#Title: R-ticulate - Intro to R----
#Author: Danney Rasco, PhD
#Affiliation: 
#     Department of Psychology, Sociology, and Social Work
#     West Texas A&M University
#
'THANK YOU to the following for supporting student travel and this event: 
    College of Education and Social Sciences
    Department of Psychology, Sociology, and Social Work
    Dr. Lisa Garza'
#
#PRE-WORKSHOP PACKAGE INSTALL
#Select (Highlight with Cursor) 
#the next 12 lines of code (Lines 16 to 22)
#Press Ctrl and Enter at the same time once selected
options(repos=c(CRAN="https://cloud.r-project.org"))
PacMan <- function(pkg){new.pkg <- pkg[!(pkg %in% installed.packages()[, "Package"])] 
if (length(new.pkg))         
install.packages(new.pkg, dependencies = TRUE)     
sapply(pkg, library, character.only = TRUE)}
pkgs <- c("apa", "apaTables", "broom", "car",
          "corrplot", "DescTools", "dplyr", "ez",
          "foreign", "ggfortify", "ggplot2", 
          "ggplotgui", "gvlma", "lavaan", "MVN", "nortest", 
          "olsrr", "purrr", "readr", "rgl", 
          "scatterplot3d", "semPlot", "semTools", "tidyr")
PacMan(pkgs)
#
#STOP SELECTION
#Once you hit Ctrl+Enter, you should see some code running in the Console window
#It will likely take a few minutes to install all packages
# It may take several minutes depending on the speed of connection
#
#Do not worry if the code is red. 
#Red is the default color for downloads.
#As long as you see "downloaded XX KB" 
# or "downloaded XX MB" with a number for XX
# it worked correctly.
#
#We will cover the remaining code during the session. 
#We look forward to meeting you soon.
#
#
#Thank you----
#Department of Psychology, Sociology, and Social Work 
#   Dr. Lisa Garza
#
#INTRO to Rstudio Windows----
#Script, Console, Environment, Files...
#
#CREATING A PROJECT FOLDER---- 
#1. Create folder on desktop or other location
#2. Label folder
#3. In R, select File>New Project
#4. Select New Directory (unless you have an existing directory from previous work in R)
#5. Select New Project
#6. Select Browse
#7. Find the location of the folder you previously created
#8. Enter the directory name (e.g., IntroToR)
#9. Hit Create Project
#
#Working with a SCRIPT----
#Open Script
#  In R, select File>New File>Rscript (or Ctrl+Shift+N)
#
#Save Script
#  File>Save As
#
#Few Basics: Comments, Foldable Sections, Running Code, etc.----
#Create comments to make code easier to read (or add reminders)
#  Start the comment with a "#" symbol
#
#Create foldable sections
#   Add four "-" marks (or more) on a comment line to create a foldable section
#   Some portions of code (e.g., functions) are automatically foldable
#
#Run a line (or lines) of code
#   To run a line of code, place your cursor on the line and hit "ctrl" and "enter"
#   To run lines of code, highlight the section of code and hit ctrl+enter
#
#Download (i.e., INSTALL) package
#  install.packages("[INSERT PACKAGE]", dependencies=TRUE)
#
#Attach package from library
install.packages("apa", dependencies = TRUE)
library(apa)
#Attach multiple packages from library
pkgs <- c("apaTables", "broom", "corrplot", "DescTools", "dplyr", "foreign", "ggfortify", "ggplot2", "ggplotgui", "gvlma", "purrr")
lapply(pkgs, library, character.only=TRUE)
#
#Learn more about a function
#  Learn more about a function by using "help([insert function])" or "??[function]"
#
#Clear console
#  ctrl + l
#
#Working directory----
#Get working directory (i.e., what location are you currently using)
getwd()
#
#Change (i.e., SET) working directory
#  Note: You can find location by right clicking folder and selecting properties
#  If you copy and paste location, flip slashes to forward
setwd("C:/Users/Danney/Desktop/IntroToR")
read.csv("DataExample.csv")
#
#OBJECT-ORIENTED CODING----
#INTRODUCE TERMINOLOGY: Object, Class, Package, Function, Library
#  Objects are assigned pieces of information (a score, vector, matrix, data frame) 
#  Objects can be of different types or CLASSES that have certain attributes
#  Packages are created to perform a number of jobs or FUNCTIONS
#  Functions are typically performing a limited number of actions on the objects
#  Packages are stored in the library - Some automatically, some you add
#
#Assign a value (10) to an object (x)
#  NOTE: In Rstudio you can create "<-" using "alt" and "-" together
x <- 10
#Check the Environment window; Should see new object (x) added
#
#Call the object in the console
x
#Add the value 20 to the object x
x+20
#Assign sum of x+20 to a new object (z)
z <-  (x+20)
#Call object z
z
#Fundamentals of data use----
#Assign vector of numbers to object (v1)
v1 <- c(100, 108, 109, 88, 79, 99, 95, 106, 103, 118, 115, 121, 130, 145, 132, 82, 93, 84, 82, 117, 98, 97, 96, 148)
#Convert vector to a data frame and call object
data1 <- as.data.frame(v1)
#
#View data in console
v1
data1
#
#Open full data frame
View(data1)
#
#Create a second variable for GPA
v2 <- c(3.82, 3.60, 3.91, 2.58, 1.92, 2.02, 1.98, 3.20, 3.21, 3.80, 3.83, 3.90, 3.87, 3.20, 4.00, 2.68, 2.87, 2.00, 2.50, 1.80, 2.92, 3.20, 3.27, 3.87)
#
#Combine two vectors
data2 <-  data.frame(v1, v2)
data2
#  Note: data.frame can create a data frame; as.data.frame coerces an existing object to a data frame
#
#Example of using data.frame
data2$ID <- (1:24)
data3 <- data.frame(ID=c(6,8,32), IQ=v1, GPA=v2)
View(data3)
#
#Add column labels for data2
names(data3) <- c("ID", "IQ","GPA")
View(data3)
head(data3, n=6)
head(data3, n=20)
#
#Save data as a CSV file
write.csv(data2, file="IQ_GPA_example2.csv", row.names=TRUE)
write.csv(data3, file="IQ_GPA_example3.csv", row.names=FALSE)
#
#Write data to other file types (e.g., Stata)
library(foreign)
?write.dta
write.dta(data3, file="IQ_GPA_example4.dta")
#
#Read data from CSV file
IQdata <- read.csv("IQ_GPA_example3.csv")
head(IQdata)
#
#EXAMPLES: Read data from other files
DataSPSS <- read.spss("FileName.sav", to.data.frame=TRUE)
??read.spss
DataStata <- read.dta("FileName.dta")
??read.dta
#
#Summarizing and Visualizing Single Continuous Data - Descriptive Stats-------
#Summary output for a single variable in data frame
summary(IQdata$IQ)
#
#Summary output for entire data frame
summary(IQdata)
#
#Look at distribution of IQ scores
hist(IQdata$GPA, main="Histogram of GPA", xlab="GPA", ylab="n")
#
#Calculate mean for IQ
IQmean <- mean(IQdata$IQ)
IQmean
#
#Estimate variance for IQ
IQvar <- var(IQdata$IQ)
IQvar
#
#Estimate the standard deviation for IQ
IQsd <- sd(IQdata$IQ)
IQsd
sqrt(IQvar)
#
#Report IQsd rounded to 2 decimal places
round(IQsd, 2)
#
#Create z score
IQdata$IQz <- 0
head(IQdata)
IQdata$IQz <- ((IQdata$IQ-IQmean)/IQsd)
View(IQdata)
round(mean(IQdata$IQz), 4)
round(sd(IQdata$IQz), 4)
#skewness and kurtosis
library(DescTools)
Skew(IQdata$IQ)
Kurt(IQdata$IQ)
#
#Recoding continuous to categorical-----------------------------------
#Convert IQ to dichotomous variable: Above 100, Equal or below 100
#1. Add variable to data frame
IQdata$IQcat <- 0
#
#2. Create values for new variable using continuous variable
IQdata$IQcat[IQdata$IQ<101] <- "Average or below"
IQdata$IQcat[IQdata$IQ>100] <- "Above average"
summary(IQdata$IQcat)
#3. Let R know variable is categorical (i.e., factor) variable
IQdata$IQcat <- factor(IQdata$IQcat)
summary(IQdata$IQcat)
#
#Look at updated data with categorical variable
View(IQdata)
#
#Summarizing and Visualizing Single Categorical Variable --------------
#Create frequency table for IQcat
IQfreq <- table(IQdata$IQcat)
IQfreq
#
#Create frequency table as data frame using wrapping
as.data.frame(table(IQdata$IQcat, dnn='IQ category'))
#
#Create bar graph for IQcat using IQfreq
barplot(IQfreq, main="Bar Graph: IQ", xlab="IQ", ylab="Frequency (n)", names.arg=c("Above Average", "Average or Below"), border="blue")
#
#Visualizing Continuous with Categorical-------------------------------
#
#Get error info - The easy way
library(dplyr)
SEdata <- IQdata %>% group_by(IQcat) %>% summarise(N=n(),mean=mean(GPA),sd=sd(GPA),se=sd/sqrt(N))
View(SEdata)
#
#Create plot with 95% CI around mean
library(ggplot2)
plot <- ggplot(data=SEdata, 
          aes(x= IQcat, y=mean))+
          geom_bar(stat="identity")
plot
plot+geom_errorbar(aes(ymin=mean-(1.96*se), ymax=mean+(1.96*se)))
#
#Make it pretty
plot2 <- ggplot(data=SEdata, aes(x=IQcat, y=mean, fill=IQcat, width=.5))+
  geom_bar(stat="identity", color="black")+
  geom_errorbar(aes(ymin=mean-(1.96*se), ymax=mean+(1.96*se)), width=.2)+
  labs(title="Average GPA by IQ", x="IQ Category", y="GPA (M)")
plot2
plot2+scale_fill_manual(values=c("#000099", "#990000"))+theme_classic()
#Use WT colors
plot2+scale_fill_manual(values=c("#660000", "#990000"))+theme_minimal()+
  theme(plot.title=element_text(hjust=0.5))
plot3 <- ggplot(data=SEdata, aes(x=IQcat, y=mean, fill=IQcat, width=.5))+
  geom_bar(stat="identity", color="black")+
  geom_errorbar(aes(ymin=mean-(1.96*se), ymax=mean+(1.96*se)), width=.2)+
  labs(title="Average GPA by IQ", x="IQ Category", y="GPA (M)")+
  scale_fill_manual(values=c("#660000", "#990000"))+theme_minimal()+theme(plot.title=element_text(size = 22, hjust=0.5))
plot3
#
#Create histogram by group
histBygrp <- ggplot(data=IQdata, aes(x=GPA, fill=IQcat))+
  geom_histogram(position='identity', alpha=0.8, binwidth=1)+
  scale_fill_brewer(palette="Dark2")+theme_classic()
histBygrp
densityBygrp <- ggplot(data=IQdata, aes(x=GPA, fill=IQcat))+
  geom_density(position='identity', alpha=0.8, adjust=0.8)+
  scale_fill_manual(values=c("#660000", "#990000"))+theme_classic()
densityBygrp
#Data Analysis and Functions in R----------------------------------------------------
#Creating a Function for One-sample z test
ztest = function(x, mu, sigma){
  Mx = mean(x)
  n = length(x)
  z = round((Mx - mu) / (sigma/sqrt(n)), 3)
  p = round((2*pnorm(-abs(z), mean = 0, sd = 1)), 3)
  statistic <- c("Sample Size (n)", "Sample Mean", "z obtained", "p (two-tailed)")
  value <- c(n, Mx, z, p)
  output <- cbind(statistic, value)
  as.data.frame(output)
}
#Using Function for One-sample z test
#Compare IQ scores in IQdata to pop. (M=100, SD=15)
ztest(IQdata$IQ, 100, 15)