library(ggplot2)

df <- read.csv("psyexperiment.csv")

# first task
ggplot(df, aes(x=agree, fill=gender)) +
  geom_bar(position="dodge") +
  xlab("Did the student agree?") +
  ylab("Number of cases") +
  ggtitle("Correlation between agreeing and gender")

# second task
p <- ggplot(df, aes(y=gender, x=agree)) +
  geom_count(color="#add8e6", show.legend = F) +
  ggtitle("Yet another correlation between agreeing and gender") +
  ylab("Students gender") +
  xlab("Did the student agree?") +
  scale_size_continuous(range = c(11, 20))
p + geom_text(data = ggplot_build(p)$data[[1]],
              aes(x,y,label=n))

# third task
female_df <- subset(df, gender == "F" & wave=="third")
ggplot(female_df, aes(x=agree, fill=track)) +
  geom_bar(position="dodge") +
  xlab("Did the student agree?") +
  ylab("Number of cases") +
  ggtitle("Correlation between agreeing and gender", subtitle="Third wave, women")

# forth task
library(reshape2)
df <- read.csv("basketball.csv")
df.m <- melt(df)
df.m <- subset(df.m, variable!="age" & variable !="body.height" & variable != "mean")
ggplot(data = df.m, aes(x=variable, y=value)) +
  geom_boxplot() +
  ggtitle("Results by tests") +
  xlab("Test numbers") +
  ylab("Results")

# fifth task
library(reshape2)
df <- read.csv("basketball.csv")
df.m <- melt(df)
df.m <- subset(df.m, variable!="age" & variable !="body.height" & variable != "mean")
ggplot(data = df.m, aes(x=variable, y=value)) +
  geom_violin(col="navy", fill="#add8e6", trim=F) +
  ggtitle("Results by tests", subtitle="Lines represent quantiles") +
  xlab("Test numbers") +
  ylab("Results")

# sixth task
library(reshape2)
df <- read.csv("basketball.csv")
df.m <- melt(df)
df.m <- subset(df.m, variable!="age" & variable !="body.height" & variable != "mean")
ggplot(df.m, aes(x=ID, y=value, group=variable, col=variable)) +
  geom_point() +
  geom_line() +
  ggtitle("Results by tests for all subjects") +
  ylab("Results") +
  xlab("Subjects") +
  scale_color_discrete("Tests")

# seventh task
library(reshape2)
df <- read.csv("basketball.csv")
df$Means <- df$mean
for(i in 1:15){
  df$Medians[i] <- median(c(df$test.1[i],df$test.2[i],df$test.3[i],df$test.4[i],df$test.5[i]))
}
df.m <- melt(df)
df.m <- subset(df.m, variable=="Means" | variable=="Medians")
ggplot(df.m, aes(x=ID, y=value, group=variable, col=variable)) +
  geom_point() +
  geom_line() +
  ggtitle("Means vs. medians for all subjects") +
  ylab("Means and meds") +
  xlab("Subjects") +
  scale_color_discrete("Measures")