Assignment 3b
Olivia Crispo and Brooke Butler
2025-10-07
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.2
## ✔ ggplot2 4.0.0 ✔ tibble 3.3.0
## ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
## ✔ purrr 1.1.0
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorscol_types_spec <- cols_only(
id = col_integer(),
voyage_id = col_integer(),
voyage_dates__imp_arrival_at_port_of_dis_sparsedate__year = col_double(),
voyage_slaves_numbers__imp_total_num_slaves_disembarked = col_double(),
voyage_slaves_numbers__imp_total_num_slaves_embarked = col_double(),
voyage_dates__length_middle_passage_days = col_double(),
voyage_dates__imp_length_home_to_disembark = col_double(),
voyage_crew__crew_first_landing = col_double(),
voyage_crew__crew_voyage_outset = col_double(),
voyage_ship__tonnage_mod = col_double(),
voyage_slaves_numbers__imp_jamaican_cash_price = col_double(),
voyage_slaves_numbers__imp_mortality_ratio = col_double(),
voyage_slaves_numbers__percentage_women_among_embarked_slaves = col_double(),
voyage_outcome__vessel_captured_outcome__name = col_character(),
voyage_ship__imputed_nationality__name = col_character(),
voyage_itinerary__imp_region_voyage_begin__name = col_character(),
voyage_ship__rig_of_vessel__name = col_character(),
voyage_itinerary__place_voyage_ended__name = col_character(),
# Force as character
voyage_dates__slave_purchase_began_sparsedate__month = col_double(),
voyage_slaves_numbers__percentage_men = col_double(),
voyage_dates__voyage_completed_sparsedate__month = col_double(),
voyage_itinerary__region_of_return__name = col_character(),
voyage_slaves_numbers__percentage_boy = col_double(),
voyage_itinerary__imp_principal_region_slave_dis__name = col_character(),
voyage_itinerary__imp_principal_region_of_slave_purchase__name = col_character(),
voyage_dates__date_departed_africa_sparsedate__month = col_double(),
voyage_dates__voyage_began_sparsedate__month = col_double(),
voyage_itinerary__imp_port_voyage_begin__name = col_character(),
voyage_dates__first_dis_of_slaves_sparsedate__month = col_double(),
voyage_itinerary__imp_broad_region_slave_dis__name = col_character(),
voyage_slaves_numbers__percentage_girl = col_double(),
voyage_outcome__particular_outcome__name = col_character(),
voyage_itinerary__imp_principal_port_slave_dis__name = col_character(),
voyage_slaves_numbers__percentage_child = col_double(),
voyage_slaves_numbers__percentage_women = col_double(),
voyage_dates__departure_last_place_of_landing_sparsedate__month = col_double(),
voyage_outcome__outcome_owner__name = col_character(),
voyage_outcome__outcome_slaves__name = col_character(),
voyage_itinerary__imp_principal_place_of_slave_purchase__name = col_character(),
voyage_outcome__resistance__name = col_character(),
voyage_slaves_numbers__percentage_male = col_double(),
voyage_slaves_numbers__percentage_female = col_double(),
voyage_itinerary__imp_broad_region_voyage_begin__name = col_character(),
voyage_itinerary__imp_broad_region_of_slave_purchase__name = col_character(),
voyage_sources = col_character(),
enslavers = col_character()
)
# Load the datasets
trans <- read_csv("https://raw.githubusercontent.com/imowerman-prog/data-3210/refs/heads/main/Data/trans-atlantic.csv",
col_types = col_types_spec)
intra <- read_csv("https://raw.githubusercontent.com/imowerman-prog/data-3210/refs/heads/main/Data/intra-american.csv",
col_types = col_types_spec)
# Define successful outcomes
successful_outcomes <- c(
"Sold slaves in Americas - subsequent fate unknown",
"Voyage completed as intended",
"Captured by pirates or privateers - after disembarkation",
"Condemned - Americas after disembarkation",
"Detained and condemned in the United States after slaves disembarked",
"Condemned in the Americas by British after slaves disembarked",
"Captured by pirates - slaves sold in Americas from another ship",
"Shipwrecked or destroyed, after disembarkation",
"Privateer captured slaves at sea and delivered for sale in America",
"Prisoners of war stole slaves during escape and carried to port of sale",
"Captives seized from vessel by Spanish officials and sold",
"Captured by Dutch - after disembarkation",
"Shipwrecked, slaves salvaged",
"Captured by slaves, recaptured and landed slaves in the Americas"
)
# Process trans-atlantic data
trans <- trans %>%
rename(
year = voyage_dates__imp_arrival_at_port_of_dis_sparsedate__year,
slaves_embarked = voyage_slaves_numbers__imp_total_num_slaves_embarked,
slaves_disembarked = voyage_slaves_numbers__imp_total_num_slaves_disembarked,
particular_outcome = voyage_outcome__particular_outcome__name,
dis_broad = voyage_itinerary__imp_broad_region_slave_dis__name,
dis_port = voyage_itinerary__imp_principal_port_slave_dis__name,
embark_broad = voyage_itinerary__imp_broad_region_of_slave_purchase__name,
embark_port = voyage_itinerary__imp_principal_place_of_slave_purchase__name
) %>%
mutate(
year = as.integer(year),
slaves_embarked = as.numeric(slaves_embarked),
slaves_disembarked = as.numeric(slaves_disembarked)
) %>%
filter(
!is.na(slaves_disembarked),
slaves_disembarked > 0,
particular_outcome %in% successful_outcomes
) %>%
mutate(
decade = floor(year / 10) * 10,
estimated_deaths = slaves_embarked - slaves_disembarked,
is_us = dis_broad == "Mainland North America" |
grepl("New Orleans", dis_port, ignore.case = TRUE),
embark_is_us = embark_broad == "Mainland North America" |
grepl("New Orleans", embark_port, ignore.case = TRUE),
source_type = "Trans-Atlantic"
)
# Process intra-american data
intra <- intra %>%
rename(
year = voyage_dates__imp_arrival_at_port_of_dis_sparsedate__year,
slaves_embarked = voyage_slaves_numbers__imp_total_num_slaves_embarked,
slaves_disembarked = voyage_slaves_numbers__imp_total_num_slaves_disembarked,
particular_outcome = voyage_outcome__particular_outcome__name,
dis_broad = voyage_itinerary__imp_broad_region_slave_dis__name,
dis_port = voyage_itinerary__imp_principal_port_slave_dis__name,
embark_broad = voyage_itinerary__imp_broad_region_of_slave_purchase__name,
embark_port = voyage_itinerary__imp_principal_place_of_slave_purchase__name
) %>%
mutate(
year = as.integer(year),
slaves_embarked = as.numeric(slaves_embarked),
slaves_disembarked = as.numeric(slaves_disembarked)
) %>%
filter(
!is.na(slaves_disembarked),
slaves_disembarked > 0,
particular_outcome %in% successful_outcomes
) %>%
mutate(
decade = floor(year / 10) * 10,
estimated_deaths = slaves_embarked - slaves_disembarked,
is_us = dis_broad == "Mainland North America" |
grepl("New Orleans", dis_port, ignore.case = TRUE),
embark_is_us = embark_broad == "Mainland North America" |
grepl("New Orleans", embark_port, ignore.case = TRUE),
source_type = "Intra-American"
)
# Combine the datasets
combined <- bind_rows(trans, intra)Question 1:Total slaves imported to the US: Filter for is_us == TRUE, sum slaves_disembarked from both datasets.
gross_us <- combined %>%
filter(is_us == TRUE) %>%
summarise(gross_us = sum(slaves_disembarked, na.rm = TRUE)) %>%
pull(gross_us)
re_exports <- combined %>%
filter(source_type == "Intra-American", embark_is_us == TRUE, is_us == FALSE) %>%
summarise(re_exports = sum(slaves_embarked, na.rm = TRUE)) %>%
pull(re_exports)
net_us <- gross_us - re_exports
cat("Gross slaves imported to the US:", gross_us, "\n")## Gross slaves imported to the US: 395919cat("Slaves re-exported from the US:", re_exports, "\n")## Slaves re-exported from the US: 7652cat("Net slaves retained in the US:", net_us, "\n")## Net slaves retained in the US: 388267Question 2: Proportion of all slaves taken from Africa: Calculate US total / total slaves_embarked from Trans-Atlantic dataset (as this represents slaves taken from Africa).
trans_total_embarked <- trans %>%
summarise(total_embarked = sum(slaves_embarked, na.rm = TRUE)) %>%
pull(total_embarked)
proportion <- net_us / trans_total_embarked
cat("Proportion of all slaves taken from Africa (net US):", proportion, "\n")## Proportion of all slaves taken from Africa (net US): 0.05107614Question 3: Graph slave imports by decade to the US: Filter for US, group by decade, sum slaves_disembarked, plot as a bar graph with ggplot2.
us_imports_by_decade <- combined %>%
filter(is_us == TRUE) %>%
group_by(decade) %>%
summarise(total_disembarked = sum(slaves_disembarked, na.rm = TRUE)) %>%
ungroup()
ggplot(us_imports_by_decade, aes(x = factor(decade), y = total_disembarked)) +
geom_bar(stat = "identity", fill = "steelblue") +
labs(
title = "Slave Imports to the US by Decade",
x = "Decade",
y = "Number of Slaves Disembarked"
) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
map_port_to_state <- function(port, region) {
case_when(
grepl("New Orleans", port, ignore.case = TRUE) ~ "Louisiana",
grepl("Charleston|Charles Town", port, ignore.case = TRUE) ~ "South Carolina",
grepl("Savannah", port, ignore.case = TRUE) ~ "Georgia",
grepl("Baltimore", port, ignore.case = TRUE) ~ "Maryland",
grepl("Norfolk|Virginia", port, ignore.case = TRUE) ~ "Virginia",
grepl("Boston", port, ignore.case = TRUE) ~ "Massachusetts",
grepl("New York", port, ignore.case = TRUE) ~ "New York",
grepl("Philadelphia", port, ignore.case = TRUE) ~ "Pennsylvania",
grepl("Rhode Island|Newport|Providence", port, ignore.case = TRUE) ~ "Rhode Island",
region == "Mainland North America" & is.na(port) ~ "Unknown US State",
TRUE ~ "Other/Unknown"
)
}Question 4: Imports to the US by decade and region/port/state: Filter for US, group by decade, dis_region, dis_port (approximate state from port/region, e.g., “New Orleans” -> “Louisiana”), sum slaves_disembarked. Use a table and faceted bar plot.
us_imports_by_decade_region_port <- combined %>%
filter(is_us == TRUE) %>%
mutate(state = map_port_to_state(dis_port, dis_broad)) %>%
group_by(decade, dis_broad, dis_port, state) %>%
summarise(total_disembarked = sum(slaves_disembarked, na.rm = TRUE), .groups = "drop") %>%
arrange(decade, dis_broad, state, dis_port)
print(us_imports_by_decade_region_port)## # A tibble: 272 × 5
## decade dis_broad dis_port state total_disembarked
## <dbl> <chr> <chr> <chr> <dbl>
## 1 1610 Mainland North America Hampton Othe… 29
## 2 1620 Mainland North America Virginia, port unspeci… Virg… 3
## 3 1630 Mainland North America Boston Mass… 7
## 4 1630 Mainland North America New York New … 53
## 5 1630 Mainland North America Virginia, port unspeci… Virg… 13
## 6 1640 Mainland North America New York New … 69
## 7 1640 Mainland North America Virginia, port unspeci… Virg… 12
## 8 1650 Mainland North America New York New … 477
## 9 1650 Mainland North America Maryland, port unspeci… Othe… 5
## 10 1650 Mainland North America Virginia, port unspeci… Virg… 125
## # ℹ 262 more rowsggplot(us_imports_by_decade_region_port, aes(x = factor(decade), y = total_disembarked)) +
geom_bar(stat = "identity", fill = "darkgreen") +
facet_wrap(~ state, scales = "free_y") +
labs(
title = "Slave Imports to the US by Decade and State",
x = "Decade",
y = "Number of Slaves Disembarked"
) +
theme_minimal() +
theme(
axis.text.x = element_text(angle = 45, hjust = 1),
strip.text = element_text(size = 10))
Question 5: Countries participating in export from Africa, by decade: From Trans-Atlantic dataset, group by decade and voyage_ship__imputed_nationality__name (as “country”), count unique voyages or sum slaves_embarked. Display in a table.
voyages_by_country_decade <- trans %>%
group_by(decade, voyage_ship__imputed_nationality__name) %>%
summarise(num_voyages = n_distinct(voyage_id), .groups = "drop") %>%
arrange(decade, desc(num_voyages)) %>%
rename(country = voyage_ship__imputed_nationality__name)
print(voyages_by_country_decade)## # A tibble: 203 × 3
## decade country num_voyages
## <dbl> <chr> <int>
## 1 1510 0 9
## 2 1510 Spain / Uruguay 8
## 3 1520 Spain / Uruguay 3
## 4 1520 0 2
## 5 1530 0 8
## 6 1530 Portugal / Brazil 2
## 7 1530 Spain / Uruguay 1
## 8 1540 0 23
## 9 1540 Portugal / Brazil 1
## 10 1550 0 27
## # ℹ 193 more rows slaves_by_country_decade <- trans %>%
group_by(decade, voyage_ship__imputed_nationality__name) %>%
summarise(total_embarked = sum(slaves_embarked, na.rm = TRUE), .groups = "drop") %>%
arrange(decade, desc(total_embarked)) %>%
rename(country = voyage_ship__imputed_nationality__name)
print(slaves_by_country_decade)## # A tibble: 203 × 3
## decade country total_embarked
## <dbl> <chr> <dbl>
## 1 1510 0 223
## 2 1510 Spain / Uruguay 144
## 3 1520 Spain / Uruguay 1043
## 4 1520 0 373
## 5 1530 0 1418
## 6 1530 Portugal / Brazil 560
## 7 1530 Spain / Uruguay 224
## 8 1540 0 7750
## 9 1540 Portugal / Brazil 160
## 10 1550 0 9390
## # ℹ 193 more rows us_imports_by_decade <- combined %>%
filter(is_us == TRUE) %>%
group_by(decade) %>%
summarise(total_disembarked = sum(slaves_disembarked, na.rm = TRUE)) %>%
ungroup()
ggplot(us_imports_by_decade, aes(x = factor(decade), y = total_disembarked)) +
geom_bar(stat = "identity", fill = "steelblue") +
labs(
title = "Slave Imports to the US by Decade",
x = "Decade",
y = "Number of Slaves Disembarked"
) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
map_port_to_state <- function(port, region) {
case_when(
grepl("New Orleans", port, ignore.case = TRUE) ~ "Louisiana",
grepl("Charleston|Charles Town", port, ignore.case = TRUE) ~ "South Carolina",
grepl("Savannah", port, ignore.case = TRUE) ~ "Georgia",
grepl("Baltimore", port, ignore.case = TRUE) ~ "Maryland",
grepl("Norfolk|Virginia", port, ignore.case = TRUE) ~ "Virginia",
grepl("Boston", port, ignore.case = TRUE) ~ "Massachusetts",
grepl("New York", port, ignore.case = TRUE) ~ "New York",
grepl("Philadelphia", port, ignore.case = TRUE) ~ "Pennsylvania",
grepl("Rhode Island|Newport|Providence", port, ignore.case = TRUE) ~ "Rhode Island",
region == "Mainland North America" & is.na(port) ~ "Unknown US State",
TRUE ~ "Other/Unknown"
)
}
us_imports_by_decade_state <- combined %>%
filter(is_us == TRUE) %>%
mutate(state = map_port_to_state(dis_port, dis_broad)) %>%
group_by(decade, state) %>%
summarise(total_disembarked = sum(slaves_disembarked, na.rm = TRUE), .groups = "drop") %>%
arrange(decade, state)
ggplot(us_imports_by_decade_state, aes(x = factor(decade), y = total_disembarked)) +
geom_bar(stat = "identity", fill = "darkgreen") +
facet_wrap(~ state, scales = "free_y") +
labs(
title = "Slave Imports to the US by Decade and State",
x = "Decade",
y = "Number of Slaves Disembarked"
) +
theme_minimal() +
theme(
axis.text.x = element_text(angle = 45, hjust = 1),
strip.text = element_text(size = 10)
)
voyages_by_country_decade <- trans %>%
group_by(decade, voyage_ship__imputed_nationality__name) %>%
summarise(num_voyages = n_distinct(voyage_id), .groups = "drop") %>%
arrange(decade, desc(num_voyages)) %>%
rename(country = voyage_ship__imputed_nationality__name)Summary of Insights
From the data, several key takeaways stand out. The net number of enslaved people retained in the U.S. was found by subtracting re-exports (intra-American trade) from total imports. This shows that while the U.S. was an important destination, it wasn’t the main one compared to regions like the Caribbean. Most arrivals were concentrated in the southern states—particularly South Carolina, Georgia, and Louisiana—with Charleston and New Orleans serving as major ports. Northern ports played a much smaller role. On the broader scale, European nations such as Britain, Portugal, and France led the trans-Atlantic trade, with differences in voyage numbers and embarkations by decade reflecting each empire’s shifting colonial and economic goals.