#    FRIEDMAN in R SCRIPT FILE.
#    www_statstutor_ac_uk Community Project.
#    Basile Marquier and Ellen Marshall, Sheffield University.
#    Reviewed by Basile Marquier, University of Sheffield.
#    Dataset: video csv.
#    Resource: FRIEDMAN in R.

#Open the diet dataset which is saved as a csv file and call it dietR.  

#If your file is saved as a standard Excel file, save it as a csv file first.

#You will need to change the command depending on where you have saved the file.


# You will need to install the 'PMCMR' package
# Click on "Packages"->"Install Package(s)" and choose 'PMCMR'


#Download the data set in .csv format and put it in a directory on your computer
#Open the file from the place it is saved on your computer
#This example refers to the memory stick where the data is stored as stcp-Rdataset-video.
videoR<-read.csv("E:\\stcp-Rdataset-video.csv",header=T,sep=",")

#You should install the 'stats' package in order to have the function friedman.

library('stats')

#LOADING THE DATA

#For me

videoR<-read.csv("stcp-Rdataset-Video.csv",header=T,sep=",")


#For MASH:

#videoR<-read.csv("D:\\video.csv",header=T,sep=",")
#Tell R we are using the diet dataset until further notice using attach.

attach(videoR) 


median(videoR[,14])

median(TotalAGen)
median(TotalBdoc)
median(TotalCOld)
median(TotalDDEMO)

boxplot(TotalAGen,TotalBdoc,TotalCOld,TotalDDEMO,names=c("A","B","C","D"))

#############################################################################################
#####################################################################
#####################################
########################################################
########################################            ASSUMPTIONS
##################################################################################
###################################################################################
#############################################################################################


# Assumption 1: One group that is measured on three or more different occasions.

# Assumption 2: Group is a random sample from the population.

# Assumtion 3: Your dependent variable should be measured at the ordinal or continuous level.

# Assumption 4: Samples DO not need to be normally distributed
#If any of the columns is not normally distributed, then we should use Friedman. 
#If all columns are normally distributed then ANOVA is preferable.




par(mfrow=c(2,2))

hist(Sample[,1], main="Histogram of TotalAGen",xlab="Value")

hist(Sample[,2], main="Histogram of TotalBdoc",xlab="Value")

hist(Sample[,3], main="Histogram of TotalCOld",xlab="Value")

hist(Sample[,4], main="Histogram of TotalDDEMO",xlab="Value")

## => most of the graphs show skewed data, in particular for AGenUnderstanding, 

## So we use a Friedman test rather than a Repeated Measures Anova


#####################################################################

########################################################
########################################            ANALYSIS
##################################################################################
###################################################################################





# The Friedman test is conducted from the table of data

# with each column representing each sample:

Sample<-matrix(c(TotalAGen,TotalBdoc,TotalCOld,TotalDDEMO),ncol=4)

#To carry out a friedman test use friedman.test(table with samples presented as columns) provided by 'stats' package.


friedman.test(Sample)

## => strong evidence against the null that the conditions are equivalent


####################### MULTIPLE COMPARISONS

library(PMCMR)

posthoc.friedman.nemenyi.test(Sample)

###  => Apart from Sample 2 vs. Sample 4, there is a strong evidence against similarity in all pairs