Plotting SF salaries using the DiagrammeR R package. Dataset is available from Kaggle Datasets.

sf_salaries.R

library(DiagrammeR)
library(DiagrammeRsvg)
library(magrittr)

# The CSV file is located inside a zip file at:
# https://www.kaggle.com/kaggle/sf-salaries/downloads/
# sf-salaries-release-2015-12-21-03-21-32.zip
salaries <- read.csv("Salaries.csv",
                     stringsAsFactors = FALSE)

# Use only salary data from 2014
salaries <- subset(salaries, Year == 2014)

# Select the `Pay` columns from the data frame
salaries_pay <- salaries[,4:6]

# Ensure that the `Pay` columns are numeric
salaries_pay$BasePay <-
  as.numeric(salaries_pay$BasePay)

salaries_pay$OvertimePay <-
  as.numeric(salaries_pay$OvertimePay)

salaries_pay$OtherPay <-
  as.numeric(salaries_pay$OtherPay)

# Select only complete cases
salaries_pay <-
  salaries_pay[which(complete.cases(salaries_pay)),]

# Take a sample of 5000 values
set.seed <- 20
salaries_pay_sample <-
  salaries_pay[sample(1:nrow(salaries_pay), 5000, FALSE),]

# Create a series of (x, y) data points as an NDF,
# make a scatterplot, and then save the file as a PDF
create_xy_pts(
  series_label = "pay",
  x = as.numeric(salaries_pay_sample$BasePay),
  y = as.numeric(salaries_pay_sample$OvertimePay) +
    as.numeric(salaries_pay_sample$OtherPay),
  line_width = 1.0,
  width = 0.05,
  height = 0.05,
  shape = "circle",
  fill_color = "#00c5cd50",
  line_color = "#00868b50") %>%
  create_xy_graph(
    x_name = "Base Pay (USD)",
    y_name = "Overtime + Other Pay (USD)",
    heading = c("#San Francisco Salaries in 2014*",
                "#####Published February 11, 2016"),
    footer = "####*Source: Kaggle Datasets (https://www.kaggle.com/kaggle/sf-salaries).",
    xy_value_labels = c("USD:K", "USD:K"),
    include_xy_minima = c(FALSE, FALSE),
    include_legend = FALSE) %>%
  export_graph("salaries_sf_scatterplot.pdf")

The graph:

Here is a piped version

library(DiagrammeR)
library(DiagrammeRsvg)
library(magrittr) 
library(dplyr)
library(readr)
# The CSV file is located inside a zip file at:
# https://www.kaggle.com/kaggle/sf-salaries/downloads/
# sf-salaries-release-2015-12-21-03-21-32.zip
all_salaries <- 
  'https://gist.githubusercontent.com/abresler/3ecbe3898017eb13e451/raw/b66d95d81c0e5f95297bcc8346469b44d620a0f4/sf_salaries.csv' %>% 
  read_csv()

# Use only salary data from 2014
salaries <-
  all_salaries %>% 
  dplyr::filter(Year == 2014)

# Select the `Pay` columns from the data frame
salaries_pay <- 
  salaries %>% 
  dplyr::select(BasePay:OtherPay) %>% 
  mutate_each(funs(as.numeric(.)), contains("Pay"))


salaries_pay %<>% 
  dplyr::filter(complete.cases(salaries_pay))    


# Select only complete cases


# Take a sample of 5000 values
set.seed <-
  20

salaries_pay_sample <-
  salaries_pay %>% 
  sample_n(5000)

# Create a series of (x, y) data points as an NDF,
# make a scatterplot, and then save the file as a PDF
create_xy_pts(
  series_label = "pay",
  x = 
    salaries_pay_sample$BasePay,
  y = 
    salaries_pay_sample$OvertimePay + 
    salaries_pay_sample$OtherPay,
  line_width = 1.0,
  width = 0.05,
  height = 0.05,
  shape = "circle",
  fill_color = "#00c5cd50",
  line_color = "#00868b50") %>%
  create_xy_graph(
    x_name = "Base Pay (USD)",
    y_name = "Overtime + Other Pay (USD)",
    heading = c("#San Francisco Salaries in 2014*",
                "#####Published February 11, 2016"),
    footer = "####*Source: Kaggle Datasets (https://www.kaggle.com/kaggle/sf-salaries).",
    xy_value_labels = c("USD:K", "USD:K"),
    include_xy_minima = c(FALSE, FALSE),
    include_legend = FALSE) %>%
  export_graph("salaries_sf_scatterplot.pdf")