r-mydata-twoway

#This script produces a first look of some hospital data

#load packages

# Install pacman ("package manager") if needed
if (!require("pacman")) install.packages("pacman")

# pacman must already be installed; then load contributed
pacman::p_load(magrittr, pacman, rio, tidyverse)

# Reading SPSS data file into R

library(foreign)    

dat <- read.spss("Demo data.sav", to.data.frame=TRUE)
attach(dat)


## Random sample of 400 patients

set.seed(1)

mydata <-dat[sample(nrow(dat), 400), ]


###Use variables education and typeofadmission. 
# make a two-way table. 

mydata %>%
  select(education, typeofadmission) %>%
  table()


###Produce frequency tables with percentages.

ct <- table(mydata$education, mydata$typeofadmission)
ct

ct %>%
  prop.table(1) %>%       # 1 is for row percentages, 2 is column
  round(2) %>%            # round to 2 figures
  multiply_by(100)


###Use row percentages and standardized residuals and describe differences between education groups in typeofadmission

# Chi-squared test and std residuals

tchi <- chisq.test(ct)
tchi

tchi$stdres

##Descriptive statistics for a numerical variable on a scale 1-5

#Summarise variable

summary(mydata$Pers_A)
typeof()   # double
print(class(mydata$Pers_A))   # numeric


#histogram

hist(mydata$Pers_A, xlab = "Pers_A", col = "blue" , border = "red", breaks = 10)
  
#boxplot to compare two variables

boxplot(Pers_A, Pers_B, data = mydata,
          xlab = "Pers_A / Pers_B", col = c("green", "yellow"), names = c("A","B"))

#scatterplot

plot(x = mydata$Pers_A, y = mydata$Pers_B, 
     xlab = "A",
     ylab = "B",
     main = "A vs B",
     col = "red"
     )
 
# Clear data
rm(list = ls())  # Removes all objects from environment

# Clear packages
p_unload(all)  # Remove all contributed packages

# Clear console
cat("\014")  # Mimics ctrl+L