police-violence

commit b70fa0516f853b81b51bf2528728938ff2c41482
parent ae9de08e7a6e88883aec8feee8564918a408b100
Author: eamoncaddigan <eamon.caddigan@gmail.com>
Date:   Wed, 23 Dec 2015 16:15:40 -0500

Had this code in the Rmd, but it was a deadend.

Diffstat:
A
victimsPerDepartment.R
 | 
25
+++++++++++++++++++++++++

1 file changed, 25 insertions(+), 0 deletions(-)
diff --git a/victimsPerDepartment.R b/victimsPerDepartment.R
@@ -0,0 +1,25 @@
+# This doesn't actually do what I want, but it's still useful code. Maybe?
+library(readxl)
+library(dplyr)
+library(tidyr)
+
+mpvData <- read_excel("MPVDatasetDownload-83zj.xlsx", sheet = 1)
+
+# Pitch bad columns, which are empty and raise errors in dplyr
+mpvData <- mpvData[, 1:19]
+
+mpvData <- mpvData %>%
+  # Only data from 2015 and agency must be specified
+  filter(`Date of injury resulting in death (month/day/year)` >=
+           as.POSIXct("2015-01-01"),
+         !is.na(`Agency responsible for death`)) %>%
+  # Split up the agencies
+  mutate(agency = sub(",? and ", ", ", `Agency responsible for death`)) %>%
+  separate(agency, paste("agency", 1:4, sep="_"), 
+           "[[:space:]]*,[[:space:]]*", fill = "right") %>%
+  # Gather back into a single column
+  gather("agency_number", "agency_name", starts_with("agency_"), na.rm  = TRUE)
+
+# Count the number of victims per department
+victimsPerDepartment <- mpvData %>%
+  count(agency_name)