##
# Libary imports
##

library(readODS)
library(tidyverse)
library(dplyr)


# read a data frame from the ods document
df <- read_ods("ironwood_data_cleaned.ods", sheet = 1)


##
# asses the tree health
##

# First, let's create a vector of condition names corresponding to the health index numbers
condition_names <- c("Healthy", "Light damage", "Medium damage", "Severe damage", "At point of death")
# Define colors for each condition
condition_colors <- c("green", "yellow", "orange", "red", "black")
# Calculate the percentage of trees in each health condition
percentage <- prop.table(table(df$Tree_Health_Index)) * 100
# Now, let's create the bar plot
barplot(percentage,
        names.arg = condition_names,
        main = "Distribution of Tree Health Index",
        xlab = "Health Index",
        ylab = "Percentage of Trees",
        ylim = c(0, max(percentage) + 10),
        col = condition_colors,
        border = "black")
# Adding a legend
legend("topright", legend = condition_names, fill = condition_colors)
# Adding a grid
#grid(nx = NULL, ny = NULL, col = "lightgray", lty = "dotted")
# Adding a box around the plot
box()
# Add labels with the percentage of trees in each bar
text(x = barplot(percentage, plot = FALSE), y = percentage, labels = paste0(round(percentage, 1), "%"), pos = 3)

##
# Perform a shapiro-wilk normality test
##

# Perform Shapiro-Wilk test
shapiro_test <- shapiro.test(df$Tree_Health_Index)
# Print the test results
print(shapiro_test)
# Check the p-value
p_value <- shapiro_test$p.value
# Interpret the results
if (p_value < 0.05) {
  print("The data is not normally distributed (reject the null hypothesis)")
} else {
  print("The data is normally distributed (fail to reject the null hypothesis)")
}

##
# Tasks:
##

# 1. find average dbh
# 2. find