This week, I used the dataset that lists every slave that lived and worked at Mount Vernon. It essentially combines every slave census into one spreadsheet.
slaves <- read.csv("~/Desktop/MountVernon/Spreadsheets/slaves.csv", stringsAsFactors = FALSE)
library(dplyr)
##
## Attaching package: 'dplyr'
##
## The following objects are masked from 'package:stats':
##
## filter, lag
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(ggplot2)
library(tidyr)
library(lubridate)
This function creates a table that only lists the children.
slave_children <- slaves %>%
select(Name, Gender, Birth.Year, Skill, Farm, Census) %>%
filter(Skill == "Child")
head(slave_children)
## Name Gender Birth.Year Skill Farm Census
## 1 Abbay A Female 1789 Child Dogue Run 1799
## 2 Adam A Male 1792 Child Muddy Hole 1799
## 3 Alce A Female 1791 Child Muddy Hole 1799
## 4 Alexander A Male 1796 Child Muddy Hole 1799
## 5 Ambrose A Male 1798 Child Union Farm 1799
## 6 Ambrose B Male 1786 Child River Farm 1786
This chart shows the number of children on each farm for 1786 and 1799. Ferry Farm evolved into Union Farm, which the chart illustrates. Every farm increased in children, especially Muddy Hole and Mansion House.
ggplot(data = slave_children, aes(x = Farm, stat = "identity")) + geom_bar() + facet_wrap(~Census) + theme(axis.text.x=element_text(angle = 90, hjust = 0))
This chart shows the number of children born on each farm during each birth year listed on the census. It is misleading, because it also shows the counts of unknown birth years. The dates are also not standardized in the original spreadsheet.
ggplot(data = slave_children, aes(x = Birth.Year, na.rm = TRUE, stat = "identity")) + geom_bar() + facet_wrap(~ Farm) + theme(axis.text.x=element_text(angle = 90, hjust = 0))
That chart led me to consider the number of children born on the entire estate over time. Of couse, this also is misleading because a number of children do not have birth years on the census.
slave_br <- slaves %>%
filter(Skill == "Child") %>%
group_by(Birth.Year) %>%
summarize(Name = n())
head(slave_br)
## Source: local data frame [6 x 2]
##
## Birth.Year Name
## 1 13
## 2 1774 2
## 3 1775 3
## 4 1777 1
## 5 1778 1
## 6 1779 1
This chart attempts to show the slave population on the estate per census, but the spreadsheet lists overlapping census dates and they’re not standardized.
slave_pop <- slaves%>%
group_by(Census)%>%
summarize(Name = n())
This table shows the number of children born to each mother.
slave_mothers <- slaves %>%
group_by(Mother) %>%
summarize(Name = n())
head(slave_mothers)
## Source: local data frame [6 x 2]
##
## Mother Name
## 1 232
## 2 Agnes A 3
## 3 Agnes B 1
## 4 Alce B 4
## 5 Alce C 9
## 6 Alce D 2
ggplot(data = slave_mothers, aes(x = Mother, y = n)) + geom_bar() + theme(axis.text.x=element_text(angle = 90, hjust = 0))
It would be interesting to compare the distribution of spouses across the estate. Slaves could not legally marry, so it’s interesting anyway that Washington listed slave spouses. With this new table, I wanted to see how many slave couples were on the same farm or on different farms. I then wanted to compare that finding to their children.
slave_spouses <- slaves %>%
select(Name, Spouse, Farm)
head(slave_spouses)
## Name Spouse Farm
## 1 Aaron A
## 2 Abbay A Dogue Run
## 3 Abram A Nancy F Union Farm
## 4 Acco A
## 5 Adam A Muddy Hole
## 6 Adam B River Farm
With this table, you can see that not every mother had a spouse. This chart makes me question the significance of slave spouses, because women who had no husband on the census still had children.
slave_fam <- slaves %>%
select(Name, Spouse, Mother, Siblings, Farm)
head(slave_fam)
## Name Spouse Mother
## 1 Aaron A
## 2 Abbay A Sall Twine
## 3 Abram A Nancy F
## 4 Acco A
## 5 Adam A Alce B
## 6 Adam B
## Siblings Farm
## 1
## 2 Barbary A, George C, Hannah A, Jesse B, Kate B, Lawrence B Dogue Run
## 3 Union Farm
## 4
## 5 George D, Cecelia B, Kate A Muddy Hole
## 6 River Farm