Projet1_mat2775.R

### ETAPE 2


##importation des donnees 
data <- read.csv("/Users/keliane/Downloads/StudentDepressionDataset.csv",header=T)
head(data,15)
str(data)
print(dim(data))

## nettoyage des donnes
sum(is.na(data))
data_clean <- na.omit(data)

              #Convertir variables categorielles
ID_fct <- factor(data$id)
gender_fct <- factor(data$Gender)
city_fct <- factor(data$City)
profession_fct <- factor(data$Profession)
Sleep_duration_fct <- factor(data$Sleep.Duration)
dietary_habits_fct <- factor(data$Dietary.Habits)
family_history_fct <- factor(data$Family.History.of.Mental.Illness)
suicidal_thoughts_fct <- factor(data$Have.you.ever.had.suicidal.thoughts)
degree_fct <- factor(data$Degree)
depression_status_fct <- factor(data$Depression)

ID_fct
gender_fct
city_fct
profession_fct
Sleep_duration_fct
dietary_habits_fct
family_history_fct
suicidal_thoughts_fct
degree_fct
depression_status_fct



### ETAPE 3 : Analyse et exploration des donnees 

##Statistiques Descriptives

            #colonne Age
mean_age <- mean(data$Age, na.rm = TRUE)
median_age <- median(data$Age, na.rm = TRUE)
variance_age <- var(data$Age, na.rm = TRUE)
sd_age <- sd(data$Age, na.rm = TRUE)
min_age <- min(data$Age, na.rm = TRUE)
max_age <- max(data$Age, na.rm = TRUE)
quartiles_age <- quantile(data$Age, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_age <- summary(data$Age)

print(summary_stats_age)


            #colonne CGPA
mean_cgpa <- mean(data$CGPA, na.rm = TRUE)
median_CGPA <- median(data$CGPA, na.rm = TRUE)
variance_CGPA <- var(data$CGPA, na.rm = TRUE)
sd_CGPA <- sd(data$CGPA, na.rm = TRUE)
min_CGPA <- min(data$CGPA, na.rm = TRUE)
max_CGPA <- max(data$CGPA, na.rm = TRUE)
quartiles_CGPA <- quantile(data$CGPA, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_CGPA <- summary(data$CGPA)

print(summary_stats_CGPA)





            #colonne Study Satisfaction
mean_Study_satisfaction <- mean(data$Study.Satisfaction, na.rm = TRUE)
median_Study_satisfaction <- median(data$Study.Satisfaction, na.rm = TRUE)
variance_Study_satisfaction <- var(data$Study.Satisfaction, na.rm = TRUE)
sd_Study_satisfaction <- sd(data$Study.Satisfaction, na.rm = TRUE)
min_Study_satisfaction <- min(data$Study.Satisfaction, na.rm = TRUE)
max_Study_satisfaction <- max(data$Study.Satisfaction, na.rm = TRUE)
quartiles_Study_satisfaction <- quantile(data$Study.Satisfaction, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_Study_satisfaction <- summary(data$Study.Satisfaction)

print(summary_stats_Study_satisfaction)


            #colonne Job Satisfaction
mean_Job_Satisfaction <- mean(data$Job.Satisfaction, na.rm = TRUE)
median_Job_Satisfaction <- median(data$Job.Satisfaction, na.rm = TRUE)
variance_Job_Satisfaction <- var(data$Job.Satisfaction, na.rm = TRUE)
sd_Job_Satisfaction <- sd(data$Job.Satisfaction, na.rm = TRUE)
min_Job_Satisfaction <- min(data$Job.Satisfaction, na.rm = TRUE)
max_Job_Satisfaction <- max(data$Job.Satisfaction, na.rm = TRUE)
quartiles_Job_Satisfaction <- quantile(data$Job.Satisfaction, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_Job_Satisfaction <- summary(data$Job.Satisfaction)

print(summary_stats_Job_Satisfaction)





              #colonne Academic Pressure
mean_Academic_Pressure <- mean(data$Academic.Pressure, na.rm = TRUE)
median_Academic_Pressure <- median(data$Academic.Pressure, na.rm = TRUE)
variance_Academic_Pressure <- var(data$Academic.Pressure, na.rm = TRUE)
sd_Academic_Pressure <- sd(data$Academic.Pressure, na.rm = TRUE)
min_Academic_Pressure <- min(data$Academic.Pressure, na.rm = TRUE)
max_Academic_Pressure <- max(data$Academic.Pressure, na.rm = TRUE)
quartiles_Academic_Pressure <- quantile(data$Academic.Pressure, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_Academic_Pressure <- summary(data$Academic.Pressure)

print(summary_stats_Academic_Pressure)


              #colonne Work Pressure
mean_Work_Pressure <- mean(data$Work.Pressure, na.rm = TRUE)
median_Work_Pressure <- median(data$Work.Pressure, na.rm = TRUE)
variance_Work_Pressure <- var(data$Work.Pressure, na.rm = TRUE)
sd_Work_Pressure <- sd(data$Work.Pressure, na.rm = TRUE)
min_Work_Pressure <- min(data$Work.Pressure, na.rm = TRUE)
max_Work_Pressure <- max(data$Work.Pressure, na.rm = TRUE)
quartiles_Work_Pressure <- quantile(data$Work.Pressure, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_Work_Pressure <- summary(data$Work.Pressure)

print(summary_stats_Work_Pressure)



              #colonne Financial Stress
mean_Financial_stress <- mean(data$Financial.Stress, na.rm = TRUE)
median_Financial_stress <- median(data$Financial.Stress, na.rm = TRUE)
variance_Financial_stress <- var(data$Financial.Stress, na.rm = TRUE)
sd_Financial_stress <- sd(data$Financial.Stress, na.rm = TRUE)
min_Financial_stress <- min(data$Financial.Stress, na.rm = TRUE)
max_Financial_stress <- max(data$Financial.Stress, na.rm = TRUE)
quartiles_Financial_stress <- quantile(data$Financial.Stress, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
summary_stats_Financial_stress <- summary(data$Financial.Stress)

print(summary_stats_Financial_stress)

             #correlation
correlation_matrix <- cor(data[, c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")], use = "complete.obs")
print(correlation_matrix)




## Visualisation

          #Données catégorielles : Diagrammes en barres (geom_bar())
library(ggplot2)
ggplot(data, aes(x = gender_fct)) +
  geom_bar(fill = "blue") +
  labs(title = "Distribution of Gender", x = "Gender", y = "Count")

ggplot(data, aes(x = city_fct)) +
  geom_bar(fill = "green") +
  labs(title = "Distribution by City", x = "City", y = "Count") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1))  # Rotation des labels si nécessaire

ggplot(data, aes(x = profession_fct)) +
  geom_bar(fill = "red") +
  labs(title = "Distribution of Profession", x = "Profession", y = "Count")

# Pour Sleep Duration
ggplot(data, aes(x = Sleep_duration_fct)) +
  geom_bar(fill = "purple") +
  labs(title = "Distribution of Sleep Duration", x = "Sleep Duration", y = "Count")

# Pour Dietary Habits
ggplot(data, aes(x = dietary_habits_fct)) +
  geom_bar(fill = "orange") +
  labs(title = "Distribution of Dietary Habits", x = "Dietary Habits", y = "Count")

# Pour Family History of Mental Illness
ggplot(data, aes(x = family_history_fct)) +
  geom_bar(fill = "cyan") +
  labs(title = "Family History of Mental Illness", x = "Family History", y = "Count")

# Pour Suicidal Thoughts
ggplot(data, aes(x = suicidal_thoughts_fct)) +
  geom_bar(fill = "magenta") +
  labs(title = "Have you ever had suicidal thoughts?", x = "Suicidal Thoughts", y = "Count")

# Pour Degree
ggplot(data, aes(x = degree_fct)) +
  geom_bar(fill = "yellow") +
  labs(title = "Distribution of Degrees", x = "Degree", y = "Count")

# Pour Depression Status
ggplot(data, aes(x = depression_status_fct)) +
  geom_bar(fill = "lightblue") +
  labs(title = "Depression Status", x = "Depression", y = "Count")


                # Données continues : Histogrammes (geom_histogram())

library(ggplot2)

# Histogramme pour la colonne 'Age'
ggplot(data, aes(x = Age)) +
  geom_histogram(bins = 30, fill = "cornflowerblue", color = "black") +
  labs(title = "Histogram of Age", x = "Age", y = "Frequency")

# Histogramme pour la colonne 'CGPA'
ggplot(data, aes(x = CGPA)) +
  geom_histogram(bins = 30, fill = "steelblue", color = "black") +
  labs(title = "Histogram of CGPA", x = "CGPA", y = "Frequency")

# Histogramme pour la colonne 'Study Satisfaction'
ggplot(data, aes(x = Study.Satisfaction)) +
  geom_histogram(bins = 30, fill = "lightblue", color = "black") +
  labs(title = "Histogram of Study Satisfaction", x = "Study Satisfaction", y = "Frequency")

# Histogramme pour la colonne 'Job Satisfaction'
ggplot(data, aes(x = Job.Satisfaction)) +
  geom_histogram(bins = 30, fill = "gray", color = "black") +
  labs(title = "Histogram of Job Satisfaction", x = "Job Satisfaction", y = "Frequency")

# Histogramme pour la colonne 'Academic Pressure'
ggplot(data, aes(x = Academic.Pressure)) +
  geom_histogram(bins = 30, fill = "darkblue", color = "black") +
  labs(title = "Histogram of Academic Pressure", x = "Academic Pressure", y = "Frequency")

# Histogramme pour la colonne 'Work Pressure'
ggplot(data, aes(x = Work.Pressure)) +
  geom_histogram(bins = 30, fill = "purple", color = "black") +
  labs(title = "Histogram of Work Pressure", x = "Work Pressure", y = "Frequency")

# Histogramme pour la colonne 'Financial Stress'
ggplot(data, aes(x = Financial.Stress)) +
  geom_histogram(bins = 30, fill = "red", color = "black") +
  labs(title = "Histogram of Financial Stress", x = "Financial Stress", y = "Frequency")




                     #Relations entre variables : Nuages de points (geom_point())
data_frame <- data.frame(Variable1 = data$Age, Variable2 = data$CGPA)
ggplot(data_frame, aes(x = Variable1, y = Variable2)) +
  geom_point(color = "darkorange") +
  theme_minimal() +
  labs(title = " Age vs CGPA", x = "Variable 1", y = "Variable 2")


data_frame <- data.frame(Variable3 = data$Study.Satisfaction, Variable4 = data$Job.Satisfaction)
ggplot(data_frame, aes(x = Variable3, y = Variable4)) +
  geom_point(color = "darkorange") +
  theme_minimal() +
  labs(title = " Study Satisfaction vs Job Satisfaction", x = "Variable 3", y = "Variable 4")


data_frame <- data.frame(Variable5 = data$Academic.Pressure, Variable6 = data$Work.Pressure)
ggplot(data_frame, aes(x = Variable5, y = Variable6)) +
  geom_point(color = "darkorange") +
  theme_minimal() +
  labs(title = " Academic Pressure vs Work Pressure", x = "Variable 5", y = "Variable 6")


data_frame <- data.frame(Variable5 = data$Academic.Pressure, Variable7 = data$Financial.Stress, Variable4=data$Job.Satisfaction)
ggplot(data_frame, aes(x = Variable5, y = Variable7, color = factor(Study.Satisfaction), size =Variable4)) +
  geom_point(color = "darkorange") +
  theme_minimal() +
  scale_color_manual(values = c("blue", "green", "red", "purple")) +  # Exemple avec des couleurs prédéfinies
  labs(title = "Age vs CGPA with Study Satisfaction and Job Satisfaction",
       x = "Variable 5",
       y = "Variable 7",
       color = "Study Satisfaction",
       size = "Job Satisfaction")

labs(title = " Academic Pressure vs Financial Stress", x = "Variable 5", y = "Variable 7")



                        #Détection des valeurs aberrantes : Boîtes à moustaches (geom_boxplot())
data_frame <- data.frame(Variable1=data$Age, Variable2=data$CGPA, Variable3=data$Study.Satisfaction, Variable4=data$Job.Satisfaction, Variable5 = data$Academic.Pressure, Variable6 = data$Financial.Stress, Variable7=data$Work.Pressure)
boxplot(data_frame, main="Données aléatoires | Graphiques en boîte comparatifs", ylab="Fréquence",
        outpch=18, outcex=1.7, outcol="red",
        col="lightblue")




### ETAPE 4 : ANALYSE STATISTIQUE

## Tendances centrales

               #Moyenne et Mediane
mean_age <- mean(data$Age, na.rm = TRUE)
median_age <- median(data$Age, na.rm = TRUE)

mean_CGPA <- mean(data$CGPA, na.rm = TRUE)
median_CGPA <- median(data$CGPA, na.rm = TRUE)

mean_Study_satisfaction <- mean(data$Study.Satisfaction, na.rm = TRUE)
median_Study_satisfaction <- median(data$Study.Satisfaction, na.rm = TRUE)

mean_Job_Satisfaction <- mean(data$Job.Satisfaction, na.rm = TRUE)
median_Job_Satisfaction <- median(data$Job.Satisfaction, na.rm = TRUE)

mean_Academic_Pressure <- mean(data$Academic.Pressure, na.rm = TRUE)
median_Academic_Pressure <- median(data$Academic.Pressure, na.rm = TRUE)

mean_Work_Pressure <- mean(data$Work.Pressure, na.rm = TRUE)
median_Work_Pressure <- median(data$Work.Pressure, na.rm = TRUE)

mean_Financial_stress <- mean(data$Financial.Stress, na.rm = TRUE)
median_Financial_stress <- median(data$Financial.Stress, na.rm = TRUE)


                     #Variance et ecart type
variance_values <- sapply(data[c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")], var, na.rm = TRUE)
print(variance_values)

sd_values <- sapply(data[c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")], sd, na.rm = TRUE)
print(sd_values)


                 #Plage Interquartile
iqr_values <- sapply(data[c("Age", "CGPA", "Study.Satisfaction","Job.Satisfaction","Academic.Pressure","Work.Pressure","Financial.Stress")], IQR, na.rm = TRUE)
print(iqr_values)


                  #Valeurs aberrantes
variables <- c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")
outliers <- sapply(data[variables], function(x) boxplot.stats(x)$out)
print(outliers)




## Analyse de la symétrie et de l’asymétrie       

library(e1071)
skewness_values <- sapply(data[c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")], skewness)
print(skewness_values)

## *COMMENTAIRES SUR RESULTATS* Les variables Age, CGPA, Study.Satisfaction, et Academic.Pressure ont des valeurs de skewness très proches de zéro, ce qui indique que leurs distributions sont relativement symétriques.


## Analyse des correlations
correlation_matrix <- cor(data[c("Age", "CGPA", "Study.Satisfaction", "Job.Satisfaction", "Academic.Pressure", "Work.Pressure", "Financial.Stress")], use = "complete.obs")
print(correlation_matrix)



## Visualisation avec nuage de points