Foreign Travel of Government Officials

library(tidyverse)
library(tidyquant)
library(modelr)
library(gridExtra)
library(grid)
library(readr)

On May 17, 2017, the Washington Post broke a story about a recorded conversation between some members of Congress where one of the participants says he thinks Vladimir Putin pays another member of Congress and the Republican Presidential Candidate. I became curious about the this and wanted to investigate the travel habits of Congressional members.

This data comes from Data for Democracy/ProPublica and it represents the travel records of government officials from 1995-2017.

travels <- read_csv("https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv",
                   col_types = cols(
                      name = col_character(),
                      honorific = col_character(),
                      Description = col_character(),
                      Quantity = col_integer(),
                      arrival_date = col_date(format = "%m/%d/%Y"),
                      departure_date = col_date(format = "%m/%d/%Y"),
                      country = col_character(),
                      table_header = col_character(), 
                      source_file = col_character(), 
                      ArrivalYear = col_integer())) %>%
  mutate(year = format(arrival_date, "%Y"))

The following named parsers don't match the column names: Description, Quantity, ArrivalYear956 parsing failures.
 row            col           expected     actual                                                   file
 986 arrival_date   date like %m/%d/%Y 1//1998    'https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv'
 986 departure_date date like %m/%d/%Y 1//1998    'https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv'
1409 arrival_date   valid date         2/29/2007  'https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv'
1443 departure_date valid date         13/13/2006 'https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv'
1646 departure_date date like %m/%d/%Y 1/5//1999  'https://query.data.world/s/7ezifc8eqig9vdazaoa1noecv'
.... .............. .................. .......... ......................................................
See problems(...) for more details.

head(travels)

dim(travels)

[1] 48237     8

How many travellers are represented?

nrow(travels %>% group_by(name) %>% summarise(GovernmentOfficials = n()) %>% na.omit())

[1] 5346

Who travelled to Russia most often?

travels %>% filter(grepl("Russia",country)) %>% group_by(name) %>% summarise(visitsToRussia = n()) %>% arrange(desc(visitsToRussia)) %>% top_n(6)

Selecting by visitsToRussia

I narrowed down the data set to those who have made more than 4 trips to Russia during their tenure.

myVectorOfStrings <- c("Rohr", "Curt Weldon", "Charles Taylor", "Sloat", "Paul Berkowitz", "Mark Gage", "Robert King" )
matchExpression <- paste(myVectorOfStrings, collapse = "|")
# [1] "foo|bar"

The table below shows of often each of the select Officials has travelled to each country represented.

travels %>% filter(grepl(matchExpression, name)) %>% group_by(name, country) %>% summarise(n = n()) %>% arrange(country) %>% top_n(10)

Selecting by n

NN <- travels %>% filter(grepl(matchExpression, name)) %>% select(name, country, year) %>% arrange(name)%>% na.omit()

The proportion of each officials’ travels to Russia in relation to all travels.

NN$namesVector <- rep(c("Amanda Sloat", "Curt Weldon","Dana Rohrabacher", "Amanda Sloat","Robert King", "Charles Taylor", "Curt Weldon", "Dana Rohrabacher", "Mark Gage", "Robert King", "Dana Rohrabacher", "Mark Gage", "Paul Berkowitz","Robert King","Dana Rohrabacher"), times = c(43,1,4,9,1, 27,88,174,1,2, 4,49, 160, 87,1))
x <- NN %>% filter(country == "Russia") %>% na.omit() %>% group_by(namesVector) %>% summarise(n = n())
ggplot(NN, aes(x = namesVector)) + geom_bar(fill = "blue", alpha = 0.5) + theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1)) + 
  geom_boxplot(data = x, aes(x = namesVector, y = n, fatten = 4)) + labs(title = "All Trips", subtitle = " Black line demarcates the number of trips to Russia") +
  labs(x="Government Official",y="Total Number of Trips")

head(NN)

NN %>% 
  ggplot(aes(x = year, fill = country)) + 
  geom_bar(position = "fill") + 
  #scale_y_continuous(limits=c(0,30)) + 
  facet_grid(namesVector ~ .,scales = "free_y")  + 
  scale_fill_manual(values=c('burlywood4',
'cadetblue',
'cadetblue1',
'cadetblue2',
'cadetblue4',
'chartreuse',
'chartreuse1',
'chartreuse2',
'chartreuse3',
'chartreuse4',
'chocolate',
'chocolate1',
'chocolate3',
'chocolate4',
'coral',
'coral1',
'coral2',
'coral3',
'coral4',
'cornflowerblue',
'cornsilk',
'cornsilk1',
'cornsilk2',
'cornsilk3',
'cornsilk4',
'cyan',
'cyan1',
'cyan2',
'cyan3',
'cyan4',
'darkblue',
'darkcyan',
'darkgoldenrod',
'darkgoldenrod1',
'darkgoldenrod3',
'darkgoldenrod4',
'darkgray',
'aliceblue',
'antiquewhite',
'antiquewhite1',
'antiquewhite2',
'antiquewhite3',
'antiquewhite4',
'aquamarine',
'aquamarine1',
'aquamarine2',
'aquamarine3',
'aquamarine4',
'azure',
'azure1',
'azure2',
'azure3',
'azure4',
'beige',
'bisque',
'bisque1',
'bisque2',
'bisque3',
'bisque4',
'blanchedalmond',
'blue',
'blue1',
'blue2',
'blue3',
'blue4',
'blueviolet',
'brown',
'brown1',
'brown2',
'brown3',
'brown4',
'burlywood',
'burlywood1',
'burlywood2',
'burlywood3',
'darkgreen',
'darkgrey',
'darkkhaki',
'darkmagenta',
'darkolivegreen',
'darkolivegreen1',
'darkolivegreen2',
'darkolivegreen3',
'darkolivegreen4',
'darkorange',
'darkorange1',
'darkorange2',
'darkorange3',
'darkorange4',
'darkorchid',
'darkorchid1',
'darkorchid2',
'darkorchid3',
'black',
'darkorchid4',
'darkred',
'darksalmon',
'darkseagreen',
'darkseagreen1',
'darkseagreen2',
'darkseagreen3',
'darkseagreen4',
'darkslateblue',
'darkslategray',
'darkslategray1',
'darkslategray2',
'darkslategray3',
'darkslategray4',
'darkslategrey',
'darkturquoise',
'darkviolet',
'deeppink',
'deeppink1',
'deeppink2',
'deeppink3',
'deeppink4',
'lightblue4',
'lightcoral',
'lightcyan',
'lightcyan1',
'lightcyan2',
'lightcyan3',
'lightcyan4',
'lightgoldenrod')) + 
  theme_tq() + 
  guides(color = F) + 
  labs(title = "How much of the Official's travel is to Russia", x = "")  + 
  labs(title = "", subtitle = "Civil Servants Travel from 1995-2016Trips to \n Russia are in black",y = "Number of Trips")  + theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1)) 

Taylor

BerkowitzRussia <- travels %>%  filter(grepl("Berkowitz",name)) %>% filter(grepl("Russia", country)) %>% arrange(desc(departure_date))
BerkowitzN <- travels %>%  filter(grepl("Berkowitz",name)) %>% filter(grepl("Nether", country)) %>% arrange(desc(departure_date))
dim(BerkowitzN)

MostRussia <- rbind.data.frame(WeldonRussia, RohrabacherRussia, TaylorRussia, GageRussia, KingRussia, SloatRussia, BerkowitzRussia)
dim(MostRussia)

MostRussia$namesVector <- rep(c("Curt Weldon", "Dana Rohrabacher","Charles Taylor","Mark Gage","Robert King","Amanda Sloat","Paul Berkowitz"), times = c(22,12,11,8,6,5,5))
MostRussia

MostRussia %>% ggplot(aes(x = year)) + geom_bar(fill = "plum4") + scale_y_continuous(limits=c(0,4)) + facet_grid(namesVector ~ .,scales = "free_y")  + scale_color_manual(values = palette_light()) + theme_tq() + guides(color = F) + labs(title = "Civil Servants and their Trips to Russia", x = "")

Finally I saw in a 2 part documentary, that there might be a connections with those in Russia to the Netherlands for the purposes of money laundering. Shown below is the proportion of trips to the Netherlands and Russia to all other countries visited.

NN %>% mutate(Country = ifelse(country == "Russia", "Russia", ifelse(country == "Netherlands","Netherlands", "Other"))) %>% 
  ggplot(aes(x = year, fill = Country)) + 
  geom_bar(position = "fill") + 
  #scale_y_continuous(limits=c(0,30)) + 
  facet_grid(namesVector ~ .,scales = "free_y")  + 
  # scale_color_manual(values = palette_dark()) + theme_tq() + 
  guides(color = F) + 
  labs(title = "Civil Servants Travel from 1995-2016", subtitle = "Trips to Russia are in light blue",y = "Number of Trips")  + theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1))