#Research question: Does the weight of Village Weavers vary across years?

#---------------------------------------------
# Using Village Weaver data
#---------------------------------------------

# Load necessary packages
library(dplyr)    # for data manipulation
library(ggplot2)  # for data visualization
library(car)      # for Levene's Test

weaver<-read.csv('villageweaver.csv')
str(weaver)

# Convert 'year' to a categorical variable (factor)
weaver$year2 <- as.factor(weaver$year)
str(weaver)

# Check ANOVA Assumption: Before performing ANOVA, we must check:
  # Normality – whether weight values are normally distributed.
  # Homogeneity of variance – whether the spread (variance) is similar across years.

# Shapiro-Wilk test for overall normality
shapiro.test(weaver$weight)

# Homogeneity of Variance (Levene’s Test)
leveneTest(weaver$weight ~ weaver$year2)

# If normally distributed, One-way ANOVA
anova_model <- aov(weaver$weight ~ weaver$year2)
summary(anova_model)

#If not normally distributed, Kruskall-Wallis
kruskal.test(weaver$weight ~ weaver$year2)

# Visualise Data (Boxplot)
ggplot(weaver, aes(x = year2, y = weight, fill = year2)) +
  geom_boxplot() +
  theme_classic() +
  labs(x = "Year",
       y = "Weight (g)")

# To remove legend
ggplot(weaver, aes(x = year2, y = weight, fill = year2)) +
  geom_boxplot() +
  theme_classic() +
  theme(legend.position = "none") +
  labs(x = "Year",
       y = "Weight (g)")

# Using geom_violin (combines boxplot and density shape, to visualize both
 # spread and distribution.)
ggplot(weaver, aes(x = year2, y = weight, fill = year2)) +
  geom_violin(trim = FALSE) +
  geom_boxplot(width = 0.2, fill = "white") +
  theme_classic() +
  theme(legend.position = "none") +
  labs(x = "Year",
       y = "Weight (g)")