How to use summarise function in dplyr with multiple conditions? - r

Question

I am working with a time-series cross-country dataset covering the period from 2003 to 2018. Each entry in the database corresponds to a protest event, the number of participants, level of engagement of the security services, and level of participant violence. I have multiple observations per year per country. I want to create a new df that counts the number of protests for each country (Count), the average number of participants (AvgParticipants), the average security services engagement (AvgSecurity), and the average level of participant violence (AvgPartViolence). Here is the code I have written thus far:
# Creating Yearly Protest Count Data
# Load packages
library(dplyr)
# Set working directory
setwd("~/Desktop/Cooptation and Protest")
# Load data
dat <- read.csv("reports.csv")
# Subset to relevant variables
dat <- dat %>%
select(cowcode, event_date, side, scope, part_violence, sec_engagement,
numparticipants)
# Convert event_date to only year
dat$event_date <- as.Date(dat$event_date)
dat$year <- as.numeric(format(dat$event_date,'%Y'))
my_summary_data <- dat %>%
group_by(year, cowcode) %>%
summarise(Count = n()) %>%
summarise(AvgSecurity = mean(sec_engagement)) %>%
summarise(AvgPartviolence = mean(part_violence))
I have no issue when I run summarise(Count = n()), but I can't get running summarise(AvgSecurity = mean(sec_engagement)) and summarise(AvgPartviolence = mean(part_violence)) to work. Any advice would be appreciated. Below are some data for your convenience.
structure(list(cowcode = c(40L, 40L, 40L, 40L, 40L, 40L), event_date = structure(c(12183,
15302, 12173, 12173, 12393, 12583), class = "Date"), side = c(0L,
1L, 0L, 0L, 0L, 0L), scope = c(0L, 0L, 0L, 0L, 0L, 0L), part_violence = c(NA_integer_,
NA_integer_, NA_integer_, NA_integer_, NA_integer_, NA_integer_
), sec_engagement = c(NA_integer_, NA_integer_, NA_integer_,
NA_integer_, NA_integer_, NA_integer_), numparticipants = c("",
"", "", "", "2000", ""), year = c(2003, 2011, 2003, 2003, 2003,
2004)), row.names = c(NA, 6L), class = "data.frame")

Answer 1

The comment has it!
library(tidyverse)
dat <- structure(list(cowcode = c(40L, 40L, 40L, 40L, 40L, 40L), event_date = structure(c(12183,
15302, 12173, 12173, 12393, 12583), class = "Date"), side = c(0L,
1L, 0L, 0L, 0L, 0L), scope = c(0L, 0L, 0L, 0L, 0L, 0L), part_violence = c(NA_integer_,
NA_integer_, NA_integer_, NA_integer_, NA_integer_, NA_integer_
), sec_engagement = c(NA_integer_, NA_integer_, NA_integer_,
NA_integer_, NA_integer_, NA_integer_), numparticipants = c("",
"", "", "", "2000", ""), year = c(2003, 2011, 2003, 2003, 2003,
2004)), row.names = c(NA, 6L), class = "data.frame")
dat$event_date <- as.Date(dat$event_date)
dat$year <- as.numeric(format(dat$event_date,'%Y'))
my_summary_data <- dat %>%
group_by(year, cowcode) %>%
summarise(Count = n(),
AvgSecurity = mean(sec_engagement),
AvgPartviolence = mean(part_violence))
my_summary_data