R Notebook

library(ggplot2)
library(dplyr)
library(car)

df = read.csv("blue light and angles wide.csv", header = T)
df2 = read.csv("blue light and angles long.csv", header = T)

summary(df)

 ComputerScore      PaperScore          X            
 Min.   :0.0000   Min.   :0.0000   Length:145        
 1st Qu.:0.2500   1st Qu.:0.2500   Class :character  
 Median :0.5000   Median :0.5000   Mode  :character  
 Mean   :0.5431   Mean   :0.5914                     
 3rd Qu.:1.0000   3rd Qu.:1.0000                     
 Max.   :1.0000   Max.   :1.0000                     

a = ggplot(df2, aes(x = Group, y = Score, fill = Group)) +
  geom_boxplot(position = position_dodge(width = 0.8), notch = F,) + # Dodge boxes for side-by-side comparison
  geom_jitter(width = 0.2, alpha = 0.5)+
  stat_summary(fun = mean, geom = "point", shape = 23, size = 3, fill = "red")+
  labs(x = "Format",
       y = "Score",
       fill = "Format") +
  theme_classic() +
  theme(axis.text.x = element_text(size = 18),   # Adjust the size value as needed
        axis.text.y = element_text(size = 18),
        axis.title.x = element_text(size = 24),
        axis.title.y = element_text(size = 24),)
a

ggsave("Blue light x angle.png", plot=a, dpi = 300, width =12, height = 12)

cor = cor(df)

Error in cor(df) : 'x' must be numeric

# Assuming df has columns 'before' and 'after'
df$differences <- df$PaperScore - df$ComputerScore

shapiro_test_result <- shapiro.test(df$differences)
print(shapiro_test_result)

    Shapiro-Wilk normality test

data:  df$differences
W = 0.93902, p-value = 6.254e-06

qqnorm(df$differences)
qqline(df$differences, col = "red")

# Paired t-test
t_test_result <- t.test(df$PaperScore, df$ComputerScore, paired = TRUE)

print(t_test_result)

    Paired t-test

data:  df$PaperScore and df$ComputerScore
t = 1.0807, df = 144, p-value = 0.2816
alternative hypothesis: true mean difference is not equal to 0
95 percent confidence interval:
 -0.04001555  0.13656728
sample estimates:
mean difference 
     0.04827586 

# Assuming df has columns 'Net.Speed.PreGinger' and 'Net.Speed.PostGinger'
wilcox_test_result <- wilcox.test(df$PaperScore, df$ComputerScore, paired = TRUE)

print(wilcox_test_result)

    Wilcoxon signed rank test with continuity correction

data:  df$PaperScore and df$ComputerScore
V = 3271, p-value = 0.227
alternative hypothesis: true location shift is not equal to 0

# Extract the W statistic
V <- wilcox_test_result$statistic

# # Calculate the rank-biserial correlation
# n1 <- sum(df2$Ginger == "Net.Speed.PreGinger")
# n2 <- sum(df2$Ginger == "Net.Speed.PostGinger")
# rank_biserial <- (2 * W) / (n1 * n2) - 1
# 
# print(rank_biserial)

n <- length(df$Net.Speed.PreGinger)
rank_biserial <- (2 * V) / (n * (n + 1)) - 1

print(rank_biserial)

  V 
Inf