Process NZ MOJ convictions

convictions.R

install.packages('dplyr')
install.packages('ggplot2')
install.packages('readr')
install.packages('ggthemes')
install.packages('stringr')
install.packages('lubridate')
install.packages('reshape2')
install.packages('grid')

library(grid)
library(dplyr)
library(reshape2)
library(ggplot2)
library(readr)
library(ggthemes)
library(lubridate)
library(stringr)

# Read convictions data into a dataframe
convictions <- read_csv("convicted-offenders-02-10-2013.csv")

# tranform the names of the dataframe to lower case
names(convictions) <- names(convictions) %>% tolower()

names(convictions)
# [1] "year"         "offence"      "sentence"     "courtcluster" "agegroup"     "gender"       "ethnicity"    "count" 

unique(convictions$year)
# [1] "2003/04" "2004/05" "2005/06" "2006/07" "2007/08" "2008/09" "2009/10" "2010/11" "2011/12" "2012/13"

unique(convictions$offence)
# [1] "Abduction, harassment and other offences against the person"                       
# [2] "Acts intended to cause injury"                                                     
# [3] "Dangerous or negligent acts endangering persons"                                   
# [4] "Fraud, deception and related offences"                                             
# [5] "Homicide and related offences"                                                     
# [6] "Illicit drug offences"                                                             
# [7] "Miscellaneous offences"                                                            
# [8] "Offences against justice procedures, government security and government operations"
# [9] "Prohibited and regulated weapons and explosives offences"                          
# [10] "Property damage and environmental pollution"                                       
# [11] "Public order offences"                                                             
# [12] "Robbery, extortion and related offences"                                           
# [13] "Sexual assault and related offences"                                               
# [14] "Theft and related offences"                                                        
# [15] "Traffic and vehicle regulatory offences"                                           
# [16] "Unlawful entry with intent/burglary, break and enter"  

unique(convictions$sentence)
# [1] "Community work, Corrections"          "Supervision by Community Corrections" "Conviction and discharge"            
# [4] "Imprisonment"                         "Preventive Detention"                 "Fine"                                
# [7] "Reparation/Restitution"               "Other"                                "Disqualification from driving"       
# [10] "Community Detention"                 "Intensive Supervision"                "Home Detention"                      
# [13] "Life imprisonment" 

unique(convictions$courtcluster)
# [1] "Northland"                     "Auckland"                      "Manukau"                       "Waikato"                      
# [5] "Bay of Plenty/Coromandel"      "Central North Island"          "East Coast/Hawke's Bay"        "Taranaki/Wanganui"            
# [9] "Manawatu/Wairarapa"            "Wellington"                    "Nelson/Marlborough/West Coast" "Canterbury"                   
# [13] "Otago/South Canterbury"        "Southland/Central Otago"   

unique(convictions$agegroup)
# [1] "17 to 19" "20 to 24" "25 to 29" "30 to 39" "40+"      "Unknown"

unique(convictions$gender)
# [1] "Female"  "Male"    "Unknown"

unique(convictions$ethnicity)
# [1] "Maori"           "European"        "Pacific Peoples" "Other"           "Unknown"

# Normalise the years from 
out <- str_split_fixed(convictions$year, "/", 2)

# Name the split
colnames(out) <- c("start_date", "end_date")

# Append start date to all items in the first column
out[,1] <- paste("30/06/", out[,1], sep = "")

# Append start date and "20" to all items in the second column
out[,2] <- paste("01/07/20", out[,2], sep = "")

# merge the normalised dates with the convictions data frame
convictions <- cbind(out, convictions)

# coerce the string field to a date type for start_date
convictions$start_date <- convictions$start_date %>% 
  as.Date(format = "%d/%m/%Y")

# coerce the string field to a date type for end_date
convictions$end_date <- convictions$start_date %>% 
  as.Date(format = "%d/%m/%Y")

# define the list of offences we are interested in
# as 16 may be too many
interesting_offences <- c(
  "Fraud, deception and related offences",
  "Homicide and related offences",
  "Illicit drug offences",
  "Robbery, extortion and related offences",
  "Sexual assault and related offences",
  "Theft and related offences",
  "Unlawful entry with intent/burglary, break and enter"
)

# filter and grab a subset
imprisonment_offences <- convictions %>% 
  filter(sentence == "Imprisonment") %>%
  filter(offence %in% interesting_offences)
  select(end_date, offence, count) %>%
  subset

# get the sums of each group
imprisonment_offences <- imprisonment_offences %>%
  group_by(end_date, offence) %>%
  summarise(amount = sum(count))



imprisionment_plot <- ggplot(imprisonment_offences, aes(x=end_date, y=amount, colour=offence)) +
  geom_line(size = 1) +
  ylim(0, max(imprisonment_offences$amount)) + theme_solarized(light = TRUE) + 
  scale_colour_solarized("red", name = "Offences") +
  theme(axis.text = element_text(size = 18), axis.title = element_text(size=18), axis.title.y = element_text(vjust=1.4)) +
  theme(legend.position = "right", legend.text = element_text(size=10), legend.title = element_text(size=10)) +
  theme(plot.title = element_text(size=20)) +
  theme(plot.margin = unit(c(10,10,10,10),"mm")) +
  labs(title = "Convictions resulting in imprisonment by offence since 2003") + ylab('Number of convictions') + xlab('Year')
imprisionment_plot

interesting_regions = c(
  # "Northland", 
  # "Bay of Plenty/Coromandel", 
  # "Manawatu/Wairarapa", 
  # "Otago/South Canterbury", 
  "Auckland", 
  # "Central North Island", 
  "Wellington", 
  # "Southland/Central Otago", 
  "Manukau", 
  # "East Coast/Hawke's Bay", 
  # "Nelson/Marlborough/West Coast", 
  # "Waikato", 
  # "Taranaki/Wanganui", 
  "Canterbury"
)

# filter and grab a subset
environmental_convictions <- convictions %>% 
  filter(offence == "Property damage and environmental pollution") %>%
  filter(courtcluster %in% interesting_regions) %>%
  select(end_date, courtcluster, count) %>%
  subset

# get the sums of each group
environmental_convictions <- environmental_convictions %>%
  group_by(end_date, courtcluster) %>%
  summarise(amount = sum(count))

environmental_plot <- ggplot(environmental_convictions, aes(x=end_date, y=amount, colour=courtcluster)) +
  geom_line(size = 1) +
  ylim(0, max(environmental_convictions$amount)) + theme_solarized(light = TRUE) + 
  scale_colour_solarized("red", name = "Court district") +
  theme(axis.text = element_text(size = 18), axis.title = element_text(size=18), axis.title.y = element_text(vjust=1.4)) +
  theme(legend.position = "right", legend.text = element_text(size=10), legend.title = element_text(size=10)) +
  theme(plot.title = element_text(size=20)) +
  theme(plot.margin = unit(c(10,10,10,10),"mm")) +
  labs(title = "Environmental pollution and property damage convictions by court region since 2003") + ylab('Number of convictions') + xlab('Year')
environmental_plot

# filter and grab a subset
drug_convictions <- convictions %>% 
  filter(offence == "Illicit drug offences") %>%
  select(end_date, ethnicity, count) %>%
  subset

# get the sums of each group
drug_convictions <- drug_convictions %>%
  group_by(end_date, ethnicity) %>%
  summarise(amount = sum(count))

drug_plot <- ggplot(drug_convictions, aes(x=end_date, y=amount, colour=ethnicity)) +
  geom_line(size = 1) +
  ylim(0, max(drug_convictions$amount)) + theme_solarized(light = TRUE) + 
  scale_colour_solarized("red", name = "Ethnicities") +
  theme(axis.text = element_text(size = 18), axis.title = element_text(size=18), axis.title.y = element_text(vjust=1.4)) +
  theme(legend.position = "right", legend.text = element_text(size=10), legend.title = element_text(size=10)) +
  theme(plot.title = element_text(size=20)) +
  theme(plot.margin = unit(c(10,10,10,10),"mm")) +
  labs(title = "Drug convictions by ethnicity since 2003") + ylab('Number of convictions') + xlab('Year')
drug_plot

interesting_sentences <- c(
  # "Life imprisonment",
  "Community work, Corrections",
  # "Supervision by Community Corrections",
  # "Conviction and discharge",
  "Imprisonment",
  # "Preventive Detention",
  "Fine",
  # "Reparation/Restitution",
  # "Other",
  # "Disqualification from driving",
  "Community Detention",
  # "Intensive Supervision",
  "Home Detention"
  
)

# filter and grab a subset
drug_sentences <- convictions %>% 
  filter(offence == "Illicit drug offences") %>%
  filter(sentence %in% interesting_sentences) %>%
  select(end_date, sentence, count) %>%
  subset

# get the sums of each group
drug_sentences <- drug_sentences %>%
  group_by(end_date, sentence) %>%
  summarise(amount = sum(count))

drug_plot <- ggplot(drug_sentences, aes(x=end_date, y=amount, colour=sentence)) +
  geom_line(size = 1) +
  ylim(0, max(drug_sentences$amount)) + theme_solarized(light = TRUE) + 
  scale_colour_solarized("red", name = "Sentence type") +
  theme(axis.text = element_text(size = 18), axis.title = element_text(size=18), axis.title.y = element_text(vjust=1.4)) +
  theme(legend.position = "right", legend.text = element_text(size=10), legend.title = element_text(size=10)) +
  theme(plot.title = element_text(size=20)) +
  theme(plot.margin = unit(c(10,10,10,10),"mm")) +
  labs(title = "Drug convictions by sentence 2003") + ylab('Number of convictions') + xlab('Year')
drug_plot


# output a CSV of all drug convictions by sentence

# filter and grab a subset
all_drug_sentences <- convictions %>% 
  filter(offence == "Illicit drug offences") %>%
  select(end_date, sentence, count) %>%
  subset

# get the sums of each group?
all_drug_sentences <- all_drug_sentences %>%
  group_by(end_date, sentence) %>%
  summarise(amount = sum(count))

# transformed dataframe from long to wide
all_drug_sentences <- all_drug_sentences %>%
  melt(id.vars = c("end_date", "sentence")) %>%
  dcast(end_date ~ sentence)

write.csv(all_drug_sentences, file = 'drug_sentences.csv', na = "", row.names = FALSE)