##Analysis of open grading weights

library(here)
here("Assessments", "AssessmentWeight.csv")


##brings in the data file
Assessments <- read.csv(file = "3. AssessmentWeights.csv", header = T, sep=",")
##prints a few lines of the data file
head(Assessments)
## 'Student' is categorical
Assessments $Student <- as.factor(Assessments$Student)

subset <- subset(Assessments, Group == "student")
head(subset)

##RQ: Are assessment categories weighted differently by students?
##run one-way anova
model = aov(Percentage.Weight~ Assessment.Category,
           data=subset)
summary(model)
TukeyHSD(model)

##Checking assumptions of the model
# 1. Homogeneity of variances
plot(model, 1)

# 2. Normality
plot(model, 2)


##summarize the data (mean, SE, etc.)
library(Rmisc)
summary = summarySE(Assessments,
                measurevar="Percentage.Weight",
                groupvars=c("Group", "Assessment.Category"))
summary

##visualize the data
# 1. Open jpeg file
jpeg("Figure2.jpg", width = 550, height = 400)

# 2. Create the plot
ggboxplot(Assessments, x = "Assessment.Category", y = "Percentage.Weight",
    ylab = "Weight (%)", xlab = "Assessment", 
   add = "jitter", color = "Group", palette = c("#E69F00", "#56B4E9"))

# 3. Close the file
dev.off()