IS607 - Project 2 - Religion Income (Dataset 2 of 3)

Description of the Project

Choose any three of the “wide” datasets identified in the Week 6 Discussion items.

• Create a .CSV file that includes all of the information included in the dataset. You’re encouraged to use a “wide” structure similar to how the information appears in the discussion item.
• Read the information from your .CSV file into R, and use tidyr and dplyr as needed to tidy and transform your data.
• Perform the analysis requested in the discussion item.

Dataset Used: Religion Income

Suggested analysis: Analyze the income by religion

• 1. Group by income type, what religions have the highest and lowest percentage income?
• 2. What religion/s have the highest and lowest percentage overall?
• 3. Group by religion, what income types have the highest and lowest percentage?

Steps

Attach libraries, load and show raw data

library(tidyr)
library(dplyr)
library(stringr)
library(knitr)
library(DT)

rawdata = read.csv("https://raw.githubusercontent.com/L-Velasco/Fall16_IS607/master/Project%202_Religion_Income.csv", stringsAsFactors = FALSE)

Prefix the different income types to maintain the hierarchy

names(rawdata) <- c("Religion","1. Less than $30,000","2. $30,000-$49,999","3. $50,000-$99,999","4. $100,000 or more","Sample_Size")

datatable(rawdata)

Tidy the data

Used select() to exclude the Sample Size variable, gather() to make the different types of income be part of observation and arrange() to sort the observations in descending order according to percentage per income type.

tidydata <- rawdata %>%
  select(-Sample_Size) %>% 
  gather("Income","pct",2:5) %>% 
  arrange(Income, desc(as.numeric(str_replace_all(pct,"%",""))))

## Warning in combine_vars(vars, ind_list): '.Random.seed' is not an integer
## vector but of type 'NULL', so ignored

datatable(tidydata)

Transform To Analyze the data

#1. Group by income type, what religions have the highest and lowest percentage income?

income_data <- tidydata %>%
  group_by(Income) %>% 
  slice(c(1,n())) %>%
  mutate(High_Low = c("Highest","Lowest"), Rel = str_c(Religion,pct, sep=" ")) %>% 
  select(Income, Rel, High_Low) %>% 
  spread(High_Low, Rel)
kable(income_data)

Income	Highest	Lowest
1. Less than $30,000	Historically Black Protestant 53%	Jewish 16%
2. $30,000-$49,999	Jehovah’s Witness 25%	Hindu 13%
3. $50,000-$99,999	Orthodox Christian 36%	Historically Black Protestant 17%
4. $100,000 or more	Jewish 44%	Jehovah’s Witness 4%

#2. What Religion/s have the highest and lowest percentage overall?

income_data_overall <- tidydata %>%
  slice(c(1,n())) %>% 
  select(Religion, pct, Income)
income_data_overall$Income <- str_replace(income_data_overall$Income,str_sub(income_data_overall$Income,1,3),"")
kable(income_data_overall)

Religion	pct	Income
Historically Black Protestant	53%	Less than $30,000
Jehovah’s Witness	4%	$100,000 or more

#3. Group by religion, what income types have the highest and lowest percentage?

religion_data <- tidydata %>%
  arrange(Religion, desc(as.numeric(str_replace_all(pct,"%",""))),desc(Income)) %>% 
  group_by(Religion) %>% 
  slice(c(1,n()))

religion_data$Income <- str_replace(religion_data$Income,str_sub(religion_data$Income,1,3),"")

religion_data <- religion_data %>%
  mutate(High_Low = c("Highest","Lowest"), Inc = str_c(pct,"in",Income, sep=" ")) %>% 
  select(Religion, Inc, High_Low) %>%  
  spread(High_Low, Inc)

kable(religion_data)

Religion	Highest	Lowest
Buddhist	36% in Less than $30,000	13% in $100,000 or more
Catholic	36% in Less than $30,000	19% in $30,000-$49,999
Evangelical Protestant	35% in Less than $30,000	14% in $100,000 or more
Hindu	36% in $100,000 or more	13% in $30,000-$49,999
Historically Black Protestant	53% in Less than $30,000	8% in $100,000 or more
Jehovah’s Witness	48% in Less than $30,000	4% in $100,000 or more
Jewish	44% in $100,000 or more	15% in $30,000-$49,999
Mainline Protestant	29% in Less than $30,000	20% in $30,000-$49,999
Mormon	33% in $50,000-$99,999	20% in $30,000-$49,999
Muslim	34% in Less than $30,000	17% in $30,000-$49,999
Orthodox Christian	36% in $50,000-$99,999	17% in $30,000-$49,999
Unaffiliated (religious “nones”)	33% in Less than $30,000	20% in $30,000-$49,999