Introduction

This analysis seeks to explore differences between respondents’ use of hallucinogens and use of stimulants. Stimulants, sometimes called “uppers” has an effect on users by increasing alertness, elevating blood pressure and increasing heart rate and respiration. On the contrary, hallucinogens affect a person’s perception, sensation, thinking, self-awareness, and emotion. With this analysis, the investigative approach is to highlight whether a person’s use of stimulants and/or hallucinogens reflect a differentiated finding related to use of other drugs, education attainment, employment metrics, and perception of overall health.

Data

The 2016 National Survey on Drug Use and Health (NSDUH) measures:

Use of illegal drugs, prescription drugs, alcohol, and tobacco mental disorders, treatment, and co-occurring substance use and mental disorders

The data provides estimates of substance use and mental illness at the national, state, and sub-state levels. NSDUH data also help to identify the extent of substance use and mental illness among different sub-groups, estimate trends over time, and determine the need for treatment services.

This report will analyze responses to the NSDUH Survey in order to answer the research question outlined above.

Variables

In order to gather information for this analysis, participants responses to questions in the NSDUH survey will be used.

Illicit Drugs

The variables pertaining to illicit drugs asked respondents to specify if they have or have not used any of the 10 drugs as listed:

PainRelievers

Hallucinogens

Marijuana

Inhalants

Crack

Methamphetamine

Cocaine

Tranquilizers

Heroin

Stimulants

Demographics

To profile information related to demographic, the following variables were used:

  • Educationlvl: What education category best describes you?

  • WkStat: Employment status

  • Income: What is your household income?

  • WkSkip: During the past 30 days, how many whole days did you miss from work because you just didn’t want to be there?

  • OvHealth: Would you say your health in general is excellent, very good, good, fair, or poor?

Analysis Plan

By profiling the differences between hallucinogen users and stimulant users, we will be able to answer questions as related to the following categories:

Profiling Drug Use(Category 1)
To profile stimulant and hallucinogen users, we will answer questions using frequency distribution, cross tabulations and a bar graph.

Posed Question: Are stimulant and hallucinogen users more likely to use other drugs that recruit the same or greater feeling?

  • Frequency distribution will help in precisely getting an esitmate of the population we seek to examine.

  • Cross tabulations allow for the ability to visualize the proportional breakdown of those who use other drugs along with stimulant and/or hallucinogen.

  • A bar graph will grant the ability to visualize the population of drug users as a whole to easily compare against one another.

Education (Category 2)
To see how education varies by drug use, we will use a bar graph to assist in our analysis.

Posed Question: Is there a difference in the highest/lowest reported education level by hallucinogen users, stimulant users, those who use both stimulant and hallucinogen, and those who use neither?

  • A bar graph will help in establishing a clear visualization of what groups contains the greatest proportional cases in comparisonn to others.

Employment (Category 3)
To see how employment related metrics vary by drug use, we will use bar graphs and tables to assist in answering our questions.

Posed Question: Given that stimulants are associated with performance, are the users of such drug similarly reported in the highest income category and have the low reported average of skipping work?

  • The use of a bar graph will allow for the ease of combining income categories to futher include a greater range of repesentation.

  • A table will grant the ability to use continuous variable to extract the average days respondents skipped worked.

Perception of Health(Category 4)
To see how health perception differs according to stimulant users and hallucinogen users, we will use a bar graph.

Posed Question: Is there a striking health difference between those who use stimulant or hallucinogen?

  • The use of a bar graph will allow the efficiency of identifying the categorical group of health perception in additon with the combination of two groups or more for ease of representation.

Data Prep

library(dplyr)
library(readr)
library(ggplot2)
library(knitr)
library(tidyr)



Data <- read.csv("~/Downloads/SOC333_NSDUH_2016(1).csv")

DrugData <- Data%>%
rename("Marijuana"=mrjmon, "Crack"=crkmon, "Cocaine" = cocmon, "Heroin"=hermon, "Hallucinogens" = hallucmon, "Inhalants"=inhalmon, "Methamphetamine"=methammon, "PainRelievers"=pnrnmmon, "Tranquilizers"=trqnmmon, "Stimulants"=stmnmmon)%>%
mutate(Income=ifelse(IRPINC3==1, "Less than 10k", 
ifelse(IRPINC3==2, "$10-29k", 
ifelse(IRPINC3==3, "$10-29k", 
ifelse(IRPINC3==4, "$30k-49k", 
ifelse(IRPINC3==5, "$30k-49k", 
ifelse(IRPINC3==6, "More than 50k", 
ifelse(IRPINC3==7, "More than 50k", NA))))))),
Income = factor(Income, levels = c ("Less than 10k", "$10-29k", "$30k-49k", "More than 50k")),
Educationlvl=ifelse(eduhighcat==2, "HS Grad", 
             ifelse(eduhighcat==3, "Some coll/Assoc Degree", 
             ifelse(eduhighcat==4, "College Graduate", NA))),
Educationlvl= factor(Educationlvl, levels = c("HS Grad", "Some coll/Assoc Degree", "College Graduate")),
WkSkip=ifelse(wrkskipmo>85, NA, wrkskipmo), 
WkSick=ifelse(wrksickmo>85, NA, wrksickmo),
WkStat=ifelse(IRWRKSTAT18==1,"Full-Time", 
       ifelse(IRWRKSTAT18==2,"Part-Time", 
       ifelse(IRWRKSTAT18==3,"Unemployed",NA))), 
StimulantOrHallucinogen=ifelse(Stimulants==1 & Hallucinogens==1, "Both",
                        ifelse(Stimulants==0 & Hallucinogens==0, "Neither", 
                        ifelse(Stimulants==1, "Stimulant User",
                        ifelse(Hallucinogens==1,"Hallucinogen User", NA)))), 
StimulantOrHallucinogen= factor(StimulantOrHallucinogen, levels =c("Stimulant User", "Hallucinogen User", "Both", "Neither")),
OvHealth=ifelse(health==1, "Excellent - Very Good", 
                ifelse(health==2, "Excellent - Very Good", 
                ifelse(health==3,"Very Good - Fair", 
                ifelse(health==4,"Very Good - Fair", 
                ifelse(health==5, "Poor", NA))))), 
OvHealth=factor(OvHealth, levels=c("Excellent - Very Good", "Very Good - Fair", "Poor")))

Profiling Drug Use (Frequency distribution)

The freuqency table below displays the metrics and population of those reported to be a user of stimulant, hallucinogen, both, or neither.

  • Those who reported to not be a user of either stimulant or hallucinogen represents almost 6k of the drug user population, while the hallucinogen and stimulant user population is shown to be about 400.
kable(table(DrugData$StimulantOrHallucinogen))
Var1 Freq
Stimulant User 445
Hallucinogen User 364
Both 70
Neither 56267

Profiling Drug Use (Table)

To futher profile the population of stimulant and hallucinogen users, we have broken the categories down to highlight what other drugs respondents are taking.

Hallucinogen User

  • Large majority reported also using marijuana at a proportion of 78%, followed by cocaine (20%), painrelievers (13%), and tranquilizers (13%). Specific to hallucinogen users, they are shown to be the highest group reported to use inhalants.

Stimulant User

  • Also reveal a large proportion use marjuana at 55%, followed by painrelievers (20%), tranquilizers (16%), and cocaine (11%).

Use Hallucinogen and Stimulant

  • Across the board, collectively, large proportion use marijuana at 83%.

Spotlight Findings: As related to the type of drug, users of hallucinogen were shown to use other drugs that produced similar euphoric outcome such as inhalants by 8%.

DrugTable <-DrugData%>%
  select(StimulantOrHallucinogen, Marijuana, Cocaine, Crack, Heroin, Hallucinogens, Inhalants, Methamphetamine, PainRelievers, Tranquilizers, Stimulants)%>%
  tidyr::gather(Drug, Used, -StimulantOrHallucinogen)%>%
  group_by(StimulantOrHallucinogen, Drug)%>%
  summarize(percentUsed=paste0(100*round(mean(Used, na.rm = TRUE), 2),"%"))%>%
spread(StimulantOrHallucinogen,percentUsed)%>%
  select(Drug, `Hallucinogen User`, `Stimulant User`, Both, Neither)

kable(DrugTable)
Drug Hallucinogen User Stimulant User Both Neither
Cocaine 20% 11% 31% 0%
Crack 2% 2% 0% 0%
Hallucinogens 100% 0% 100% 0%
Heroin 2% 3% 4% 0%
Inhalants 8% 1% 7% 0%
Marijuana 78% 55% 83% 10%
Methamphetamine 6% 5% 14% 0%
PainRelievers 13% 20% 41% 1%
Stimulants 0% 100% 100% 0%
Tranquilizers 13% 16% 40% 1%

Display Previous Table in a More Visual Manner

DrugData%>%
  select(StimulantOrHallucinogen, Marijuana, Cocaine, Crack, Heroin, Hallucinogens, Inhalants, Methamphetamine, PainRelievers, Tranquilizers, Stimulants)%>%
  tidyr::gather(Drug, Used, -StimulantOrHallucinogen)%>%
  group_by(StimulantOrHallucinogen, Drug)%>%
  summarize(percentUsed=100*round(mean(Used), 2))%>%
 ggplot()+
  geom_col(aes(x=reorder(Drug,percentUsed), y=percentUsed, fill=StimulantOrHallucinogen), colour="black")+
  facet_wrap(~StimulantOrHallucinogen)+
  coord_flip()

Profiling Education (Bar Graph)

Further analysis to explore the educational difference between stimulant and hallucinogen users.

  • Those reported to not use stimulant or hallucinogen shows the largest proportion of graduating college by 30%.

  • For those who take both stimulant and hallucinogen, they’re shown to be the largest proportion to have attended college and/or earned an Associates degree by 54%.

Spotlight Finding: Specific to those who take stimulant or hallucinogen, an inverse relationship is shown regarding the education level pertaining to HS Grads and College Grads for these groups. Stimulant users have a higher representation of being college graduates at 23%, while hallucinogen users represents the highest of the two groups for being HS graduates at 32%.

SchoolData <- DrugData%>%
  filter(!is.na(Educationlvl))%>%
  group_by(StimulantOrHallucinogen, Educationlvl)%>%
   summarise(n=n())%>%
   mutate(pct=n/sum(n))
ggplot(SchoolData)+
  geom_col(aes(x=Educationlvl,y=pct, fill=StimulantOrHallucinogen),stat="count", colour="black")+ 
  facet_wrap(~StimulantOrHallucinogen)+
  geom_text(aes(x=Educationlvl, y=pct, label=paste0(round(pct*100),"%")), position=position_dodge(0), vjust=-0.5, color="black", size=4)+
  theme(axis.text.x = element_text(angle = 60, hjust = 1))

Profiling Employment: Work Status (Bar Graph)

Employment metrics such as work status, income and average rate of skipping work has been used to further explore the differences between hallucinogen and stimulant users.

  • Stimulant users show the lowest proportion of being unemployed by 8%, while those who use both hallucinogen and stimulant represent the greatest unemployment rate at 18%.

  • Those who reported to not take hallucinogen or stimulant show the highest rate of full-time employment at 69%, followed by stimulant users at 61%.

Spotlight Finding:

With the amalgamation of both stimulant and hallucinogen users, there is no clear sign that proves that stimulant users outwork hallucinogen users, given that rates of full-time employment are closely similar. On the contrary, those that take both stimulant and hallucinogen shows the highest unemployment rates by 18%.

 WkStat2<- DrugData%>%
filter(!is.na(WkStat))%>%
  group_by(StimulantOrHallucinogen, WkStat)%>%
   summarise(n=n())%>%
   mutate(pct=n/sum(n))

 ggplot(WkStat2)+
  geom_col(aes(x=WkStat,y=pct, fill=StimulantOrHallucinogen),stat="count", colour="black")+
   geom_text(aes(x=WkStat, y=pct, label=paste0(round(pct*100), "%")),position = position_dodge(0),vjust = -0.5, color="black", size=4)+
   
  facet_wrap(~StimulantOrHallucinogen)

Profiling Employment: Income (Bar Graph)

To expand on profiling employment in comparison to stimulant and hallucinogen users, analyzing the income of respondents will help in showcasing the economic dichtomomy between the groups using a bar graph for ease of comparison.

  • All groups are closely represented in the lower economic category of earning less than 10k.

  • For those who take neither stimulant or hallucinogen are greatly represented in the upper income earners of 30k and more.

Spotlight Finding:

Hallucinogen users or stimulant users do not display a drast difference of income in terms of 10k - 49k, the difference is shown in the upper categories of earning more than 50k, where stimulant users are represented by 8% and hallucinogen users by 4%.

IncomeData <- DrugData%>%
 
 group_by(StimulantOrHallucinogen, Income)%>%
   summarise(n=n())%>%
   mutate(pct=n/sum(n))
ggplot(IncomeData)+
  geom_col(aes(x=Income,y=pct, fill=StimulantOrHallucinogen),stat="count", colour="black")+ 
  facet_wrap(~StimulantOrHallucinogen)+
  geom_text(aes(x=Income, y=pct, label=paste0(round(pct*100), "%")),position = position_dodge(0),vjust = -0.5, color="black", size=4)+
  theme(axis.text.x = element_text(angle = 60, hjust = 1))

Profiling Employment: Average Rate of Skipping Work (Table)

The average rates of skipping work while comparing stimulant and hallucinogen users will help in adding leverage to our analysis.

  • Those that take neither stimulant or hallucinogen reported an average rate of skipping work by 37%.

  • Across the board, hallucinogen users surpass other groups with the rate of skipping work about 119% of the time.

WkSkip2<-DrugData%>%
  group_by(StimulantOrHallucinogen)%>%
  summarise(AvgWkSkip=round(mean(WkSkip, na.rm=TRUE),2))

kable(WkSkip2)
StimulantOrHallucinogen AvgWkSkip
Stimulant User 0.66
Hallucinogen User 1.19
Both 0.93
Neither 0.37

Profiling Overall Health (Bar Graph)

The use of overall health will allow us to formulate an insight to what extent stimulant and/or hallucinogen users report their health.

  • As a whole, there is no drastic difference in overall health amongst the groups. Stimulant users slighly have a higher rating in comparison to other groups by 65%.
HealthData <- DrugData%>%
  filter(!is.na(OvHealth))%>%
  group_by(StimulantOrHallucinogen, OvHealth)%>%
   summarise(n=n())%>%
   mutate(pct=n/sum(n))
ggplot(HealthData)+
  geom_col(aes(x=OvHealth,y=pct, fill=StimulantOrHallucinogen),stat="count", colour="black")+ 
  facet_wrap(~StimulantOrHallucinogen)+
  geom_text(aes(x=OvHealth, y=pct, label=paste0(round(pct*100), "%")),position = position_dodge(0),vjust = -0.5, color="black", size=4)+
  theme(axis.text.x = element_text(angle = 60, hjust = 1))

Conslusion

Profiling Drug Use(Category 1):
Are stimulant and hallucinogen users more likely to use other drugs that recruit the same or greater feeling?

  • Users of hallucinogen were shown to use other drugs that produced similar euphoric outcome such as inhalants by 8%.

Education(Category 2):
Is there a difference in highest/lowest reported education level by hallucinogen users, stimulant users, those who use both stimulant and hallucinogen, and those who use neither?

  • Stimulant users have a higher representation of being college graduates at 23%, while hallucinogen users represents the highest of the two groups for being HS graduates at 32%.

Employment (Category 3)
Given that stimulants are associated with performance, are the users of such drug similarly reported in the highest income category and have the low reported average of skipping work?

  • No difference difference in income between hallucinogen users and stimulant users for 10k - 49k income. Rather, the difference is shown in the upper categories of earning more than 50k, where stimulant users are represented by 8% and hallucinogen users by 4%.

Perception of Health(Category 4)
Is there a striking health difference between those who use stimulant or hallucinogen?

  • With all groups taken into consideration, there is no drastic difference in overall health amongst one another.