw_rings: if the challenge featured throwing rings
precision_throw_sandbags: if the challenge featured throwing sandbags
strength: If the challenge has a strength based
balance: If the challenge contains a balancing element. My refer to the player balancing on something or the player balancing an object on something e.g. The Ball Drop
balance_beam: If the challenge featured a balance beam of similar they were required to balance on
balance_ball: If the challenge featured balancing a ball on something
food: If the challenge contains a food element e.g. the food challenge, biting off chunks of meat
knowledge: If the challenge contains a knowledge component e.g. Q and A about the location
memory: If the challenge contains a memory element e.g. memorising a sequence of items
fire: If the challenge contains an element of fire making / maintaining
water: If the challenge is held, in part, in the water
water_swim: If castaways had to swim in the challenge
water_paddling: If castwways were required to paddle a boat or similar
obstacle_blindfolded: If the challenge required castaways to be blindfolded
obstacle_cargo_net: If the challenge featured a cargo net
obstacle_chopping: If castaways were required to chop a rope or similar
obstacle_combination_lock: If the challenge feature a combination lock
obstacle_digging: If the challenge involved digging
obstacle_knots: If the challenge involved untying knots
obstacle_padlocks: If the challenge featured opening padlocks
mud: If the challenge required castaways to get covered in mud

Details

This data set contains the name, description, and descriptive features for each challenge where it is known. Challenges can go by different names so have included the unique name and the recurring challenge name. These are taken directly from the Survivor Wiki. Sometimes there can be variations made on the challenge but go but the same name, or the challenge is integrated with a longer obstacle. In these cases the challenge may share the same recurring challenge name but have a different challenge name. Even if they share the same names the description could be different.

The features of each challenge have been determined largely through string searches of key words that describe the challenge. It may not be 100% accurate due to the different and inconsistent descriptions but in most part they will provide a good basis for analysis.

If any descriptive features need altering please let me know in the issues.

For updated data please see the git version.

Source

https://survivor.fandom.com/wiki/Category:Challenges https://survivor.fandom.com/wiki/Main_Page

Examples

library(dplyr)
library(tidyr)
challenge_description

Challenge Results

Description

A dataset detailing the challenges played including reward and immunity challenges.

Usage

challenge_results

Format

This data frame contains the following columns

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
n_boots: The number of boots that there have been in the game e.g. if n_boots == 2 there have been 2 boots in the game so far and there are N-2 castaways left in the game
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
castaway: Name of castaway. Generally this is the name they were most commonly referred to or nickname e.g. no one called Coach, Benjamin. He was simply Coach
outcome_type: Whether the challenge is individual or tribal. Some individual reward challenges may involve multiple castaways as the winner gets to choose who they bring along
tribe: Current tribe the castaway is on
tribe_status: The status of the tribe e.g. original, swapped, merged, etc. See details for more
challenge_type: The challenge type e.g. immunity, reward, etc
challenge_id: Primary key to the challenge_description data set which contains features of the challenge
result: Result of challenge
result_notes: Additional notes about the result of the challenge
order_of_finish: Order of finish for tribal challenges. Useful when there are 3 or more tribes to see who actually came first, second and who lost the challenge.
chosen_for_reward: If after the reward challenge the castaway was chosen to participate in the reward
sit_out: TRUE if they sat out of the challenge or FALSE if they participate
team: Team allocation when they are split into teams
sog_id: Stage of game ID for joining to boot_mapping and vote_history

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series) https://survivor.fandom.com/wiki/Main_Page

Examples

library(dplyr)
library(tidyr)
challenge_results %>%
  filter(season == 40)

Challenge Summary

Description

A dataset summarising challenge_results

Usage

challenge_summary

Format

This data frame contains the following columns

category: The category of the challenge e.g. tribal, individual, individual immunity, duel, etc. This makes it easy to split out the difference types of challenges and avoid complications such as 'Team / Individual' challenges where there is a dependent outcome structure. Join to challenge_results using challenge_id, version_season and castaway_id
version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
challenge_id: Primary key to the challenge_description data set which contains features of the challenge
challenge_type: The challenge type e.g. immunity, reward, etc
outcome_type: Whether the challenge is individual or tribal. Some individual reward challenges may involve multiple castaways as the winner gets to choose who they bring along
tribe: Current tribe the castaway is on
castaway: Name of castaway. Generally this is the name they were most commonly referred to or nickname e.g. no one called Coach, Benjamin. He was simply Coach
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
n_entities: Number of entities competing for the win e.g. the number of tribes, teams, or people.
n_winners: Number of winners (or winning entities) e.g. if there are two tribes there is only one winning tribe, if there are three tribes like the new era there are two winning tribes and one that goes to tribal council.
won: number of challenges won

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series) https://survivor.fandom.com/wiki/Main_Page

Examples

library(dplyr)
library(tidyr)
challenge_summary %>%
  filter(version_season == 46)

Confessionals

Description

A dataset containing the count of confessionals per castaway per episode. A confessional is when the castaway is speaking directly to the camera about their game.

Usage

confessionals

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
castaway: Name of the castaway
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
confessional_count: The count of confessionals for the castaway during the episode
confessional_time: The total time for all confessionals for the episode for each castaway
exp_count: The expected confessional counts. See details.
exp_time: The expected confessional time. See details.

Details

Confessional data has been counted by contributors of the survivoR R package and consolidated with external sources. The aim is to establish consistency in confessional counts in the absence of official sources. Given the subjective nature of the counts and the potential for clerical error no single source is more valid than another. Therefore, it is reasonable to average across all sources.

In the case of double or extended episodes, if the episode only has one title it is considered a single episode. This means the average number of confessionals per person is likely to be higher for this episode given it's length. If there are two episode titles the confessionals are counted for the appropriate episode. This is to ensure consistency across all other datasets.

In the case of recap episodes, this episode is left blank.

The fields exp_count and exp_time are the expected values given the game events. For example players that attend tribal council, find advantages, go on rewards, and if it's their boot episode typically get more confessionals - we should expect them to get more as well. This enables analysis of the observed and expected confessionals and those that received more or fewer than expected.

If you also count confessionals, please get in touch and I'll add them into the package.

Episodes

Description

A dataset containing details for each episode

Usage

episodes

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: Season number
episode_number_overall: The cumulative episode number
episode: Episode number for the season
episode_title: Episode title
episode_label: A standardised episode label
episode_date: Date the episode aired
episode_length: Episode length in minutes
viewers: Number of viewers (millions) who tuned in
imdb_rating: IMDb rating for the episode on a scale of 0-10
n_ratings: The number of ratings submitted to IMDb
episode_summary: Description of the episode from wikipedia

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series)

Filter Alive

Description

Filters a given dataset to those that are still alive in the game at the start or end of a user specified episode.

Usage

filter_alive(df, .ep = NULL, .at = "end")

Arguments

df

Input data frame. Must have version_season

.ep

Episode. This will filter the castaways that are still alive at either the start or end of the episode.

.at

Either 'start' or 'end' to filter those who are still alive in the game.

Value

A data frame filtered to castaways who are alive.

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_us(47) |>
  filter_alive(12) |>
  group_by(castaway) |>
  summarise(n = sum(confessional_count))

Filter final `n`

Description

Filters to the final n players e.g. the final 4.

Usage

filter_final_n(df, .final_n)

Arguments

df

Input data frame. Must have version_season

.final_n

An integer to represent the final n.

Value

A data frame filtered to only the final n

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_us(47) |>
  filter_final_n(6) |>
  group_by(castaway) |>
  summarise(n = sum(confessional_count))

Filter to finalists

Description

Filters a data set to the finalists of a given season.

Usage

filter_finalist(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame filtered to the finalists

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_finalist()

Filter to jury

Description

Filters a data set to the jury members of a given season.

Usage

filter_jury(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame filtered to the jury members

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_jury()

Filter to the new era seasons

Description

Filters a data set to all New Era seasons.

Usage

filter_new_era(df)

Arguments

df

Data frame. Must include version and season.

Value

A data frame filtered to the New Era seasons.

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_new_era() |>
  distinct(version_season)

Filter to US seasons

Description

Filter a data set to a specified set of US season or list of seasons. A shorthand version of filter_vs() for the US seasons.

Usage

filter_us(df, .season = NULL)

Arguments

df

Data frame. Must include version and season.

.season

Season or vector of seasons. If NULL if will filter to all US seasons.

Value

Data frame filtered to the specified US seasons

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_us(47)

Filter version season

Description

Filters a data set to a specified version season or list of version seasons.

Usage

filter_vs(df, .vs)

Arguments

df

Data frame. Must have version_season

.vs

Version season.

Value

Data frame filtered to the specified version seasons

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_vs("US47")

Filter to winners

Description

Filters a data set to the winners of a given season.

Usage

filter_winner(df)

Arguments

df

Data frame. Requires version_season and castaway_id.

Value

A data frame filtered to the winners

Examples


library(survivoR)
library(dplyr)

confessionals |>
  filter_winner()

Get cast for a season

Description

For a given season (or seasons) the function will return a data frame of the cast.

Usage

get_cast(.vs)

Arguments

.vs

Version season. Can be a vector of version_season values.

Value

A data frame

Examples

library(survivoR)

get_cast("US47")

Castaway images

Description

Returns the URL for the image of the specified castaways by their castaway_id and season / version they were in

Usage

get_castaway_image(castaway_ids, version_season)

Arguments

castaway_ids

Castaway ID

version_season

Version season key for the season they played

Value

Character vector of URLs

Examples

library(dplyr)

survivoR::castaways %>%
  filter(version_season == "US42") %>%
  mutate(castaway_image = get_castaway_image(castaway_id, version_season))

Confessional time

Description

Takes the output of the times recorded from the Shiny app and aggregates to the final confessional times and confessional counts. confessional_time is the total duration in seconds for the episode. confessional_count is the number of confessionals recorded to be at least 10 seconds apart.

Usage

get_confessional_timing(x, .vs, .episode, .mda = 3)

Arguments

x

Either a data frame or path(s) to the csv file containing all the time stamps from the Shiny app

.vs

Version season

.episode

Episode

.mda

Missing duration adjustment (MDA) in seconds. If either start or stop is missing from the records, the missing value is imputed with a 3 second adjustment by default.

Value

data frame

Examples

# After running app and recording confessionals, run...
# Example from a saved timing file

library(readr)

path <- system.file(package = "survivoR", "extdata/US4412.csv")
df_us4412 <- read_csv(path)
get_confessional_timing(df_us4412, .vs = "US44", .episode = 12)

Journeys

Description

Details on who went on Journeys, what they won or if they lost their vote.

Usage

journeys

Format

This data frame contains the following columns:

version: Country code for the version of the show
season: The season number
version_season: Version season key
episode: Episode
sog_id: Stage of game ID
castaway_id: Castaway ID
castaway: Castaway
reward: The thing they won (or lost)
lost_vote: Logical. If they lost their vote
game_played: The game they played on the journey
chose_to_play: If they chose to play or not
event: The event that occured e.g. risked vote, lost vote

Jury votes

Description

A dataset containing details on the final jury votes to determine the winner for each season

Usage

jury_votes

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
castaway: Name of the castaway
finalist: The finalists for which a vote can be placed
vote: Vote. 0-1 variable for easy summation
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
finalist_id: The ID of the finalist for which a vote can be placed. Consistent with castaway ID

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series)

Examples

library(dplyr)
jury_votes %>%
  filter(season == 40) %>%
  group_by(finalist) %>%
  summarise(votes = sum(vote))

Launch Confessional App

Description

Launches the confessional timing app in either a browser or viewer. Default is set to browser. The user is required to provide a path for which the time stamps are recorded.

Usage

launch_confessional_app(browser = TRUE, path = NULL, write = TRUE)

Arguments

browser

Open in browser instead of viewer. Default TRUE

path

Parent directory for output files. Default is a sub-folder 'confessional-timing' in the current working directory.

write

Write to disc. Default TRUE.

Value

An active R shiny application

Examples

## Only run this example in interactive R sessions

if(interactive()) {

  # launch app
  # launch_confessional_app()

}

Read episode transcripts

Description

Read the episode transcripts from Github. File is large and not explicitly part of the package. Data is update by Matt Stiles.

Usage

load_episode_transcripts()

Value

A data frame of episode transcripts

Examples

# Run
# load_episode_transcripts()
# to load all transcripts

Screen Time

Description

A dataset summarising the screen time of contestants on the TV show Survivor. Currently only contains Season 1-4 and 42.

Usage

screen_time

Format

This data frame contains the following columns:

version_season: Version season key
episode: Episode number
castaway_id: ID of the castaway (primary key). Also includes two special IDs of host (i.e. Jeff Probst) or unknown (the image detection couldn't identify the face with sufficient accuracy)
screen_time: Estimated screen time for the individual in seconds.

Details

Individuals' screen time is calculated, at a high-level, via the following process:

Frames are sampled from episodes on a 1 second time interval
MTCNN detects the human faces within each frame
VGGFace2 converts each detected face into a 512d vector space
A training set of labelled images (1 for each contestant + 3 for Jeff Probst) is processed in the same way to determine where they sit in the vector space. TODO: This could be made more accurate by increasing the number of training images per contestant.
The Euclidean distance is calculated for the faces detected in the frame to each of the contestants in the season (+Jeff). If the minimum distance is greater than 1.2 the face is labelled as "unknown". TODO: Review how robust this distance cutoff truly is - currently based on manual review of Season 42.
A multi-class SVM is trained on the training set to label faces. For any face not identified as "unknown", the vector embedding is run into this model and a label is generated.
All labelled faces are aggregated together, with an assumption of 1 full second of screen time each time a face is seen.

Season palettes

Description

A dataset containing palettes generated from the season logos

Usage

season_palettes

Format

This nested data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
palette: The season palette

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series)

Season summary

Description

A dataset containing a summary of all seasons of Survivor

Usage

season_summary

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: Season number
season_name: Season name
n_cast: Number of cast in the season
n_tribes: Number of starting tribes
n_finalists: Number of finalists
n_jury: Number of jury members
location: Location of the season
country: Country the season was held
tribe_setup: Initial setup of the tribe e.g. heroes vs Healers vs Hustlers
full_name: Full name of the winner
winner_id: ID for the winner of the season (primary key)
winner: Winner of the season
runner_ups: Runner ups for the season. Either one or two runner ups as a string
final_vote: Final vote allocation. See the jury_votes data set for better aggregation of this data
timeslot: Timeslot of the show in the US
premiered: Date the first episode aired
ended: Date the season ended
filming_started: Date the filming of the season started
filming_ended: Date the filming ended (39 or 42 days after the start)
viewers_premiere: Number of viewers (millions) who tuned in for the premier
viewers_finale: Number of viewers (millions) who tuned in for the finale
viewers_reunion: Number of viewers (millions) who tuned in for the reunion
viewers_mean: Average number of viewers (millions) who tuned in over the season
rank: Season rank

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series) https://survivor.fandom.com/wiki/Main_Page

Still alive

Description

Finds the set of players that are still alive at either the start or end of an episode, or given a set number of boots.

Usage

still_alive(.vs, .ep = NULL, .n_boots = NULL, .at = "end")

Arguments

.vs

Version season

.ep

Episode to evaluate who is alive.

.n_boots

Number of boots

.at

Either 'start' or 'end'. If 'start' the flag will indicate who is alive at the start of the episode. If 'end' it will indicate who is alive at the end of the episode i.e. after tribal council.

Value

Data frame

Examples


library(survivoR)
library(dplyr)

# at the end of the episode
still_alive("US47", 12)

# at the start of the episode
still_alive("US47", 12, .at = "start")

Survivor Auction

Description

A dataset showing who attended the Survivor Auction during the seasons they were held. survivor_auction is at the castaway level and includes all castaways whether or not they purchased an item and auction_details is at the item level.

Usage

survivor_auction

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
n_boots: The number of boots so far in the game
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU (TBA).
castaway: Name of castaway. Generally this is the name they were most commonly referred to or nickname e.g. no one called Coach, Benjamin. He was simply Coach
tribe_status: The status of the tribe e.g. original, swapped, merged, etc. See details for more
tribe: Tribe name
currency: Currency
total: Total amount either given to or found by the castaway

Source

https://survivor.fandom.com/wiki/Main_Page

Survivor season colour palette

Description

ggplot2 scales for each season of Survivor.

Usage

survivor_pal(season = NULL, scale_type = "d", reverse = FALSE, ...)

scale_fill_survivor(season = NULL, scale_type = "d", reverse = FALSE, ...)

scale_colour_survivor(season = NULL, scale_type = "d", reverse = FALSE, ...)

Arguments

season

Season number

scale_type

Discrete or continuous. Input d or c.

reverse

Logical. Reverse the palette?

...

Other arguments passed on to methods.

Details

Palettes are created from the logo for the season.

Value

Scale functions for ggplot2

Examples

library(ggplot2)
library(dplyr)
mpg %>%
  ggplot(aes(x = displ, fill = manufacturer)) +
  geom_histogram(colour = "black") +
  scale_fill_survivor(40)

Tribe colours

Description

A dataset containing the tribe colours for each season

Usage

tribe_colours

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
tribe: Tribe name
tribe_colour: Colour of the tribe
tribe_status: Tribe status e.g. original, swapped or merged. In the instance where a tribe is formed at the swap by splitting 2 tribes into 3, the 3rd tribe will be labelled 'swapped'

Source

https://survivor.fandom.com/wiki/Tribe

Examples

library(ggplot2)
library(dplyr)
library(forcats)
df <- tribe_colours %>%
  group_by(season) %>%
  mutate(
    xmin = 1,
    xmax = 2,
    ymin = 1:n(),
    ymax = ymin + 1
  ) %>%
  ungroup() %>%
  mutate(
    font_colour = ifelse(tribe_colour == "#000000", "white", "black")
  )
ggplot() +
  geom_rect(data = df,
    mapping = aes(xmin = xmin, xmax = xmax, ymin = ymin, ymax = ymax),
    fill = df$tribe_colour) +
  geom_text(data = df,
    mapping = aes(x = xmin+0.5, y = ymin+0.5, label = tribe),
    colour = df$font_colour) +
  theme_void() +
  facet_wrap(~season, scales = "free_y")

Tribe mapping

Description

A mapping for castaways to tribes for each day (day being the day of the tribal council) This is useful for observing who is on what tribe throughout the game.

Usage

tribe_mapping

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
day: The day of the tribal council
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
castaway: Name of the castaway
tribe: Name of the tribe the castaway was on
tribe_status: The status of the tribe e.g. original, swapped, merged, etc. See details for more

Details

Each season by episode and day holds a complete list of castaways still in the game and which tribe they are on. Moving through each day you can observe the changes in the tribe. For example the first day has all castaways mapped to their original tribe. The next day has the same minus the castaway just voted out. This is useful for observing the changes in tribe make either due to castaways being voted off the island, tribe swaps, who is on Redemption Island and Edge of Extinction.

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series) https://survivor.fandom.com/wiki/Main_Page

Tribes colour palette

Description

To create scale functions for ggplot. Given a season of Survivor, a palette is created from the tribe colours for that season including the merged tribe.

Usage

tribes_pal(season = NULL, scale_type = "d", reverse = FALSE, tribe = NULL, ...)

scale_fill_tribes(season = NULL, scale_type = "d", reverse = FALSE, ...)

scale_colour_tribes(season = NULL, scale_type = "d", reverse = FALSE, ...)

Arguments

season

Season number

scale_type

Discrete or continuous. Input d or c.

reverse

Logical. Reverse the palette?

tribe

Tribe names. Default NULL

...

Other arguments passed on to methods.

Details

If it is intended the colours will correspond to the tribes e.g. a stacked bar chart of votes given to each finalist and the colour corresponds to their original tribe (as in the example below), the tribe vector needs to be passed to the scale function (for now). If no tribe vector is given it will simply treat the tribe colours as a colour palette.

Value

Scale functions for ggplot2

Examples

library(ggplot2)
library(stringr)
library(dplyr)
library(glue)
ssn <- 35
labels <- castaways %>%
  filter(
    season == ssn,
    str_detect(result, "Sole|unner")
  ) %>%
  select(castaway, original_tribe) %>%
  mutate(label = glue("{castaway} ({original_tribe})")) %>%
  select(label, castaway)
jury_votes %>%
  filter(season == ssn) %>%
  left_join(
    castaways %>%
      filter(season == ssn) %>%
      select(castaway, original_tribe),
    by = "castaway"
  ) %>%
  group_by(finalist, original_tribe) %>%
  summarise(votes = sum(vote)) %>%
  left_join(labels, by = c("finalist" = "castaway")) %>% {
    ggplot(., aes(x = label, y = votes, fill = original_tribe)) +
      geom_bar(stat = "identity", width = 0.5) +
      scale_fill_tribes(ssn, tribe = .$original_tribe) +
      theme_minimal() +
      labs(
        x = "Finalist (original tribe)",
        y = "Votes",
        fill = "Original\ntribe",
        title = "Votes received by each finalist"
      )
 }

Viewers

Description

A dataset containing the viewer history for each season and episode

Usage

viewers

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: Season number
episode_number_overall: The cumulative episode number
episode: Episode number for the season
episode_title: Episode title
episode_label: A standardised episode label
episode_date: Date the episode aired
episode_length: Episode length in minutes
viewers: Number of viewers (millions) who tuned in
imdb_rating: IMDb rating for the episode on a scale of 0-10
n_ratings: The number of ratings submitted to IMDb

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series)

Vote history

Description

A dataset containing details on the vote history for each season

Usage

vote_history

Format

This data frame contains the following columns:

version: Country code for the version of the show
version_season: Version season key
season: The season number
episode: Episode number
day: Day the tribal council took place
tribe_status: The status of the tribe e.g. original, swapped, merged, etc. See details for more
tribe: Tribe name
castaway: Name of the castaway
immunity: Type of immunity held by the castaway at the time of the vote e.g. individual, hidden (see details for hidden immunity data)
vote: The castaway for which the vote was cast
vote_event: Extra details on the vote e.g. Won or lost the fire challenge, played an extra vote, etc
vote_event_outcome: The outcome of the vote event
split_vote: If there was a decision to split the vote this records who the vote was split with. Helps to identify successful boots
nullified: Was the vote nullified by a hidden immunity idol? Logical
tie: If the set of votes resulted in a tie. Logical
voted_out: The castaway who was voted out
order: Boot order. Order in which castaway was voted out e.g. 5 is the 5th person voted of the island
vote_order: In the case of ties this indicates the order the votes took place
castaway_id: ID of the castaway (primary key). Consistent across seasons and name changes e.g. Amber Brkich / Amber Mariano. The first two letters reference the country of the version played e.g. US, AU.
vote_id: ID of the castaway voted for
voted_out_id: ID of the castaway voted_out
sog_id: Stage of game ID for joining to boot_mapping and challenge_results
challenge_id: Primary key to the challenge_description data set which contains features of the challenge. The helps map the immunity challenge which result in the tribal.

Details

This data frame contains a complete history of votes cast across all seasons of Survivor. While there are consistent events across the seasons there are some unique events such as the 'mutiny' in Survivor: Cook Islands (season 13) or the 'Outcasts' in Survivor: Pearl Islands (season 7). For maintaining a standard, whenever there has been a change in tribe for the castaways it has been recorded as swapped. swapped is used as the term since 'the tribe swap' is a typical recurring milestone in each season of Survivor. Subsequent changes are recorded with a trailing digit e.g. swapped2. This includes absorbed tribes e.g. Stephanie was 'absorbed' in Survivor: Palau (season 10) and when 3 tribes are reduced to 2. These cases are still considered 'swapped' to indicate a change in tribe status.

Some events result in a castaway attending tribal but not voting. These are recorded as

Win: The castaway won the fire challenge
Lose: The castaway lost the fire challenge
None: The castaway did not cast a vote. This may be due to a vote steal or some other means
Immune: The castaway did not vote but were immune from the vote

Where a castaway has immunity == 'hidden' this means that player is protected by a hidden immunity idol. It may not necessarily mean they played the idol, the idol may have been played for them. While the nullified votes data is complete the immunity data does not include those who had immunity but did not receive a vote. This is a TODO.

In the case where the 'steal a vote' advantage was played, there is a second row for the castaway that stole the vote. The castaway who had their vote stolen are is recorded as None.

Many castaways have been medically evacuated, quit or left the game for some other reason. In these cases where no votes were cast there is a skip in the order variable. Since no votes were cast there is nothing to record on this data frame. The correct order in which castaways departed the island is recorded on castaways.

In the case of a tie, voted_out is recorded as tie to indicate no one was voted off the island in that instance. The re-vote is recorded with vote_order = 2 to indicate this is the second round of voting. In the case of a second tie voted_out is recorded as tie2. The third step is either a draw of rocks, fire challenge or countback (in the early days of survivor). In these cases vote is recorded as the colour of the rock drawn, result of the fire challenge or 'countback'.

Source

https://en.wikipedia.org/wiki/Survivor_(American_TV_series)

Examples

# The number of times Tony voted for each castaway in Survivor: Winners at War
library(dplyr)
vote_history %>%
  filter(
    season == 40,
    castaway == "Tony"
  ) %>%
  count(vote)