TidyVerse - Extension to Aaron Zalki’s assignment
Don Padmapepruma
12/1/2019
Raw Drug-Use Data
Background on the Data
I am tidying a fivethirtyeight dataset on drug use Github. I am transforming the columns of drugs (3 to 28) into rows, in order to perform analysis on the different drugs used by different age groups. It is important to note that any column header in the dataset with the word ‘use’ is referring to a percentage of those in an age group who have used the specific drug in the past 12 months. Any column header in the dataset with the word ‘frequency’ is referring to the median number of times a user in an age group used the specific drug in the past 12 months. For this analysis, I am only interested in the percentage the drug is used among different age groups.
Loading Data
library(RCurl)## Loading required package: bitopsraw_drug <- read.csv(text=getURL("https://raw.githubusercontent.com/fivethirtyeight/data/master/drug-use-by-age/drug-use-by-age.csv"), header=TRUE, stringsAsFactors = FALSE)
head (raw_drug)## age n alcohol.use alcohol.frequency marijuana.use marijuana.frequency
## 1 12 2798 3.9 3 1.1 4
## 2 13 2757 8.5 6 3.4 15
## 3 14 2792 18.1 5 8.7 24
## 4 15 2956 29.2 6 14.5 25
## 5 16 3058 40.1 10 22.5 30
## 6 17 3038 49.3 13 28.0 36
## cocaine.use cocaine.frequency crack.use crack.frequency heroin.use
## 1 0.1 5.0 0.0 - 0.1
## 2 0.1 1.0 0.0 3.0 0.0
## 3 0.1 5.5 0.0 - 0.1
## 4 0.5 4.0 0.1 9.5 0.2
## 5 1.0 7.0 0.0 1.0 0.1
## 6 2.0 5.0 0.1 21.0 0.1
## heroin.frequency hallucinogen.use hallucinogen.frequency inhalant.use
## 1 35.5 0.2 52 1.6
## 2 - 0.6 6 2.5
## 3 2.0 1.6 3 2.6
## 4 1.0 2.1 4 2.5
## 5 66.5 3.4 3 3.0
## 6 64.0 4.8 3 2.0
## inhalant.frequency pain.releiver.use pain.releiver.frequency
## 1 19.0 2.0 36
## 2 12.0 2.4 14
## 3 5.0 3.9 12
## 4 5.5 5.5 10
## 5 3.0 6.2 7
## 6 4.0 8.5 9
## oxycontin.use oxycontin.frequency tranquilizer.use
## 1 0.1 24.5 0.2
## 2 0.1 41.0 0.3
## 3 0.4 4.5 0.9
## 4 0.8 3.0 2.0
## 5 1.1 4.0 2.4
## 6 1.4 6.0 3.5
## tranquilizer.frequency stimulant.use stimulant.frequency meth.use
## 1 52.0 0.2 2.0 0.0
## 2 25.5 0.3 4.0 0.1
## 3 5.0 0.8 12.0 0.1
## 4 4.5 1.5 6.0 0.3
## 5 11.0 1.8 9.5 0.3
## 6 7.0 2.8 9.0 0.6
## meth.frequency sedative.use sedative.frequency
## 1 - 0.2 13.0
## 2 5.0 0.1 19.0
## 3 24.0 0.2 16.5
## 4 10.5 0.4 30.0
## 5 36.0 0.2 3.0
## 6 48.0 0.5 6.5Tidy & Transform
I am using the packages tidyr and dplyr to change the raw data to its final form.
library(tidyr)##
## Attaching package: 'tidyr'## The following object is masked from 'package:RCurl':
##
## completelibrary(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, uniondrug_gather <- gather(raw_drug, "Drug", "Percent", 3:28)
drug_gather$Percent <- as.numeric(drug_gather$Percent)## Warning: NAs introduced by coerciondrug_separate <- separate(drug_gather, "Drug", c("Drug", "Usage"))## Warning: Expected 2 pieces. Additional pieces discarded in 34 rows [239,
## 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254,
## 255, 256, 257, 258, ...].drug_filter <- filter(drug_separate, Usage=='use')
names (drug_filter) <- c("Age", "Size", "Drug", "Usage", "Percent")
head(drug_filter)## Age Size Drug Usage Percent
## 1 12 2798 alcohol use 3.9
## 2 13 2757 alcohol use 8.5
## 3 14 2792 alcohol use 18.1
## 4 15 2956 alcohol use 29.2
## 5 16 3058 alcohol use 40.1
## 6 17 3038 alcohol use 49.3Analysis
Teenagers
I am using the package ggplot to illustrate the differences found in drug use between teenagers and young adults.
library(ggplot2)
teen <- ggplot(subset(drug_filter, Age %in% c("13","14","15", "16", "17", "18","19")),
aes(x=Drug, y=Percent, fill=Age)) + geom_bar(stat="identity", position=position_dodge()) + theme(axis.text.x = element_text(angle = 90, hjust = 1))
teen
Young Adults
young_adults <- ggplot(subset(drug_filter, Age %in% c("18","19","20","21","22-23","24-25")),
aes(x=Drug, y=Percent, fill=Age)) + geom_bar(stat="identity", position=position_dodge()) + theme(axis.text.x = element_text(angle = 90, hjust = 1))
young_adults
Task 2 by Don P.
Introduction
I will be working on the original dataset for my part of the extention. The data set Drug Use by Age is part of the article How Baby Boomers Get High It covers 13 drugs across 17 age groups. Source of the data: National Survey on Drug Use and Health from the Substance Abuse and Mental Health Data Archive
raw_drug## age n alcohol.use alcohol.frequency marijuana.use
## 1 12 2798 3.9 3 1.1
## 2 13 2757 8.5 6 3.4
## 3 14 2792 18.1 5 8.7
## 4 15 2956 29.2 6 14.5
## 5 16 3058 40.1 10 22.5
## 6 17 3038 49.3 13 28.0
## 7 18 2469 58.7 24 33.7
## 8 19 2223 64.6 36 33.4
## 9 20 2271 69.7 48 34.0
## 10 21 2354 83.2 52 33.0
## 11 22-23 4707 84.2 52 28.4
## 12 24-25 4591 83.1 52 24.9
## 13 26-29 2628 80.7 52 20.8
## 14 30-34 2864 77.5 52 16.4
## 15 35-49 7391 75.0 52 10.4
## 16 50-64 3923 67.2 52 7.3
## 17 65+ 2448 49.3 52 1.2
## marijuana.frequency cocaine.use cocaine.frequency crack.use
## 1 4 0.1 5.0 0.0
## 2 15 0.1 1.0 0.0
## 3 24 0.1 5.5 0.0
## 4 25 0.5 4.0 0.1
## 5 30 1.0 7.0 0.0
## 6 36 2.0 5.0 0.1
## 7 52 3.2 5.0 0.4
## 8 60 4.1 5.5 0.5
## 9 60 4.9 8.0 0.6
## 10 52 4.8 5.0 0.5
## 11 52 4.5 5.0 0.5
## 12 60 4.0 6.0 0.5
## 13 52 3.2 5.0 0.4
## 14 72 2.1 8.0 0.5
## 15 48 1.5 15.0 0.5
## 16 52 0.9 36.0 0.4
## 17 36 0.0 - 0.0
## crack.frequency heroin.use heroin.frequency hallucinogen.use
## 1 - 0.1 35.5 0.2
## 2 3.0 0.0 - 0.6
## 3 - 0.1 2.0 1.6
## 4 9.5 0.2 1.0 2.1
## 5 1.0 0.1 66.5 3.4
## 6 21.0 0.1 64.0 4.8
## 7 10.0 0.4 46.0 7.0
## 8 2.0 0.5 180.0 8.6
## 9 5.0 0.9 45.0 7.4
## 10 17.0 0.6 30.0 6.3
## 11 5.0 1.1 57.5 5.2
## 12 6.0 0.7 88.0 4.5
## 13 6.0 0.6 50.0 3.2
## 14 15.0 0.4 66.0 1.8
## 15 48.0 0.1 280.0 0.6
## 16 62.0 0.1 41.0 0.3
## 17 - 0.0 120.0 0.1
## hallucinogen.frequency inhalant.use inhalant.frequency
## 1 52 1.6 19.0
## 2 6 2.5 12.0
## 3 3 2.6 5.0
## 4 4 2.5 5.5
## 5 3 3.0 3.0
## 6 3 2.0 4.0
## 7 4 1.8 4.0
## 8 3 1.4 3.0
## 9 2 1.5 4.0
## 10 4 1.4 2.0
## 11 3 1.0 4.0
## 12 2 0.8 2.0
## 13 3 0.6 4.0
## 14 2 0.4 3.5
## 15 3 0.3 10.0
## 16 44 0.2 13.5
## 17 2 0.0 -
## pain.releiver.use pain.releiver.frequency oxycontin.use
## 1 2.0 36 0.1
## 2 2.4 14 0.1
## 3 3.9 12 0.4
## 4 5.5 10 0.8
## 5 6.2 7 1.1
## 6 8.5 9 1.4
## 7 9.2 12 1.7
## 8 9.4 12 1.5
## 9 10.0 10 1.7
## 10 9.0 15 1.3
## 11 10.0 15 1.7
## 12 9.0 15 1.3
## 13 8.3 13 1.2
## 14 5.9 22 0.9
## 15 4.2 12 0.3
## 16 2.5 12 0.4
## 17 0.6 24 0.0
## oxycontin.frequency tranquilizer.use tranquilizer.frequency
## 1 24.5 0.2 52.0
## 2 41.0 0.3 25.5
## 3 4.5 0.9 5.0
## 4 3.0 2.0 4.5
## 5 4.0 2.4 11.0
## 6 6.0 3.5 7.0
## 7 7.0 4.9 12.0
## 8 7.5 4.2 4.5
## 9 12.0 5.4 10.0
## 10 13.5 3.9 7.0
## 11 17.5 4.4 12.0
## 12 20.0 4.3 10.0
## 13 13.5 4.2 10.0
## 14 46.0 3.6 8.0
## 15 12.0 1.9 6.0
## 16 5.0 1.4 10.0
## 17 - 0.2 5.0
## stimulant.use stimulant.frequency meth.use meth.frequency sedative.use
## 1 0.2 2.0 0.0 - 0.2
## 2 0.3 4.0 0.1 5.0 0.1
## 3 0.8 12.0 0.1 24.0 0.2
## 4 1.5 6.0 0.3 10.5 0.4
## 5 1.8 9.5 0.3 36.0 0.2
## 6 2.8 9.0 0.6 48.0 0.5
## 7 3.0 8.0 0.5 12.0 0.4
## 8 3.3 6.0 0.4 105.0 0.3
## 9 4.0 12.0 0.9 12.0 0.5
## 10 4.1 10.0 0.6 2.0 0.3
## 11 3.6 10.0 0.6 46.0 0.2
## 12 2.6 10.0 0.7 21.0 0.2
## 13 2.3 7.0 0.6 30.0 0.4
## 14 1.4 12.0 0.4 54.0 0.4
## 15 0.6 24.0 0.2 104.0 0.3
## 16 0.3 24.0 0.2 30.0 0.2
## 17 0.0 364.0 0.0 - 0.0
## sedative.frequency
## 1 13.0
## 2 19.0
## 3 16.5
## 4 30.0
## 5 3.0
## 6 6.5
## 7 10.0
## 8 6.0
## 9 4.0
## 10 9.0
## 11 52.0
## 12 17.5
## 13 4.0
## 14 10.0
## 15 10.0
## 16 104.0
## 17 15.0tidying the datafram
I decided to rename some columns using rename() function.
drug_usage <- raw_drug %>%
rename(sample.size = n, alcohol = alcohol.use, marijuana = marijuana.use, cocaine = cocaine.use,crack = crack.use, heroin = heroin.use, hallucinogen = hallucinogen.use, inhalant = inhalant.use, pain.releiver = pain.releiver.use, oxycontin = oxycontin.use, tranquilizer=tranquilizer.use, stimulant=stimulant.use, meth=meth.use, sedative=sedative.use)
head(drug_usage)## age sample.size alcohol alcohol.frequency marijuana marijuana.frequency
## 1 12 2798 3.9 3 1.1 4
## 2 13 2757 8.5 6 3.4 15
## 3 14 2792 18.1 5 8.7 24
## 4 15 2956 29.2 6 14.5 25
## 5 16 3058 40.1 10 22.5 30
## 6 17 3038 49.3 13 28.0 36
## cocaine cocaine.frequency crack crack.frequency heroin heroin.frequency
## 1 0.1 5.0 0.0 - 0.1 35.5
## 2 0.1 1.0 0.0 3.0 0.0 -
## 3 0.1 5.5 0.0 - 0.1 2.0
## 4 0.5 4.0 0.1 9.5 0.2 1.0
## 5 1.0 7.0 0.0 1.0 0.1 66.5
## 6 2.0 5.0 0.1 21.0 0.1 64.0
## hallucinogen hallucinogen.frequency inhalant inhalant.frequency
## 1 0.2 52 1.6 19.0
## 2 0.6 6 2.5 12.0
## 3 1.6 3 2.6 5.0
## 4 2.1 4 2.5 5.5
## 5 3.4 3 3.0 3.0
## 6 4.8 3 2.0 4.0
## pain.releiver pain.releiver.frequency oxycontin oxycontin.frequency
## 1 2.0 36 0.1 24.5
## 2 2.4 14 0.1 41.0
## 3 3.9 12 0.4 4.5
## 4 5.5 10 0.8 3.0
## 5 6.2 7 1.1 4.0
## 6 8.5 9 1.4 6.0
## tranquilizer tranquilizer.frequency stimulant stimulant.frequency meth
## 1 0.2 52.0 0.2 2.0 0.0
## 2 0.3 25.5 0.3 4.0 0.1
## 3 0.9 5.0 0.8 12.0 0.1
## 4 2.0 4.5 1.5 6.0 0.3
## 5 2.4 11.0 1.8 9.5 0.3
## 6 3.5 7.0 2.8 9.0 0.6
## meth.frequency sedative sedative.frequency
## 1 - 0.2 13.0
## 2 5.0 0.1 19.0
## 3 24.0 0.2 16.5
## 4 10.5 0.4 30.0
## 5 36.0 0.2 3.0
## 6 48.0 0.5 6.5As my second step I got rid of drug frequency columns using select() function as I only wanted to keep the drug usage column for my further analysis.
library(dplyr)
drug_usage <- drug_usage %>%
select(-c(4, 6,8,10,12,14,16,18, 20, 22,24,26,28))
head(drug_usage)## age sample.size alcohol marijuana cocaine crack heroin hallucinogen
## 1 12 2798 3.9 1.1 0.1 0.0 0.1 0.2
## 2 13 2757 8.5 3.4 0.1 0.0 0.0 0.6
## 3 14 2792 18.1 8.7 0.1 0.0 0.1 1.6
## 4 15 2956 29.2 14.5 0.5 0.1 0.2 2.1
## 5 16 3058 40.1 22.5 1.0 0.0 0.1 3.4
## 6 17 3038 49.3 28.0 2.0 0.1 0.1 4.8
## inhalant pain.releiver oxycontin tranquilizer stimulant meth sedative
## 1 1.6 2.0 0.1 0.2 0.2 0.0 0.2
## 2 2.5 2.4 0.1 0.3 0.3 0.1 0.1
## 3 2.6 3.9 0.4 0.9 0.8 0.1 0.2
## 4 2.5 5.5 0.8 2.0 1.5 0.3 0.4
## 5 3.0 6.2 1.1 2.4 1.8 0.3 0.2
## 6 2.0 8.5 1.4 3.5 2.8 0.6 0.5For my next step I used gather() function to make my wide data into key-value pairs of Drug and Percent
tidy_drug_usage <- drug_usage %>%
gather("Drug", "Percent", c(3:15))
head(tidy_drug_usage)## age sample.size Drug Percent
## 1 12 2798 alcohol 3.9
## 2 13 2757 alcohol 8.5
## 3 14 2792 alcohol 18.1
## 4 15 2956 alcohol 29.2
## 5 16 3058 alcohol 40.1
## 6 17 3038 alcohol 49.3Visualization
For the easy visualization of the data that I just tidied, I am using age variable in three categories.
1. Teens: 13-19
2. Young Adults: 20-25
3. Adults: 26-65+
Teens <- ggplot(subset(drug_filter, Age %in% c("13","14","15","16","17","18", "19")),
aes(x=Drug, y=Percent, fill=Age)) + geom_bar(stat="identity", position=position_dodge()) + theme(axis.text.x = element_text(angle = 90, hjust = 1))+
coord_flip()
Teens
library(ggplot2)
young_adults <- ggplot(subset(drug_filter, Age %in% c("20","21","22-23","24-25")),
aes(x=Drug, y=Percent, fill=Age)) + geom_bar(stat="identity", position=position_dodge()) + theme(axis.text.x = element_text(angle = 90, hjust = 1))+
coord_flip()
young_adults
Adults <- ggplot(subset(drug_filter, Age %in% c("26-29","30-34","35-49","50-64","65+")),
aes(x=Drug, y=Percent, fill=Age)) + geom_bar(stat="identity", position=position_dodge()) + theme(axis.text.x = element_text(angle = 90, hjust = 1))+
coord_flip()
Adults
conclution: Just by looking at these plots we can conclude that for all three age categories, alcohol is the highest consumed drug. And Marijuana makes the second highest in the list for all three age categories.